J01 1 <#FROWN:J01\>Cosmology, Clustering and Superclustering

J01 2 William C. Saslaw

J01 3 On large scales, the probability that galaxies occupy a given J01 4 size volume of space is not random, like the toss of a coin, but is J01 5 related to the presence of nearby galaxies. Positions of galaxies J01 6 depend on one another. Details of this dependence may provide J01 7 important clues to the origin and evolution of the universe. Modern J01 8 systematic searches for these clues started in the 1930s with Edwin J01 9 P. Hubble's galaxy counts, accelerated in the 1950s with the J01 10 development of new statistical techniques, and surged in the 1970s J01 11 and 1980s as computers became powerful enough to handle large J01 12 amounts of data and calculate complex simulations of clustering J01 13 physics.

J01 14 There are two main descriptions of galaxy clustering. The first J01 15 is essentially pictorial, derived by searching the sky for J01 16 filaments, voids, and overdense regions. Galaxies in projected J01 17 high-density regions having similar redshift distances are likely J01 18 to form a physically related cluster. If such a cluster has been J01 19 bound gravitationally for a large fraction of the present Hubble J01 20 time, it evolves fairly independently of its surrounding galaxies. J01 21 Occasionally several contiguous large clusters form a supercluster. J01 22 This may result partly from the initial positioning of clusters and J01 23 partly from their subsequent motions.

J01 24 Most galaxies are not in large physically bound clusters. They J01 25 are, however, clustered in a more general statistical sense. Rather J01 26 than examining individual clusters, this second description looks J01 27 for statistical departures from a Poisson distribution (which is J01 28 the uniform random spatial distribution of objects whose positions J01 29 are entirely independent of one another, i.e., totally J01 30 uncorrelated). In a Poisson distribution there will be some regions J01 31 where galaxies are strongly clustered just by chance, and other J01 32 regions which are unusually empty, also by chance. The observed J01 33 numbers and statistics of such regions may then be compared with J01 34 those expected from various theories.

J01 35 STATISTICAL MEASURES OF CLUSTERING

J01 36 The most useful and informative statistics can be measured J01 37 objectively from observations and related analytically to physical J01 38 processes of clustering and computer experiments. Low-order J01 39 correlation functions are one example. The two-point correlation J01 40 function <*_>xi<*/>(r) in its simplest form for a homogeneous J01 41 isotropic system is defined by

J01 42 formula

J01 43 Here n(r)dr is the average number of galaxies between J01 44 radial distance r and r+dr from any given galaxy. The J01 45 overall average number density of galaxies in the entire system, or J01 46 over a very large volume of the universe, is <*_>unch<*/>. For a J01 47 random Poisson distribution, <*_>xi<*/>(r)=0; so n(r) is J01 48 determined just by <*_>unch<*/> and geometry. Therefore J01 49 <*_>xi<*/>(r) helps measure departures from the Poisson state. J01 50 Higher-order correlation functions use the relative positions of J01 51 three or more galaxies for a more refined description that, J01 52 unfortunately, is harder to measure observationally. The first J01 53 accurate observations of <*_>xi<*/>(r) for galaxies, made by J01 54 H. Totsuji and T. Kihara in 1969, gave a power law of the form J01 55 formula on scales formula (for a Hubble constant of J01 56 50 km s-1 Mpc-1). Large clusters of galaxies J01 57 are observed to have a similar two-point correlation function if J01 58 each cluster is represented by a single point, but this result is J01 59 much more uncertain. On small scales <*_>xi<*/>(r)>>1, so the J01 60 observed clustering is highly nonlinear; that is, correlations J01 61 dominate for r <*_>unch<*/>10 Mpc.

J01 62 Another observed simple objective clustering statistic is the J01 63 distribution function f(N). This is the probability for J01 64 finding N galaxies in a volume of size V, or projected J01 65 onto the sky in an area of size A. If the distribution is J01 66 statistically homogeneous then it will not depend significantly on J01 67 the shape of the volumes or areas, provided they are sufficiently J01 68 large and numerous to give a fair average sample. Recent analyses J01 69 of the area counts of galaxies show that they have a distribution J01 70 of the form

J01 71 formula

J01 72 where <*_>unch<*/>=<*_>unch<*/>V is the average number in a J01 73 volume V for an average number density <*_>unch<*/>. The J01 74 quantity b is a measure of clustering and is related to J01 75 gravitational correlations. The observed value of b is J01 76 0.70<*_>unch<*/>0.05 for galaxies whose separations are typically J01 77 1-10 Mpc. For large clusters with separations of J01 78 <*_>approximate-sign<*/>10-50 Mpc, b=0.3<*_>unch<*/>0.1. For a J01 79 sample of faint radio sources with separations <*_>unch<*/>50 Mpc, J01 80 b=0.0, which is a random Poisson distribution. The f(N) J01 81 distribution for N=0 gives the probability that a region is a J01 82 void with no galaxies at all.

J01 83 Other statistics applied to galaxy clustering include minimal J01 84 spanning trees (the shortest line connecting all the galaxies in a J01 85 sample), topological patterns formed by contour maps of regions J01 86 with the same density, and multifractal analyses (a single fractal J01 87 dimension does not adequately describe galaxy clustering), which J01 88 are related to how the average number density of galaxies around a J01 89 given galaxy changes with distance from the galaxy. These other J01 90 statistics also yield valuable information. Unlike J01 91 <*_>xi<*/>(r) and f(N), however, they have not yet been J01 92 related generally to an underlying dynamical theory. Some J01 93 specialized computer experiments have examined their behavior.

J01 94 THEORIES OF CLUSTERING

J01 95 To understand the observed statistics we need to know the J01 96 initial conditions for clustering as well as a physical theory for J01 97 its subsequent evolution. Initial conditions may indicate J01 98 properties of the early universe before galaxies formed and perhaps J01 99 even close to the Big Bang. Some possibilities are that galaxy J01 100 clustering started from a random Poisson distribution, or from a J01 101 state with local clustering or from large-scale coherent J01 102 structures. No clear observational evidence for any particular J01 103 initial state has been found. On small scales the clustering J01 104 processes themselves tend to destroy this evidence, while on large J01 105 scales it is difficult to detect.

J01 106 Different types and distributions of dark matter may also be J01 107 important for forming galaxies and clusters. For example, massive J01 108 neutrinos, other weakly interacting massive particles, cosmic J01 109 strings, quark nuggets, or other currently speculative objects of J01 110 various high-energy theories may influence galaxy clustering if J01 111 they exist in sufficient quantity.

J01 112 All known forms of matter gravitate and gravitation promotes J01 113 clustering. Therefore, astronomers have developed analytical J01 114 theories and examined many computer simulations to describe the J01 115 gravitational clustering of galaxies. Results for different models J01 116 are then compared with <*_>xi <*/>(r) and f(N). The J01 117 models generally differ in their initial conditions, the role of J01 118 dark matter, and the time available for clustering.

J01 119 Computer simulations calculate the gravitational orbits of many J01 120 thousands of particles, each one representing a galaxy, in the J01 121 background of the expanding universe and any dark matter present. J01 122 The orbits are found either by integrating the thousands of J01 123 equations of motion - each with thousands of terms - directly, or J01 124 by averaging the gravitational forces in different ways to simplify J01 125 the problem. Averaging sacrifices detailed information in order to J01 126 include a larger number of galaxies.

J01 127 Computer models which start with strong structure on scales of J01 128 tens of megaparsecs frequently do not agree with the observed J01 129 correlation and distribution functions. Those that do, often agree J01 130 only for a short span of their evolution. On the other hand, models J01 131 with fairly homogeneous initial distributions, such as an J01 132 uncorrelated Poisson state, evolve gravitationally to agree J01 133 reasonably well with the observations and remain in agreement as J01 134 they continue to evolve. In other words, they relax into the J01 135 observed state and remain there rather than just pass through it. J01 136 This may make them more aesthetically pleasing, although it does J01 137 not guarantee they are correct. For example, some models in which J01 138 galaxies have formed and clustered very recently may conflict with J01 139 the uniformity of the cosmic microwave background.

J01 140 figure&caption

J01 141 Gravitational theory predicted the observed form of f(N) J01 142 given earlier for relaxed statistically homogeneous clustering of J01 143 point masses in a slowly expanding universe. The value of b is J01 144 essentially the ratio of gravitational correlation energy J01 145 (representing departures from a uniform Poisson distribution) to J01 146 the kinetic energy of galaxies' peculiar velocities (representing J01 147 departures from the Hubble flow). Computer experiments such as the J01 148 example in Fig. 1 show that this relaxed state is a very good J01 149 description for universes which start with no or little large-scale J01 150 correlation, have formula, and have expanded by more than J01 151 several times their initial radius. As differences with these J01 152 conditions become greater, agreement with the observed f(N) J01 153 decreases. These conditions also lead to two-point correlation J01 154 functions in reasonable agreement with the observations provided, J01 155 for example, that clustering started at redshift J01 156 z<*_>approximate-sign<*/>8 in <*_>OMEGA<*/>0=1 J01 157 models and z<*_>approximate-sign<*/>30 in J01 158 <*_>OMEGA<*/>0=0.1 models. Therefore, gravitational J01 159 clustering starting from fairly simple initial conditions seems J01 160 likely to account for the objective statistical evidence now J01 161 available. These include large underdense regions and filamentary J01 162 structures, some of which could have formed just by chance J01 163 concentration of independently clustering regions or their J01 164 boundaries. When more subtle statistics are developed further and J01 165 related to dynamical evolution, perhaps they will reveal clear J01 166 evidence for other processes such as primordial explosions, or J01 167 large-scale initial structures.

J01 168 Cosmology, Cosmic Strings

J01 169 Neil Turok

J01 170 One of the most active areas of current research in physics and J01 171 astronomy is the search for a theory of the formation of structure J01 172 in the universe.

J01 173 Historically, this is a result of the success of two different J01 174 theories and the attempt to combine them. In astronomy, the Hot Big J01 175 Bang model of the universe has three big successes. It successfully J01 176 explains the expansion of the universe, the relic microwave J01 177 background radiation, and the abundances of the light elements J01 178 today. The weakness of the standard Hot Big Bang model is that it J01 179 says nothing about how structure in the universe (galaxies and J01 180 cluster of galaxies) could have originated.

J01 181 In high-energy physics, the idea that the underlying theory of J01 182 particles and their interactions has a high degree of symmetry J01 183 which is broken at low energies forms the basis of the J01 184 Weinberg-Salam model of the electroweak interactions. Over the last J01 185 decade many predictions of this model have been confirmed, the J01 186 discovery of the W and Z particles being the most J01 187 dramatic. Based on the idea of symmetry breaking, theories which J01 188 unify all the forces except gravity (grand unified theories), and J01 189 theories including gravity (superstring theories) have been J01 190 developed. Unfortunately, at present there are many different J01 191 theories, and few ways of testing them.

J01 192 One idea, which emerged from particle physics in the early J01 193 1980s, was that the same process which broke the symmetry between J01 194 the particles and forces might break the spatial symmetry of the J01 195 universe, producing the structure we see today. This is physically J01 196 a very reasonable idea. In fact, similar processes happen in J01 197 everyday substances. Most liquids are quite homogeneous and J01 198 isotropic, which is not surprising, because there is nothing in the J01 199 description of atoms and their interactions that singles out a J01 200 particular direction or place in space as different from any other. J01 201 Cool the liquid, however, and it freezes. The crystal structure of J01 202 the solid picks a particular direction; but in different regions J01 203 different directions are chosen. The result is that (unless the J01 204 process happens very slowly) the solid is formed full of defects J01 205 where there is a mismatch between neighboring crystalline J01 206 regions.

J01 207 figure&caption

J01 208 Cosmic strings are very similar to these defects. They are J01 209 predicted to occur by some grand unified theories and superstring J01 210 theories. In the very early universe, at high temperature the J01 211 fields in these theories are random, just as the atoms of a liquid. J01 212 The density is quite uniform. As the universe cools below a certain J01 213 temperature, some of the fields 'freeze.' As they do so, defects J01 214 are formed, just as in solids.

J01 215 In different theories, the defects may be at points, along J01 216 lines, or in sheets. Cosmic strings are the linelike defects: They J01 217 have special properties which make them well suited to forming J01 218 structure later in the universe. For topological reasons they J01 219 cannot have ends; they must form closed loops or continue on J01 220 forever. The only way to change the length of a string is if it J01 221 crosses itself and reconnects the other way (Fig. 1), chopping off J01 222 a loop. This means that if one starts with some strings which J01 223 wander right across the universe, there is no way to get rid of J01 224 them completely. At best one can progressively chop more and more J01 225 of the long string off into loops, and the loops can then radiate J01 226 away (see Fig. 1). Thus some of the strings formed at very early J01 227 times (around 10-34 s after the Big Bang in most models J01 228 predicting strings) survive right up to today.

J01 229 J02 1 <#FROWN:J02\> We have attempted to make these bibliographies as J02 2 complete and comprehensive as possible. The intensity and J02 3 broadening compilations cover the published literature form J02 4 November 1983 through June 1990, and they include several earlier J02 5 references overlooked in the previous compilations. The J02 6 pressure-induced line shift compilation includes all references J02 7 published through June 1990 that were available to us. We have also J02 8 added a number of more recent references as they came to our J02 9 attention in the course of preparing the present bibliography.

J02 10 II. INTENSITIES

J02 11 Detailed discussions of intensity parameters, units, and J02 12 measurement techniques have been given in Refs. [1-2] and will not J02 13 be repeated here. However, in order to properly interpret the J02 14 values given in Tables II and III and in the individual references, J02 15 we must discuss the difference between the vibrational band J02 16 intensity, S<*_>nu<*/>, and the integrated band intensity J02 17 SBand. S<*_>nu<*/> is related to the squared J02 18 vibrational transition moment by

J02 19 formula

J02 20 where h is Planck's constant, c is the speed of J02 21 light, <*_>nu<*/>0 is the wavenumber of the band center, J02 22 N is the total number of molecules of the absorbing gas per J02 23 cubic centimeter per atmosphere, Q<*_>nu<*/> is the J02 24 vibrational partition function, and formula is the J02 25 vibrational transition moment. SBand is defined as J02 26 the summation of individual line intensities J02 27 SBA for all possible rotational J02 28 transitions within the vibrational band; that is,

J02 29 formula

J02 30 where RBA is the rotational factor, J02 31 given by

J02 32 formula

J02 33 where S'AB is the dimensionless quantity called J02 34 the line strength by Herzberg [3], EA is J02 35 the energy of the lower state A, k is Boltzmann's J02 36 constant, T is the gas temperature in degrees Kelvin, J02 37 <*_>nu<*/>AB is the wavenumber of the transition J02 38 from lower state A to upper state B, and J02 39 Qr is the rotational partition function. In Eqs. J02 40 (1-3) we are using the approximation formula, where Q J02 41 is the total internal partition sum; this approximation is valid J02 42 except at very high temperatures.

J02 43 The F-factor in Eq. (2) accounts for the effects of J02 44 centrifugal distortion, Fermi- and Coriolis-type interactions, and J02 45 other perturbations. For a rigid rotor, the F-factor would be J02 46 1, and S<*_>nu<*/> and SBand would be J02 47 equal (assuming all possible rotational transitions are included in J02 48 the summation). However, if the F-factor is significantly J02 49 different from 1 for most of the rotational transitions, as the J02 50 case for many CO2 bands, SBand is usually J02 51 larger than S<*_>nu<*/>. SBand J02 52 values determined from the summation of individual line-intensity J02 53 measurements using high-resolution techniques are more appropriate J02 54 for comparison to low-resolution band intensity measurements (e.g., J02 55 [4]). Therefore, where a given reference reports both J02 56 S<*_>nu<*/> and SBand values, (e.g., [5]) J02 57 we have chosen to include only the SBand values in J02 58 Table 2. Unless the reported intensity is designated as J02 59 S<*_>nu<*/> or SBand in the footnotes to J02 60 the table, the reader should obtain more detailed information from J02 61 the original reference.

J02 62 Table I gives the various units in which band intensities have J02 63 been reported in our literature survey, along with the J02 64 multiplicative factors used to convert these units to our standard J02 65 units (cm-2 atm-1 at 300K) for comparison J02 66 purposes. However, because these multiplicative factors do not J02 67 account for the temperature dependence of the partition functions J02 68 or the exponential factors shown in Eq. (3), the converted values J02 69 may not represent the true band intensity at 300K, particularly for J02 70 transitions whose lower state is not the ground state. Goldman, J02 71 Dang-Nhu and Bouanich [6] give an excellent brief discussion of the J02 72 temperature dependence of line intensities.

J02 73 Several of the references reporting intensity measurements of J02 74 bands of CHCl3, CDCl3, CHCl2F, J02 75 CHClF2, CHD2F, CHF3, and J02 76 CDF3 give the results not as S<*_>nu<*/> or J02 77 SBand as defined in Eqs. (1-3), but as a quantity J02 78 denoted as G. This G is identical to the <*_>GAMMA<*/> J02 79 originally defined by Golike et al. [7] and is usually J02 80 reported in units of (length)2 mole-1 or J02 81 (length)2 molecule-1. As previously stated in J02 82 [2], G (or <*_>GAMMA<*/>) may be related to J02 83 SBand approximately by SBand = J02 84 G<*_>nu<*/>0, where <*_>nu<*/>0 is the J02 85 wavenumber of the observed band center.

J02 86 To compare measurements by investigators in different J02 87 laboratories and apply their results to atmospheric or J02 88 astrophysical measurements, we must also understand the corrections J02 89 applied for the isotopic composition of the measured gas samples. J02 90 Unfortunately, in many papers, the isotopic composition is not J02 91 specified, and the interpretation of 'natural abundance' is left to J02 92 the reader. Johns [8] discussed in detail the difference between J02 93 the 'natural abundance' of 13C in atmospheric J02 94 CO2 and that in commercially supplied CO2. In J02 95 our conversion of reported intensities to a consistent set of J02 96 units, we have attempted to rescale each intensity value to J02 97 represent the band intensity for a 100% pure sample of a single J02 98 isotope. Where no specific isotopic abundance information has been J02 99 given in the original reference, we have used the 1986 HITRAN J02 100 abundances [9].

J02 101 III. COLLISION BROADENING

J02 102 For most of the range of pressures typically encountered in the J02 103 terrestrial and planetary atmospheres, infrared vibration-rotation J02 104 lines have predominantly the Lorentz, or collision-broadened line J02 105 shape, in which the width of the line is linearly dependent on the J02 106 gas pressure (for a fixed temperature). At very low gas pressures, J02 107 the line shapes follow the Doppler profile, where the line width J02 108 depends only on the temperature, mass of the molecule, and J02 109 wavenumber of the transition. In the intermediate range where the J02 110 Doppler and Lorentz profiles both contribute significantly, the J02 111 line shape is most often modeled by the Voigt profile, which is the J02 112 convolution of these two profiles. Detailed expressions for these J02 113 three line profiles have been given in [2] and will not be repeated J02 114 here. The dependence of the collision-broadened halfwidth on J02 115 pressure and temperature is usually expressed as

J02 116 formula

J02 117 where bL(p,T) is the collision-broadened J02 118 halfwidth of the line at measured pressure p and temperature J02 119 T, b0L(T0) is the J02 120 collision-broadened halfwidth of the line at the reference pressure J02 121 (usually 1 atm) and temperature T0 (usually 296 K). J02 122 According to the classical theory, the exponent n should have J02 123 the value 0.5; however, many of the experimental results cited in J02 124 Table IV of the present compilation and in [2], show quite J02 125 different values for n. The experimentally determined values J02 126 of this exponent appear to vary according to the absorbing and J02 127 perturbing gases and the vibrational and rotational quantum J02 128 numbers.

J02 129 In recent years measurements of Lorentz broadening coefficients J02 130 and their temperature dependence have become more numerous. There J02 131 have been more measurements of individual line widths within a J02 132 band, and very few studies reporting simply an average broadening J02 133 coefficient (or n value) for an entire band. Recognizing that J02 134 temperatures in the terrestrial upper atmosphere and in planetary J02 135 atmospheres are quite different from ambient values in the J02 136 laboratory, several investigators have pursued measurements of J02 137 collision-broadened gas-phase spectra at low temperatures J02 138 down to about 150K. Other investigators, driven by requirements to J02 139 measure the spectra of combustion exhaust gases in situ, J02 140 have recorded laboratory collision-broadened spectra at high J02 141 temperatures.

J02 142 Another are of increased interest has been the examination of J02 143 line shapes that depart from the usual three forms (Lorentz, J02 144 Doppler, Voigt) most commonly used to model laboratory, J02 145 atmospheric, planetary, or astronomical spectra. Collisional J02 146 narrowing is observed for spectral lines whose Lorentz halfwidth J02 147 bL is very small. The Galatry line shape [10] is J02 148 usually used to model the profiles of absorption lines affected by J02 149 collisional narrowing in addition to Doppler and Lorentz J02 150 broadening. Other phenomena that have been examined extensively J02 151 include non-Lorentzian profiles in the far wings of very strong J02 152 absorption lines (particularly for H2O and CO2) J02 153 and line mixing (also called line coupling) in regions of J02 154 very closely spaced lines such as Q-branches. References J02 155 for laboratory studies of collisional narrowing, far-wing line J02 156 shapes, and line mixing have also been included in Table IV.

J02 157 IV. PRESSURE-INDUCED LINE SHIFTS

J02 158 Shifts in the positions of vibration-rotation lines due to J02 159 collisions with other gases have become a concern for certain types J02 160 of infrared measurements. Since ambient pressures in the J02 161 terrestrial and planetary atmospheres range from several Torr to J02 162 several atmospheres, pressure-induced line shifts must be J02 163 considered in spectroscopic studies of these atmospheres. Knowledge J02 164 of line shifts is also important for certain transitions of J02 165 CO2, CH4, and other gases that are used for J02 166 stabilization of laser frequencies.

J02 167 Prior to the mid-1980s, very few measurements of J02 168 pressure-induced line shifts appeared in the literature; most of J02 169 these were for diatomic molecules such as CO, HBr, HCl, HF, J02 170 H2, and HD. More recent studies have provided extensive J02 171 information on shifts in transitions of larger molecules, including J02 172 CO2, H2O, N2O, O3, J02 173 NH3, and CH4. However, some of these studies J02 174 report measurements of shifts for only a small number of lines J02 175 within a band.

J02 176 Theoretical prediction of pressure-induced line shifts has been J02 177 successful only for selected types of molecules such as polar J02 178 diatomic molecules [11, 12]. Adequate models for pressure-induced J02 179 shifts in bands of other molecules of interest for atmospheric J02 180 studies, such as O3 or CH4, have not yet J02 181 appeared. Only a very few measurements of pressure-induced line J02 182 shifts at temperatures far above or below room temperature have J02 183 been reported, and the form of the temperature dependence of the J02 184 shifts is not certain. A number of investigators, including J02 185 Grossmann and Browell [13, 14], have modeled pressure-induced line J02 186 shifts at low temperatures using a relation similar to that used J02 187 for the Lorentz halfwidths:

J02 188 formula

J02 189 where <*_>delta<*/>(T) is the shift coefficient (in J02 190 cm-1/atm) at the measured temperature T, J02 191 <*_>delta<*/>(T0) is the shift coefficient at the J02 192 reference temperature T0, and the exponent n' J02 193 is empirically determined. However, the above expression is not J02 194 valid in cases where the line shift changes sign with temperature. J02 195 Other investigators, such as Houdeau and Boulet [15], have J02 196 developed more rigorous theoretical models to account for the J02 197 temperature dependence of both halfwidths and shifts.

J02 198 V. EXPLANATION OF TABLES AND TABLE REFERENCES

J02 199 The tables continue in the format of Refs. [1-2] with one J02 200 exception. For carbon dioxide and its isotopic variants, we have J02 201 adopted the vibrational band notation of Rothhman and Young [16], J02 202 which is also used in the HITRAN and GEISA spectroscopic line J02 203 parameters compilations [9, 17]. This notation is widely used by J02 204 numerous investigators. Briefly, each vibrational level of J02 205 CO2 is designated by an integer whose five digits J02 206 correspond to the sequence formula, where J02 207 <*_>nu<*/>1, <*_>nu<*/>2, and J02 208 <*_>nu<*/>3 are the vibrational quantum numbers, l J02 209 is the degeneracy index of the <*_>nu<*/>2 vibrational J02 210 mode, and r indicates the level's ranking in a Fermi resonance J02 211 polyad. Thus, the ground state is designated by 00001, the J02 212 <*_>nu<*/>12 level by 01101, the J02 213 <*_>nu<*/>1 level by 10001, and the J02 214 2<*_>nu<*/>02 level by 10002. For each J02 215 transition the level designations are given with the upper level J02 216 first. For example, the 12C16O2 laser J02 217 transition at 961 cm-1, designated J02 218 <*_>nu<*/>3-<*_>nu<*/>1 in the previous J02 219 compilations, is indicated as 00011-10001 in the present work, and J02 220 the formula laser transition at 1064 cm-1 is J02 221 indicated as 00011-10002.

J02 222 The literature values and references for intensities of J02 223 vibration-rotation transitions are given in Tables II and III. All J02 224 of these values are for electric dipole allowed transitions J02 225 measured in the gas phase with a few exceptions. These include the J02 226 'forbidden' 03301-00001 band of J02 227 12C16O2 at 2003 cm-1, J02 228 liquid-phase measurements of several N2O bands J02 229 that were published along with gas-phase measurements, and J02 230 quadrupole and electrically-induced dipole vibration-rotation J02 231 transitions of H2 and N2. Purely rotational and J02 232 electronic transitions are not included in the compilation.

J02 233 As in the previous compilations, Table II contains the data for J02 234 molecules with 2,3,4, and 5 atoms. The molecular data are sorted J02 235 first by number of atoms, then by molecular formula in alphabetical J02 236 order, then by isotopic form in order of increasing molecular J02 237 weight. If the isotopic forms are not individually spectified, the J02 238 reader should assume that the data refer to the most abundant J02 239 isotope of the molecule. The first page of Table II serves as an J02 240 index to the succeeding pages of the table. The bands for each J02 241 molecule are presented in order of increasing wave number, with the J02 242 assignments and band centers (in cm-1) given in the first J02 243 column of the table. In the second column, headed Reference J02 244 Value, are listed the integrated band intensity values as J02 245 reported in the original references. Any unusual features of the J02 246 reported measurements are indicated by footnotes in the relevant J02 247 sections of the table. J02 248 J02 249 J03 1 <#FROWN:J03\>If this is indeed the case the data in Fig. 5 may be J03 2 skewed upward as r/R decreases, since they could still J03 3 reflect, in part, the highly aligning nature of the converging J03 4 extensional flow at the tube entrance.

J03 5 At the highest shear rate studied in this work, 1270 J03 6 s-1, a value of 0.66 was found for S11, J03 7 indicating essentially perfect fiber alignment (S11 J03 8 = 2/3).

J03 9 A priori predictions may be obtained using Eqs. J03 10 (3)-(8) together with the parameters given in Table III. That shown J03 11 for S11 in Fig. 5 is seen to portray the data quite J03 12 well within their experimental uncertainty, and this was equally J03 13 true at the other conditions studied. This is the most gratifying J03 14 aspect of the analysis: the ability to make at least approximate J03 15 predictions of fiber orientation using only the measured apparent J03 16 viscosity function.

J03 17 (2) The parameter S12, as noted earlier, is a J03 18 measure of the skewness of the orientations about the flow J03 19 direction. Intuitively one would expect that, for fibers aligned in J03 20 a shearing flow, this parameter should be very close to zero, and J03 21 all 14 sets of experimental data, including that shown in Fig. 5, J03 22 bear this out. The theoretical prediction is thus simply J03 23 incorrect.

J03 24 (3) The components S22 and S33 J03 25 are smaller than S11, as expected, and are J03 26 consequently of lesser interest. The theoretical predictions are J03 27 seen to be of the same general magnitude as the data but the J03 28 agreement is not at all exact. Again, both data and theory show J03 29 very little change with shear rate.

J03 30 In conclusion, while further evaluation of the theory will be J03 31 needed before fiber orientations can be predicted with confidence J03 32 the orientation parameter of greatest interest, S11, J03 33 and its variation with shear rate, appear to be predicted well. J03 34 Further, very low levels of shearing rate appear to be sufficient J03 35 for substantial fiber alignment in these concentrated systems.

J03 36 Flow behavior and morphology of extruded polymers

J03 37 Die swell and melt fracture are ordinary phenomena which occur J03 38 during the extrusion of unfilled polymers from dies of any J03 39 geometry. Die swell generally increases with extrusion rate, and J03 40 melt fracture occurs at a shear stress near 0.1 MPa, as reported by J03 41 Ramamurthy (1986) and others. For extrusion form a tube, die swell J03 42 changes the diameter of the extruded filament and melt fracture may J03 43 produce surface matteness as well as both longitudinal and radial J03 44 ripples. Neither die swell nor melt fracture, however, J03 45 significantly changes the shape of a cross section of the extruded J03 46 filament from the circular shape induced by the die, except at very J03 47 high stress levels well beyond those at which incipient J03 48 irregularities are formed.

J03 49 The extrusion of fiber-filled polyethylenes and polypropylenes J03 50 from a circular die involves reduced die swell (as compared to the J03 51 unfilled polymers) and the occurrence of gross surface J03 52 irregularities not related to the phenomenon of melt fracture. The J03 53 photographs in Figs. 6 and 7 illustrate the surface irregularities J03 54 which occur during the extrusion of polypropylene containing 10 and J03 55 40 wt % glass fibers. These polymers were extruded at J03 56 230<*_>degree<*/>C from a 0.05 in. i.d. by 2 in. long capillary at J03 57 the apparent shear rates shown in the figures. Similar morphologies J03 58 were observed for the fiber-filled polyethylenes.

J03 59 The photographs in Figs. 6 and 7 show that extraordinary J03 60 surface irregularities and longitudinal variations in the diameters J03 61 of the filaments occurred at the lowest extrusion rates. Numerous J03 62 fibers protruded from the surface of the filament giving them a J03 63 fuzzy appearance, and the orientations of the protruding fibers J03 64 seemed to be random. As extrusion rates increased, the surface J03 65 irregularities faded and the diameters of the filaments became much J03 66 more uniform. The measured shear stresses for the filaments in J03 67 Figs. 6 and 7 obtained at 4.6 and 115 s -1 were well J03 68 below the melt fracture stress of 0.1 MPa, and the shear stresses J03 69 at 1272 s-1 were only slightly above the critical value. J03 70 Melt fracture, therefore, did not seem to be a contributing factor J03 71 in the observed surface irregularities, nor was any indication of J03 72 sudden changes in surface irregularity observed at the critical J03 73 stress of 0.1 MPa.

J03 74 figures&captures

J03 75 Wu (1979), Crowson et al. (1980), and Knuttson and J03 76 White (1981) have reported similar surface morphologies for J03 77 fiber-filled polymer melts. Knuttson and White reported that the J03 78 surface of fiber-reinforced polycarbonate extrudates was J03 79 significantly rougher under all conditions than that of the J03 80 unfilled polymer, but the smoothness of the extrudates increased J03 81 with extrusion rate. They attributed the increased smoothness to J03 82 increased fiber orientation in the direction of flow, although we J03 83 have noted only modest changes for our materials. Crowson et J03 84 al. observed a similar improvement in smoothness with J03 85 increased extrusion rate, and they found that shear flow produced a J03 86 decrease in fiber alignment parallel to the flow direction. Wu in J03 87 his study of glass fiber-filled poly(ethylene terephthalate), J03 88 observed three distinct regimes of extrudate surface morphology: a J03 89 smooth surface at low shear rates, irregular surface at medium J03 90 shear rates, and a somewhat smoother extrudate at high rates. Wu J03 91 suggested that the observed morphologies arose from normal stress J03 92 effects and rotation of fibers in the velocity field. Thus we see J03 93 that a comprehensive and consistent understanding of physical J03 94 concepts leading to such extrudate morphologies has not been J03 95 developed.

J03 96 Cross sections of the extruded glass fiber-filled J03 97 polypropylenes are shown in Figs. 8 and 9 for fiber loadings of 10 J03 98 and 40 wt %, respectively. The cross sections in these figures are J03 99 digitized representations obtained from photomicrographs. The J03 100 sections are perpendicular to the axis of the filaments, and the J03 101 shear rates correspond to those in Figs. 6 and 7. The points in J03 102 Figs. 8 and 9 represent the locations of individual, well-aligned J03 103 fibers. The most striking observation to be made from Figs. 8 and 9 J03 104 is that the shapes of the cross sections are extremely distorted at J03 105 low shear rates and become largely circular at high rates.

J03 106 figures&captures

J03 107 The same shear rate dependence of morphology was observed for J03 108 suspensions extruded from capillaries with lengths of 1, 2, and 3 J03 109 in. The distortion disappeared, however, when the materials were J03 110 extruded through an orifice. Cross sections of extrudates of the 10 J03 111 wt % suspension in polypropylene obtained from the orifice are J03 112 shown in Fig. 10. The orifice had the same diameter (0.05 in.) as J03 113 the capillaries for which surface irregularities were observed, and J03 114 it retained the 90<*_>degree<*/> convergence at the entrance. For J03 115 the orifice cross sections in Fig. 10, there appears to be slight J03 116 reversal in the distortion of the cross-sectional shape; i.e., the J03 117 shape is quite circular and smooth at the lowest rate and becomes J03 118 more jagged at the highest rate.

J03 119 figure&caption

J03 120 Factors affecting surface morphology

J03 121 A number of factors were considered for the development of a J03 122 plausible mechanism for the observed surface morphologies. As J03 123 pointed out in the previous section, melt fracture did not appear J03 124 to play a role since the shear stresses at which distortion J03 125 occurred were too low. Other factors considered by Becraft (1988) J03 126 were die swell, thermal connection of the polymer, surface effects J03 127 between the polymer and metal, flexing of fibers near the surface J03 128 of the extruded filaments, expansion of gas bubbles in the J03 129 material, regions of local fiber alignment, stress variations J03 130 resulting from fiber concentration inhomogeneities, and velocity J03 131 profile rearrangements at the tube exit. The more interesting and J03 132 pertinent of these considerations are as follows.

J03 133 Die swell

J03 134 Die swell was significantly reduced in the fiber-filled J03 135 polymers and it was unlikely that the large distortions observed in J03 136 the extrudates resulted from swelling of the polymer. Die swell J03 137 increased with extrusion rate even for the filled polymers. J03 138 Therefore, if swelling were the relevant mechanism, extrudate J03 139 distortion should have increased with shear rate, not decreased as J03 140 observed.

J03 141 Fiber flexing at the surface of extrudates

J03 142 The glass fibers in the polyethylenes and polypropylenes are J03 143 somewhat flexible, and the flexibility increases with temperature. J03 144 This creates the opportunity, with a large enough fiber flexural J03 145 modulus, for misaligned or bent fibers near the surface of the J03 146 extruded polymers to 'spring out' while the material is still in a J03 147 molten state. During this straightening process the fiber may J03 148 entrain polymer and conceivably cause the nonuniformities observed J03 149 in the cross sections. The photograph in Fig. 11 illustrates J03 150 entrainment of polymer by a fiber protruding from the surface. This J03 151 photograph was taken at the surface of a distorted extrudate using J03 152 polarized light and a magnification of 90x. An envelope of polymer J03 153 surrounds the protruding fiber and a surface irregularity has been J03 154 created by the entrained polymer. Protruding fibers in all of the J03 155 samples showed similar entrainment of polymer.

J03 156 To pursue this possible origin of the surface morphology the J03 157 dynamics of flow from the tube was observed using a Panasonic VHS J03 158 camcorder (model PV200). In the recorded images at low shear rates, J03 159 fibers were visibly springing out from the surface of the J03 160 extrudates immediately after the material emerged from the J03 161 capillary. All fiber motion appeared to be complete within one or J03 162 two diameters of the tube exit. It was not possible, however, to J03 163 determine from the flow visualization whether the springing out of J03 164 fibers was a consequence of long fiber flexibility or of rotation J03 165 of rigid fibers. A random side-to-side motion of the exiting J03 166 filaments was also evident, and was produced by forces originating J03 167 at or prior to the tube exit. At higher shear rates, the frequency J03 168 of the side-to-side motion increased dramatically, and the movement J03 169 resembled that of a vibrating string although with no obvious J03 170 periodicity.

J03 171 Fiber flexing and 'spring back' of bent fibers implies the J03 172 occurrence of bent fibers within the polymer melt during flow into J03 173 and through the tube. No such bent fibers were ever found in any of J03 174 these extrudates, possibly because cooling of the emerging J03 175 extrudate in the air was slower than the rate of springback of the J03 176 fibers. To check, the extrudates were also extruded into a cold J03 177 glycol-water bath held within one diameter of the capillary exit (a J03 178 closer positioning was not possible). Although perfect quenching J03 179 could not be achieved significant differences between quenched and J03 180 unquenched extrudates could be found. The quenched extrudates had J03 181 only much shorter fibers protruding from the surface, and J03 182 examination of quenched sections under polarized light revealed J03 183 numerous bent fibers within the frozen polymer. Clearly, then, J03 184 fiber flexing contributes to the observed extrudate J03 185 irregularities.

J03 186 Fiber concentration inhomogeneities

J03 187 Evidence of local variations in fiber concentration appears in J03 188 the cross section of Figs. 8 -10. Close examination of these cross J03 189 sections reveals that there are regions that seem to be completely J03 190 free of fibers, and other regions which have high densities of J03 191 fibers. Such variations in fiber concentration may cause J03 192 point-to-point variations in viscosity and a variable deformation J03 193 rate profile across the diameter of the molten polymer as it flows. J03 194 The effects would be greatest at low deformation rates when the J03 195 local viscosity variations due to differences in fiber J03 196 concentration (Figs.2 - 4) are greatest; at increasing shear rates J03 197 the viscosity differences diminish and a smoother extrudate would J03 198 appear, in attractive agreement with that observation. Radial fiber J03 199 migrations were reported for poly(ethylene terephthalate)-glass J03 200 systems by Wu (1979) but, in the present work on J03 201 polypropylene-glass fiber systems careful measurements revealed no J03 202 such large-scale changes. However, local concentration fluctuations J03 203 were indeed of large magnitude: in the 10 wt % material the J03 204 volume fractions of fibers varied between 0 and 0.15 and, for J03 205 the 40 wt % system, between 0.09 and 0.35. Using the data of Kitano J03 206 et al. (1981) to estimate the local viscosity variations J03 207 tenfold changes are computed. To determine more definitively the J03 208 probable effects of large viscosity variations on extrudate J03 209 morphology, an experiment was conducted using a 50/50 wt % blend of J03 210 high density polyethylene and polypropylene. The viscosities of J03 211 these two immiscible polymers varied by nearly an order of J03 212 magnitude at low deformation rates and a temperature of 180 J03 213 <*_>degree<*/>C. The blend was prepared by chopping extruded J03 214 strands of the individual polymers into fragments 1 mm or less in J03 215 length and mechanically mixing an equal mass of each of the solid J03 216 fragments. The blend was extruded through a 2 in. long capillary J03 217 with an L/D of 40, and cross sections of the extrudates were J03 218 examined for distortion. J03 219 J04 1 <#FROWN:J04\>We also derive, for the case of random sphere J04 2 percolation, an exact Smoluchowski equation describing the J04 3 scaling behavior of the mean cluster densities. In Sec. III we J04 4 focus on the function nc(k), which gives the mean J04 5 number of clusters, per unit volume, containing exactly k J04 6 particles. We obtain an exact differential equation for this J04 7 quantity and discuss its solution. We also give an efficient J04 8 general numerical method for solving this equation and, thus, for J04 9 obtaining cluster-size distributions. In Sec. IV we extend these J04 10 methods to clustering in ionic systems. In Sec. V, we show how to J04 11 calculate the mean volume of a cluster as a function of cluster J04 12 size. To do this, we first extend earlier research by Fanti et J04 13 al. [20] to give a recipe for calculating the mean volume of a J04 14 cluster. We then use the ghost-field method developed in previous J04 15 sections to obtain the full distribution of mean cluster volumes. J04 16 Section VI gives our conclusions. In the Appendix, we give a closed J04 17 set of Ornstein-Zernike equations that are satisfied by the J04 18 asymmetric (ghost-field-dependent) connectedness functions. We J04 19 discuss the difficulties in obtaining the mean-spherical J04 20 approximation (MSA) for this system. Finally, we discuss the J04 21 relationship between this system of equations and similar systems J04 22 encountered in other random media problems, including correlations J04 23 of a liquid in porous media and structure of the gel phase in a J04 24 percolating medium.

J04 25 II. GENERATING FUNCTION FORMALISM FOR CLUSTER J04 26 STATISTICS

J04 27 In this section we show that including a color field, or ghost J04 28 field, in a continuum percolation model allows one to develop J04 29 generating functions for the basic quantities describing such a J04 30 model. We develop these generating functions using the J04 31 continuum-Potts-model formalism already applied to percolation J04 32 without a ghost field. Certain basic relations between these J04 33 generating functions are also developed. In particular, we show J04 34 that the virial theorem for percolation models provides an exact J04 35 equation of Smoluchowski type describing the growth and coalescence J04 36 of clusters as the density is increased.

J04 37 The one-state limit of the continuum Potts model [21-23] gives J04 38 a very general correlated, continuum percolation model. We have J04 39 previously [12] discussed in detail this relationship, which is the J04 40 continuum generalization of the mapping originally developed for J04 41 lattice percolation by Fortuin and Kastelyn [14]. Here we add a J04 42 ghost field to the Hamiltonian in order to develop the basic J04 43 quantities describing this model as generating functions. The ghost J04 44 field [16, 18] is an artificial external field of constant J04 45 magnitude that acts on particles in all but one species (we choose J04 46 species 1 for definiteness). Since the correspondence between J04 47 Potts-model quantities and percolation quantities has been J04 48 developed in detail previously, we do not repeat this work here. We J04 49 rather focus on the additional structure created by including the J04 50 ghost field in the CPM.

J04 51 The asymmetric CPM is a generalization of the Widom-Rowlinson J04 52 model [24]; it describes a mixture of s different species J04 53 which can undergo phase separation at sufficiently low J04 54 temperatures. In the one-state limit, this phase-separation J04 55 transition becomes a percolation transition. This mapping exhibits J04 56 continuum percolation as an analytic continuation of a many-body J04 57 system of a standard type, and allows the methods of liquid-state J04 58 physics to be applied to it. The Hamiltonian of this system is

J04 59 formula.

J04 60 The first term is a repulsive interaction [we will take J04 61 <*_>nu<*/>(x)>0] acting only between particles of different species J04 62 (we use the symbol <*_>sigma<*/>i to denote the species J04 63 of particle i). The second term is a species-independent J04 64 potential acting between each pair of particles. In the one-species J04 65 limit, the thermodynamic quantities describing this model become J04 66 the basic quantities describing a very general model of clustering J04 67 and percolation in the continuum. This model consists of particles J04 68 with interparticle correlations induced by the potential J04 69 <*_>phi<*/>(x) and bonds connecting each pair of particles J04 70 with a separation-dependent bond probability J04 71 pb(x) given by

J04 72 formula.

J04 73 The last term in (2.1) is a ghost field which acts equally on J04 74 particles of all species except those in one distinguished J04 75 species (we choose species 1 for definiteness; henceforth we will J04 76 use subscripts i and j to denote species other than J04 77 1). This term can be absorbed into the fugacities characterizing J04 78 different species; if particles of species 1 have activity J04 79 z1, then particles of species i have J04 80 activity

J04 81 formula,

J04 82 with H=<*_>beta<*/>h, <*_>beta<*/>=1/kT the inverse J04 83 temperature and h the ghost field from (2.1). If particles in J04 84 species 1 are present with density <*_>rho<*/>1, those in J04 85 every other species have density <*_>rho<*/>i, a J04 86 nontrivial function of the fugacities J04 87 z1,zi. The one-state limits of J04 88 these quantities are

J04 89 formula,

J04 90 where <*_>unch<*/> is the percolation density, while

J04 91 formula,

J04 92 where nc(K) is the density, or number per unit J04 93 volume, of clusters containing exactly k particles. The J04 94 quantity <*_>unch<*/>i(H) is a generating function for J04 95 the cluster densities nc(k). Furthermore, if the J04 96 field H were pure imaginary, we could recover the individual J04 97 cluster densities from the function <*_>unch<*/>i(H) by J04 98 performing an inverse Fourier transform. The convergence of the J04 99 series (2.5) for purely imaginary values of H follows from the J04 100 exponential decay, as a function of cluster size k, of the J04 101 {nc(k)}. We will make this program explicit in Sec. J04 102 III. Here, we will assume that inversions [25] of this kind are in J04 103 fact possible and focus on presenting formal results. That is, we J04 104 will give the other basic generating functions provided by the CPM J04 105 mapping and state the results of performing the inverse transform J04 106 just described. Note that, according to (2.5), setting H to J04 107 zero givesformula [we treat here only the regime below the J04 108 percolation transition, and thus all particles are contained in J04 109 finite clusters; see, however, the discussion below Eq. (A18) in J04 110 the Appendix].

J04 111 The other low-order moments of the generating function are also J04 112 given by basic CPM quantities. The one-state limit of the J04 113 pressure is

J04 114 formula.

J04 115 We can also define the H-dependent version of the mean J04 116 cluster size,

J04 117 formula.

J04 118 We give an integral equation satisfied by this function below. J04 119 The one-species limits of the distribution functions are also J04 120 generating functions for useful quantities, as we now show. As J04 121 before, we will use subscripts i and j to denote two J04 122 different species other than species 1, which we have J04 123 distinguished by applying a ghost field. In the one-species limit J04 124 s<*_>arrow<*/>1, the CPM distribution functions give connectedness J04 125 and blocking functions, denoted gc(x,H) and J04 126 gb(x,H), respectively, according to

J04 127 formulae.

J04 128 The zero-field limits of these functions, which we write simply J04 129 as gb(x) and gc(x), respectively, J04 130 have a direct physical interpretation: The connectedness function J04 131 gc(x) is the probability density associated with J04 132 finding two particles with separation x both in the same J04 133 cluster. Similarly, the blocking function gb(x) is J04 134 the probability density associated with finding two particles with J04 135 separation x but in different clusters. These two functions J04 136 are related by

J04 137 formula,

J04 138 with gt(x) the thermal correlation function J04 139 associated with the potential interaction <*_>phi<*/>(x). For J04 140 the nonzero H field, the H-dependent CPM J04 141 distribution functions, in the s<*_>arrow<*/>1 limit, give J04 142 generating functions for blocking and connectedness distribution J04 143 functions restricted to clusters of specified size, which we denote J04 144 formula, formula, according to

J04 145 formulae.

J04 146 Here formula is the probability density associated with J04 147 finding two particles separated by a distance x, but in J04 148 different clusters, one of size m, one of size n. Also, J04 149 formula is the probability density associated with finding J04 150 two particles separated by distance x and both in the same J04 151 m-particle cluster. An explicit system of Ornstein-Zernike J04 152 equations obeyed by the correlation functions of an asymmetric CPM J04 153 is presented in the Appendix. Our strategy is to solve these J04 154 equations in the s<*_>arrow<*/>1 limit, and use their J04 155 solutions, as indicated in (2.11) and (2.12) to recover the J04 156 H-dependent connectedness functions. The function J04 157 gij is itself useful as a generating function. It is J04 158 valuable as an intermediate quantity because it obeys the relation J04 159 formula. It is also important because it obeys the J04 160 Ornstein-Zernike equation

J04 161 formula.

J04 162 Here the symbol <*_>unch<*/> denotes a convolution. A basic J04 163 equation relating the above generating functions is the virial J04 164 theorem for the CPM, and its s<*_>arrow<*/>1 limit, the virial J04 165 theorem for percolation. This theorem has been developed [12] for J04 166 the general correlated percolation model defined below Eq. (2.1). J04 167 However, for simplicity, we discuss here only the case of random J04 168 sphere percolation, in which randomly located particles are J04 169 directly connected if and only if they are closer together than a J04 170 fixed distance a. The virial theorem for the zero-field case J04 171 of this model is

J04 172 formula.

J04 173 For the model of sphere percolation, nc depends J04 174 only on the dimensionless variable formula. Thus we can J04 175 rewrite

J04 176 formula.

J04 177 This equation can be given a direct physical interpretation by J04 178 identifying the particle diameter a with the time variable in J04 179 a particular growth process. Specifically, we imagine randomly J04 180 distributed particles with zero diameter at time t=0, with J04 181 their diameters growing at a constant rate so as to remain J04 182 proportional to the time. As the particles grow, they overlap, and J04 183 form successively larger clusters. Now we focus on a specific J04 184 particle at the moment when its diameter has just grown to size J04 185 a. The quantity formula gives the density of particles J04 186 that are in different clusters from the particle in question, and J04 187 are about to become directly connected to it. The factor J04 188 4<*_>pi<*/>a2 gives the total surface area on which this J04 189 can happen, while the factor of 1/2 prevents double counting. Such J04 190 coalescence events are precisely the ones that decrease the total J04 191 number of clusters.

J04 192 We can extend this result to show that the cluster densities J04 193 {nc(k)} for random continuum percolation obey an J04 194 exact equation of Smoluchowski type. To see this, we write J04 195 down the virial theorem for our asymmetric Potts model, J04 196 analytically continue the ghost field H to purely imaginary J04 197 values, and then perform an inverse Fourier transform. It is J04 198 convenient to express the result in terms of reaction rates J04 199 describing the fusion and breakup of clusters. To do this, we J04 200 define an H-dependent blocking function J04 201 gb(x,H) in terms of the function J04 202 <*_>rho<*/>b(x,H) already defined:

J04 203 formula.

J04 204 This function, evaluated at contact, is a generating function J04 205 for reaction rates according to

J04 206 formula.

J04 207 In terms of these quantities, we can write the following J04 208 equation of Smoluchowski type for the cluster densities:

J04 209 formula.

J04 210 We pause to note the connection between this equation and the J04 211 form of Smoluchowski equation which is furnished by a mean-field J04 212 treatment. If we replace the function gb(x,H) on the J04 213 right-hand side (RHS) of Eq. (2.16) by the function J04 214 gb(x), i.e., if we ignore the H dependence of J04 215 this function, the summations over the subscripts m and n J04 216 in the terms on the RHS of Eq. (2.18) will be eliminated. If we set J04 217 m=0 and n=0 in these terms, a mean-field J04 218 Smoluchowski equation results. The probabilistic interpretation of J04 219 such an equation is immediate: The first term on the RHS gives the J04 220 rate at which pairs of smaller clusters coalesce to give k J04 221 clusters. The second term on the RHS gives the rate at which k J04 222 clusters are removed by fusing into larger clusters. Equation J04 223 (2.18) thus has the form of a generalized Smoluchowski equation J04 224 [26]; we emphasize that this equation is exact. Here the contact J04 225 values of blocking functions play the role of rate constants. These J04 226 contact values can be derived for random sphere percolation by J04 227 using an asymmetric version of the scaled-particle theory for J04 228 percolation [13]. In the general case a variety of integral J04 229 equation methods have been developed for calculating these J04 230 quantities, as we discuss in Sec. III and also in the Appendix.

J04 231 III. GENERAL ALGORITHM FOR CLUSTER-SIZE J04 232 DISTRIBUTIONS

J04 233 In this section we discuss the cluster-size distribution in a J04 234 very general correlated continuum percolation model. We first use J04 235 the formalism of Sec. II to develop a differential equation for the J04 236 quantity formula, which, as we have shown, is the J04 237 generating function for the cluster densities nc(k). J04 238 The RHS of this differential equation involves the volume integral J04 239 of the H-dependent connectedness function J04 240 <*_>rho<*/>c(x,H). Numerical solution of this equation is J04 241 computationally demanding; we discuss efficient methods for solving J04 242 this equation.

J04 243 The analog, for percolation, of the compressibility theorem is J04 244 the theorem giving the mean cluster size as an integral over the J04 245 two-point connectedness function. The probabilistic argument that J04 246 gives this theorem has a form specific to k clusters which we J04 247 now describe. J04 248 J05 1 <#FROWN:J05\>Thus, each finger is related by a rotation of 3 x J05 2 32<*_>degree<*/> and a translation of about 10 A<*_>circlet<*/> (3 J05 3 x 3.3 A<*_>circlet<*/>) along the DNA axis. Unlike the recognition J05 4 helix of the helix-turn-helix motif, the <*_>alpha<*/> helix in the J05 5 zinc finger is tipped at an angle away from the major groove. The J05 6 <*_>beta<*/> sheet is on the back of the helix away from the base J05 7 pairs and is shifted towards one side of the major groove. The two J05 8 strands of the <*_>beta<*/> sheet have different roles. The first J05 9 <*_>beta<*/> strand does not make contact with the DNA, whereas the J05 10 second <*_>beta<*/> strand is in contact with the sugar phosphate J05 11 backbone along one strand of DNA.

J05 12 The zinc finger peptide makes 11 hydrogen bonds with the bases J05 13 in the major groove. Six amino acid side chains interact with the J05 14 base pairs, two from each zinc finger. However, five of the six J05 15 residues that are binding to the nucleic acids are arginine J05 16 residues. One of these arginines immediately precedes the J05 17 <*_>alpha<*/> helix in each of the three fingers, and it also J05 18 includes the residues from either the second, the third, or the J05 19 sixth residues in the <*_>alpha<*/> helix. All of these form J05 20 hydrogen bonds with the G-rich strand of the consensus binding J05 21 site. Figure 7 shows the interactions of the various arginine J05 22 residues and one histidine residue from the three fingers. It can J05 23 be seen that five of the six interactions are those postulated at J05 24 an earlier time in an analysis of protein-nucleic acid J05 25 recognition (Fig. 3).

J05 26 Each zinc finger interacts with a subset of three bases. All of J05 27 them interact with the first base in the subset. However, two of J05 28 them interact with the third base of the subset and one with the J05 29 second base. The detailed interactions are listed in Table II and J05 30 are also shown in Fig. 7. Some of the arginine residues binding to J05 31 guanine are also stabilized by aspartic acids that occur as the J05 32 second residue in the <*_>alpha<*/> helices. It is likely that this J05 33 side-chain interaction helps to stabilize the binding of the J05 34 arginine to the guanine residue, although it is not found in all of J05 35 the fingers. The recognition system is relatively simple. The J05 36 residue immediately preceding the <*_>alpha<*/> helix contacts the J05 37 third base on the primary strand of the substrate at the 5' end. J05 38 The third residue on the <*_>alpha<*/> helix can contact the second J05 39 base on the primary strand, and the sixth residue can contact the J05 40 first base. In this structure, each zinc finger came in contact J05 41 with only two bases of each three-base subset. It is not known J05 42 whether it would be possible to have other zinc fingers in which J05 43 all three bases are recognized.

J05 44 The DNA is essentially in the form of a B-type helix with small J05 45 distortions. There are small changes in base-pair twist going from J05 46 one site to the next, although the overall conformation is similar J05 47 to normal B-DNA. Each of the three fingers binds in a similar J05 48 orientation and has similar contacts with the three base pairs of J05 49 the DNA. In a formal sense, the relationship of one zinc finger to J05 50 the next is in the form of a translational rotation or a screw J05 51 operation that tracks the inside of the major groove of the double J05 52 helix.

J05 53 Even though the zinc finger uses the <*_>alpha<*/> helix for J05 54 recognition, it has several unique features that differentiate it J05 55 from the helix-turn-helix interactions. First, of all, the zinc J05 56 finger complex is formed out of modular units that can be repeated J05 57 a large number of times. This entails the possibility of J05 58 recognizing very long stretches of DNA by simply having a larger J05 59 number of zinc fingers. As mentioned, some proteins have very large J05 60 numbers of zinc fingers and may actually use this. The second J05 61 characteristic is that the contacts seem to be largely with one J05 62 strand only of DNA, in this particular case with the purine-rich or J05 63 guanine-rich strand. The recognition depends largely on interaction J05 64 with the bases, and there are fewer hydrogen bonds with the DNA J05 65 backbone than are seen in the other structures.

J05 66 Even though studies of other protein-DNA complexes appear not J05 67 to have a recognition code, the zinc finger complex appears to have J05 68 a recognition code which is largely based on the arginine-guanine J05 69 contacts, at least for the Zif complex. It remains to be seen J05 70 whether these types of contacts will be found more specifically J05 71 when the structures of more zinc fingers have been done. Solution J05 72 of this structure will make it possible to synthesize new zinc J05 73 finger binding domains with different nucleotide binding J05 74 specificities. Thus, it will be possible to explore the full gamut J05 75 of interactions found in this widely used recognition motif in J05 76 eukaryotic systems.

J05 77 THE GLUCOCORTICOID RECEPTOR ALSO CONTAINS ZINC IONS

J05 78 The glucocorticoid receptor has the property of binding a J05 79 hormone, such as estrogen or another steroid, and is then J05 80 translocated from the cytoplasm to the nucleus, where it binds to J05 81 specific DNA sequences, called glucocorticoid response J05 82 elements (GREs). The binding affects transcription of the J05 83 genes. A large number of these exist; they include receptors for J05 84 steroid hormones, retinoids, vitamin D, thyroid hormones, and J05 85 others. Members of this superfamily have a common amino acid J05 86 sequence organization with discrete domains that are used for J05 87 binding DNA as well as zinc. All of these nuclear receptors are J05 88 characterized by a pattern containing eight cystenes and, in the J05 89 glucocorticoid receptor, these cystenes coordinate two zinc ions in J05 90 a tetrahedral manner. The structure of the glucocorticoid receptor J05 91 bound to DNA has been determined recently by Luisi et al. J05 92 (1991). Unlike the typical zinc fingers, the glucocorticoid J05 93 receptor forms a distinct globular binding domain and does not J05 94 occur in a long series of modular units, as is often found in the J05 95 typical zinc finger DNA binding.

J05 96 The structure of the glucocorticoid receptor bound to DNA J05 97 reveals that the receptor dimerizes onto a DNA molecule that J05 98 contains two repeats of the glucocorticoid response element J05 99 sequence, each with the major groove facing in the same direction. J05 100 Each of the proteins forms a compact globular structure in which J05 101 the two zinc ions serve to nucleate the formation of a conformation J05 102 in which an <*_>alpha<*/> helix is positioned in the major groove J05 103 of B-DNA and thereby has sequence-specific binding. The J05 104 glucocorticoid receptor conformation may be looked upon as a J05 105 conformation similar in some respects to the helix-turn-helix J05 106 conformation, except that zinc ions are used in maintaining the J05 107 stable fold of the protein rather than the helix interactions found J05 108 in the helix-turn-helix system. A number of interactions are found J05 109 between the glucocorticoid receptor and the DNA. However, three of J05 110 them are interactions with bases that are important for determining J05 111 sequence specificity. One of the most important of these is an J05 112 arginine 466 that binds to guanine 4 using the system of J05 113 arginine-guanine interactions, which has been described above using J05 114 two hydrogen bonds. Another hydrophobic interaction involves valine J05 115 462 interacting with the methyl group of thymine 5 while a lysine J05 116 461 forms a single hydrogen bond to N7 of guanine 7 as well as to a J05 117 water molecule, which in turn binds to O6 of guanine and O4 of J05 118 thymine in an adjacent base pair. If arginine 466 is replaced by J05 119 lysine or glycine, the protein no longer functions in J05 120 vivo. Arginine 466 and lysine 461 are absolutely conserved in J05 121 the superfamily of nuclear receptors; their targeted bases, guanine J05 122 4 and guanine 7, also occur consistently in all the known sequences J05 123 of the hormone response elements (Luisi et al., 1991).

J05 124 The major difference between the zinc-containing glucocorticoid J05 125 response element and the traditional zinc finger is the fact that J05 126 the latter conformation is stabilized individually by an extensive J05 127 hydrophobic core as well as by the zinc ion. Furthermore, it J05 128 assumes this conformation independent of the presence or absence of J05 129 DNA. In contrast, experiments with the glucocorticoid receptor show J05 130 that it only condenses as a dimer in the presence of the DNA. The J05 131 dimerization is stabilized both by the DNA as well as by contacts J05 132 between the protein.

J05 133 The arginine-guanine interaction, which played so predominant a J05 134 part in five of the six interactions seen in the three modules of J05 135 the traditional zinc finger structure, also plays a role in J05 136 interactions with the glucocorticoid response element. However, in J05 137 this case, only one of the three interactions that are J05 138 sequence-determining involves the arginine-guanine interaction.

J05 139 ECO RI ENDONUCLEASE BINDING TO DNA

J05 140 Restriction endonucleases are very important tools in molecular J05 141 biology since they cleave DNA molecules at specific sequences. One J05 142 of the widely used enzymes was obtained from E. coli and J05 143 is called Eco RI endonuclease. It cleaves DNA at a specific J05 144 double-stranded sequence (d(GAATTC)). Eco RI contains 276 amino J05 145 acids, and it has been crystallized with a fragment of DNA J05 146 containing 13 base pairs. The solved structure revealed a complex J05 147 interaction between a globular protein and a DNA double helix J05 148 (McClarin et al., 1986; Kim et al., 1986). The J05 149 DNA recognition motif consists of a parallel bundle of four J05 150 <*_>alpha<*/> helices penetrating the major groove of the DNA. J05 151 There, amino acids at the end of the <*_>alpha<*/> helix interact J05 152 with the bases in the major groove. Although <*_>alpha<*/> helices J05 153 are employed, this motif differs from the interaction seen both in J05 154 the helix-turn-helix proteins and the zinc fingers. In this case, a J05 155 cluster of very long <*_>alpha<*/>-helical segments interact with J05 156 the DNA at their ends. In addition, a segment of extended J05 157 polypeptide chain runs along the major groove of the DNA roughly J05 158 parallel to the DNA backbone. This is anchored at one end by one of J05 159 the recognition helices, and it has several contacts with bases. J05 160 Among the interactions that are described is one involving arginine J05 161 200 binding to guanine in a manner similar to that described in J05 162 Fig. 3.

J05 163 We do not know whether this structure is likely to be a general J05 164 structure for the recognition of DNA by restriction endonucleases. J05 165 However, one of the interesting projects arising from solution of J05 166 this protein-DNA complex is the possibility of modifying side J05 167 chains to alter recognition modes so that one might be able to make J05 168 restriction enzymes with altered cleavage specificities using the J05 169 Eco RI framework for carrying this out. Further work will be J05 170 necessary before we know whether this is a general recognition J05 171 motif for other enzymes as well. However, it is important to J05 172 emphasize that the mode of interaction is quite distinct from that J05 173 seen in any other protein-nucleic acid cocrystal.

J05 174 <*_>beta<*/> SHEET DNA BINDING PROTEINS

J05 175 The methionine repressor controls its own gene as well as J05 176 structural genes for enzymes involved in the synthesis of J05 177 methionine. It is a protein with 104 amino acids and forms stable J05 178 dimers in solution. The structure had been determined by Phillips J05 179 and colleagues, and it consists of two highly intertwined monomers J05 180 that form a two-stranded antiparallel <*_>beta<*/> sheet with one J05 181 strand coming from each monomer (Rafferty et al., 1989). J05 182 This <*_>beta<*/> sheet forms a protrusion on the surface of the J05 183 molecule. A similar structure has been deduced for the Arc J05 184 repressor based on NMR studies (Kaptain et al., 1985). J05 185 Phillips has also solved the structure of the Met repressor bound J05 186 to a synthetic DNA fragment containing 18 base pairs (S. Phillips, J05 187 personal communication). The structure of the Met dimer is not J05 188 changed greatly by binding to the DNA, which is largely in the B J05 189 conformation. The two-stranded <*_>beta<*/> sheet of the repressor J05 190 is found in the major groove of the DNA with side chains from the J05 191 <*_>beta<*/> strands interacting with base pairs within the J05 192 operator sequences. These interactions are the base J05 193 sequence-specific interactions. The DNA itself is somewhat kinked J05 194 in the center of the operator sequence. That has the effect of J05 195 narrowing the major groove slightly so that it can form closer J05 196 bonding to the side chains of the <*_>beta<*/> sheet.

J05 197 The two-stranded <*_>beta<*/> sheet is thus another DNA binding J05 198 motif which, unlike the others mentioned above, does not use an J05 199 <*_>alpha<*/> helix for recognition but rather an extended J05 200 polypeptide chain.

J05 201 The listing of protein structural motifs that are involved in J05 202 recognizing DNA sequences (Table I) is necessarily incomplete. J05 203 J06 1 <#FROWN:J06\>II. EXPERIMENT

J06 2 The molecular beam time-of-flight mass spectrometer (TOFMS) J06 3 used in this study is identical to that in Paper I, to which the J06 4 reader is referred. A given J06 5 C6H6-(CH3OH)n neutral J06 6 cluster size is selected for photoionization using resonance J06 7 enhancement through the S0-S1 transitions of J06 8 the C6H6 chromophore in the cluster. Resonant J06 9 two-photon ionization scans utilize the unfocused, doubled output J06 10 of an excimer-pumped dye laser. Typical peak UV laser powers are J06 11 3x105 W/cm2. Laser power studies indicate that J06 12 the observed product ions result from two-photon processes. J06 13 Relative product yields are determined from scans over the resonant J06 14 features of the reactant complex while simultaneously monitoring J06 15 ion signals from all relevant product mass channels using a 100 MHz J06 16 digital oscilloscope.

J06 17 III. RESULTS

J06 18 In Paper I, we reported on the spectroscopy of neutral J06 19 C6H6-(CH3OH)n clusters. In J06 20 that case, attention was focused on R2PI spectra taken monitoring J06 21 unreactive ion mass channels J06 22 [C6H6-(CH3OH)n]+. J06 23 Figures 1-3 present a series of scans over the same J06 24 610 region including scans monitoring mass J06 25 channels which arise from intracluster ion-molecule chemistry. As J06 26 is readily apparent from these spectra, resonant two-photon ionized J06 27 C6H6-(CH3OH)n clusters with J06 28 n<*_>unch<*/>3 [Eq. (1)] react by dissociative electron J06 29 transfer (DET) to form (CH3HO)n+ [Eq. J06 30 (3)], while those with n<*_>unch<*/>4 also undergo J06 31 dissociative proton transfer (DPT) to form J06 32 H+(CH3OH)n ions [Eq. (4)]. This J06 33 intracluster ion chemistry is completely absent from the resonantly J06 34 photoionized C6H6-(H2O)n J06 35 clusters and from J06 36 C6H6-(CH3OH)n with J06 37 n<*_>unch<*/>2 [Fig. 1(b)]. In all cases, the ion chemistry J06 38 occurs in competition with fragmentation of the cluster [Eq. (2)] J06 39 via loss of one (or sometimes two) methanol molecules.

J06 40 formulae

J06 41 The dissociative electron transfer channel to form J06 42 [(CH3OH)n+] is quite unexpected since J06 43 only protonated methanol cluster ions are observed in either J06 44 electron bombardment or photoionized pure methanol clusters. While J06 45 the resolution of our reflectron TOFMS is easily capable of J06 46 distinguishing protonated from unprotonated clusters J06 47 (formula), we have as an additional check carried out J06 48 resonant two-photon ionization (R2PI) scans using CH3OD J06 49 which confirm that the major product is J06 50 [(CH3OD)n+] and not J06 51 [(CH3OD)nD]+ or J06 52 [(CH3OD)nH]+.

J06 53 Figures 4 and 5 present the relative product yields for the J06 54 observed fragmentation, DET, and DPT channels following resonant J06 55 ionization through the 610 and formula J06 56 transitions of the 1:2, 1:3, 1:4, and 1:5 clusters. These yields J06 57 are obtained by integrating the peak intensities of the J06 58 610 or formula ion signals in the J06 59 relevant product channels in order to avoid nonresonant J06 60 contributions from the signal. The measurements at formula J06 61 increase the maximum internal energy in the cluster ion (determined J06 62 by the photon energy) by 5.2 kcal/mol from those using the J06 63 610 transition as the intermediate state.

J06 64 figure&caption

J06 65 IV. DISCUSSION

J06 66 One factor which plays a major role in determining the presence J06 67 and efficiency of a given product channel is the energetic J06 68 threshold for the channel relative to the internal energy of the J06 69 photoionized reactant cluster. In photoionization, a distribution J06 70 of ion internal energies is produced which reflect both direct J06 71 ionization and autoionization processes in the cluster. In the free J06 72 C6H6 molecule, R2PI through the J06 73 610 transition reaches 8 kcal/mol above the J06 74 ionization threshold. In that case, the largely J06 75 <*_>DELTA<*/><*_>nu<*/>=0 Franck-Condon factors between the J06 76 S1 state of the neutral and the ground states of the J06 77 ion result in the electron taking away most of the excess energy as J06 78 kinetic energy.

J06 79 figure&caption

J06 80 In clusters, on the other hand, the nature and strength of the J06 81 intermolecular forces change significantly upon ionization, often J06 82 leading to very different lowest-energy structures for the neutral J06 83 and ionic clusters. This is particularly true for the present J06 84 clusters containing polar methanol molecules as solvent, where the J06 85 good Franck-Condon factors between the S1 state and J06 86 the ion are to regions of the ionic potential energy surface far J06 87 above the adiabatic ionization threshold for the cluster. As we saw J06 88 in Paper I, the efficient fragmentation of the photoionized J06 89 clusters is one result of the high ion internal energies produced. J06 90 Here, since the same photoionization process also initiates the J06 91 intracluster ion chemistry, we expect to form a distribution of J06 92 reactant cluster ion internal energies which favors ion internal J06 93 energies near the maximum allowed by the photon energy.

J06 94 Figures 6-8 present energy level diagrams reflecting our best J06 95 estimates of the energies of fragmentation, DET, and DPT product J06 96 thresholds relative to the maximum ion energies produced by J06 97 photoionization. Experimentally observed product channels are J06 98 highlighted by placing them in boxes. In the figures, the zero of J06 99 the energy level scales is taken to be the energy of the J06 100 formula asymptote. Table I collects the heats of formation J06 101 of the relevant species. The thermochemistry of the protonated J06 102 methanol clusters is known with good accuracy by virtue of several J06 103 studies of these clusters. The neutral cluster binding energies are J06 104 not known from experiment. We estimate them using the calculations J06 105 of Paper I after approximate correction for zero point energy J06 106 effects.

J06 107 figure&caption

J06 108 An upper bound for the energy of the DET channel [Eq. (3)] on J06 109 our relative scale is determined by the threshold for reaction J06 110 (5)

J06 111 formula

J06 112 whose thermochemistry is determined by well-known heats of J06 113 formation for all species. If the CH3O radical is formed J06 114 instead of CH2OH, the upper bound for DET would be 10 J06 115 kcal/mol higher. We assume that the observation of J06 116 [(CH3OH)n]+ ions in the present J06 117 experiment arises because there is an energetic barrier to breakup J06 118 of this ion to formula. The shaded region in the diagrams J06 119 place some reasonable bounds on the height of such a barrier (0-10 J06 120 kcal/mol) and thus loosely brackets the threshold for DET.

J06 121 figure&caption

J06 122 A. The 1:2 cluster

J06 123 As Fig. 4 indicates, the only observed product channel in J06 124 one-color R2PI of the 1:2 cluster is fragmentation via loss of a J06 125 single CH3OH to form formula. Fragmentation is J06 126 86% <*_>unch<*/> 5% efficient using the 61 level of the J06 127 S1state as the intermediate state and 80% J06 128 <*_>unch<*/> 10% efficient at 00, despite our use of J06 129 laser powers for which three-photon contributions to the ion J06 130 signals are negligible. The small change in the percentage of J06 131 fragmentation accompanying the 3 kcal/mol increase in the J06 132 two-photon energy confirms the notion that most of the cluster ions J06 133 have an internal energy near the two-photon energy in Fig. 4, well J06 134 in excess of the formula dissociation asymptote in both the J06 135 O0 and 61 scans. Loss of CH3OH from J06 136 the ionic cluster occurs on a time scale fast compared to movement J06 137 of the cluster ion in the ion source (t<1<*_>mu<*/>s), since J06 138 no asymmetry is observed in the arrival time profile of the parent J06 139 or fragment. Fragmentation occurs almost exclusively by loss of a J06 140 methanol monomer rather than a methanol dimer, despite the fact J06 141 that the latter channel should be accessible to many of the J06 142 clusters. This favoring of evaporative loss of monomer units has J06 143 been a trademark of cluster fragmentation in many types of J06 144 clusters.

J06 145 figures&captions

J06 146 The lack of ion-molecule chemistry in the formula J06 147 cluster appears to be a direct consequence of energetic J06 148 constraints. Dissociative electron transfer, which is the J06 149 dominating reaction channel in higher clusters, is predicted to be J06 150 endothermic in R2PI through 61 (Fig. 6). This is the case J06 151 because the ionization potential of CH3OH is some 37 J06 152 kcal/mol above that for C6H6 (Table I), so that J06 153 even with the stabilization of CH3OH+ provided J06 154 by a second methanol molecule, formation of J06 155 [(CH3OH)2]+ is still endothermic. J06 156 Similarly, the high proton affinity of C6H5 J06 157 (which clearly precludes proton transfer of J06 158 C6H6+ to CH3OH) is nearly J06 159 thermoneutral for transfer to (CH3OH)2 (see J06 160 Table I and Fig. 6). Hence it is not surprising that no reaction J06 161 products beside fragmentations are observed in the 1:2 clusters.

J06 162 B. The 1:3 cluster

J06 163 The formula cluster ion is the smallest sized J06 164 1:n cluster to undergo intrascluster ion-molecule J06 165 chemistry. The appearance of J06 166 [(CH3OH)3)]+ (i.e., mass 96) is J06 167 consistent with the energy level diagram of Fig. 7 which shows that J06 168 the asymptote for DET is now well below the maximum ion internal J06 169 energy produced in R2PI. Nevertheless, DET is still only a minor J06 170 channel (6%) which competes only poorly with fragmentation via loss J06 171 of CH3OH (94%).

J06 172 1. The formation of J06 173 [(CH3OH)3]+

J06 174 The formation of a reaction product at mass 96 is notable for J06 175 being exclusively the unprotonated methanol cluster ion. As J06 176 mentioned earlier, when pure methanol clusters are either electron J06 177 bombardment or photoionized, only protonated cluster ions are J06 178 observed due to exothermic ion-molecule reactions of the type

J06 179 formulae

J06 180 The bimolecular analogs of these reactions

J06 181 formula

J06 182 are exothermic by 24 and 14 kcal/mol, respectively. Both these J06 183 facts suggest that the initially unprotonated cluster ions would be J06 184 inherently unstable with respect to loss of CH2OH or J06 185 CH3O. However, in recent experiments by Vaidyanathan J06 186 et al., electron bombardment ionization of J06 187 Ar/CH3OH heteroclusters has successfully produced J06 188 significant quantities of unprotonated formula ions via J06 189 intracluster Penning ionization involving high-lying states of the J06 190 argon neutrals. The present study offers a second example of the J06 191 formation of unprotonated methanol cluster ions, this time mediated J06 192 by C6H6. The attachment of the methanol J06 193 clusters to a C6H6 chromophore offers an J06 194 extremely gentle means of producing unprotonated J06 195 Mn+ clusters by photoionizing the cluster with J06 196 only 9.60 eV energy via the C6H6 chromophore. J06 197 In the case of the 1:3 cluster, the unprotonated J06 198 M3+ products can be formed completely free from J06 199 interference from protonated clusters. This mechanism thus provides J06 200 a route producing the novel Mn+ cluster ions J06 201 for subsequent spectroscopic and mass spectrometric study.

J06 202 The formation of the DET product M3+ (M = J06 203 methanol) provides supporting evidence for the neutral J06 204 C6H6-M3 cluster possessing a J06 205 structure composed of hydrogen-bonded methanols all attached to the J06 206 same side of the benzene ring (Paper I). It seems unlikely that DET J06 207 to form M3+ would occur from an initial J06 208 geometry in which methanol molecules are on both sides of the J06 209 benzene ring rather than as an aggregate on the same side.

J06 210 The structure of the mass 96 ion is not determined in the J06 211 present work. However, it seems likely that the J06 212 M3+ cluster exists either as formula or J06 213 as formula in which proton transfer within the methanol J06 214 cluster has occurred with cluster energy insufficient to allow its J06 215 breakup to form the protonated M2H+ ion. Loss J06 216 of the C6H6 molecule form the cluster provides J06 217 the means by which the M3+ ion is stabilized J06 218 below its M2H + CH2OH/CH3O J06 219 dissociation asymptote.

J06 220 The only energetically allowed channel which competes J06 221 successfully with DET is fragmentation to form formula, J06 222 which dominates the product distribution (94%). No parent 1:3 J06 223 cluster ions are observed in our experiment. Thus the observed J06 224 processes are the following:

J06 225 formulae

J06 226 One could imagine two limiting cases for the competition J06 227 between DET and fragmentation. In one limit, the energy dependence J06 228 of the rate constants is either small enough or similar enough in J06 229 the two channels that the observed product distribution directly J06 230 reflects the relative magnitudes of the rate constants for J06 231 fragmentation and DET; i.e., the product yields are kinetically J06 232 controlled. In this case, the rate of fragmentation would provide J06 233 and 'internal clock' for the DET reaction if it could be determined J06 234 by other means.

J06 235 In the second limit, the observed product distribution will J06 236 reflect energetic constraints rather than kinetics. This would J06 237 occur if kfrag(E) is much greater then J06 238 kDET(E) for energies where fragmentation can occur. J06 239 Then all cluster ions with internal energies above the J06 240 fragmentation threshold would undergo fragmentation, while the J06 241 remaining, lower energy cluster ions would react via DET. More J06 242 complete control over the ion internal energies produced or J06 243 detected (e.g., via photoelectron-photoion coincidence J06 244 measurements) will be required before the energy dependences of the J06 245 reaction rates can be determined unambiguously.

J06 246 2. Reactions which are not observed

J06 247 The formation of unprotonated M3+ ions is J06 248 unusual in a second respect - it occurs to the exclusion of several J06 249 other energetically open channels, most notably those involving J06 250 dissociative proton transfer [Eq. (4)]. Careful searches for the J06 251 DPT products formula and formula, or the J06 252 CH2OH loss channel to produce formula place upper J06 253 bounds on these channels at less than 1% of the 1:2 fragment J06 254 channel (Fig. 4), nearly ten times less than the observed yield of J06 255 DET products.

J06 256 Again, energetics could play a significant role in suppressing J06 257 the proton transfer channels if the rate constants for the product J06 258 channels were sensitively dependent on their exothermicity. J06 259 J07 1 <#FROWN:J07\>Let's briefly examine what this might do. If the J07 2 acceleration is a, the change in the potential energy is J07 3 formula and the total kinetic energy gained is J07 4 formula. Here the rate of growth is<*_>tau<*/>-1 J07 5 and <*_>delta<*/>z is the displacement. Then the growth time J07 6 scale is given by

J07 7 formula

J07 8 where k<*_>approximate-sign<*/>z-1 is the J07 9 wave number for the disturbance (assuming that the wavelength is of J07 10 order z). The instability is guaranteed if the density J07 11 gradient is in the opposite direction to the local acceleration. J07 12 Now by noting that any acceleration is locally equivalent to J07 13 gravitation (this is after all the basis of general relativity), we J07 14 can replace a by <*_>DELTA<*/>P, the pressure jump across J07 15 the shock front at the wind-wind interface. We could also use J07 16 geff, the effective gravity including radiation J07 17 pressure. Let's concentrate on the first choice. Across a shock, J07 18 which moves with a speed v<*_>SIGMA<*/> into the J07 19 surrounding gas, the pressure is given by

J07 20 formula

J07 21 For a perfect gas this is larger on the postshocked side and J07 22 the pressure jump is formula. The interface is unstable to J07 23 the formation of knots and blobs. As we have already discussed, as J07 24 these rise, they tend to show induced vorticity and develop the J07 25 characteristically mushroom-shaped structures so J07 26 <}_><-|>familar<+|>familiar<}/> from both violent explosions (like J07 27 nuclear blasts) and simulations of supernova envelopes. This same J07 28 condition will certainly prevail during the formation of planetary J07 29 nebulae, and the wind-wind collision should generate stringy and J07 30 bloblike structures because of this instability. If optically J07 31 thick, they will shadow the material in the slow wind, and the J07 32 result is that low ionization regions should be mixed into the H II J07 33 region. One expectation is that in any of the blastlike expansions, J07 34 be they Wolf-Rayet winds, planetary nebulae, or supernova remnants, J07 35 it should be possible to see the effects of the wind collision J07 36 preserved in these structures.

J07 37 8.7 Accretion Disks in Astrophysics

J07 38 8.7.1 Some Observational Motivations

J07 39 We have invoked binary systems frequently in this chapter, and J07 40 it is therefore appropriate to end it with a discussion of a most J07 41 interesting confluence of rotation and viscosity, namely accretion J07 42 disks around massive objects. But first, before launching into an J07 43 extended expos<*_>e-acute<*/> of the properties of the disks J07 44 themselves, let's examine the conditions under which rotating J07 45 accretion flows may arise.

J07 46 For many binaries, especially ones of long period, this is J07 47 precisely what is observed. The more evolved star really is the J07 48 more massive. In the case of Algol (<*_>beta<*/> Persei), an J07 49 extremely well-studied star and the first discovered eclipsing J07 50 binary, just the opposite is observed. Algol consists of a G giant J07 51 and a B main sequence star. The mass ratio is J07 52 <*_>approximate-sign<*/>3, but in favor of the main sequence B J07 53 star. The orbital period is short, less than 3 days. the light J07 54 curve data and modeling the equilibrium shapes of the stars show J07 55 that the red subgiant completely fills its Roche surface. This is J07 56 the limiting surface for tidal interaction, given approximately J07 57 by

J07 58 formula

J07 59 where a is the semimajor axis given by Kepler's law J07 60 formula for circular orbits, q is the mass ratio, and J07 61 M is the total mass of the system. The reason for the peculiar J07 62 mass of the red giant is that it has been significantly altered by J07 63 mass loss from the binary system and by mass transfer onto the main J07 64 sequence star. The fact that this is still going on, in both this J07 65 and related semidetached systems (the term comes from the fact J07 66 that only one of the stars is in contact with the critical J07 67 surface), means that accretion flows onto the companion have played J07 68 a role in the orbital dynamics and that this has fed back into the J07 69 stellar evolution through the alteration of the mass and boundary J07 70 conditions on the stars. The observation, for a number of these J07 71 stars, of emission lines which are formed in a Keplerian disk J07 72 surrounding one of the components adds fuel to the argument, J07 73 although in Algol it does not appear that an extensive accretion J07 74 disk is observed.

J07 75 8.7.2 Flow through the Inner Lagrangian Point

J07 76 First, a binary system, that is, a close system, is one that is J07 77 not spherically symmetric. The presence of the companion star, as J07 78 well as the rotation of the mass-losing star due to spin-orbit J07 79 coupling, produces immediate departures from sphericity. To see J07 80 what happens to the mass transfer at the inner Lagrangian point, J07 81 also called L1, we need to consider the flow of J07 82 material in a potential that switches sign at some point in the J07 83 flow (see Fig. 8.4).

J07 84 Let's look back at the spherical case for a moment. The mass J07 85 loss is driven by the combined effect of pressure gradient and J07 86 retardation due to gravity. At some point, where the outward J07 87 driving becomes strong enough relative to gravity, the material J07 88 coasts at the sound speed and then accelerates as the gravitational J07 89 acceleration continues to fall off. In other words, the reason the J07 90 effective gravity vanishes is that at some point, g is J07 91 balanced by <*_>unch<*/>p. But what about other possible J07 92 cases? We've already seen that if g is balanced by J07 93 <*_>unch<*/>prad, or at some point g(1 - J07 94 <*_>GAMMA<*/>) vanishes, then the material becomes supersonic and a J07 95 strong wind. Both of these depend on the presence of a pressure J07 96 gradient to do the job of producing a sign change in the J07 97 acceleration. The alternative is to say that the gravity itself J07 98 reverses sign, something impossible for a single star but normal J07 99 for a close binary. Put differently, imagine that the potential is J07 100 taken to be

J07 101 formula

J07 102 in the vicinity of the L1 point. The problem can J07 103 be rendered one dimensional if we assume that we look at the flow J07 104 only in the vicinity of the Lagrangian point and that the system is J07 105 not so rapidly rotating that the Coriolis deviation of the flow is J07 106 large compared with rectilinear flow. This means that J07 107 <*_>OMEGA<*/><u', where the prime denotes the spatial J07 108 derivative of the velocity.

J07 109 figure&caption

J07 110 The fact that the equation of motion can be written as

J07 111 formula

J07 112 means that we can apply the same condition that we had in the J07 113 spherical Parker wind solution. Take a look at the flow through a J07 114 small region around the L1 point. The equipotentials J07 115 on either side of L1, along the line of centers, J07 116 give a local critical point to the flow. This is because there is a J07 117 local maximum in the gravitational field. Perpendicular to the line J07 118 of centers the gradient has a local minimum. (See Fig. 8.5.) The J07 119 L1 point is a saddle point in the gravitational J07 120 potential; the gravitational acceleration changes sing on crossing J07 121 this point. For the case of one-dimensional flow, this has the same J07 122 effect as the changing gravitational acceleration relative to the J07 123 pressure gradient. The critical condition for stream formation is J07 124 similar to the Parker solution; that is, u = as J07 125 at L1. Thereafter, as the gravitational field J07 126 increases toward the secondary, the flow accelerates. As material J07 127 is forced through this point, it has the same effect as the passage J07 128 through the geff = 0 point in a spherical wind. The J07 129 pressure gradient does not vanish, so the material is accelerated J07 130 and the sound speed is reached, after which the flow is ballistic J07 131 toward the secondary. Stream formation is important because it J07 132 transfers material with high specific angular momentum toward the J07 133 accreting star. Once in the vicinity of the companion, the matter J07 134 forms an accretion disk, the details of which we shall now J07 135 discuss.

J07 136 figure&caption

J07 137 8.7.3 Some Consequences of Mass Transfer

J07 138 The observation of tidal distortion leads immediately to J07 139 important hydrodynamic consequences. The existence of the Roche J07 140 surface in general, and the limiting radius in particular, is the J07 141 result of the three-body problem. The inner Lagrangian point, J07 142 L1, is the point along the line of centers at which J07 143 the effective gravitational acceleration vanishes. However, since J07 144 the pressure does not vanish even though the gravitational J07 145 acceleration does, matter will be forced to exit through this J07 146 region and begin to flow to the other star. In effect, we have set J07 147 up the de Laval nozzle problem from Chapter 1. The variation in the J07 148 effective gravity acts like the nozzle (although without many J07 149 material walls) to accelerate the flow through the sonic point and J07 150 eventually to hypersonic speeds. The matter carries some net J07 151 angular momentum because the L1 point is generally J07 152 not at the center of mass, and therefore the deviation of the flow J07 153 and its acceleration toward the companion produce an accretion J07 154 disk.

J07 155 The stream must eventually rid itself of this excess angular J07 156 momentum before it can accrete onto the companion even for direct J07 157 impact. Several mechanisms are available, probably all of which J07 158 operate somewhere in the universe. One is turbulent viscosity. That J07 159 is the one we shall mainly deal with here. Another is magnetic J07 160 breaking. If the material forms a disk that becomes J07 161 Kelvin-Helmholtz unstable at the boundary of a stellar J07 162 magnetosphere, blobs may be formed that accrete onto the companion. J07 163 The process is certainly not well understood but can be simulated J07 164 for neutron star accretion and is well established as a scenario. J07 165 For direct impact, the stream may submerge and pump angular J07 166 momentum into a deeply generated boundary layer.

J07 167 The final mechanism is spiral shocks. Since the matter falls J07 168 into the disk with excess angular momentum and drives spiral waves J07 169 in the disk, these may form stable circulating structures that J07 170 serve to deviate the flow onto the companion and dissipate energy J07 171 and momentum. (See Fig. 8.6.) Presently, however, the details of J07 172 the accretion process are the most schematic parts of accretion J07 173 disk theory. This is a pity, because these details contain J07 174 virtually all of the essential physics.

J07 175 If the mass-accreting star is a compact object, like a white J07 176 dwarf, neutron star, or black hole, the disks reach temperatures J07 177 considerably higher than for main sequence stars. The simple reason J07 178 is that the gravitational well around the accreting object is very J07 179 deep and the energy source for the dynamics is consequently J07 180 greater. Any dissipative processes fed by the global circulation J07 181 will therefore find a very large reservoir of energy which can be J07 182 tapped, and the resulting emission of radiation can take place in J07 183 the ultraviolet or even the x-ray. It was the latter wavelength J07 184 region, observed with satellites like Uhuru and Einstein, that J07 185 first signaled the presence of accretion disks around neutron stars J07 186 like Hercules X-1 = HZ Her and black holes like Cygnus X-1 = HD J07 187 226868. The added discovery that the optical counterparts of these J07 188 and other galactic x-ray sources are spectroscopic binaries and the J07 189 observation of optical emission lines which tracked the compact J07 190 star clinch the argument for accretion powering the radiation.

J07 191 figure&caption

J07 192 The observation of strong emission lines arising in a very J07 193 compact region in the cores of active galaxies like quasars and J07 194 Seyferts also indicates that accretion can occur on a scale of J07 195 extremely massive but otherwise 'single' collapsed objects. These J07 196 galaxies have emission line widths indicative of velocities of J07 197 order 1000 to 104 km s-1 coming from a region J07 198 less than 1 pc across. The temperature of the regions, indicated by J07 199 the presence of extremely high radiation rates for x-rays, argues J07 200 that accretion flow around a black hole is the likely source of the J07 201 observed luminosity.

J07 202 In light of these observations, and because the range of J07 203 physics required for an understanding of such flows touches on J07 204 virtually all aspects of astrophysical hydrodynamics, we will J07 205 discuss accretion disk theory at some length.

J07 206 8.7.4 Heating the Disks: Dissipation and Viscous J07 207 Torques

J07 208 We first take up the question of the effect of the generation J07 209 of energy by the shearing within the disk due to its differential J07 210 rotation. Recall that the viscous energy dissipation rate is given J07 211 by

J07 212 formula

J07 213 where Tij is the stress tensor and J07 214 <*_>sigma<*/>ij is the shear. The shear for an J07 215 axi-symmetric system is given by formula. The J07 216 negative sign in the second term comes from formula. In J07 217 light of the previous discussion, the disk is Keplerian, and since J07 218 the angular frequency is given by formula, the shear is J07 219 formula. Since the shear varies with radius for such a J07 220 disk, so does the rate of energy generation. J07 221 J08 1 <#FROWN:J08\>Global Warming and Potential Changes in J08 2 Host-Parasite and Disease-Vector Relationships

J08 3 ANDREW DOBSON AND ROBIN CARPER

J08 4 I. INTRODUCTION

J08 5 Parasitology has always been a discipline in which purely J08 6 academic studies of the evolution of parasites and their life J08 7 cycles have progressed as a necessary complement to the study of J08 8 the pathology and control of the major tropical diseases of humans J08 9 and their livestock. Indeed, the most striking feature of J08 10 parasitology is the diversity of parasites in the warm tropical J08 11 regions of the world and the frightening levels of debilitation and J08 12 misery they cause. Determining how long-term climatic changes will J08 13 affect the distributions of different parasites and pathogens at J08 14 first seems a daunting task that almost defies quantification. J08 15 Nevertheless, as parasitologists have always been concerned with J08 16 the influence of climatological effects on different parasite J08 17 species, it is possible to begin to speculate on the ways that J08 18 global warming might affect the distributions of some specific J08 19 tropical diseases. Similarly, the study of parasite population J08 20 dynamics has developed within a solid theoretical framework J08 21 (Anderson and May 1979, May and Anderson 1979). This permits the J08 22 development of quantitative speculation in more general studies J08 23 concerned with how parasite-host interactions may respond to J08 24 perturbation.

J08 25 This chapter addresses both general questions about the J08 26 response of parasite-host systems to long-term climatic changes and J08 27 the specific response of one particular pathogen, Trypanosoma, J08 28 to the changes in climate predicted for the next hundred years.

J08 29 A. Macroparasites and Microparasites

J08 30 Current estimates suggest that parasitism of one form or J08 31 another may be the most common life-history strategy in at least J08 32 three of the five major phylogenetic kingdoms (May 1988, Toft J08 33 1986). The enormous array of pathogens that infect humans and other J08 34 animals may be conveniently divided on epidemiological grounds into J08 35 microparasites and macroparasites (Anderson and May 1979, May and J08 36 Anderson 1979). The former include the viruses, bacteria, and fungi J08 37 and are characterized by their ability to reproduce directly within J08 38 individual hosts, their small size and relatively short duration of J08 39 infection, and the production of an immune response in infected and J08 40 recovered individuals. Mathematical models examining the dynamics J08 41 of microparasites divide the host population into susceptible, J08 42 infected, and recovered classes. In contrast, the macroparasites J08 43 (the parasitic helminths and arthropods) do not multiply directly J08 44 within an infected individual but instead produce infective stages J08 45 that usually pass out of the host before transmission to another J08 46 host. Macroparasites tend to produce a limited immune response in J08 47 infected hosts; they are relatively long-lived and usually visible J08 48 to the naked eye. Mathematical models of the population dynamics of J08 49 macroparasites have to consider the statistical distribution of J08 50 parasites within the host population.

J08 51 B. Direct and Indirect Life Cycles

J08 52 A second division of parasite life histories distinguishes J08 53 between those species with monoxenic life cycles and those with J08 54 heteroxenic life cycles. The former produce infective stages that J08 55 can directly infect another susceptible definitive host individual. J08 56 Heteroxenic species utilize a number of intermediate hosts or J08 57 vectors in their transmission between definitive hosts. The J08 58 evolution of complex heteroxenic life cycles permits parasite J08 59 species to colonize hosts from a wide range of ephemeral and J08 60 permanent environments, while also permitting them to exploit host J08 61 populations at lower population densities than would be possible J08 62 with simple direct transmission (Anderson 1988, Dobson 1988, J08 63 Mackiewicz 1988, Shoop 1988). However, heteroxenic life cycles J08 64 essentially confine the parasite to areas where the distribution of J08 65 all the hosts in the life cycle overlap. Shifts in the distribution J08 66 of these host species due to climatic changes, will therefore be J08 67 important in determining the areas where parasites may persist and J08 68 areas where parasites may be able to colonize new hosts.

J08 69 C. Aquatic and Terrestrial Hosts

J08 70 Climatic changes are likely to have different effects on J08 71 aquatic and terrestrial environments (chapter 24). The heteroxenic J08 72 life cycles of some parasite species often allow them to utilize J08 73 hosts sequentially from either type of habitat. It is thus J08 74 important to determine the different responses of the terrestrial J08 75 and aquatic stages of a parasite's life cycle to climatic change. J08 76 That, along with an examination of other parasite responses to J08 77 climatic change, demands a quantitative framework within which to J08 78 discuss parasite life-history strategies.

J08 79 II. PARASITE LIFE-HISTORY STRATEGIES

J08 80 The complexities of parasite host population dynamics may be J08 81 reduced by the derivation of expressions that describe the most J08 82 important epidemiological features of a parasite's life cycle J08 83 (Anderson and May 1979, May and Anderson 1979, Dobson 1988). Three J08 84 parameters are important in describing the dynamics of a pathogen: J08 85 the rate it will spread in a population, the threshold number of J08 86 hosts required for the parasite to establish, and the mean levels J08 87 of infection for the parasite in the host population.

J08 88 Basic reproductive rate of a parasite, Ro: The basic J08 89 reproductive rate, Ro, of a microparasite may be formally J08 90 defined as the number of new infections that a solitary infected J08 91 individual is able to produce in a population of susceptible hosts J08 92 (Anderson and May 1979). In contrast, Ro for a macroparasite J08 93 is defined as the number of daughters that are established in a J08 94 host population following the introduction of a solitary fertilized J08 95 female worm. In both cases the resultant expression for Ro J08 96 usually consists of a term for the rates of parasite transmission J08 97 divided by an expression for the rate of mortality of the parasite J08 98 in each stage in the life cycle (Dobson 1989). Increases in host J08 99 population size or rates of transmission tend to increase Ro, J08 100 and increases in parasite virulence or other sources of parasite J08 101 mortality tend to reduce the spread of the pathogen through the J08 102 population.

J08 103 Threshold for establishment, HT: The J08 104 threshold for establishment of a parasite, HT, is J08 105 the minimum number of hosts required to sustain an infection of the J08 106 pathogen. An expression for HT, may be obtained by J08 107 rearranging the expression for Ro to find the population J08 108 density at which Ro equals unity. This may be done for both J08 109 micro- and macroparasites with either simple or complex life J08 110 cycles. The resultant expressions suggest that changes in the J08 111 parameters that tend to increase Ro tend to reduce J08 112 HT, and vice versa. Although many virulent species J08 113 require large populations to sustain J08 114 <}_><-|>themselvles<+|>themselves<}/>, reductions in the mortality J08 115 rate of transmission stages may allow parasites to compensate for J08 116 increased virulence and maintain infections in populations J08 117 previously too small to sustain them.

J08 118 Mean prevalence and burden at equilibrium: It is also J08 119 possible to derive expressions for the levels of prevalence J08 120 (proportion of the hosts infected) and incidence (mean parasite J08 121 burden) of parasites in the host populations. In general, J08 122 parameters that tend to increase Ro also tend to give J08 123 increases in the proportion of hosts infected by a microparasite J08 124 and increases in the mean levels of abundance of any particular J08 125 macroparasite (Anderson and May 1979, May and Anderson 1979, Dobson J08 126 1988). Most important, increases in the size of the host population J08 127 usually lead to increases in the prevalence and incidence of the J08 128 parasite population (fig. 16.1).

J08 129 These expressions, which characterize the most important J08 130 features of a parasite's interaction with its host at the J08 131 population level, can be used to ascertain how parasites with J08 132 different life cycles will respond to long-term climatic changes. J08 133 This may best be undertaken by determining which stages of the life J08 134 cycles are most susceptible to climatic variation and by J08 135 quantifying the response of those stages to climatic change.

J08 136 figure&caption

J08 137 III. EFFECT OF TEMPERATURE ON PARASITE TRANSMISSION J08 138 RATES

J08 139 The physiology of adult parasites is intimately linked with the J08 140 physiology of their hosts. Providing the hosts can withstand J08 141 environmental changes, it seems unlikely that the J08 142 within-host component of the parasite life cycle will be J08 143 significantly affected. However, any form of increased stress on J08 144 the host may lead to increase in rates of parasite-induced host J08 145 mortality (Esch et al. 1975). In the absence of data from the J08 146 specific experimental studies that could throw considerable light J08 147 on these relationships, this study will concentrate on the effect J08 148 of changes in meteorological factors on the free-living infective J08 149 stages of different groups of parasites.

J08 150 A. Parasites with Aquatic Transmission Stages

J08 151 Several detailed laboratory studies have examined the effect of J08 152 temperature on the transmission success of parasites with aquatic J08 153 infective stages. The parasitic trematodes are probably the most J08 154 important class of parasites to utilize an aquatic stage for at J08 155 least part of their life cycle. The data presented in figure 16.2 J08 156 are for an echinostome species that is a parasite of ducks. J08 157 Increased temperature leads to increased mortality of the larval J08 158 infective stages of the parasite. It also leads to increased J08 159 infectivity of the larval stage. The interaction between larval J08 160 infectivity and survival means that net transmission efficiency J08 161 peaks at some intermediate temperature but remains relatively J08 162 efficient over a broad range of values J08 163 (16<*_>degree<*/>-36<*_>degree<*/>C for Echinostoma liei J08 164 cercariae; fig. 16.2). These synergistic interactions between the J08 165 different physiological processes determining survival and J08 166 infectivity allow the aquatic parasites to infect hosts at a J08 167 relatively constant rate over the entire spectrum of water J08 168 temperatures that they are likely to experience in their natural J08 169 habitats (Evans 1985).

J08 170 B. Poikilothermic Hosts

J08 171 The effect of temperature on the developmental rate of J08 172 parasites in both aquatic and terrestrial hosts has been examined J08 173 for several of the major parasites of humans in the tropics. In J08 174 contrast to the effect on transmission efficiency, increases in J08 175 temperature usually lead to reduced development times for parasites J08 176 that utilize poikilothermic hosts (fig. 16.3). As with many J08 177 physiological processes, a 10<*_>degree<*/> increase in temperature J08 178 seems to lead to a halving of the developmental time. This may J08 179 allow parasite populations to build up rapidly following increases J08 180 in temperature.

J08 181 C. Parasite Populations in Thermal Cooling Streams

J08 182 The expressions for Ro and HT, derived in J08 183 the first part of this chapter, suggest that increases in J08 184 transmission efficiency and reductions in development time induced J08 185 by temperature changes allow parasites to establish in smaller J08 186 populations and grow at more rapid rates. This is observed to some J08 187 extent in a pair of long-term studies that compare the parasite J08 188 burdens of mosquito fish (Gambuis affinis) populations in J08 189 artificially heated and control sections of the Savannah River in J08 190 South Carolina. The data for the trematode Ornithodiplostomum J08 191 ptychocheilus show significant differences between heated and J08 192 ambient sites during the earlier period of the study when J08 193 temperature differences were most pronounced. Infection by the J08 194 parasites starts several months earlier each year in the thermally J08 195 altered sites (fig. 16.4). However, infection rates decline in the J08 196 summer in the artificially heated sites when populations of hosts J08 197 decline in response to high water temperatures (Camp et al. 1982). J08 198 This effect may be compounded by the movement of the waterfowl that J08 199 act as definitive hosts for the parasite. These birds tend to J08 200 prefer the warmer water in winter and cooler water in the summer. J08 201 Similar but less clearly defined patterns are observed in the data J08 202 for Diplostomum scheuringi from the same site (Aho et al. J08 203 1982).

J08 204 These studies illustrate the important role of host population J08 205 density in the response of a parasite's transmission rate to J08 206 thermal stress, while also demonstrating the ability of parasites J08 207 to capitalize on improved opportunities for transmission and to J08 208 establish whenever opportunities arise. Obviously the data are open J08 209 to several interpretations, but they do emphasize the importance of J08 210 long-term experiments in determining the possible effects of global J08 211 warming on the distribution of parasites.

J08 212 D. Terrestrial Hosts

J08 213 The survival rates of the infective stages of the parasites of J08 214 most terrestrial species tend to decrease with increasing J08 215 temperature (fig. 16.5a). Although little evidence is available to J08 216 determine how the infectivity of these larvae is affected by J08 217 temperature, rates of larval development tend to increase with J08 218 increasing temperatures (fig.16.5b). These two processes again J08 219 interact synergistically - as an increase in temperature depresses J08 220 survival, development speeds up - allowing the parasite to J08 221 establish at a broad range of environmental temperatures. In J08 222 contrast to parasites that utilize aquatic hosts, parasites of J08 223 terrestrial hosts have transmission stages that are susceptible to J08 224 reduced humidity, and these stages are highly susceptible to J08 225 desiccation (Wallace 1961). To compensate for reduced opportunities J08 226 for transmission during periods of severely adverse climate, J08 227 parasites of terrestrial hosts have evolved adaptations such as J08 228 hypobiosis, the ability to remain in a state of arrested J08 229 development within the relatively protected environment provided by J08 230 their hosts until such time as transmission through the external J08 231 environment proves more effective. J08 232 J09 1 <#FROWN:J09\>EXTRACELLULAR SIGNAL-REGULATED KINASES IN T J09 2 CELLS

J09 3 Anti-CD3 and 4<*_>beta<*/>-Phorbol 12-Myristate J09 4 13-Acetate-Induced Phosphorylation and Activation

J09 5 CHARLES E. WHITEHURST, TERI G. BOULTON, MELANIE H. COBB, AND J09 6 THOMAS D. GEPPERT

J09 7 Extracellular signal-regulated kinases (ERK) 1 and 2 are J09 8 growth factor- and cytokine-sensitive serine/threonine kinases that J09 9 are known to phosphorylate microtubule-associated protein 2 and J09 10 myelin basic protein. The current studies examined whether ERK1 J09 11 and/or ERK2 was present in T cells and whether they were J09 12 phoshorylated and activated as a consequence of T cell activation. J09 13 The data demonstrated that both ERK1 and ERK2 were present in J09 14 Jurkat cells and peripheral blood T cells. In T cells, ERK2 was J09 15 more prevalent than ERK1. The concentrations of ERK1 and ERK2 were J09 16 not altered by stimulating the cells for 16 h with immobilized J09 17 anti-CD3 mAb or anti-CD3 mAb and phorbol myristate acetate. mAb to J09 18 CD3 and phorbol myristate acetate stimulated an increase in ERK1 J09 19 and ERK2 MBP kinase activity. Anti-CD3 mAb triggered an increase J09 20 their phosphate content which was detectable at 2 min but reached a J09 21 maximum at 5 min. A portion of the increase in phosphate was caused J09 22 by an increase in phosphotyrosine. We also examined the rate of J09 23 ERK2 degradation. ERK2 was stable for up to 36 h, and its J09 24 degradation was unaffected by the activation state of the cells. J09 25 The data demonstrate that ERK1 and ERK2 are part of an J09 26 anti-CD3 mAb-stimulated signal transduction cascade that is J09 27 downstream of protein kinase C and, therefore, suggest that these J09 28 kinases play an important role in T cell activation.

J09 29 T cell activation is triggered by an interaction between the J09 30 TCR and a complex formed by the antigenic peptide and a class I or J09 31 II MHC molecule. The recognition of Ag by the TCR leads to a J09 32 variety of early biochemical changes (1-4). Among the earliest J09 33 occurring within seconds of engagement of the TCR, is the J09 34 phosphorylation of a variety of substrates on tyrosine residues J09 35 (5). The kinase or kinases responsible for these first J09 36 phosphorylation events have not been identified, although a member J09 37 of the src family of tyrosine kinase, fyn, appears to be physically J09 38 associated with the CD3 complex (6). The subsequent steps in the J09 39 cascade of reactions have not been well characterized. Temporally, J09 40 activation of tyrosine phosphorylation precedes activation of J09 41 phospholipase C (PLC) (5). Further, inhibitors of tyrosine kinases J09 42 block the activation of PLC, demonstrating that tyrosine J09 43 phosphorylation is necessary even for the stimulation of PLC (5, J09 44 7). As in other systems, PLC releases two products, inositol J09 45 phosphates, that elevate intracellular calcium, and diacylglycerol, J09 46 that activates and translocates protein kinase C (PKC) to the J09 47 membrane, where it is in proximity to membrane substrates (2-4, 8, J09 48 9). In addition to these events triggered by engagement of the TCR J09 49 complex, cross-linking CD4 or CD8 molecules on T cells leads to the J09 50 activation of a tyrosine kinase, p561ck, or the src J09 51 family (10, 11). Based on burgeoning evidence from studies of T J09 52 cell activation and by analogy to other receptor signaling systems, J09 53 it is believed that these early biochemical phenomena result in the J09 54 activation of additional cascades of protein kinases which are J09 55 responsible for the subsequent cellular events. In other systems, J09 56 the vast majority of these subsequent phosphorylations occur on J09 57 serine and threonine residues.

J09 58 Of the data suggesting that other kinases are stimulated in T J09 59 cells, the most convincing is for a MAP2 kinase activity that is J09 60 increased by cross-linking the TCR complex (12-14) or by the J09 61 phorbol ester, PMA (13). The stimulation of this serine/threonine J09 62 kinase activity by a mAb to the TCR complex is diminished but not J09 63 eliminated by an inhibitor of PKC (H7) or by prior depletion of PKC J09 64 (13). Cross-linking CD4 on T cells stimulates the tyrosine J09 65 phosphorylation of a 43-kDa protein that co-migrates with a protein J09 66 with MAP2 kinase-like activity (12, 14). Cross-linking CD3 together J09 67 with CD4 results in a greater increase in MAP2 kinase activity than J09 68 cross-linking CD3 alone (14). This combination of stimuli also J09 69 delivers a more effective activation signal than cross-linking the J09 70 TCR complex alone (15-17), indicating a correlation between MAP2 J09 71 kinase activity and T cell activation. Taken together these J09 72 findings suggest that MAP2 kinase plays an important role in T cell J09 73 activation.

J09 74 Recently, cDNA have been cloned and sequenced from human, rat, J09 75 mouse, and Xenopus libraries which encode two proteins with J09 76 insulin and nerve growth factor-sensitive MAP2/MBP kinase activity J09 77 (18-23). These two proteins are 90% identical within their J09 78 catalytic domains and have been named extracellular J09 79 signal-regulated kinase (ERK) 1 and 2 because of the wide variety J09 80 of extra-cellular signals which stimulate their activity J09 81 (13, 14, 18, 20, 24-32). Moreover, they are highly conserved across J09 82 species (19). The activation of these two serine/threonine kinases J09 83 is correlated with increases in the phosphorylation of their J09 84 tyrosine and threonine residues (20). Phosphatases specific for J09 85 either tyrosine (CD45) or serine/threonine (2a) residues partially J09 86 inactivate ERK1 and ERK2 (25), indicating that full activation of J09 87 these kinases requires both tyrosine and serine/threonine J09 88 phosphorylation.

J09 89 The current studies were carried out, therefore, to determine J09 90 whether ERK1 and/or ERK2 is present in T cells and to determine J09 91 whether they are phosphorylated and/or activated as a consequence J09 92 of activation. We found that both ERK1 and ERK2 are present in J09 93 Jurkat cells and peripheral blood T cells and that ERK2 is more J09 94 abundant. The concentration of both ERKs was not altered by J09 95 stimulating the cells for 16 h with immobilized anti-CD3 mAb or J09 96 anti-CD3 and PMA. Both anti-CD3 and PMA stimulated the J09 97 phosphorylation of ERK1 and ERK2 in Jurkat cells and induced an J09 98 increase in ERK1 and ERK2 MBP kinase activity. The data suggest J09 99 that ERK1 and ERK2 may play an important role in T cell activation. J09 100 Moreover, the size, kinetics of activation, and substrate J09 101 specificity of ERK1 and ERK2 are similar to properties of kinases J09 102 with anti-CD3-inducible MAP2 kinase-like activity noted in previous J09 103 reports. Thus, the studies strongly support the hypothesis that J09 104 ERK1 and ERK2 are the kinases responsible for this activity.

J09 105 MATERIALS AND METHODS

J09 106 Reagents, mAb, and mitogens. PMA (Sigma Chemical Co., J09 107 St. Louis, MO) and phorbol 12,13-dibutyrate (PDB; Sigma) were J09 108 dissolved in ethanol and added to the cultures at a final J09 109 concentration of 10 ng/ml for PMA and 20 ng/ml for phorbol J09 110 12,13-dibutyrate, and 0.1% ethanol. The mAb used were OKT3 and 64.1 J09 111 (American Type Culture Collection, Rockville, MD), IgG2a mAb J09 112 directed at the CD3 molecular complex on mature T cells; TS1/18, an J09 113 IgG1 mAb directed at the <*_>beta<*/>-chain of CD18; and P1.17, an J09 114 IgG2a mAb or irrelevant specificity. All mAb were purified over a J09 115 column of Sepharose 4B coupled with staphylococcal protein A and J09 116 used at final concentrations as indicated in the figure legends. J09 117 PHA (Wellcome Reagents Division, Burroughs Wellcome Co., Research J09 118 Triangle Park, NC) was used at a final concentration of 0.5 J09 119 <*_>mu<*/>g/ml. Orthophosphate (H3PO4; 285Ci/mg J09 120 P) and 35S-labeled amino acids (Tran35S-label; J09 121 >1000 Ci/mmol) were purchased from ICN Biomedicals, Inc., Costa J09 122 Mesa, CA.

J09 123 Cell preparation. PBMC were obtained from healthy J09 124 adult volunteers by centrifugation of heparin-treated venous blood J09 125 over sodium diatrizoate/Ficoll gradients (Isolymph; Gallard J09 126 Schlesinger Chemical Manufacturing Corp., Carle Place, NY). Cells J09 127 were washed once in HBSS and twice in saline before further J09 128 processing. T cells were prepared from PBMC by isolating the cells J09 129 forming rosettes with neuraminidase-treated SRBC and passing them J09 130 over a nylon wool column to deplete B cells and macrophages. Jurkat J09 131 cells, a malignant T cell line, were generously provided by Dr. J09 132 Arthur Weiss (33).

J09 133 Antipeptide antisera. Antisera raised to peptides at J09 134 the carboxyl terminus (837; IFQETARFQPGAPEAP) and subdomain XI J09 135 (691; KRITVEEALAHPYLEQYYDPTDE) of rat ERK1 had characteristics J09 136 described previously (25). Antiserum A249 was raised to a peptide J09 137 at the carboxyl terminus of rat ERK2 (ELIFEETARFQPGYRS) using the J09 138 same methodologies described previously (25). The sequences of J09 139 human ERK1 and ERK2 are identical to the rat sequences in the J09 140 region of the peptides used to derive 691 and A249 (19). The J09 141 sequence of the carboxyl terminus of human ERK1 differs slightly J09 142 from that of rat ERK1 (IFQETARFQPGVLEAP).

J09 143 Preparation of cell lysates and immunoblotting. After J09 144 T cells or Jurkat cells were treated with the various stimuli as J09 145 indicated in the figure legends the cells were homogenized or J09 146 sonicated in homogenization buffer (20 mM Tris, pH 7.5, 20 mM J09 147 p-nitrophenyl phosphate, 1 mM EGTA, 50 mM NaF, 50 <*_>mu<*/>M J09 148 sodium vanadate, 3 <*_>mu<*/>g/ml aprotinin, 1 mM leupeptin, 1 mM J09 149 PMSF, and 10 mM iodoacetamide), the lysates clarified by J09 150 centrifugation, and the soluble protein concentrations were J09 151 determined using the bicinchoninic acid/Cu2+ reagent J09 152 (Micro BCA kit; Pierce Chemical Co., Rockford, IL). Equivalent J09 153 amounts of protein/lane were analyzed by electrophoresis on 10 to J09 154 14% SDS-polyacrylamide gels, transferred to nitrocellulose or J09 155 Immobilon-P (Millipore), and immunoblotted with a mAb to J09 156 phospho-tyrosine (hybridoma 4G10; Upstate Biotechnology J09 157 Ind., Lake Placid, NY) or antiserum 691 or A249. Reactive proteins J09 158 were visualized with goat anti-rabbit or mouse IgG (Cappel) J09 159 conjugated to horseradish peroxidase and 4-chloro-1-napthol, or J09 160 with chemiluminescent procedure per manufacturers' specifications J09 161 (Amersham Corp., Arlington Heights, IL). For chemiluminescent J09 162 detection, the films were preflashed before exposure.

J09 163 Radiolabeling of Jurkat cells and J09 164 immunoprecipitations. Jurkat cells (40 x 106 J09 165 cells/ml) were preincubated in phosphate-free RPMI 1640 containing J09 166 1% BSA for 50 to 60 min at 37<*_>degree<*/>C, J09 167 [32P]orthophosphate was added (1.0 mCi/ml), and the cells J09 168 were incubated an additional 50 to 60 min at 37<*_>degree<*/>C. J09 169 After washing twice in ice-cold 0.9% NaCl, the cells were J09 170 resuspended in RPMI 1640, stimulated as described in the figure J09 171 legends, pelleted, and lysed. For immunoprecipitation with A249, J09 172 the cells were lysed in homogenization buffer containing 1% NP-40 J09 173 and 2 mM MgCl2, pH 8.0, for 10 min at 4<*_>degree<*/>C. J09 174 Insoluble debris was removed by centrifugation. For J09 175 immunoprecipitation with 837 the proteins were denatured either by J09 176 boiling the lysates in SDS (0.2%) or by lysing the cells directly J09 177 in 1% SDS. With the latter technique the lysate was later diluted J09 178 so that the concentration of SDS was <0.2% for immunoprecipitation. J09 179 For immunoprecipitations, the antisera were either added directly J09 180 to the supernatants (10 <*_>mu<*/>l/sample), incubated for at least J09 181 1 h at 4<*_>degree<*/>C, and then precipitated with SPA-agarose (30 J09 182 <*_>mu<*/>l of packed SPA-agarose/sample) for at least 1 hour at J09 183 4<*_>degree<*/>C, or alternatively, saturating amounts of the J09 184 antisera were preabsorbed to SPA-agarose, and this complex was J09 185 added to the lysates (30 <*_>mu<*/>l of packed preabsorbed J09 186 SPA-agarose/sample and incubated in the same manner. The results of J09 187 using these two methods were comparable. The immunoprecipitates J09 188 were washed four times in homogenization buffer containing 1% J09 189 NP-40, 0.1% SDS, and 150 mM NaCl, boiled in gel loading buffer J09 190 containing <*_>beta<*/>-ME, and analyzed by SDS-PAGE and J09 191 autoradiograhy. To correlate the molecular weights of the proteins J09 192 immunoprecipitated by anti-serum 837 with those detected by J09 193 antiserum 691 on an immunoblot, 837 immunoprecipitates or a whole J09 194 cell lysate were electrophoresed on each side of a molecular weight J09 195 standard on an SDS-PAGE gel. The gel was then sliced in half down J09 196 the middle of the lane containing molecular weight standards, and J09 197 the half of the gel containing the 837 immunoprecipitates was dried J09 198 and used for autoradiography. The proteins on the other half of the J09 199 gel were transferred to a polyvinylidene difluoride membrane for J09 200 immunoblotting with antiserum 691. Exposures from the J09 201 autoradiograph and immunoblot were then aligned using the molecular J09 202 weight markers. Sequential immunoprecipitations were performed by J09 203 immunoprecipitating with A249 as described earlier, boiling the J09 204 immunoprecipitate in 1% SDS, and then diluting the supernatant to J09 205 0.2% SDS using 1% NP-40 homogenization buffer. 837 J09 206 immunoprecipitates were then performed as described above.

J09 207 Chromatography. Jurkat cells (200 x 106 J09 208 cells/group) were stimulated for 5 min as indicated in the figure J09 209 legends, pelleted, and then resuspended in 1 ml of homogenization J09 210 buffer at 4<*_>degree<*/>C (20 mM Tris, pH 7.5, 1 mM EGTA, 50 mM J09 211 NaF, 20 mM p-nitrophenyl phosphate, 1 mM sodium vanadate, 1 mM J09 212 PMSF). The cells were sonicated with two 10-s bursts using a J09 213 Vibracell (model ASI; Sonics and Materials Inc., Danbury, CT) and J09 214 the cellular debris pelleted by centrifugation for 10 min at 15,000 J09 215 x g. The supernatants were stored at -80<*_>degree<*/>C. For J09 216 chromatography, the sonicates were thawed on ice, diluted with 3 J09 217 vol. of water, and loaded onto a Mono Q HR5/5 column equilibrated J09 218 with buffer A (50 mM <*_>beta<*/>glycerophosphate, pH 7.3, 1 mM J09 219 EGTA, 0.1 mM sodium orthovanadate, and 0.1 <*_>mu<*/>M pepstatin). J09 220 J10 1 <#FROWN:J10\>Such trade-offs would make genetic load and the cost J10 2 of natural selection greater than they would be with functional J10 3 independence.

J10 4 Wallace (1987, 1989; Reeve et al. 1988) recently J10 5 renewed his effort to lay the problem to rest with arguments and J10 6 evidence that the traditional genetic-load argument, such as mine J10 7 above, is based on faulty logic and a misunderstanding of the J10 8 dynamics of viability selection during the culling of an age J10 9 cohort. He emphasizes that much culling must take place in every J10 10 population because of the universal Malthusian factor of J10 11 over-production of offspring. He envisions a world in which an J10 12 individual dying as a result of some genetic deficiency is thereby J10 13 making room for a better endowed individual. If one does not die J10 14 the other one would. Given that populations do remain finite, the J10 15 argument fits the facts. Yet the conceptual problem remains, J10 16 especially in low-fecundity species like our own, because the J10 17 genetic-load arithmetic makes us expect far more culling than is J10 18 actually found, and far more than the population could bear.

J10 19 Wallace's (1987) illustrative model of cohort culling is the J10 20 competition between seedlings in an experimental tray (e.g. Schmidt J10 21 and Ehrhardt 1990). The space available will permit only a limited J10 22 plant biomass to develop, and this will be produced by the small J10 23 number of successful contenders. The great majority will do very J10 24 little growing and gradually die out. The possibly small fraction J10 25 that survives to maturity will be enormously variable in size and J10 26 fitness. This result seems to be unaffected by levels of genetic J10 27 load in the seeds used. If a thousand viable seeds are sown in one J10 28 tray and a thousand with 90% lethal genotypes in another, the two J10 29 trays may produce about the same number of ultimate survivors, J10 30 total biomass, and phenotypic fitness variation.

J10 31 I would suggest another kind of experiment as more relevant to J10 32 the problem Haldane had in mind. Sow only 100 of the viable seeds J10 33 in the first tray and 100 with a high incidence of genetic load in J10 34 the other, and also in each tray sow 900 seeds of competing J10 35 species. I would expect the 100 viables to win much more J10 36 representation in their tray than the ten viables and 90 lethals in J10 37 the other, and this result would be a closer parallel to what J10 38 usually takes place in the culling of a plant cohort in nature. For J10 39 most animals, Wallace's experimental model is even less realistic. J10 40 Only sessile invertebrates meet intense and inescapable competition J10 41 from near neighbors. Wallace's seedling experiment would be broadly J10 42 applicable to animal populations only with competition for social J10 43 status, West Eberhard's (1983) social selection, of which J10 44 sexual selection would be a special case. Social status is a J10 45 resource that can seldom be appropriated by a member of a different J10 46 species.

J10 47 The central problem with Wallace's model, which he calls J10 48 soft selection, is that it implies unrealistically strong J10 49 density dependence with an age cohort. Most populations in nature J10 50 are extremely sparse. The tendency for field ecologists to study J10 51 organisms that are abundant enough to study may greatly bias our J10 52 impressions. Populations in nature are seldom dense enough to cause J10 53 any obvious resource depression (Tilman 1982). The most convincing J10 54 examples of resource depression result from exploitation by many J10 55 species, such as the reduction of marine invertebrate biomass on J10 56 mudflats from concerted onslaughts of many species of migratory J10 57 bird (Schneider 1978). Even the extraordinarily dense populations J10 58 commonly studied by field ecologists, e.g. by Andrewartha and Birch J10 59 (1954), usually show numerical changes that look like random J10 60 fluctuation and seldom give clear evidence of density effects in J10 61 short-term studies. Individual survival must be mainly a matter of J10 62 chance, partly a matter of many kinds of adaptive performance, and J10 63 only to a minor degree affected by density-dependent competition J10 64 with conspecifics.

J10 65 Near neighbors, sessile or motile, will often be of different J10 66 species in diverse natural communities, and the death of one J10 67 individual will often allow the survival of a member of a competing J10 68 species. In such situations we most clearly confront the challenge J10 69 of genetic load in relation to population survival, the problem J10 70 that worried Haldane. If too large a dose of its population's J10 71 genetic load causes one individual to die, it is likely to mean J10 72 that the abundance of that species will be reduced by one. A J10 73 reduced genetic load would make it more likely for the population J10 74 to survive in competition with other species. Every (1-s) that J10 75 enters into a fitness calculation means a finite deficiency in some J10 76 sort of adaptive performance. Any such deficiency implies not only J10 77 adverse selection within a population, but also a decreased J10 78 representation of the population in the community. Dudash's (1990) J10 79 experiments nicely confirm this expectation. Fitness differences J10 80 between her inbred and outbred seedlings were much greater in the J10 81 field than in greenhouse monoculture. The expectation is that the J10 82 natural populations that are still available for study should be J10 83 those that have extremely low levels of genetic load, and this is J10 84 not what is found.

J10 85 Wallace (1989) claims that Haldane and others have been J10 86 needlessly worried about a mere "computational J10 87 artifact." They arbitrarily assign a fitness of 1 to a J10 88 favored genotype and of 1-s to an unfavored competitor. If J10 89 instead we used 1+s and 1 we would not calculate such low J10 90 fitness values for so many multi-locus genotypes. This is true, but J10 91 the change is merely cosmetic. We would still get the same J10 92 variation in fitness and be faced with the same problem of how fit J10 93 the average individual can possibly be. Also, the traditional J10 94 notation is more realistic. A rare favorable mutation may be said J10 95 to have a fitness of 1+s, but this implies a deficiency in the J10 96 ancestral gene pool. If a mutation can improve some character by J10 97 some fraction s, that character must have been suboptimal. How J10 98 could the ancestral population have survived with suboptimal J10 99 genotypes at a large number of loci if there were competing J10 100 populations with a lower genetic load?

J10 101 I think the time has come for renewed discussion and J10 102 experimental attack on Haldane's dilemma.

J10 103 10.2 Paradoxes of sexuality

J10 104 Sexual reproduction by its existence and in many special J10 105 aspects is a complex of puzzles on which many books have been J10 106 written (e.g. Bradbury and Andersson 1987; Stearns 1987; Michod and J10 107 Levin 1988). The main theoretical challenge is in the cost of J10 108 meiosis, but this is a matter already getting attention from J10 109 able investigators. I can do no better than refer readers to J10 110 Maynard Smith (1984b), Eberhard (1985), Felsenstein J10 111 (1985a), Bierzychudek (1989), Hamilton et al. J10 112 (1990), Parts II and III of Stearns (1987) and Chapters 4-9 in J10 113 Michod and Levin (1988). Another major challenge is in resolving J10 114 the data of life-history diversity in the frequency and the J10 115 developmental and ecological correlates of sexual phases. The J10 116 problem here is not so much logical as logistic. The diversity is J10 117 overwhelming in relation to the time and money that a few thousand J10 118 interested biologists can devote to it.

J10 119 Of the many recombination-related difficulties that I could J10 120 discuss I will echo Maynard Smith (1988a) and choose the one J10 121 that best serves as a kind of text-book illustration of an J10 122 evolutionary anomaly, the absence of sexual reproduction throughout J10 123 the rotifer order Bdelloidea. This is anomalous because it clearly J10 124 violates the principle of Muller's ratchet, which seems a logically J10 125 tight line of reasoning from well established premises (but see J10 126 Gabriel (1989)). Muller (1964) was the first to recognize that an J10 127 asexual lineage "incorporates a kind of ratchet mechanism, J10 128 such that it can never get to contain, in any of its lines, a load J10 129 of mutation smaller than that already existing in its at present J10 130 least loaded lines." It can acquire a higher load of J10 131 mutation simply by the occurrence of a new one in a least loaded J10 132 line. So exclusively asexual reproduction leads inevitably to a J10 133 degeneration of the genome, in the sense of its being ever more J10 134 ruled by chemical stability, and ever less informative as to what J10 135 has succeeded in the past. This must always lead to rapid J10 136 extinction on an evolutionary time scale. For a recent quantitative J10 137 study of Muller's ratchet, see G. Bell (1988).

J10 138 Muller's ratchet explains the phylogenetic distribution of J10 139 asexual species in most major types of eukaryotes. There is a fair J10 140 number of exclusively clonal species, but never any entirely clonal J10 141 genera or higher categories. Asexual species arise from time to J10 142 time, but Muller's ratchet must lead them to extinction long before J10 143 they can produce any appreciable taxonomic diversification. The J10 144 loss of sexuality seems to be a classic example of an evolutionary J10 145 step that is opposed by clade selection (Van Valen 1975). J10 146 Unfortunately the general rule of conformity to expectations of J10 147 Muller's ratchet has some exceptions. The whole rotifer order J10 148 Bdelloidea (Meglitch (1967) calls them a class), with its several J10 149 families and many genera and species, is composed entirely of J10 150 parthenogenetic females (Pennak 1978). There are also a few other J10 151 noteworthy violators of the theory, such as the freshwater J10 152 gastrotrich order Chaetonotoidea (Meglitch 1967; Pennak 1978).

J10 153 Another difficulty that surely deserves more attention is the J10 154 scarcity of adaptively flexible sex determination (for a J10 155 comprehensive review, see Bull, (1983)). Sex determination in most J10 156 animals is genetic and is fixed at conception. Only a few have sex J10 157 determination as a facultative response to information perceived J10 158 during development. A neatly understandable example (Conover and J10 159 Heins 1987) is provided by a fish, the Atlantic silverside, which J10 160 spawns every spring in shallow waters along the Atlantic coast of J10 161 the United States and Canada. Eggs spawned early in the season J10 162 become females; those later on become males. Temperature provides J10 163 the cue, so that development below a certain threshold causes J10 164 female development, warmer water male development. This mechanism J10 165 gives females a longer growing season and larger size for the J10 166 following spring's spawning. The close relationship between size J10 167 and fecundity in fishes makes large size more important to female J10 168 fitness than to male.

J10 169 This environmental sex determination is adaptive in a way that J10 170 extend the size-advantage model used to explain the occurrence of J10 171 protandry vs. protogyny among sequential hermaphrodites (Ghiselin J10 172 1969; Warner 1975). The silverside is almost entirely semelparous, J10 173 and this would rule out sequential hermaphroditism as a viable life J10 174 history. As predicted (Conover and Heins 1987), the temperature J10 175 threshold that determines sex varies as expected of combined J10 176 optimizing and frequency-dependent selection. It is lower in J10 177 northern, higher in southern parts of the range, so that the sexes J10 178 are nearly equally abundant all along the coast. Experiments by J10 179 Conover and Van Voorhees (1990) show that the threshold can be J10 180 changed by selection in the laboratory.

J10 181 Besides the greater dependence of reproductive success on size J10 182 in females than in males in most animals, it is possible to think J10 183 of many other ways in which it may be more adaptive for an J10 184 individual in a given situation to be male or female, and to J10 185 identify cues that would predict the situation during development. J10 186 A clear example would be the stochastically varying sex ratios of J10 187 many social groups. In a pond in which most of the frogs happen to J10 188 be of sex A, it would pay a tadpole to develop into a member of sex J10 189 B. It would also pay a parent to bias sex determination away from J10 190 whatever is the majority in previously produced young (Taylor and J10 191 Sauer 1980). This would avoid what might be called Baptista's J10 192 Burden. Having Bianca instead of a son after having had J10 193 Katharina was something of a challenge to his fitness. Yet despite J10 194 the advantage that can be envisioned in alternating sons and J10 195 daughters, each sex determination is a largely independent event in J10 196 most animal populations (Williams 1979; Huck et al. J10 197 1990).

J10 198 Facultative sex determination would only be expected in groups J10 199 in which useful cues can be perceived prior to any major J10 200 developmental commitment to maleness or femaleness, requirements J10 201 discussed in detail by Bull (1983), Charnov and Bull (1977), and J10 202 Korpelainen (1990). These conditions must surely obtain in many J10 203 diploid insect populations. Why is adaptive sex determination, such J10 204 as that found in the silverside, not widespread in many groups of J10 205 insects? J10 206 J11 1 <#FROWN:J11\>The argument for conversation is strongest when J11 2 functional tests, such as gene replacement or in vitro J11 3 complementation, can be applied. Most often, though, we must rely J11 4 on sequence comparisons. But do these sequence comparisons monitor J11 5 adaptive evolution, or do they monitor genetic drift? In the case J11 6 of p34 there is extensive sequence identity throughout the molecule J11 7 between yeast and humans, which diverged more than a billion years J11 8 ago. In the case of the conserved regulators of p34, cyclin, cdc25, J11 9 and weel, the identity is limited to a small portion of the J11 10 molecules and there is extensive divergence in other domains. Yet J11 11 in the case of cdc25, for example, despite the large sequence J11 12 divergence, the human molecule will complement yeast mutants (Sadhu J11 13 et al. 1990) and in the case of weel, frog can also complement J11 14 yeast (Booher and Kirschner, unpublished).In an in vitro system, J11 15 sea urchin cyclin, despite its large divergence in most of the J11 16 molecule from the frog cyclin, will complement a deficiency of frog J11 17 cyclin. These results suggest that functional divergence of these J11 18 important regulating genes has been minimal, while sequence J11 19 divergence has been extensive. Since cdc25, weel, and the p34 J11 20 kinase fully complement deletions of these genes in species that J11 21 diverged more than a billion years ago, we can conclude that no J11 22 important yeast functions are missing in the human protein. The J11 23 reciprocal experiment is not possible in humans but may soon be J11 24 possible in mouse (Thomas and Capecchi 1990). Therefore, while J11 25 these sequences have apparently drifted extensively, they do not J11 26 appear to have evolved functionally very much.

J11 27 With the help of genetic tests, the list of highly conserved J11 28 cellular functions has continued to grow. In some cases J11 29 phylogenetic barriers have emerged, but for many systems they are J11 30 minor and easily overcome. For example, the <*_>beta<*/>-adrenergic J11 31 receptor that normally responds to catecholamines in heart muscle J11 32 will not function in yeast to replace a related receptor that J11 33 responds to mating pheromones. However, addition of one more J11 34 element to the signaling system, the mammalian G<*_>alpha<*/> J11 35 protein, will allow the yeast cell to respond to catecholamines and J11 36 undergo the mating response (King et al. 1991). The obvious J11 37 conservation of DNA structure has been matched by the conservation J11 38 of histones, transcription factors, splicing enzymes, J11 39 ribonucleoprotein complexes, and nuclear pores. The well-known J11 40 conservation of the protein synthesis machinery has been extended J11 41 to protein secretion, including components of the endoplasmic J11 42 reticulum and Golgi. The major cytoskeletal proteins such as J11 43 tubulin and actin have been conserved and serve similar functions J11 44 in many organisms, as do the metabolic enzymes. The signaling and J11 45 regulating molecules such as ras and hormone receptors, the J11 46 regulating kinases such as protein kinase A, protein kinase C, S6 J11 47 kinase - all have been found in every eukariotyc cell. There have J11 48 been, of course, new inventions, expansion and specialization of J11 49 each repertoire, but, as we shall see, many of the new demands of J11 50 specialized cell types have come about by usurpation of existing J11 51 components. Viewed as a computer, we would have to say that the J11 52 basic hardware is similar in all eukaryotic cells; if anything has J11 53 changed, it is the software.

J11 54 Software Changes in the Cell Cycle

J11 55 Though the proteins that regulate the cell cycle may be nearly J11 56 identical in all organisms, the strategy for regulating the cell J11 57 cycle is not. In the frog egg the accumulation of cyclin to a J11 58 threshold initiates mitosis, and this process is independent of any J11 59 transcriptional control (Murray and Kirschner)1989b). In the J11 60 Drosophila embryo after cellularization, cyclin accumulation J11 61 is also required for mitosis, but it is not the regulator. In this J11 62 case control of the mitotic process is under control of the mitotic J11 63 activator cdc25, and its expression is under transcriptional J11 64 control (Edgar and O'Farrell 1990). Recent studies of cyclins J11 65 specific to the G1/S transition in yeast have shown that their J11 66 accumulation is under transcriptional control, but also may be J11 67 under posttranslational control (Wittenberg et al. 1990; I. J11 68 Herskowitz, pers. comm.). In cleaving frog eggs and sea urchin J11 69 eggs, cyclin accumulation is completely unregulated. However, in J11 70 the case of the G1/S cyclin in yeast, the accumulation is tied to J11 71 the whole pathway of the mating pheromone response as well as to J11 72 other less well understood pathways involving cell size and J11 73 nutrition.

J11 74 In the few well-studied cases of cell cycle control we can see J11 75 both conservation and divergence. The major components are highly J11 76 conserved and most are functionally interchangeable. The basic J11 77 reaction pathway involving cyclins, p34 kinase, and other kinases J11 78 and phosphatases is also identical. Yet the rate limiting steps and J11 79 their linkage to other processes are different in different cells. J11 80 This has enabled the cell cycle control mechanisms to respond J11 81 transcriptionally to spatial signals, to be linked to extracellular J11 82 cues, to be coupled to various homeostatic mechanisms, or to J11 83 operate nearly autonomously during the rapid cleavages in the early J11 84 embryo. As with modern computers, the architecture of the machine J11 85 allows for many software applications. One might even say that J11 86 computer hardware has evolved to allow for greater software J11 87 flexibility. As we shall see, much the same can be said for J11 88 cellular mechanisms.

J11 89 Divergent Pathways of Photoreception

J11 90 The cephalopod eye and the vertebrate eye are exquisite J11 91 examples of convergent evolution. The anatomy suggests that the J11 92 origins are totally independent. The vertebrate eye develops as an J11 93 outgrowth of the brain; the cephalopod and insect eye develops as a J11 94 peripheral ectodermal structure that grows into the brain (Young J11 95 1974). The topology of the nerves and photoreceptors is reversed. J11 96 In the vertebrate eye, light passes through the nerves to the J11 97 photoreceptor; in the cephalopod eye, light impinges directly on J11 98 the photoreceptors. Is this anatomical convergence reflected in a J11 99 totally separate origin of the biochemistry of photoreception?

J11 100 The key event in photoreception, the photoisomerization of J11 101 retinal-dehyde, has been widely used. In prokaryotes, where J11 102 it is part of the proton pump, and in eukaryotes, where it is used J11 103 as a photoreceptor, retinaldehyde, which is chemically the same in J11 104 all systems, is bound to an integral membrane protein called opsin, J11 105 whose polypeptide chain spans the plasma membrane seven times. J11 106 There is no sequence homology between the prokaryotic opsins and J11 107 the eukaryotic opsins, though overall structural similarities in J11 108 the positions of the amino and carboxyl ends and the number of J11 109 transmembrance helices suggest that at one time these proteins J11 110 could have had a common origin (Henderson and Schertler 1990).

J11 111 In eukaryotes, whether cephalopods or mammals, opsin is a J11 112 7-membrane spanning protein, and all such proteins are receptors J11 113 that are thought to couple to intracellular GTP binding proteins J11 114 called G-proteins. This widespread family of membrane protein J11 115 receptors includes the receptors for the mating pheromones in J11 116 yeast, the cAMP receptor in slime molds, and the serotonin and J11 117 <*_>beta<*/>-adrenergic receptor in mammals (King et al. 1991). The J11 118 receptors catalyze the exchange of GTP for GDP on the J11 119 heterotrimeric G protein. Binding of GTP causes dissociation of the J11 120 trimeric G protein into G<*_>alpha<*/> and J11 121 G<*_>beta<*/>G<*_>gamma<*/>; these subunits interact with other J11 122 cellular enzymes and regulate their functions. The invertebrate J11 123 opsins, which are 7-membrane spanning integral membrane proteins, J11 124 have clear sequence similarity to the vertebrate opsins (Yokoyama J11 125 and Yokoyama 1989). In the central region of the molecule there is J11 126 also a very strong similarity on the nucleic acid level, and J11 127 throughout the molecule there is extensive similarity with a few J11 128 insertions or deletions. There is no question that rhodopsin, the J11 129 primary unit of photoreception, has evolved from a common J11 130 precursor.

J11 131 The vertebrate opsins are known to couple to a heterotrimeric G J11 132 protein called transducin, which in its GTP form activates directly J11 133 a cGMP phosphodiesterase. In the vertebrate photoreceptor, J11 134 increased levels of cGMP open a Na+ channel leading to J11 135 increased neurotransmitter release. Therefore, the action of light J11 136 causes a drop in cGMP and an inhibition of transmitter release that J11 137 inactivates an inhibitory neuron, which ultimately leads to J11 138 elevated electrical activity in the brain (Stryer 1988). The J11 139 vertebrate photoreception system also has a means of adaptation J11 140 that desensitizes the receptor after stimulation. It involves the J11 141 binding of a small protein, called <*_>beta<*/>-arrestin, to the J11 142 cytoplasmic domain of the receptor after a period of activation J11 143 (Bennett and Sitaramayya 1988).

J11 144 In invertebrates, although the initial coupling of opsin to J11 145 signal transmission are similar, the complete pathway is designed J11 146 differently. Drosophila is known to contain G-proteins J11 147 (Guillen et al. 1990), and the structure of invertebrate opsin J11 148 strongly suggests that the receptor couples to G-proteins; the J11 149 exact G-protein that couples to Drosophila rhodopsin is not J11 150 known. Like vertebrates, Drosophila contains a J11 151 <*_>beta<*/>-arrestin molecule that is highly conserved, suggesting J11 152 that Drosophila rhodopsin contains the same desensitization J11 153 system as mammals (Smith et al. 1990). However, the next part of J11 154 the pathway seems divergent. G proteins are known to couple to J11 155 several second messenger systems, and the best evidence suggests J11 156 that G protein in invertebrates (Drosophila and the horseshoe J11 157 crab, Limulus) couples to a different second messenger system J11 158 from that affecting cGMP phosphodiesterase (Suss et al. 1989; J11 159 J.Brown, pers. comm.). Genetic approaches can be useful in J11 160 delineating this second messenger pathway. Recently, J11 161 Drosophila mutants have been obtained that have J11 162 morphologically normal cells that do not respond to light. The gene J11 163 that is defective in one of these mutants has been cloned and shown J11 164 to have strong similarity phospholipase C, an enzyme involved in J11 165 cell signaling (Bloomquist et al. 1988). There is evidence that J11 166 Ca++ release, mediated by inositol triphosphate, occurs J11 167 during light stimulation, which suggests that in the invertebrate J11 168 photoreceptors the G protein linked to opsins may activate J11 169 phospholipase C and signal either Ca ++ pathways via J11 170 inositol triphosphate or protein kinase C via diacylglycerol. It is J11 171 also possible that the unknown G protein signals some other second J11 172 messenger pathway. Downstream of this signaling system there is an J11 173 increase (as opposed to the decrease in vertebrate photoreceptors) J11 174 in a nonselective cation channel leading to a depolarization J11 175 and secretion. Thus the invertebrate system uses the same visual J11 176 pigment, an evolutionarily related receptor, a very similar J11 177 desensitization system; but most likely it couples this receptor to J11 178 a different G-protein-mediated system to produce the opposite J11 179 electrophysiological result from the one that occurs in J11 180 vertebrates. In the end the brain still gets the signal.

J11 181 The lessons of the comparative physiology of vertebrate and J11 182 invertebrate photoreceptors is that the basic components have been J11 183 highly conserved but their linkage has developed differently. The J11 184 basic input of photons is the same; the output hyperpolarization or J11 185 depolarization of the photoreceptor cell is completely different. J11 186 In between there has been a high degree of conservation: J11 187 retinaldehyde, 7-membrane spanning receptors, G proteins, J11 188 <*_>beta<*/>-arrestin, phospholipase C, nonselective cation J11 189 channels; but the circuitry is different. The evolutionary J11 190 invention was not in the types of proteins but in software for J11 191 linking signaling and responding pathways together.

J11 192 New Components and Their Evolutionary Value

J11 193 Not all the remodeling of the eukaryotic cell is the equivalent J11 194 of rearranging the furniture. There are, of course, new genes whose J11 195 expression facilitated rapid evolutionary change. In the computer J11 196 analogy these are the hardware improvements, which often provide J11 197 new capacities for software innovations. As we shall see, some of J11 198 these new genes may have persisted underutilized for extended J11 199 periods of time, until the appropriate software mechanisms were J11 200 developed to make use of them. In most cases the origins of these J11 201 genes is traceable to more primitive structures that were stitched J11 202 together by gene duplication and exon shuffling, but in some cases J11 203 there is little clue as to their origins. It seems likely that some J11 204 of these specific genes are crucial for major branches of J11 205 macroevolution. Although one can tabulate many genes that would J11 206 qualify as a "great moment in evolution", I will discuss only two J11 207 structures dependent on new genes that are important for the major J11 208 radiations within the vertebrates: myelin and feathers.

J11 209 The biophysical features of nerve conduction explained by cable J11 210 theory show that the rate and efficiency of nerve conduction J11 211 increase with the diameter of the nerve fiber and with the decrease J11 212 in the capacitance of the plasma membrane. To process complex J11 213 information or to respond quickly to a predator or to capture food, J11 214 rapid nerve conduction is obviously advantageous. J11 215 J12 1 <#FROWN:J12\>Of those animals which had positive isolations from J12 2 the oviducts, 68.8% had isolations from both oviducts.

J12 3 Since a relatively high inoculating dose of chlamydiae was used J12 4 to infect the guinea pigs, we wanted to determine if ascending J12 5 infection into the uterus and oviducts would develop with lower J12 6 numbers of organisms. Thus, a dose-response experiment was J12 7 performed in which animals were inoculated with J12 8 104-108 IFU of GPIC (Table 1). When animals J12 9 were sacrificed on days 7 or 9, no dose-related differences were J12 10 found in the number of animals acquiring infection in the J12 11 oviducts.

J12 12 Histopathologic Analysis of Endometrium

J12 13 After it was established that ascending infection was a common J12 14 occurrence in the genital tract, we examined the various endometria J12 15 and oviducts for pathologic changes associated with infection. The J12 16 histopathology of the exo-and endocervix has been previously J12 17 described. The data presented is from a pool of animals killed at J12 18 various times after infection and is based on the total number of a J12 19 given specimen examined, rather than the total number of animals, J12 20 i.e., two uterine horns and two oviducts per animal. The percentage J12 21 of total endometrial tissues with specific pathologic changes is J12 22 presented in Figure 2. Only 1 in 12 animals showed inflammation by J12 23 day 3. Acute inflammation was the most prevalent finding at all J12 24 time points from day 7-12, peaking at day 9. Peak infiltrates with J12 25 lymphocytes and plasma cells were also seen at day 9 although the J12 26 percentage of animals showing these findings was less than those J12 27 showing acute inflammation. Fibrosis of endometrial stroma was J12 28 virtually unseen. The percentage of animals displaying an J12 29 inflammatory response decreased by day 12, and with the exception J12 30 of a single animal on day 20 (not shown), no histopathology was J12 31 identified in the endometrium after day 12 even including specimes J12 32 examined on days 30 and 75-85. Of interest is the observation that J12 33 on day 7, only 43.6% of the uterine horns were positive for J12 34 pathology whereas 63.8% were isolation positive. Similarly, on day J12 35 9, 57.7% were positive for pathology with 66.7% positive for J12 36 isolation.

J12 37 caption&table

J12 38 To further characterize the pathologic findings, we J12 39 semiquantified the morphologic findings in endometria showing J12 40 abnormalities (Figure 3). Animals were included if any pathologic J12 41 parameter was positive. The predominant pathologic finding was J12 42 acute inflammation particularly on days 7 and 9. Polymorphonuclear J12 43 leukocytes infiltrated the glandular surface epithelium, filled the J12 44 endometrial gland lumens, and were scattered throughout the J12 45 superficial stroma. Chronic inflammation was also present at all J12 46 timepoints, but in lesser quantity. The lymphocytes were arranged J12 47 in loose aggregates in the superficial stroma with occasional J12 48 transformed lymphocytes identified (Figure 4). Plasma cell J12 49 infiltrates were seen from days 5 through 12 but always in smaller J12 50 numbers than either polymorphonuclear leukocytes or lymphocytes. J12 51 Plasma cells were scattered throughout the endometrial stroma in a J12 52 patchy distribution.

J12 53 figure&caption

J12 54 In 12 uninfected control animals, an occasional solitary J12 55 lymphocyte aggregate was seen in either the uterine fundus or J12 56 horns. Infrequently, one or two endometrial glands contained a few J12 57 scattered polymorphonuclear leukocytes. There were no plasma cells, J12 58 fibrosis, or erosions identifed in any control animal. Thus, the J12 59 pathology described earlier in infected animals was obviously a J12 60 result of the infection and not associated with a normal resident J12 61 endometrial response.

J12 62 Histopathologic Analysis of Oviduct and Mesosalpinx

J12 63 Figure 5 illustrates the percentage of mesosalpingeal tissue J12 64 and oviducts that showed infiltrates of polymorpho-nuclear J12 65 leukocytes, lymphocytes or plasma cells and/or fibrosis. Because J12 66 many animals had only unilateral pathology, the data is presented J12 67 based on the total number of tissues examined. The number of J12 68 specimens in either the mesosalpinx or oviduct that had pathologic J12 69 changes was low at 7 days after infection, when the isolation of J12 70 organisms from the same specimens was maximum. Nevertheless, by 9 J12 71 days after infection, the number of samples with pathologic changes J12 72 had doubled although this number was never as great as the total J12 73 number of specimens from which chlamydiae were isolated. Thus, some J12 74 tissues were isolation positive but did not have detectable J12 75 pathology.

J12 76 figure&caption

J12 77 Early in the infection (days 5-12), acute and chronic J12 78 inflammatory responses as well as plasma cell infiltration were J12 79 common in both the mesosalpinx and oviducts. By day 30, the acute J12 80 inflammatory and plasma cell responses had markedly diminished in J12 81 both tissues; however, lymphocytic infiltrates and fibrosis were J12 82 still evident in the mesosalpinx. Pathologic changes in the J12 83 oviducts at day 30 and beyond were minimal. The reactions in the J12 84 mesosalpinx continued to decrease, but 21% of the tissues still had J12 85 obvious fibrosis 75-85 days after infection, and 19% had ongoing J12 86 chronic inflammation.

J12 87 Semiquantification of the morphologic findings in the specimens J12 88 showing abnormalities is presented in Figure 6. The early stage of J12 89 the infection was characterized by an acute inflammatory reaction J12 90 in both the mesosalpinx and oviducts (Figure 7). Chronic J12 91 inflammation was also present early but did not reach its peak J12 92 level until day 12 in the mesosalpinx as did plasma cell J12 93 infiltration (Figure 8). The appearance of plasma cells J12 94 corresponded to the development of antibody that normally is J12 95 detectable about day 10. The development of fibrosis was primarily J12 96 restricted to the mesosalpinx and was maximum at day 12 although it J12 97 persisted and was still obvious as late as 75-85 days after J12 98 infection. Tubal dilatation (hydrosalpinx) was apparent in 12% of J12 99 the observed oviducts in the 75-85 day period, and in some cases, J12 100 was marked with tubal diameters as great as 1 cm.

J12 101 figures&captions

J12 102 Oviduct and mesosalpingeal tissues were also stained with J12 103 guinea pig anti-GPIC antibodies followed by peroxidase-labeled J12 104 rabbit anti-guinea pig IgG to visualize and localize chlamydial J12 105 antigen. Chlamydial antigen and inclusion bodies were commonly J12 106 detected in the 7-12 day period in the epithelial cells of the J12 107 oviduct (Figure 9).

J12 108 figure&caption

J12 109 Oviducts from uninfected animals were also examined to J12 110 determine whether any inflammatory infiltrates were normally J12 111 present. No acute or chronic inflammation was identified in any of J12 112 12 control animals nor were any plasma cells, fibrosis, or erosions J12 113 found.

J12 114 Discussion J12 115 In this study, we describe a model for chlamydial genital J12 116 infection in which ascending infection to the endometrium and J12 117 oviducts routinely occurs as a result of vaginal inoculation of the J12 118 chlamydial agent. Other than the primate, which is limited in J12 119 usefulness by expense, the guinea pig:GPIC model represents the J12 120 only animal model in which ascending infection from a vaginal J12 121 inoculation, analogous to the human situation, can be commonly J12 122 demonstrated, even though GPIC is a member of the C. J12 123 psittaci, it has been found to elicit ocular and genital J12 124 infections remarkably paralleling the corresponding human disease. J12 125 GPIC primarily infects superficial epithelial cells of the cervix J12 126 and epithelial cells of the male urethra, but most significantly, J12 127 the infection can be transmitted sexually in guinea pigs. Moreover, J12 128 newborns of infected mothers acquire a conjunctival infection by J12 129 passage through the birth canal and can develop a pneumonia typical J12 130 of chlamydial pneumonia of the newborn when inoculated J12 131 intranasally. Hormonal effects on chlamydial infection can and have J12 132 been effectively studied in the guinea pig, since of all the J12 133 rodents, the reproductive system of the female guinea pig most J12 134 closely resembles the human with regard to their long estrous cycle J12 135 (17 days), spontaneous ovulation, and actively secreting corpus J12 136 luteum. Immunologically, guinea pigs develop both cell-mediated and J12 137 humoral immune responses to GPIC. Analogous to humans, guinea pigs J12 138 also produce significant antibody responses to the major outer J12 139 membrane protein (39 kDa), the chlamydial GroEL (57 kDa), the J12 140 Omp2protein (60 kDa), and lipopoly-saccharide of GPIC. Also J12 141 comparable to humans is the short immunity to reinfection that J12 142 occurs, with animals becoming susceptible to reinfection as early J12 143 as 2 months after the resolution of a primary infection.

J12 144 caption&figure

J12 145 Previously, we have only noted the develoment of upper tract J12 146 disease when animals have been manipulated either by treatment with J12 147 cyclophosphamide or estradiol. However, in those studies, the early J12 148 timepoints of the infection were not carefully evaluated or studied J12 149 in a large number of animals. In the current investigation, we J12 150 analyzed the tissues early after infection and found that in a high J12 151 percentage of guinea pigs, chlamydiae can be isolated form the J12 152 oviducts within 1 week of vaginal inoculation. The presence of J12 153 organisms in the oviducts is limited in duration, with J12 154 disappearance from the oviducts concomitant with the resolution of J12 155 cervical infection and the development of both cell-mediated J12 156 immunity and serum and secretion antibodies.

J12 157 caption&figure

J12 158 The ascending nature of the infection is confirmed by the fact J12 159 that no chlamydiae could be isolated from the oviducts on day 3 and J12 160 only a few on day 5 despite the presence of organisms in the cervix J12 161 of virtually all animals. Isolation from the uterus paralleled that J12 162 of the oviducts although animals became positive earlier in the J12 163 uterus. Thus, several days were required for the organisms to reach J12 164 both the endometrium and the oviducts. However, it is significant J12 165 that organisms could be recovered from the endometrium and oviducts J12 166 of almost 80% of the guinea pigs assessed at day 7. Moreover, the J12 167 appearance of the bacteria in the oviducts by day 7 also represents J12 168 a remarkably rapid ascending infection for a non-motile, slowly J12 169 growing organism. If one estrapolates these data to the human J12 170 situation, they suggest that a much higher number of women develop J12 171 upper tract infection than previously believed and that this may J12 172 occur quickly after infection. As stated earlier, Jones et al have J12 173 reported evidence in support of these data in cases of chlamydial J12 174 infections in women.

J12 175 However, it is interesting that not all animals that acquire J12 176 tubal infection go on to develop tubal pathology. Although 78% of J12 177 the animals had tubal infection on day 7, only 45% were found to J12 178 have pathologic changes in the oviducts on day 9, the time at which J12 179 maximum pathology was noted. Swenson et al using the direct J12 180 injection model with MoPn also observed that not all injected J12 181 animals developed salpingitis. A similar phenomenon was also noted J12 182 in our study with regard to isolation and pathologic changes in the J12 183 endometrium. Significantly, it has been recently reported that J12 184 chlamydiae could also be isolated from the fallopian tubes of women J12 185 without laparoscopic evidence for salpingitis. These data would J12 186 suggest that, in some cases, the immune response may be J12 187 sufficiently rapid in producing those effector functions that can J12 188 resolve the infection. Antibody can be detected in genital J12 189 secretions of guinea pigs as early as 10 days after infection, and J12 190 we have found even earlier appearance of antibody on some J12 191 occasions. Cell-mediated immunity, which is also required for J12 192 resolution of a chlamydial genital infection, can also be present J12 193 as early as 10 days.

J12 194 An alternative explanation for the variation in incidence of J12 195 pathology is that other physiologic factors may be affecting their J12 196 development. We have previously reported that when guinea pigs are J12 197 treated with estradiol in either pharmacologic or physiologic J12 198 doses, a markedly enhanced infection is noted in the cervix with a J12 199 significantly increased number of animals developing J12 200 hydro-salpinx. Moreover, the infection is prolonged when J12 201 compared with untreated controls. Sweet et al have also noted that J12 202 the onset of acute salpingitis in women occurred significantly more J12 203 often within 7 days following the beginning of menses than at other J12 204 times in their menstrual cycle. In addition, it has been J12 205 well-described that treatment with oral contraceptives does J12 206 increase the number of individuals from whom chlamydiae can be J12 207 isolated. Although hormonal changes are not the only possible J12 208 factor in the variance seen in tubal pathology, they certainly may J12 209 have some role based on available data. Since this model resembles J12 210 humans in that not all individuals develop overt salpingitis, it J12 211 will be useful to investigate those endogenous or exogenous factors J12 212 that alter the incidence of salpingitis. An important point, J12 213 however, is that a high number of individuals do have organisms in J12 214 the oviducts, and the development of overt salpingitis may be J12 215 dependent on factors that prevent elimination of the bacteria from J12 216 the oviducts or mediate the development of pathologic changes. J12 217 Furthermore, the presence of organisms in the oviducts may not J12 218 necessarily mandate the production of disease with harsh J12 219 sequelae.

J12 220 Finally, it was found that the pathologic changes occurring as J12 221 a result of chlamydial genital infection in the guinea pig with the J12 222 GPIC agent were remarkably parallel to that in human chlamydial J12 223 endometritis and salpingitis. J12 224 J13 1 <#FROWN:J13\>Between Bone Tissue-Type Comparison

J13 2 The second step in the tissue-type comparison was to analyze J13 3 interbone variation using the bone sites selected as least and most J13 4 variable - the midshaft femur and vertebral body, respectively. J13 5 Waldron (1989) reported that between bones of the same skeleton, J13 6 elemental levels can vary by as much as a factor of two. J13 7 Elaborating upon the method employed by Tanaka et al. (1981) to J13 8 compare vertebra to other bones, the Nubian and modern samples were J13 9 compared for the degree of dissimilarity between cortical and J13 10 cancellous tissues using a modification of the two-sample Student's J13 11 t-test (Greene, 1973). Greene first developed this J13 12 modified t-test as a method for comparing hominoid species for J13 13 differences in the degree of sexual dimorphism expressed in the J13 14 dentition. the method was further revised by Greene (1989) for J13 15 comparisons "between populations and comparisons between J13 16 generations within populations" (p.121). This modification J13 17 was used in the present investigation as follows:

J13 18 formula

J13 19 where subscript: 1=Nubian midshaft femur

J13 20 2=Nubian vertebral body

J13 21 3=Modern midshaft femur

J13 22 4=Modern vertebral body

J13 23 Table 17 lists the results of these comparisons. Barium, J13 24 copper, iron, potassium, magnesium, manganese, and zinc show J13 25 significant (p<0.05) tissue-type differences between Nubian J13 26 and modern samples. Of these, barium, iron, magnesium, manganese, J13 27 and zinc are more variable in the Nubians than the moderns. J13 28 Conversely, copper and potassium are more variable in the modern J13 29 samples.

J13 30 caption&table

J13 31 The more variable elements in the Nubian bone may indicate J13 32 enrichment from the soil. Except for zinc, each is found at higher J13 33 concentrations in the soil than in the bone. The reduced J13 34 variability of copper and potassium in the Nubian bone (compared to J13 35 the moderns) may indicate depletion of these elements into the J13 36 soil.

J13 37 Summary of Element Selection

J13 38 Seven tests were employed to identify five measures of J13 39 diagenesis among the elements - 1) range overlap between the modern J13 40 and Nubian samples; 2) variability among the elements using CV; 3) J13 41 antagonistic/synergistic interactions between the elements using J13 42 multi-element correlations; 4) analysis of bone contamination from J13 43 elements in the soil; and 5) variation between tissue-types to J13 44 assess enrichment/depletion of elements in bone. Based on these J13 45 measures, elements were divided into those minimally, moderately J13 46 and highly affected by diagenesis.

J13 47 Nubian means for the midshaft femur and for all sites combined J13 48 were overlayed on the modern distributions for each element. Only J13 49 one element fell outside the modern range - boron. Manganese and J13 50 iron varied for the combined site mean, but the Nubian femur mean J13 51 was within the modern distribution.

J13 52 Coefficients of variation were metrically ranked for the J13 53 Nubians and moderns across all sites combined and for the midshaft J13 54 femur alone. In all four rankings, calcium, phosphorus, magnesium, J13 55 sodium, and strontium were the least variable 100% of the time. J13 56 Vanadium was among the six least variable 87% of the time. The J13 57 three most variable elements across all four rankings were boron, J13 58 manganese, and zinc.

J13 59 When elements were ranked across all bone sites ordinally, J13 60 again calcium, phosphorus, sodium, magnesium, and strontium were J13 61 the least variable for both the Nubian and modern samples. Boron J13 62 and manganese were the most variable.

J13 63 All of the multi-element correlations noted for the Nubians J13 64 could be explained using the modern matrix or the literature, J13 65 except for zinc/phosphorus. The modern and Nubian patterns matched J13 66 for calcium/phosphorus, nickel/vanadium, potassium/sodium, and J13 67 boron/lead. The Nubians agreed with the literature for those J13 68 elements believed to be diagenetic - copper, iron, and manganese. J13 69 The strong strontium/barium and sodium/magnesium correlations were J13 70 also expected based on published studies (Buikstra et al., J13 71 1989).

J13 72 Comparison of the Nubian bones to associated soil values and to J13 73 the modern sample showed that calcium, phosphorus, strontium, and J13 74 sodium were not affected by enrichment or depletion from the J13 75 surrounding soil. Boron, iron, and manganese showed a pattern J13 76 indicative of enrichment, and potassium of depletion. Barium, J13 77 copper, magnesium, nickel, lead, and zinc were indeterminate J13 78 because while Nubian bone levels were below soil values, they were J13 79 equivalent to the modern samples.

J13 80 Tissue-type comparisons within and between bones also showed J13 81 distinct patterns. The within-bone comparison between cortical and J13 82 cancellous bone showed that elements concentrated in the same J13 83 tissues for both the femur and humerus. When the percent variation J13 84 between tissues was compared, calcium, phosphorus, strontium, J13 85 magnesium, sodium, and potassium were found in highest J13 86 concentrations in cortical bone. The latter three, significantly J13 87 so. All other elements concentrated in cancellous bone. According J13 88 to the literature (Buikstra et al., 1989; Waldron, 1989), elements J13 89 concentrating in cancellous bone are more likely the result of J13 90 diagenesis than those in cortical bone.

J13 91 The between-bone tissue-type comparison showed degree of J13 92 dissimilarity for the least and most variable bone sites in the J13 93 Nubian skeleton. Barium, iron, magnesium, manganese, and zinc were J13 94 more variable among the Nubian bones, possibly indicating J13 95 enrichment. Copper and potassium, were found to vary less in the J13 96 Nubians than moderns, indicating potential depletion.

J13 97 Other than the between bone comparison, for each measure, J13 98 calcium, phosphorus, sodium, and strontium appeared the least J13 99 affected by diagenesis. Magnesium also rated high in each test J13 100 except the soil and between-bone comparisons. Boron and manganese J13 101 were consistently ranked the most variable. Based on these J13 102 comparisons, calcium, phosphorus, magnesium, strontium, and sodium J13 103 are considered minimally affected; barium, copper, nickel, J13 104 vanadium, iron, potassium, and zinc, moderately affected; and, J13 105 boron and manganese, highly affected.

J13 106 SUMMARY OF DIAGENESIS EVALUATION

J13 107 Analysis of the degree of diagenesis was conducted along J13 108 multiple lines of inquiry. First, the quality of the bone J13 109 composition was assessed. Then bone sites were compared to J13 110 determine which were the least affected in the depositional J13 111 environment. Finally, elements were tested for alteration and J13 112 grouped into those minimally, moderately, and highly affected by J13 113 diagenesis.

J13 114 Bone preservation proved exceptional by all measures except J13 115 %ash. However, this discrepancy is very likely the result of J13 116 experimenter error. Of the bone sites, the midshaft femur proved J13 117 the least altered, based on a rank-ordering of coefficients of J13 118 variation for both Nubian and modern samples. Among the elements, J13 119 calcium, phosphorus, magnesium, sodium, and strontium were judged J13 120 minimally affected by diagenesis. Boron and manganese were the most J13 121 affected, and all other elements were moderately altered.

J13 122 This analysis employed virtually all the conventional methods J13 123 suggested in the literature for assessing diagenesis (Price, 1989; J13 124 Buikstra et al., 1989). Several traditional tests were modified in J13 125 this study, and others were introduced for future use. Whereas J13 126 diagenesis cannot be measured directly, the present analysis J13 127 demonstrates a strong circumstantial argument for selection of the J13 128 least affected bone and elements in the Nubian remains.

J13 129 CHAPTER 5:

J13 130 RESULTS AND DISCUSSION II:

J13 131 BIOCULTURAL RECONSTRUCTION

J13 132 Numerous biocultural reconstructions of ancient human J13 133 populations have been conducted utilizing chemical analyses (for J13 134 reviews of this literature, see Klepinger 1984; Price et al, 1985; J13 135 Gilbert, 1975; Schoeninger 1979; Sillen and Kavanagh, 1982; Price, J13 136 1989; Buikstra et al., 1989). Most promote an interdisciplinary J13 137 approach to the use of elemental variation.

J13 138 A principal advantage of the Nubian remains to elemental J13 139 analysis is the rich biocultural context within which the elemental J13 140 data can be interpreted. Price et al (1985) and Blakely and Beck J13 141 (1981) recommended that chemical analyses be conducted in concert J13 142 with nutritional and paleopathological investigations of ancient J13 143 remains, the later stating that elemental research is "of J13 144 little utility when ... findings cannot be corroborated by other J13 145 sources of information, such as demographic data or pathological J13 146 diagnoses" (p.422). Martin and Armelagos (1985) concurred, J13 147 advising that elemental analyses will not realize their full J13 148 potential "until they are used in conjunction with other J13 149 techniques such as gross and microscopic analyses. A thorough J13 150 understanding of nutritional deficiencies and health in prehistory J13 151 will require an examination of anatomy, pathology, histology, and J13 152 chemistry in a systematic analysis using multiple J13 153 indicators" (p.527).

J13 154 The biocultural analysis of elemental variation at Kulubnarti J13 155 will proceed along two lines of inquiry. First, a general J13 156 assessment by age and sex will be conducted. Then given the J13 157 importance of independent lines of confirmatory evidence noted by J13 158 the authors quoted above, the second phase of the investigation J13 159 will focus on previous studies of aging, nutrition, and disease at J13 160 Kulubnarti. Based on the results of Chapter 4, only the femur will J13 161 be used in these analyses. Also, given the substantial evidence of J13 162 their idagenetic alteration, boron and manganese will be excluded J13 163 from further consideration.

J13 164 BIOCULTURAL DIMENSIONS AT KULUBNARTI

J13 165 As previously discussed, the people of the Batn el J13 166 Hajar lived in villages of perhaps a dozen households J13 167 dependent on small-scale farming for their livelihood. Agriculture J13 168 was intensified where possible by simple irrigation systems and J13 169 retaining walls which protected alluvial soils. Staple crops J13 170 included sorghum, dates, millet, barley, beans, lentils, peas, and J13 171 a small amount of wheat. Coprolite analysis also revealed that some J13 172 fish and crocodile were consumed (Cummings, 1988). In addition, a J13 173 few cattle, sheep and pigs were kept, but animal protein appears to J13 174 have been a minor part of the Nubian diet (Carlson et al., 1974; J13 175 Adams, 1977).

J13 176 From the standpoint of dietary variation within the Kulubnarti J13 177 population, the archaeological (Adams, 1977) and biological records J13 178 (Van Gerven et al., 1981) overwhelmingly support a single J13 179 interpretation: Kulubnarti was an egalitarian community of J13 180 household producers and consumers. The few communal activities that J13 181 existed were limited to production and maintenance of the village J13 182 saquia (waterwheel) and the irrigation ditches and retaining J13 183 walls. There is no indication of a political or economic elite with J13 184 preferential access to critical resources.

J13 185 BIOCULTURAL RECONSTRUCTIONS UTILIZING ELEMENTAL J13 186 ANALYSIS

J13 187 Elemental analyses in the past have been applied to questions J13 188 of: differential access to food resources related to status (Brown, J13 189 1973; Schoeninger, 1979; Lambert et al., 1979; Blakely and Beck, J13 190 1981; Hatch and Geidel, 1985), sex (Brown, 1973; Schoeninger, 1979; J13 191 Lambert et al., 1979; Price et al., 1986); changes in subsistence J13 192 methods (Lambert et al, 1979; Gilbert, 1975; Jaworoski et al, 1985; J13 193 Katzenberg, 1984; Price and Kavanagh, 1982; Sillen 1981; J13 194 Schoeninger 1982); relative contributions of plant versus animal J13 195 resources (Lambert et al, 1983; Price et al., 1986); contributions J13 196 of marine versus terrestrial resources (Connor and Slaughter, J13 197 1984); patterns of weaning (Sillen and Smith, 1985); and, residence J13 198 patterns (Ericson, 1985). However, inclusion of many of these J13 199 parameters in the present investigation was not possible.

J13 200 For example, the Kulubnarti Nubians were egalitarian, J13 201 therefore, evidence for preferential access to food resources based J13 202 on political or economic status was absent. Given that only J13 203 individuals from the Feudal period were examined, no analysis of J13 204 diachronic change in subsistence patterns was possible. Also, an J13 205 examination of marine versus terrestrial resources could not be J13 206 conducted because the Batn el Hajar is land-locked. J13 207 Patterns of weaning could not be assessed because only adults were J13 208 studied. And finally, analysis of residence patterns requires J13 209 examination of elemental concentrations in the teeth and bone to J13 210 determine childhood versus adult strontium levels. Teeth were not J13 211 studied in the present investigation.

J13 212 Lacking evidence for either political/economic stratification J13 213 or temporal change, elemental variation related to nutrition and J13 214 disease at Kulubnarti was analyzed from the demographic J13 215 perspectives of age and sex. Price (1985) took a similar approach J13 216 in his analysis of a prehistoric Amerindian population stating that J13 217 "Late Archaic groups are generally regarded as egalitarian J13 218 so that dietary differences associated with rank, status or J13 219 position do not play a major role" (p 450). Outlined below J13 220 is a review of the literature pertaining to sex and age-related J13 221 variation in elemental concentrations, followed by the pertinent J13 222 findings of the present investigation.

J13 223 Prior to an analysis of elemental variation by age, sex, diet J13 224 or disease however, it is important to describe features of those J13 225 elements that most often appear in such reconstructions. These J13 226 include strontium, barium, magnesium, and zinc.

J13 227 Strontium J13 228 Strontium is an alkaline earth metal with an uneven J13 229 distribution throughout the lithosphere (Odum, 1951). There are J13 230 four naturally occurring isotopes and 14 radionuclides. Of the J13 231 latter, two are the result of nuclear fission, which explains the J13 232 vast body of knowledge amassed for this element.

J13 233 Strontium has a similar ionization energy, ionic size, and J13 234 electron configuration to calcium, and therefore behaves in a J13 235 similar fashion. Both enter the foodchain at the plant level J13 236 through soil and water, and the amount found in plants is directly J13 237 proportional to the concentrations in the local environment. Plants J13 238 do not discriminate between calcium and strontium in absorption, J13 239 and therefore have a higher strontium concentration than that found J13 240 at any other trophic level. J13 241 J14 1 <#FROWN:J14\>In adults, the small number of patients and their J14 2 uniformly poor performance status has not allowed clear J14 3 identification of the high-risk group. However, certain clinical J14 4 features, as well as immunologic and morphologic features, of adult J14 5 ALL have been associated with poorer clinical performance.

J14 6 Clinical Features

J14 7 Sex, age, initial blast count, initial platelet count, and J14 8 organ involvement are features that identify high-risk patients. J14 9 Males do more poorly than females. Patients over 40 years of age J14 10 have a shorter mean survival time than those below 40. J14 11 Hepatosplenomegaly and/or lymphadenopathy at presentation has been J14 12 associated with shorter remission duration, although the presence J14 13 of isolated splenomegaly may indeed not be a high-risk factor.

J14 14 The white blood cell count at presentation is clearly most J14 15 important in predicting remission duration. Blast counts in excess J14 16 of 100,000 cell/mm3 were a dire prognostic sign in many J14 17 series. Additionally, statistically significant longer remissions J14 18 were demonstrated in patients presenting with an initial blast J14 19 count of 10,000 cells/mm3 or less. Survival is also J14 20 adversely affected by decreased platelet counts at presentation: J14 21 survival was significantly worse in patients with presenting J14 22 platelet counts below 50,000/mm3. Finally, the presence J14 23 of central nervous system involvement at the time of diagnosis was J14 24 associated with a significantly shorter survival.

J14 25 table&caption

J14 26 Morphologic and Immunologic Subtypes

J14 27 The L3 B cell ALL has a very poor prognosis in most J14 28 series. In rank order of increasingly good prognosis, the J14 29 immunologic subtypes may be listed as follows: B cell ALL, T cell J14 30 ALL, null cell ALL, and common ALL. Whether these immunologic J14 31 subtypes are independent prognostic variables or whether they J14 32 reflect differences in clinical presentation remains unclear at the J14 33 present time.

J14 34 Treatment J14 35 Chemotherapy with cytotoxic agents remains the cornerstone of J14 36 the treatment of adult ALL. The present-day approach has been J14 37 developed as a result of multiple clinical trials in both pediatric J14 38 and adult patients. The therapy is divided into remission induction J14 39 (restoration of the normal hematologic state with less than 5% J14 40 blasts in the bone marrow aspirate), early intensification once J14 41 remission is obtained, <}_><-|>maintanence<+|>maintenance<}/> J14 42 therapy to maintain the leukemia-free state, and central nervous J14 43 system prophylaxis or treatment.

J14 44 Induction and Early Intensification

J14 45 A comparison of various induction regimens is shown in Table J14 46 4-5. Initial complete remission rates with vincristine and J14 47 prednisone in adults with ALL were approximately 40%-50%, and the J14 48 remissions were of short duration. it was shown in a number of J14 49 trials that the addition of daunorubicin or doxorubicin to the J14 50 induction regimen increased the complete remission rate to 70%-80%. J14 51 The addition of L-asparaginase to the vincristine and prednisone J14 52 combination yielded similar success.

J14 53 Cancer and Leukemia Group b (CALGB) showed that the addition of J14 54 L-asparaginase after induction with vincristine and prednisone J14 55 resulted in more long-term remissions than the simultaneous J14 56 administration of the three drugs or than vincristine and J14 57 prednisone used alone. Lister et al,. using vincristine, J14 58 prednisone, daunorubicin, and L-asparaginase in combination J14 59 reported a complete remission rate of 71%.

J14 60 The concept of early intensification has been sparked by the J14 61 disappointing duration of remission in the adult population. Gee et J14 62 al. reported a mean remission duration of 24 months with a program J14 63 incorporating cytosine arabinoside, 6-thioguanine, L-asparaginase, J14 64 and carmustine after induction with vincristine, prednisone, and J14 65 daunorubicin. Trials in the pediatric population by Aur et al. and J14 66 Muriel-Sackman et al., using cytosine arabinoside, J14 67 cyclophosphamide, and methotrexate intensification, showed no J14 68 improvement in remission duration.

J14 69 It appears safe to conclude that the approach to adult ALL J14 70 should include intensive induction with vincristine, prednisone, J14 71 and an anthracycline, with the addition of L-asparaginase soon J14 72 after. An example of such a regimen is shown in Table 4-6. The role J14 73 of early and late intensification remains unclear.

J14 74 table&caption

J14 75 Maintenance J14 76 The need for effective maintenance therapy in ALL is evidenced J14 77 by the high relapse rates in patients in whom a complete marrow J14 78 remission is obtained. Short remission durations were noted in J14 79 studies using short intensive induction courses of chemotherapy. J14 80 Similarly, patients in complete remission who discontinue therapy J14 81 relapse quickly. The most commonly used maintenance schedule has J14 82 been methotrexate and 6-mercaptopurine. Attempts to improve J14 83 maintenance duration with the addition of intermittent vincristine, J14 84 prednisone, cytosine arabinoside, and cyclophosphamide have not J14 85 proved successful. Sallan et al. reported improvement in duration J14 86 of remissions with the addition of doxorubicin to vincristine, J14 87 methotrexate, 6-mercaptopurine, and prednisone. However, it is too J14 88 early to know whether this improvement will stand the test of time. J14 89 As there are so few patients remaining in complete remission, the J14 90 question of duration of maintenance therapy remains unanswered in J14 91 adults. In children, it appears that therapy can safely be stopped J14 92 after 3 years of relapse-free maintenance therapy.

J14 93 Relapse J14 94 Hematologic relapse occurs in 70%-80% of adult patients with J14 95 ALL in whom a complete remission is obtained. Once in relapse, J14 96 reinduction with vincristine, prednisone, anthracyclines, J14 97 L-asparaginase, or cytosine arabinoside is successful in 80% of J14 98 children and 50%-70% of adults. These remissions are usually short, J14 99 especially in patients having relapsed during maintenance J14 100 chemotherapy. New chemotherapeutic agents, such as m-AMSA, VP-16 J14 101 and VM-26, are being used in refractory ALL with variable success. J14 102 High-dose therapy with either methotrexate or cytosine arabinoside J14 103 is also being used.

J14 104 Central Nervous System Prophylaxis and Treatment

J14 105 With improvement in the treatment of ALL in children, the J14 106 incidence of central nervous system involvement by leukemic J14 107 infiltration became significant, occurring in 50%-60% of patients J14 108 in hematologic remission. The problem of central nervous system J14 109 involvement was realized as early as 1965, when Frei et al. of J14 110 CALGB instituted intrathecal methotrexate in the induction regimen J14 111 for childhood ALL. This led to a series of trials over the years to J14 112 examine the question of central nervous system prophylaxis. The St. J14 113 Jude's group showed a decrease in incidence to less than 10% with J14 114 the institution of whole-brain irradiation therapy in intrathecal J14 115 methotrexate. The same group showed similar results using 2400 rads J14 116 craniospinal irradiation. This group and others have shown that the J14 117 goal of central nervous system prophylaxis can be achieved with J14 118 whole-brain irradiation therapy and intrathecal methotrexate, J14 119 avoiding spinal irradiation, which causes bone marrow suppression J14 120 and a higher incidence of complications. Attempts to avoid central J14 121 nervous system disease by incorporating drugs that cross the J14 122 blood-brain barrier have been ineffective.

J14 123 The problem of central nervous system leukemia was observed in J14 124 the adult population as well, with an incidence of at least J14 125 40%-50%. Omura et al. showed that 2400 rads whole-brain J14 126 radiotherapy combined with five doses of intrathecal methotrexate J14 127 in doses of 10 mg/m2 (15 mg maximum dose) decreased J14 128 central nervous system incidence from 11/34 in an untreated group J14 129 to 3/28 in a treated group. Unfortunately, no difference in the J14 130 duration of hematologic remission nor survival could be seen J14 131 between the two groups. The L2 protocol series from J14 132 memorial Hospital, which incorporated intrathecal methotrexate J14 133 without radiotherapy, had a 66% incidence of central nervous system J14 134 disease in patients with white counts above 25,000/mm3. J14 135 It seems that the combination of whole-brain radiotherapy and J14 136 intrathecal methotrexate should be advocated at this time. J14 137 Investigations into intraventricular drug delivery via Ommaya J14 138 reservoir are currently under way.

J14 139 In patients with documented central nervous system leukemia, J14 140 treatment is approached by a combination of radiotherapy and J14 141 intrathecal methotrexate or cytoxine arabinoside. Clinical J14 142 presentation of central nervous system leukemia is variable. J14 143 Cranial nerve palsies are common due to leukemic infiltration of J14 144 nerve sheaths at presentation. Diagnosis depends on cytologic J14 145 examination of cytocentrifuge preparations of cerebrospinal fluid. J14 146 The CNS relapse rates are high in patients in whom intrathecal J14 147 therapy is discontinued once central nervous system relapse has J14 148 occurred.

J14 149 New Approaches to ALL

J14 150 Despite advances in remission induction rates, the ability to J14 151 cure ALL with conventional chemotherapy in the majority of adults J14 152 remains elusive. New approaches have emerged to improve results in J14 153 these patients.

J14 154 Bone Marrow Transplantation

J14 155 Bone marrow transplantation of 22 patients with HLA-compatible J14 156 donors, aged 4-22 years, in second or subsequent remission, has J14 157 resulted in long-term survival: 50% remain in remission at 15-35 J14 158 months after transplant. Only 18% of those transplanted in relapse J14 159 have survived. Therefore, it appears that transplant of patients in J14 160 second or subsequent remission is preferable to transplant once J14 161 disease is refractory. When transplantation was compared with J14 162 conventional chemotherapy in patients in second or subsequent J14 163 remission, 11 of 24 of the transplanted group remained disease free J14 164 at 17-55 months, whereas only 1 child of 21 treated with J14 165 conventional chemotherapy remained in remission for more than 2 J14 166 years. Certainly, bone marrow transplantation is not available for J14 167 all patients. However, if available, one should screen family J14 168 members for HLA-compatible donors and should consider transplant J14 169 for patients under 30 years old during second remission or possibly J14 170 in first remission if they have 'high-risk' disease (i.e., a high J14 171 white count at presentation, and null, B, or T cell disease).

J14 172 Serotherapy and Specific Therapy

J14 173 With the expansion of immunologic techniques and subtyping of J14 174 ALL, passive serotherapy with anti-CALLA antibodies has been J14 175 attempted in patients with CALLA-positive ALL. Initial responses J14 176 have been noted; however, these have been very short lived. The J14 177 role of immunologic manipulation in the therapy of ALL remains J14 178 experimental and speculative. Finally, specific selective therapy J14 179 has been reported with 2-deoxycoformycin, an inhibitor of adenosine J14 180 deaminase in a patient with T cell ALL.

J14 181 Outlook J14 182 ALL remains a fatal disease in 70%-80% of affected adults. It J14 183 has been seen that the favorable results in adult ALL have more or J14 184 less been confined to adolescents and young adults, with the J14 185 'older' population faring less well.

J14 186 The challenges seem clear. The role of both early and late J14 187 intensification requires further study. Furthermore, more precise J14 188 designations of poor prognostic signs need to be made in the adult J14 189 population so that more intensive treatment may be attempted. The J14 190 immunologic approach to therapy is in its infancy and clearly needs J14 191 development. It is hoped that these advances will come quickly and J14 192 improvement in long-term survival will be realized.

J14 193 Chronic Lymphocytic Leukemia

J14 194 Chronic lymphocytic leukemia (CLL) was described by Dameshek as J14 195 "an accumulative disease of immunologically incompetent J14 196 lymphocytes." This concise definition describes the J14 197 majority of patients with CLL, who present with physical findings J14 198 and laboratory data that are consistent with Dameshek's theory. CLL J14 199 is a disease of older persons: less than 10% of patients present J14 200 before the age of 50 years. Men are about two times as likely to J14 201 develop CLL as are women. As in all forms of leukemia, CLL is more J14 202 common in the white than in the black population.

J14 203 The diagnosis of CLL is based on the minimal requirement of an J14 204 absolute and sustained lymphocytosis in the peripheral blood of no J14 205 less than 15,000 cells/mm3. The bone marrow is J14 206 hypercellular and more than 40% of the cells are lymphocytes. The J14 207 lymphocytes in peripheral blood and marrow are of the small, J14 208 well-differentiated type (Figure 4-4). Slowly enlarging lymph nodes J14 209 and gradual enlargement of the liver and spleen due to the J14 210 accumulation of neoplastic lymphocytes may occur early or late in J14 211 the course of the disease. The immunologic incompetence of the J14 212 expanding lymphocyte population expresses itself in J14 213 hypo-gammaglobulinemia with predisposition to infections, J14 214 and in the emergence of such autoimmune phenomena as the production J14 215 of antibodies against host red blood cells.

J14 216 Biology

J14 217 Lymphocytes in Chronic Lymphocytic Leukemia

J14 218 CLL primarily represents a clonal proliferation of B J14 219 lymphocytes (Figure 4-2). The malignant cells have immunoglobulin J14 220 molecules on their surfaces bearing the same idiotype and light J14 221 chain type. The most common surface immunoglobulins are IgM and J14 222 IgD. The finding of two heavy chain classes does not preclude the J14 223 concept of monoclonality, as the IgD and IgM have the same idiotype J14 224 specificity and presumably reflect a stage of differentiation of J14 225 normal B lymphocytes. CLL lymphocytes also express the Ia antigen, J14 226 the receptor for C'3, and the receptor for the Fc portion of J14 227 immunoglobulin.

J14 228 Studies of the clonal origin of CLL have used J14 229 glucose-6-phosphate dehydrogenase (G6PD) as a marker in patients J14 230 who are heterozygous at this locus. Skin tissue from these patients J14 231 manifested both A and B G6PD types. However, as one would predict, J14 232 the CLL B lymphocytes manifested only one type of G6PD. In contrast J14 233 to the neoplastic B lymphocytes, the T lymphocytes, granulocytes, J14 234 erythrocytes, and platelets displayed both enzyme types in J14 235 proportions similar to those found in skin. These findings indicate J14 236 that CLL involves only committed B lymphocyte progenitors. J14 237 J15 1 <#FROWN:J15\>Molecular Mechanisms of Drug Addiction

J15 2 Eric J. Nestler

J15 3 Laboratory of Molecular Psychiatry, Departments of Psychiatry J15 4 and Pharmacology, Yale University School of Medicine, Connecticut J15 5 Mental Health Center, New Haven, Connecticut 06508

J15 6 Drug addiction has afflicted mankind for centuries, yet the J15 7 mechanisms by which particular drugs lead to addiction, and the J15 8 genetic factors that make some individuals particularly vulnerable J15 9 to addiction, have remained elusive. From a clinical perspective, J15 10 drug abuse continues to exact enormous human and financial costs on J15 11 society, yet all currently available treatments for drug addiction J15 12 are notoriously ineffective. The search for a better understanding J15 13 of the neurobiological mechanisms underlying the addictive actions J15 14 of drugs of abuse and of the genetic factors that contribute to J15 15 addiction should be given a high priority, as this should result in J15 16 crucial advances in our ability to treat and prevent drug J15 17 addiction.

J15 18 From the basic neuroscience perspective, study of the J15 19 neurobiology of drug addiction offers a novel opportunity to J15 20 establish the biological basis of a complex and clinically relevant J15 21 behavioral abnormality. Many prominent aspects of drug addiction in J15 22 people can be clearly reproduced in laboratory animals, in striking J15 23 contrast to most other forms of neuropsychiatric illness, such as J15 24 psychotic and affective disorders, animal models for which are much J15 25 harder to interpret. Advances made in the study of drug addiction J15 26 should provide important insights into mechanisms underlying some J15 27 of these other disorders.

J15 28 Three terms related to drug abuse are used commonly: J15 29 tolerance, dependence, and addiction. Tolerance J15 30 represents a reduced effect upon repeated exposure to a drug at a J15 31 constant dose, or the need for an increased dose to maintain the J15 32 same effect. Dependence is defined as the need for continued J15 33 exposure to a drug so as to avoid a withdrawal syndrome (physical J15 34 and/or psychological disturbances) when the drug is withdrawn. J15 35 Dependence is considered a priori to result from adaptive changes J15 36 that develop in body tissues in response to repeated drug exposure. J15 37 The traditional distinction between physical and psychological J15 38 dependence is somewhat artificial, since both are mediated by J15 39 neural mechanisms, possibly even similar neural mechanisms, as will J15 40 be seen below. Addiction is defined as the compulsive use of a drug J15 41 despite adverse consequences. In the past, physical dependence was J15 42 part of the definition of addicton. However, the requirement for J15 43 physical dependence as a neccessary or sufficient aspect of drug J15 44 addiction is no longer considered valid. Many drugs with no abuse J15 45 potential, for example, <*_>beta<*/>-adrenergic antagonists, J15 46 clonidine, and tricyclic antidepressants, can produce marked J15 47 physical symptoms on withdrawal. On the other hand, many J15 48 unquestionably severe abusers of some drugs have little or no J15 49 physical withdrawal syndrome upon cessation of drug exposure (e.g., J15 50 most marijuana or cocaine users). Similarly, not all drugs of abuse J15 51 produce tolerance to all of their effects.

J15 52 This article reviews the results of recent research efforts J15 53 that have begun to characterize the neurobiological basis of J15 54 compulsive drug use. Its major focus is on opiates and cocaine, J15 55 since the addictive mechanisms underlying the actions of these J15 56 drugs are the best understood.

J15 57 Cellular site of drug addiction

J15 58 The discovery of endogenous opiate receptors in the 1970s J15 59 raised the possibility that opiate addiction might be mediated by J15 60 changes in these receptors. However, a decade of research has J15 61 failed to identify consistent changes in the number of opiate J15 62 receptors, or changes in their affinity for opiate ligands, under J15 63 conditions of opiate addiction (Loh and Smith, 1990). Changes in J15 64 levels of endogenous opioid peptides also do not appear to explain J15 65 prominent aspects of opiate tolerance and dependence. The discovery J15 66 that cocaine and other addictive psychostimulants acutely inhibit J15 67 the reuptake or stimulate the release of monoamines throughout the J15 68 brain has focused study of their addictive mechanisms on the J15 69 regulation of monoamine neurotransmitters and their receptors. J15 70 These studies too have been disappointing because it has been J15 71 difficult to demonstrate consistent long-term changes in specific J15 72 neurotransmitter or receptor systems in brain regions thought to J15 73 underlie psychostimulant addiction (see Clouet et al., 1988; J15 74 Liebman and Cooper, 1989; Peris et al., 1990).

J15 75 The failure to account for important aspects of opiate and J15 76 psychostimulant addictions in terms of regulation of J15 77 neuro-transmitters and receptors has shifted attention to J15 78 postreceptor mechanisms. Most types of neurotransmitter receptors J15 79 present in brain produce most of their physiological responses in J15 80 target neurons through a complex cascade of intracellular J15 81 messengers. These intracellular messengers include G-proteins J15 82 (Simon et al., 1991), which couple the receptors to intracellular J15 83 effector systems, and the intracellular effector systems J15 84 themselves, which include second messengers, protein kinases and J15 85 protein phosphatases, and phosphoproteins (Nestler and Greengard, J15 86 1984, 1989). Regulation of these intracellular messenger pathways J15 87 mediates the effects of the neurotransmitter-receptor systems on J15 88 diverse aspects of neuronal function, including gene expression. J15 89 Given that many important aspects of drug addiction develop J15 90 gradually and progressively in response to continued drug exposure, J15 91 and can persist for a long time after drug withdrawal, it is likely J15 92 that the regulation of neuronal gene expression is of particular J15 93 relevance to addiction.

J15 94 In recent years, the increasing knowledge of intracellular J15 95 messenger pathways has provided an experimental framework for J15 96 studies of the molecular mechanisms underlying drug addiction. J15 97 These investigations have demonstrated that changes in the activity J15 98 of G-proteins and the cAMP second messenger and protein J15 99 phosphorylation pathway mediate important aspects of opiate, and J15 100 possibly cocaine, addiction in a number of drug-responsive J15 101 brain regions.

J15 102 Molecular mechanisms underlying opiate tolerance, J15 103 dependence, and withdrawal: studies in the locus coeruleus

J15 104 The locus coeruleus (LC) of the rat has served for many years J15 105 as a useful model of opiate action. The LC is the largest J15 106 noradrenergic nucleus in brain, located bilaterally on the floor of J15 107 the fourth ventricle in the anterior pons. It is particularly J15 108 suited for biochemical and molecular investigations, as it is a J15 109 relatively homogeneous brain region that has been extensively J15 110 characterized anatomically and electrophysiologically.

J15 111 Pharmacological and behavioral studies have indicated that J15 112 modulation of LC neuronal firing rates contributes to physical J15 113 aspects of opiate addiction, namely, physical dependence and J15 114 withdrawal, in several mammalian species, including primates (see J15 115 Redmond and Krystal, 1984; Rasmussen et al., 1990). The importance J15 116 of the LC in mediating opiate addiction is highlighted by a recent J15 117 study that examined the effects of local injection of an opiate J15 118 receptor antagonist into various brain regions of opiate-dependent J15 119 rats (Maldonado et al., 1992). The most severe opiate withdrawal J15 120 syndrome was produced by antagonist injections into the LC, which, J15 121 in fact, elicited a withdrawal syndrome even more severe than that J15 122 seen following intracerebroventricular administration.

J15 123 Acute opiate action in the LC

J15 124 The mechanism of acute opiate action in the LC, based on J15 125 electrophysiological and biochemical studies, is well established J15 126 and is shown schematically in Figure 1 (top). Acutely, opiates J15 127 decrease the firing rate of LC neurons via activation of an inward J15 128 rectifying K+ channel (Aghajanian and Wang, 1987; North J15 129 et al., 1987) and inhibition of a slowly depolarizing, nonspecific J15 130 cation channel (Aghajanian and Wang, 1987; M. Alreja and G. K. J15 131 Aghajanian, unpublished observations). Both actions are mediated J15 132 via pertussis toxin-sensitive G-proteins (i.e., Gi and/or J15 133 Go) (Aghajanian and Wang, 1986; North et al., 1987), and J15 134 inhibition of the nonspecific cation channel is mediated by reduced J15 135 neuronal levels of cAMP and activated cAMP-dependent protein kinase J15 136 (Aghajanian and Wang, 1987; Wang and Aghajanian, 1990; Alreja and J15 137 Aghajanian, 1991). Opiates acutely inhibit adenylate cyclase J15 138 activity in the LC (Duman et al., 1988; Beitner et al., 1989), as J15 139 is the case in many other brain regions (see Childers, 1991), and J15 140 inhibit cAMP-dependent protein phosphorylation (Guitart and J15 141 Nestler, 1989). Such regulation of protein phosphorylation J15 142 presumably mediates the effects of opiates on the nonspecific J15 143 cation channel through the phosphorylation of the channel itself or J15 144 some associated protein. Opiate regulation of protein J15 145 phosphorylation also probably mediates the effects of opiates on J15 146 many other aspects of LC neuronal function, including some of the J15 147 initial steps underlying longer-term changes associated with J15 148 addiction.

J15 149 Chronic opiate action in the LC

J15 150 Upon chronic opiate treatment, LC neurons develop tolerance to J15 151 the acute inhibitory actions of opiates, as neuronal firing rates J15 152 recover toward pretreatment levels (Aghajanian, 1978; Andrade et J15 153 al., 1983; Christie et al., 1987). The neurons also become J15 154 dependent on opiates after chronic exposure, in that abrupt J15 155 cessation of opiate treatment, for example, by administration of an J15 156 opiate receptor antagonist, leads to an elevation in LC firing J15 157 rates manyfold above pretreatment levels (Aghajanian, 1978; J15 158 Rasmussen et al., 1990).

J15 159 The tolerance and dependence exhibited by LC neurons during J15 160 chronic opiate exposure occur in the absence of detectable changes J15 161 in opiate receptors or opiate-regulated ion channels themselves J15 162 (see Christie et al., 1987; Loh and Smith, 1990). This raises the J15 163 possibility that intracellular messenger pathways may be involved. J15 164 Indeed, over the past several years, it has been demonstrated that J15 165 chronic administration of opiates leads to a dramatic upregulation J15 166 of the cAMP system at every major step between receptor and J15 167 physiological response (Fig. 1, bottom). Chronic opiate treatment J15 168 increases levels of Gi<*_>alpha<*/> and J15 169 Go<*_>alpha<*/> (the active subunits of the G-proteins J15 170 Gi and Go) (Nestler et al., 1989), adenylate J15 171 cyclase (Duman et al., 1988), cAMP-dependent protein kinase J15 172 (Nestler and Tallman, 1988), and a number of MARPPs (morphine- and J15 173 cAMP-regulated phosphoproteins) (Guitart and Nestler, 1989). Among J15 174 these MARPPs is tyrosine hydroxylase (TH) (Guitart et al., 1990), J15 175 the rate-limiting enzyme in the biosynthesis of catecholamines. J15 176 These various intracellular adaptations to chronic opiate treatment J15 177 are mediated via persistent activation of opiate receptors: the J15 178 adaptations are blocked by concomitant treatment of rats with J15 179 naltrexone, an opiate receptor antagonist, and are not produced by J15 180 a single morphine injection.

J15 181 Direct evidence for a functional role of an upregulated J15 182 cAMP system in opiate addiction in the LC

J15 183 The upregulated or 'hypertrophied' cAMP system in the LC can be J15 184 viewed as a compensatory, homeostatic response of LC neurons to the J15 185 inhibition devolving from chronic opiate treatment (Fig. 1). J15 186 According to this view, opiate upregulation of the cAMP system J15 187 increases the intrinsic excitability of LC neurons and thereby J15 188 accounts, at least in part, for opiate tolerance, dependence, and J15 189 withdrawal exhibited by these neurons (Nestler, 1990). In the J15 190 opiate-tolerant/dependent state, the combined presence of the J15 191 opiate and the upregulated cAMP system would return LC firing rates J15 192 toward pretreatment levels, whereas removal of the opiates would J15 193 leave the upregulated cAMP system unopposed, leading to withdrawal J15 194 activation of the neurons. This model, which is similar to one J15 195 proposed previously based on studies of cultured neuroblastoma x J15 196 glioma cells (Sharma et al., 1975; Collier, 1980), is supported by J15 197 several lines of evidence.

J15 198 First, cAMP and agents that elevate cAMP levels excite LC J15 199 neurons via the activation of cAMP-dependent protein kinase and J15 200 subsequent activation of the nonspecific cation channel (Wang and J15 201 Aghajanian, 1990). In fact, the spontaneous firing rate of LC J15 202 neurons requires an active cAMP system and the opening of the J15 203 nonspecific cation channel (Alreja and Aghajanian, 1991). Second, J15 204 the time course by which certain components of the upregulated cAMP J15 205 system revert to normal levels during naltrexone-precipitated J15 206 opiate withdrawal parallels the rapid, early phase of the time J15 207 course of recovery of LC neuronal firing rates and of various J15 208 behavioral signs during withdrawal (Rasmussen at al., 1990). Third, J15 209 upon bath application of naltrexone, LC neurons in brain slices J15 210 obtained from morphine-dependent animals exhibit J15 211 spontaneous firing rates more than twofold higher compared to LC J15 212 neurons in slices from normal animals (Fig. 2A) (Kogan et al., J15 213 1992). Earlier studies failed to detect such withdrawal activation J15 214 of morphine-dependent LC neurons in vitro, possibly due J15 215 to the poor condition of the brain slices used and the small number J15 216 and nonrandom samples of neurons examined (Andrade et al., 1983; J15 217 Christie et al., 1987). Since most major afferents to the LC are J15 218 severed in the brain slice preparation, the results establish that J15 219 an increased intrinsic excitability caused by chronic opiate J15 220 exposure contributes to opiate dependence in these cells. Fourth, J15 221 LC neurons from morphine-dependent animals show a greater maximal J15 222 responsiveness to cAMP analogs in vitro (Fig. 2B) J15 223 (Kogan et al., 1992). Taken together, these results provide strong J15 224 evidence to support the view that the opiate-induced upregulation J15 225 of the cAMP system represents one mechanism by which opiates J15 226 produce addictive changes in LC neurons.

J15 227 Molecular mechanisms underlying opiate upregulation of the J15 228 cAMP system in the LC

J15 229 One of the central questions raised by these studies concerns J15 230 the molecular mechanisms by which chronic opiate administration J15 231 leads to upregulation of the cAMP system in LC neurons. J15 232 J16 1 <#FROWN:J16\>Spliced RNA of Woodchuck Hepatitis Virus

J16 2 C. WALTER OGSTON AND DOLORES G. RAZMAN

J16 3 Polymerase chain reaction was used to investigate RNA J16 4 splicing in liver of woodchucks infected with woodchuck hepatitis J16 5 virus (WHV). Two spliced species were detected, and the splice J16 6 junctions were sequenced. The larger spliced RNA has an intron of J16 7 1300 nucleotides, and the smaller spliced sequence shows an J16 8 additional downstream intron of 1104 nucleotides. We did not detect J16 9 singly spliced sequences from which the smaller intron alone was J16 10 removed. Control experiments showed that spliced sequences are J16 11 present in both RNA and DNA in infected liver, showing that the J16 12 viral reverse transcriptase can use spliced RNA as template. J16 13 Spliced sequences were detected also in virion DNA prepared from J16 14 serum. The upstream intron produces a reading frame that fuses the J16 15 core to the polymerase polypeptide, while the downstream intron J16 16 causes an inframe deletion in the polymerase open reading frame. J16 17 Whereas the splicing patterns in WHV are superficially similar to J16 18 those reported recently in hepatitis B virus, we detected no J16 19 obvious homology in the coding capacity of spliced RNAs from these J16 20 two viruses.

J16 21 INTRODUCTION J16 22 Hepatitis B virus (HBV), the prototype of the hepadnavirus J16 23 family, causes serious human disease and has a complex and J16 24 interesting mode of replication. Hepadnaviruses make efficient use J16 25 of a very small genome by use of overlapping reading frames, each J16 26 of which may produce several alternative peptides (Ganem and J16 27 Varmus, 1987), and their genome replication cycle involves reverse J16 28 transcription of an RNA template (Summers and Mason, 1982. We would J16 29 like to know how the peculiar features of hepadnavirus replication J16 30 relate to their natural history and pathogenesis.

J16 31 The mammalian hepadnaviruses woodchuck hepatitis virus (WHV) J16 32 and ground squirrel hepatitis virus (GSHV), which are genetically J16 33 close to HBV and similar in their pathogenesis, make the best J16 34 experimental models for our purposes. We have chosen WHV as the J16 35 object of our studies because infected woodchucks suffer from J16 36 chronic hepatitis, and the majority of persistently infected J16 37 woodchucks get hepatocellular carcinoma (HCC) that is J16 38 histologically similar to liver cancer caused by HBV in humans J16 39 (Popper et al., 1987; Korba et al., 1989).

J16 40 The manner in which hepadnavirus messenger RNA codes for J16 41 proteins remains a subject for research. Four genes have been J16 42 identified in the DNA sequence of the mammalian hepadnaviruses as J16 43 open reading frames (ORFs) diagrammed in Fig. 1. The core (C) ORF J16 44 gives rise to the major nucleocapsid polypeptide as well as the J16 45 secreted e antigen, and the surface (S) ORF gives rise to a total J16 46 of six known glycoprotein species that are incorporated in the J16 47 envelope of the virion and in related 'empty' 22-nm particles. The J16 48 large polymerase (P) ORF is inferred from amino acid sequence J16 49 homology to encode reverse transcriptase. The function of the X ORF J16 50 in virus replication is not known, although it has been shown to be J16 51 a transcriptional trans-activator in transfected J16 52 cells.

J16 53 The pattern of RNA transcription of the hepadnavirus genome has J16 54 been reviewed by Ganem and Varmus (1987). Two abundant transcripts J16 55 of 3.5 and 2.1 kb, respectively, can account for expression of the J16 56 abundantly expressed C and S ORFs. A minor transcript of 2.4 kb J16 57 encoding the pre-S1 ORF has been described in HBV-infected J16 58 chimpanzee liver (Cattaneo et al., 1984; Will et J16 59 al., 1987), but no corresponding RNA has been detected in WHV J16 60 or GSHV (Ganem and Varmus, 1987).

J16 61 The expression of the ORFs P and X remains to be accounted for, J16 62 since no transcripts have been detected in vivo that J16 63 include these ORFs at their 5' end. To solve this difficulty three J16 64 hypotheses may be considered. First, these ORFs may be expressed J16 65 from abundant mRNA by use of downstream AUG start codons (Kozak, J16 66 1986); second, there may be rare mRNAs that have 5' ends J16 67 immediately upstream of these ORFs; and third, there may be rare J16 68 spliced transcripts that express these ORFs.

J16 69 The third hypothesis above is supported by the recent reports J16 70 of spliced mRNAs in HBV-infected tissues and cells transfected with J16 71 the HBV genome (Chen et al., 1989; Su et al., J16 72 1989; Suzuki et al., 1989, 1990; Wu et al., J16 73 1991). These workers have described two introns that encompass the J16 74 region of overlap between the C and P genes of HBV. We have used J16 75 polymerase chain reaction (PCR) (Kawasaki, 1990) to look for J16 76 analogous splicing in WHV, and we now describe two species of J16 77 spliced mRNA in WHV-infected liver, both of which could encode J16 78 core-polymerase fusion proteins.

J16 79 MATERIALS AND METHODS

J16 80 The animals used in this study were naturally infected J16 81 woodchucks (Ogston et al., 1989) and woodchucks infected J16 82 as neonates (RW101, RW103, RW115, and RW116; C. W. Ogston and B. C. J16 83 Tennant, unpublished). Virus from the serum of naturally infected J16 84 woodchuck RW003 was used to initiate the infections of RW101 and J16 85 RW103, and RW013 was the source of the inoculum for RW115 and J16 86 RW116.

J16 87 Liver tissue obtained at autopsy was cooled on ice, cut into J16 88 slices 2-5-mm thick, and frozen in liquid nitrogen. RNA was J16 89 prepared by homogenization in guanidine thiocyanate followed by J16 90 cesium chloride centrifugation (Maniatis et al., 1982). J16 91 Polyadenylated RNA was purified by oligo(dT) cellulose J16 92 chromatography (Maniatis et al., 1982). The yield of RNA J16 93 was measured by absorbance at 260 nm.

J16 94 Complementary DNA was synthesized in a reaction containing 5 J16 95 <*_>mu<*/>g of RNA, 100 mM Tris, pH 8.3, 7 mM J16 96 MgCl2, 40 mM KCl, 1 mM dithiothreitol, and 0.5 J16 97 <*_>mu<*/>g dT12-18 primers (Collaborative Research), J16 98 with 8 units of AMV reverse transcriptase (U.S. Biochemical). The J16 99 reaction was incubated 2 hr at 42<*_>degree<*/>, then stopped by J16 100 heating to 100<*_>degree<*/> for 3 min. The DNA was precipitated J16 101 with ethanol, dried, and dissolved in 30 <*_>mu<*/>l of water.

J16 102 The standard PCR reaction contained 100 mM Tris; pH 8.3 J16 103 (at 25<*_>degree<*/>); 0.2 <*_>mu<*/>M each of dATP, dCTP, J16 104 dGTP, and TTP (Pharmacia); primers at 0.08 <*_>mu<*/>M; and J16 105 0.6 units of Taq DAN polymerase (Perkin-Elmer Cetus). Amplification J16 106 was carried out using a Perkin-Elmer Cetus DNA thermal cycler for J16 107 30 cycles with denaturation at 94<*_>degree<*/> for 60 sec, J16 108 annealing at 50<*_>degree<*/> for 2 min, and extension at J16 109 72<*_>degree<*/> for 30 or 60 sec. Longer extension times were used J16 110 in some reactions that were expected to produce amplification J16 111 products greater than 1000 bp.

J16 112 To generate single-stranded amplified DNA for sequencing, the J16 113 amplified DNA fragment of interest was cut from an agarose gel and J16 114 electroeluted. Asymmetric amplification was carried out using the J16 115 same pair of primers as in the original amplification, but the J16 116 concentration of one primer was reduced 32-fold (McCabe, 1990). The J16 117 product of asymmetric PCR was purified by two rounds of centrifugal J16 118 dialysis (Centricon 100, Amicon) to remove the primers and other J16 119 components from the PCR reaction, then concentrated by centrifugal J16 120 evaporation to a volume of about 7 <*_>mu<*/>l. The asymmetric DNA J16 121 was sequenced directly using the Sequenase system (U.S. J16 122 Biochemical) according to the manufacturer's directions. A 'nested' J16 123 primer internal to the original PCR fragment was used for J16 124 sequencing whenever possible.

J16 125 figure&caption

J16 126 The sequences were analyzed using the Beckman Microgenie J16 127 software package. Sequence read from the gel was first compared to J16 128 the published sequence of WHV (Galibert et al., 1982) by J16 129 dot-matrix comparison. Areas of homology so identified were J16 130 confirmed and then aligned.

J16 131 In referring to primer sequences and experimental results, WHV J16 132 map coordinates are given in nucleotides originating at the J16 133 EcoRI site, as defined by the data of Galibert et J16 134 al. (1982). The total length of this sequence is 3308 nt.

J16 135 RESULTS J16 136 PCR can be used to detect spliced sequences in cDNA by J16 137 employing primers located in different exons spanning one or more J16 138 introns. Unspliced template then gives rise to an amplification J16 139 product whose length corresopnds to the distance between the J16 140 primers on the genomic DNA, whereas amplification of spliced J16 141 template results in a shorter PCR product. Since short templates J16 142 are amplified more efficiently than long ones this method is very J16 143 sensitive even in the presence of excess unspliced template. We J16 144 chose pairs of primers, shown in Fig. 1b, that we expected to span J16 145 intron(s) analogous to those previously described in HBV.

J16 146 figure&caption

J16 147 When we amplified cDNA from WHV-infected liver using the J16 148 primers 2110(+) and 575(-), we obtained a product of approximately J16 149 560 bp, as shown in Fig. 2a, as well as the full-length product of J16 150 1773 bp. This indicates that a segment of about 1200 nucleotides J16 151 has been removed from the template. Likewise, the primers 2110(+) J16 152 and 1478(-) gave a short product of 1260 bp as well as the J16 153 full-length product of 2676 bp, indicating removal of about 1400 J16 154 nucleotides. These two results presumably imply the existence of a J16 155 single intron, since the inferred sizes of the missing segment J16 156 agree within the limits of experimental error.

J16 157 In the reaction using the primers 2110(+) and 1478(-), an J16 158 additional product of 320 bp was amplified, indicating a template J16 159 from which a total of 2300 nucleotides are missing from between the J16 160 priming sites (Fig. 2a). This will be discussed further below.

J16 161 To check the specificity of the amplification we carried out J16 162 PCR using lower concentrations of nucleotide triphoshates. As shown J16 163 in Figure 2a the short fragments described above were all amplified J16 164 when the triphosphate concentration was reduced, while the J16 165 full-length segments became less intense as expected. As a further J16 166 control for specificity we amplified WHV DNA from the clone J16 167 pWHV81-2 (Seeger et al., 1987) with both the above primer J16 168 sets. No short products were obtained from this reaction, although J16 169 the full-length sequence was amplified (data not shown).

J16 170 To determine whether these observations are general among J16 171 separate strains of WHV, we amplified cDNA from a number of J16 172 experimentally and naturally infected woodchucks (Figure 2b and J16 173 data not shown). The PCR products were the same size in all cases, J16 174 although the relative intensity of the fragments varied. We think J16 175 that this variation is due to complex interactions in the J16 176 competition between more abundant unspliced template on the one J16 177 hand, and rarer but shorter spliced template on the other hand. J16 178 This competition will also be influenced by such factors as the J16 179 degree of RNA degradation and the efficiency of reverse J16 180 transcription.

J16 181 To determine whether these results reflect authentic RNA J16 182 splicing, the short PCR products were sequenced. We found in J16 183 preliminary experiments that the best sequence data were obtained J16 184 when a different 'nested' primer was used for sequencing rather J16 185 than that used for amplification. This method also ensures that J16 186 products derived from mispriming will not contribute to the J16 187 sequence. Accordingly the primers 1961(+) and 314(-) were used to J16 188 amplify a fragment of about 400 bp derived from the first spliced J16 189 template described above, and this 400-bp fragment was purified by J16 190 preparative electrophoresis. Single-stranded DNA was then made by J16 191 asymmetric reamplification, using the primer 314(-) in excess over J16 192 1961(+) (McCabe, 1990), and the single-stranded DNA was sequenced J16 193 using the 2110(+) primers, with the result shown in Fig. 3a.

J16 194 The sequence obtained in this experiment was aligned to the J16 195 published sequence of WHV (Galibert et al., 1982) as J16 196 shown in Fig. 3b. This alignment is consistent with an intron of J16 197 1300 nucleotides between the donor site GT at bases 2151-2152 to J16 198 the acceptor AG at 141-142. To confirm the sequence of this splice J16 199 junction we generated positive-strand DNA template by amplification J16 200 with 2110(+) in excess over 314(-) and sequenced the minus strand J16 201 from the primer 314(-). The resulting complementary sequence showed J16 202 the identical splice junction (data not shown).

J16 203 figure&caption

J16 204 The smaller (320 bp) product of amplification using the primers J16 205 2110(+) and 1478(-) (Fig. 2) could result from a doubly spliced J16 206 mRNA with the 1300 nt intron and one of 1000 nt, or alternatively J16 207 from a singly spliced mRNA with an intron of about 2300 nt. We J16 208 first looked for the hypothetical second splice junction by J16 209 amplifying cDNA with the primers 176(+) and 1478(-), but detected J16 210 only the 1300-bp product derived from the unspliced sequence (data J16 211 not shown). We thought that abundant cDNA derived from the 2.1-kb J16 212 S-gene transcript might be overwhelming a rarer spliced template J16 213 sequence, so we set out to separate the putative spliced sequence J16 214 from unspliced S transcript by carrying out PCR in two steps. In J16 215 the first step we used the primers 1961(+) and 1478(-) to J16 216 selectively amplify spliced DNA, avoiding the 2.1-kb S transcript J16 217 whose 5' end is at nt 145 (Moroy et al., 1985). J16 218 J17 1 <#FROWN:J17\>CONFUSION IN THE DETERMINATION OF DEATH: J17 2 DISTINGUISHING PHILOSOPHY FROM PHYSIOLOGY

J17 3 JEFFREY R. BOTKIN and STEPHEN G. POST

J17 4 Two decades have now passed since the Report of the Ad Hoc J17 5 Committee of the Harvard Medical School to Examine the Definition J17 6 of Death. The adoption of whole brain criteria as a new standard of J17 7 death proceeded quickly, supported by the American Medical J17 8 Association, the American Bar Association, and the President's J17 9 Commission, and has been adopted into legislation in 49 states. One J17 10 of the final tasks toward the uniform adoption of the whole brain J17 11 standard is the establishment of criteria for death in young J17 12 infants. But despite this definitional consensus and its rapid J17 13 adoption, beliefs about what it means to be dead may vary J17 14 considerably in our highly pluralistic society.

J17 15 Within the medical community alone there remains a startling J17 16 amount of confusion about the determination of death. The J17 17 redefinition of death has given rise to a distinct tension between J17 18 the new definition of death based on brain criteria and common J17 19 perceptions of the nature of death. Physicians, nurses, and others J17 20 at the bedside may 'know' that a person is dead by established J17 21 criteria, yet life still may be perceived as long as the heart J17 22 continues to beat and the skin remains warm to touch. Many J17 23 clinicians can rattle off criteria for obscure conditions or long J17 24 differentials for anomalous lab results, yet they cannot offer J17 25 specific criteria for death. More importantly, it is obvious that J17 26 many health care professionals, including physicians, simply do not J17 27 believe patients are dead when their brains alone have ceased to J17 28 function.

J17 29 Such confusion in the medical professions was clearly J17 30 documented by Youngner, et al. This study documented that only 35 J17 31 percent of physicians or nurses who dealt routinely with critically J17 32 ill patients could correctly identify the legal and medical J17 33 criteria for determining death. Further, 58 percent of the J17 34 respondents did not apply a consistent concept of death when asked J17 35 to justify their determination of death in hypothetical cases. The J17 36 wider public is confused as well - a confusion no doubt fostered by J17 37 news reports of 'brain dead' patients who are kept 'alive' by J17 38 'life-support' technology.

J17 39 We commonly perceive the presence of death when the body is J17 40 cold and pale, when breathing and heart cease. Little confusion J17 41 about the presence of death arises in such circumstances. But J17 42 contemporary technology has fostered confusion by forcing us to J17 43 recognize the ambiguous nature of the moment of death. We can J17 44 now restart functions such as breathing or heartbeat when once they J17 45 would have been irreversibly lost. Certainly someone is not really J17 46 dead with the cessation of heartbeat if the person can be revived. J17 47 Yet despite common language about being 'brought back to life,' we J17 48 like to think of death as a final and irreversible state, so we J17 49 must now speak in terms of an 'irreversible cessation' of J17 50 functions. However, what constitutes 'irreversible' will no doubt J17 51 change with technology. This influence of technology on the moment J17 52 of death indicates that the determination of this moment cannot be J17 53 made independently of the cultural environment in which death J17 54 occurs. Additionally, the contemporary ability to entirely replace J17 55 cardio-respiratory functions with machines further blurs the J17 56 boundaries between life and death. It was, of course, this latter J17 57 technological development that was the primary force behind the J17 58 adoption of the whole brain standard of death.

J17 59 The basic argument we wish to make here is that the moment of J17 60 death is not a specific physiologic event amenable to scientific J17 61 determination. Rather, it is a moment defined by philosophic J17 62 concepts - concepts that speak to what it means to be alive. Since J17 63 such philosophic contentions defy objective proof, the moment of J17 64 death must be seen as an event fixed by social consensus. Two J17 65 implications of this realization will be discussed: first, that J17 66 education for physicians and other medical professionals must J17 67 address philosophic concepts if confusion about death is to be J17 68 reduced; and second, that our pluralistic society must address the J17 69 demand for alternative definitions of death for those who reject J17 70 the philosophic foundations of the prevailing standard.

J17 71 First we wish to look critically at the three competing J17 72 standards of death. By 'standard' we mean a set of criteria for J17 73 determining death that is based on an underlying philosophic J17 74 concept. The three standards of death are the traditional J17 75 heart/lung standard, the whole brain standard, and the higher brain J17 76 standard. Current law recognizes the combination of the heart/lung J17 77 or whole brain standards, but for the purposes of clarity, these J17 78 will be discussed separately. In this discussion we will focus on J17 79 our common perceptions of death and how these correlate with the J17 80 philosophic concepts of death that underlie the three standards.

J17 81 Advocates of the higher brain death standard argue that death J17 82 occurs when cognitive function irreversibly ceases. Cognition is J17 83 considered to be the essential element in personal identity, and J17 84 thus the loss of cognition heralds the death of the individual. J17 85 Under this standard, death would be an event occurring when the J17 86 relevant brain activities cease. The philosophic appeal of this J17 87 approach is that it accurately defines what most of us feel is of J17 88 primary importance in human life - when all cognitive capacities J17 89 are lost, we 'might as well be dead' figuratively, and, it is J17 90 argued, literally. In addition, cognition is embodied in the J17 91 cerebral hemispheres and so is potentially measurable by technical J17 92 means (although it is not so at the present time).

J17 93 From the perspective of many observers, the problem with the J17 94 higher brain standard is the perception of continued 'life' despite J17 95 the presence of death so defined. Individuals who have permanently J17 96 lost all higher brain functions but retain brain stem functions J17 97 will still breathe spontaneously, remain warm, and may have J17 98 sleep/wake cycles and spontaneous movement. These patients simply J17 99 do not look dead by our more intuitive understanding of the term. J17 100 The conflict and confusion that arises here is due to a distinct J17 101 and important difference between the death of the individual J17 102 and the death of the organism. When cognition is irreversibly J17 103 lost, but other physiologic processes continue, the human J17 104 individual may be dead, but the human organism is not.

J17 105 Which is the correct focus for our determination of death - J17 106 death of the human individual or death of the human organism? There J17 107 are reasoned arguments for each. One distinct advantage of J17 108 determining death in terms of the organism is that such a J17 109 definition could be applied across the spectrum of life. Since J17 110 death is a term that is relevant to all life forms, it must J17 111 therefore have some consistent biologic meaning. When we speak of J17 112 death in other forms of life, we refer to the death of the organism J17 113 - 'personhood' or cognition do not have any logical correlates in J17 114 the majority of life forms. We may choose to define death in humans J17 115 in terms of the loss of a single, key function (like cognition), J17 116 but this will leave us with the task of defining such key functions J17 117 for other life forms as well. Is there a single key function in J17 118 fungi, barnacles, or bacteria on which our recognition of life J17 119 hinges?

J17 120 Even in the simplest of life forms, the definition of life has J17 121 proven remarkably difficult and has been the subject of much J17 122 debate. Orstan has provided a brief review of the definitions of J17 123 life offered in Western thought and concludes: "It does not J17 124 take much effort to realize that these definitions are little more J17 125 than arbitrary lists of things the living organisms do". J17 126 Thus, we would suggest that the common perception of death in an J17 127 organism requires the cessation of a host of functions that J17 128 comprise the perception of life. It is quite unlikely that any J17 129 nonhuman life forms would be considered dead as long as there was J17 130 spontaneous movement, respiration, and the flow of vital fluids.

J17 131 So while the higher brain standard has the intellectual appeal J17 132 of specifying the loss of a key function as the moment of J17 133 transition between life and death, it will remain unconvincing for J17 134 those whose intuitive or intellectual beliefs require that death J17 135 signify the death of the human organism. Thus the higher brain J17 136 standard is not 'wrong,' but if the popular perception of death is J17 137 grounded in a concept of a living organism, the appeal of the J17 138 higher brain standard may be limited as a basis for public J17 139 policy.

J17 140 J17 141 The Whole Brain Standard

J17 142 The current standard for the determination of death reads as J17 143 follows:

J17 144 An individual who has sustained either (1) irreversible J17 145 cessation of circulatory and respiratory functions, or (2) J17 146 irreversible cessation of all functions of the entire brain, J17 147 including the brain stem, is dead.

J17 148 The second half of the current standard represents the whole J17 149 brain standard. The key concept for the whole brain standard is the J17 150 loss of integration of the organism. Clearly some cellular and J17 151 organ function continues for a period of time after brain or heart J17 152 and lung functions cease; but the argument is that we do not care J17 153 about isolated organ or cell function, we care about the integrated J17 154 function of the organism. When the organism as a coordinated whole J17 155 has irreversibly ceased to function, the argument goes, the J17 156 organism is dead.

J17 157 The brain is the logical locus for this integrative activity. J17 158 Loss of whole brain activity leads to the loss of consciousness, J17 159 respiratory arrest, no spontaneous movement, and no interaction J17 160 with the environment. Without 'life support,' cardiac arrest soon J17 161 develops and loss of all organ and cellular functions follow. Under J17 162 this standard, the event of death is determined as the time at J17 163 which there is irreversible loss of whole brain function.

J17 164 The whole brain standard has several advantages; it can be J17 165 measured with reliability, and irreversibility can be assured J17 166 within reasonable limits. In addition, loss of integration is J17 167 potentially applicable to other life forms (although it may be less J17 168 clear how integration is determined in organisms without central J17 169 nervous systems). Finally, loss of whole brain function produces an J17 170 individual with many of the qualities we usually associate with J17 171 death: lack of spontaneous breathing, movement, and response to the J17 172 environment.

J17 173 A principal difficulty with this standard is that individuals J17 174 who lack whole brain function while on life support retain J17 175 significant attributes of life: their skin may be warm and soft, J17 176 they have a heartbeat, food continues to be metabolized, and waste J17 177 products are excreted. Individuals without whole brain function may J17 178 even gestate fetuses and bear children. It is precisely these J17 179 attributes that lead many in the public and the medical profession J17 180 to believe that such individuals are not 'really dead.'

J17 181 The counterargument is that these life-like attributes of whole J17 182 brain-dead patients do not count, since they do not reflect J17 183 any integrated function in the organism, but exist only by J17 184 technical intervention. This counterargument only makes sense if we J17 185 can define what we mean by integration. According to Webster, J17 186 integrate means: (1) "to make or become whole or J17 187 complete," (2) "to bring (parts) together into a J17 188 whole." Clearly an individual without brain function is not J17 189 integrated; however, an individual without function in any J17 190 major organ is not integrated. Barney Clarke was not intrinsically J17 191 integrated once the artificial heart was implanted, yet he was J17 192 obviously alive by virtue of technical support.

J17 193 Youngner and Bartlett make this argument with a hypothetical J17 194 case of a man who has had brain stem function surgically destroyed J17 195 by a diabolical physician. The patient is then hooked to technical J17 196 support that maintains bodily functions as well as cerebral J17 197 functions: In essence, this hypothetical patient has been afforded J17 198 a mechanical brain stem. Youngner and Bartlett note that the J17 199 patient has "entirely lost the innate and spontaneous J17 200 ability to integrate essential body systems," yet is J17 201 obviously alive, at least as long as consciousness remains. It is J17 202 irrelevant to our perception of life that brain stem functions are J17 203 performed mechanically.

J17 204 When an individual has lost all brain function yet is J17 205 maintained on a ventilator, the heart, bowels, kidneys, and other J17 206 organ systems may continue to function normally, at least J17 207 temporarily, and these subsystems work in coordination. The J17 208 individual without brain function is not wholly integrated, but J17 209 neither is he or she wholly disintegrated. J17 210 J18 1 <#FROWN:J18\>We have thus shown that the Hilbert transform J18 2 determines a bounded operator from Lp into itself, J18 3 for p of the form 2n. By the Riesz Interpolation Theorem, J18 4 it follows then that the Hilbert transform determines a bounded J18 5 operator from Lp into itself, for formula. J18 6 The proof can now be completed by appealing to Exercise 6.21 for J18 7 the cases 1 J18 8 EXERCISE 6.24. (a) Show that any constant formula J18 9 will satisfy inequality (6.4). Supposing that f is supported J18 10 in the interval [-a,a], show that any constant formula will J18 11 satisfy inequality (6.5) if formula.

J18 12 (b) Establish Equation (6.6) by integrating around a large J18 13 square contour.

J18 14 (c) Let m be the characteristic function of an open J18 15 interval (a,b), where formula. Prove that the multiplier J18 16 M, corresponding to m determines a bounded operator on J18 17 every Lp for 1< p < <*_>infinity<*/>. HINT: J18 18 Write m as a finite linear combination of translates of - J18 19 isgn.

J18 20 (d) Let m be the characteristic function of the set J18 21 formula. Verify that the multiplier M corresponding to J18 22 m has no bounded extension to any Lp space for J18 23 p <*_>unch<*/> 2.

J18 24 CHAPTER VII

J18 25 AXIOMS FOR A MATHEMATICAL MODEL OF EXPERIMENTAL SCIENCE

J18 26 This chapter is a diversion from the main subject of this book, J18 27 and it can be skipped without affecting the material that follows. J18 28 However, we believe that the naive approach taken in this chapter J18 29 toward the axiomatizing of experimental science serves as a good J18 30 motivation for the mathematical theory developed in the following J18 31 four chapters.

J18 32 We describe here a set of axioms, first introduced by G.W. J18 33 Mackey, to model experimental investigation of a system in nature. J18 34 We suppose that we are studying a phenomenon in terms of various J18 35 observations of it that we might make. We postulate that there J18 36 exists a nonempty set S of what we shall call the possible J18 37 states of the system, and we postulate that there is a J18 38 nonempty set O of what we shall call the possible J18 39 observables of the system. We give two examples.

J18 40 (1) Suppose we are investigating a system that consists of a J18 41 single physical particle in motion on an infinite straight line. J18 42 Newtonian mechanics (f = ma) tells us that the system is J18 43 completely determined for all future time by the current position J18 44 and velocity, i.e., by two real numbers. Hence, the states of this J18 45 system might well be identified with points in the plane. Two of J18 46 the (many) possible observables of this system can be described as J18 47 position and velocity observables. We imagine that there is a J18 48 device which indicates where the particle is and another device J18 49 that indicates its velocity. More realistically, we might have many J18 50 yes/no devices that answer the observational questions: 'Is the J18 51 particle between a and b?' 'Is the velocity of the J18 52 particle between c and d?'

J18 53 Quantum mechanical models of this single particle are different J18 54 from the Newtonian one. They begin by assuming that the (pure) J18 55 states of this one-particle system are identifiable with certain J18 56 square-integrable functions and the observables are identified with J18 57 certain linear transformations. This model seems quite mysterious J18 58 to most mathematicians, and Mackey's axioms form one attempt at J18 59 justifying it.

J18 60 (2) Next, let us imagine that we are investigating a system in J18 61 which three electrical circuits are in a black box and are open or J18 62 closed according to some process of which we are not certain. The J18 63 states of this system might well be described as all triples of 0's J18 64 and 1's (0 for open and 1 for closed). Suppose that we have only J18 65 the following four devices for observing this system. First, we can J18 66 press a button b0 and determine how many of the J18 67 three circuits are closed. However, when we press this button, it J18 68 has the effect of opening all three circuits, so that we have no J18 69 hope of learning exactly which of the three were closed. (Making J18 70 the observation actually affects the system.) In addition, we have J18 71 three other buttons b1, b2, J18 72 b3, bi, telling whether circuit J18 73 i is open or closed. Again, when we press button J18 74 bi, all three circuits are opened, so that we have J18 75 no way of determining if any of the ciricuits other than the J18 76 ith was closed. This is a simple example in which certain J18 77 simultaneous observations appear to be impossible, e.g., J18 78 determining whether circuits 1 and 2 are both closed.

J18 79 The axioms we introduce are concerned with the concept of J18 80 interpreting what it means to make a certain observation of the J18 81 system when the system is in a given state. The result of such an J18 82 observation should be a real number, with some probability, J18 83 depending on the state and on the observable.

J18 84 AXIOM 1. To each state <*_>alpha<*/> <*_>EPSILON<*/> J18 85 S and observable A <*_>EPSILON<*/> O there J18 86 corresponds a Borel probability measure J18 87 <*_>mu<*/><*_>alpha<*/>, A on R.

J18 88 REMARK. The probability J18 89 measure<*_>mu<*/><*_>alpha<*/>, A contains the J18 90 information about the probability that the observation A will J18 91 result in a certain value, when the system is in the state J18 92 <*_>alpha<*/>.

J18 93 EXERCISE 7.1. Write out in words, from probability J18 94 theory, what the following symbols mean.

J18 95 (a) <*_>mu<*/><*_>alpha<*/>,A([3,5]) = 0.9.

J18 96 (b) <*_>mu<*/><*_>alpha<*/>,A({0}) = 1.

J18 97 AXIOM 2. (a) If A, B are observables for J18 98 which formula for every state <*_>alpha<*/> <*_>EPSILON<*/> J18 99 S, then A = B.

J18 100 (b) If <*_>alpha<*/>, <*_>beta<*/> are states for which J18 101 formula for every observable A <*_>EPSILON<*/> O, J18 102 then <*_>alpha<*/> = <*_>beta<*/>.

J18 103 EXERCISE 7.2. Discuss the intuitive legitimacy of J18 104 Axiom 2.

J18 105 AXIOM 3. If <*_>alpha<*/>1, ... J18 106 <*_>alpha<*/>n are states, and t1, ... J18 107 tn are nonnegative real numbers for which J18 108 formula, then there exists a state <*_>alpha<*/> for J18 109 which

J18 110 formula

J18 111 for every observable A. This axiom can be interpreted as J18 112 asserting that the set S of states is closed under convex J18 113 combinations. If the <*_>alpha<*/>i's are not all J18 114 identical, we call this state <*_>alpha<*/> a mixed state J18 115 and we write formula.

J18 116 We say that a state <*_>alpha<*/> <*_>EPSILON<*/> S is a J18 117 pure state if it is not a mixture of other states. That J18 118 is, if formula, with each ti > 0 and J18 119 formula, then <*_>alpha<*/>i = <*_>alpha<*/> for J18 120 all i.

J18 121 EXERCISE 7.3. Discuss the intuitive legitimacy of J18 122 Axiom 3. Think of a physical system, like a beaker of water, for J18 123 which there are what we can interpret as pure states and mixed J18 124 states.

J18 125 AXIOM 4. If A is an observable, and J18 126 formula is a Borel function, then there exists an J18 127 observable B such that

J18 128 formula

J18 129 for every state <*_>alpha<*/> and every Borel set J18 130 formula. We denote this observable B by f(A).

J18 131 EXERCISE 7.4. Discuss the intuitive legitimacy of J18 132 Axiom 4. Show that, when the system is in the state <*_>alpha<*/> J18 133 and the observable A results in a value t with J18 134 probability p, the observable B = f(A) results in J18 135 the value f(t) with the same probability p.

J18 136 EXERCISE 7.5. (a) Prove that there exists an J18 137 observable A such that formula for every state J18 138 <*_>alpha<*/>. That is, A is an observable that is nonnegative J18 139 with probability 1 independent of the state of the system. HINT: J18 140 Use formula for example.

J18 141 (b) Given a real number t, show that there exists an J18 142 observable A such that formula for every state J18 143 <*_>alpha<*/>. That is, A is an observable that equals t J18 144 with probability 1, independent of the state of the system.

J18 145 (c) Show that the set of observables is closed under scalar J18 146 multiplication. That is, if A is an observable and c is a J18 147 nonzero real number, then there exists an observable B such J18 148 that

J18 149 formula

J18 150 We may then write B = cA.

J18 151 (d) If A and B are observables, does there have to be J18 152 an observable C that we could think of as the sum A + J18 153 B?

J18 154 (e) In what way must we alter the descriptions of the systems J18 155 in Example 1 and Example 2 in order to incorporate these first four J18 156 axioms (particularly Axioms 3 and 4)?

J18 157 DEFINITION. We say that two observables A and B J18 158 are compatible, pairwise compatible, or J18 159 simultaneously observable if there exists an observable J18 160 C and Borel functions f and g such that A = J18 161 f(C) and B = g(C). A sequence {Ai} is J18 162 called mutually compatible if there exists an observable J18 163 C and Borel functions {fi} such that J18 164 formula for all i.

J18 165 EXERCISE 7.6. Is there a difference between a J18 166 sequence {Ai} of observables being pairwise J18 167 compatible and being mutually compatible? In particular, is it J18 168 possible that there could exist observables A, B, C, J18 169 such that A and B are compatible, B and C are J18 170 compatible, A and C are compatible, and yet A, J18 171 B, C are not mutually compatible? HINT: Try to modify J18 172 Example 2.

J18 173 EXERCISE 7.7. (a) If A, B are observables, J18 174 what should it mean to say that an observable C is the sum J18 175 A + B of A and B? Discuss why we do not J18 176 hypothesize that there always exists such an observable C.

J18 177 (b) If A and B are compatible, can we prove that J18 178 there exists an observable C that can be regarded as A + J18 179 B?

J18 180 DEFINITION. An observable q is called a question J18 181 or a yes/no observable if, for each state <*_>alpha<*/>, J18 182 the measure <*_>mu<*/><*_>alpha<*/>,q is supported on the J18 183 two numbers 0 and 1. We say that the result of observing q, J18 184 when the system is in the state <*_>alpha<*/>, is 'yes' with J18 185 probability <*_>mu<*/><*_>alpha<*/>,q({1}), and it is J18 186 'no' with probability <*_>mu<*/><*_>alpha<*/>,q({0}).

J18 187 THEOREM 7.1. Let A be an observable.

J18 188 (1) For each Borel subset E in R, the observable J18 189 <*_>chi<*/>E(A) is a question.

J18 190 (2) If g is a real-valued Borel function on R, for J18 191 which g(A) is a question, then there exists a Borel set E such that J18 192 g(A) = <*_>chi<*/>E(A).

J18 193 (Note that condition 2 does not assert that g necessarily J18 194 equals <*_>chi<*/>E.)

J18 195 PROOF. For each Borel set E, we have

J18 196 formula,

J18 197 and

J18 198 formula,

J18 199 which proves that formula is supported on the two J18 200 points 0 and 1 for every <*_>alpha<*/>, whence J18 201 <*_>chi<*/>E(A) is a question and so part 1 is J18 202 proved.

J18 203 Given a g for which q = g(A) is a question, set J18 204 E = g-1({1}), and observe that for any J18 205 <*_>alpha<*/> <*_>EPSILON<*/> S we have

J18 206 formula

J18 207 Since both q and <*_>chi<*/>E(A) are questions, J18 208 it follows from the preceding paragraph that

J18 209 formula

J18 210 showing that

J18 211 formula

J18 212 for every state <*_>alpha<*/>. Then, by Axiom 2, we have J18 213 that

J18 214 formula

J18 215 We now define some mathematical structure on the set Q of J18 216 all questions. This set will form the fundamental ingredient of our J18 217 model.

J18 218 DEFINITION. Let Q denote the set of all questions. J18 219 For each question q <*_>EPSILON<*/> Q, define a J18 220 real-valued function mq on the set S of states J18 221 by

J18 222 formula

J18 223 If p and q are questions, we say that formula J18 224 if formula for all <*_>alpha<*/> <*_>EPSILON<*/> S.

J18 225 If p, q and r are questions, for which J18 226 formula, we say that p and q are summable J18 227 and that r is the sum of p and q. We then write J18 228 r = p + q. More generally, if {qi} is a J18 229 countable (finite or infinite) set of questions, we say the J18 230 qi's are summable if there exists a question J18 231 q such that

J18 232 formula

J18 233 for every <*_>alpha<*/> <*_>EPSILON<*/> S. In such a case, J18 234 we write formula.

J18 235 Finally, a countable set {qi} is called mutually J18 236 summable if every subset of the qi's is J18 237 summable.

J18 238 REMARK. As mentioned above, the set Q will turn out J18 239 to be the fundamental ingredient of our model, in the sense that J18 240 everything else will be described in terms of Q.

J18 241 THEOREM 7.2.

J18 242 (1) The set Q is a partially ordered set with respect to J18 243 the ordering <*_>unch<*/> defined above.

J18 244 (2) There exists a question q1 <*_>EPSILON<*/> J18 245 Q, which we shall often simply call 1, for which J18 246 formula for every q <*_>EPSILON<*/> Q. That is, Q has a J18 247 maximum element q1.

J18 248 (3) There exists a question q0 <*_>EPSILON<*/> J18 249 Q, which we shall often simply call 0, for which formula J18 250 for every q <*_>EPSILON<*/> Q. That is, Q has a minimum element J18 251 q0.

J18 252 (4) For each question q, there exists a question J18 253 <*_>unch<*/> such that

J18 254 formula

J18 255 That is, every question has a complementary question.

J18 256 PROOF. That Q is a partially ordered set is clear.

J18 257 J19 1 <#FROWN:J19\>Consider the diagram

J19 2 formula

J19 3 where <*_>pi<*/> is orthogonal projection (with respect to the J19 4 Hodge inner product), i is the isomorphism

J19 5 formula,

J19 6 and

J19 7 formula.

J19 8 Using the Hodge decomposition theorem, it is not hard to see J19 9 that this diagram commutes, and that <*_>pi<*/> is an isomorphism J19 10 for <*_>rho<*/> near 1. Hence, for such <*_>rho<*/>, det J19 11 <*_>nu<*/>) can be identified with the determinant of J19 12 D'<*_>rho<*/>.

J19 13 The lagrangian formula gives rise to a smooth section J19 14 of formula near <*_>rho<*/> = 1. For formula, this J19 15 section coincides with formula. For <*_>rho<*/> = 1, it J19 16 coincides with formula. Similar things are true if J19 17 W1 is replaced by W2 or if 1 is J19 18 replaced with another representation in P. Thus we have found J19 19 the desired extensions of det1(<*_>nu<*/>) and J19 20 det1(<*_>eta<*/>j).

J19 21 As a notational contrivance, let us define, for J19 22 formula,

J19 23 formula

J19 24 If formula, define

J19 25 formula.

J19 26 So, abusing notation slightly, det1(<*_>nu<*/>) is a J19 27 smooth bundle over all of <*_>unch<*/> and J19 28 det1(<*_>eta<*/>j) is a smooth section defined J19 29 over all of <*_>unch<*/>j. This will simplify notation in J19 30 what follows.

J19 31 (1.15) Proposition. The first Chern class J19 32 c1 (det(<*_>nu<*/>)) of det(<*_>nu<*/>) is J19 33 represented by a multiple <*_>omega<*/> of J19 34 <*_>omega<*/>B. If formula is a symplectic J19 35 2-torus corresponding to a 2-dimensional symplectic summand of J19 36 H1(F; R), then formula.

J19 37 Proof: Let A be the space of connections on J19 38 formula. The first step is to compute the curvature the J19 39 universal S01 bundle formula over J19 40 A x F (cf. [AS]).

J19 41 Identify A with formula via formula. J19 42 Identify TAA with formula. Let J19 43 formula and formula. Define the connection 1-form J19 44 <*_>theta<*/> by

J19 45 formula

J19 46 The connection <*_>theta<*/> has the property that over the J19 47 surface {A} x F it restricts to the connection A on J19 48 formula. The curvature of <*_>theta<*/> is formula. J19 49 Thus, for formula,

J19 50 formula

J19 51 Let G0 be the group of smooth maps J19 52 formula such that g(x0) = 1 for some fixed J19 53 formula. G0 acts freely on U. This J19 54 action preserves the connection <*_>theta<*/>, and so leads to a J19 55 connection <*_>theta<*/>b on formula. Recall that J19 56 <*_>unch<*/> can be identified with F/G0, where J19 57 formula is the space of flat connection on formula. J19 58 Thus <*_>theta<*/>b restricts to a connection on J19 59 <*_>unch<*/> x F.

J19 60 The representation formula associates to J19 61 formula bundle E over <*_>unch<*/> x F with J19 62 curvature

J19 63 formula

J19 64 Hence formula and formula is represented by

J19 65 formula

J19 66 Thus the Chern character of E is represented by

J19 67 formula

J19 68 E can be thought of as a family of flat unches J19 69 bundles over F parameterized by <*_>unch<*/>. Corresponding to J19 70 each flat connection A is the operator DA, J19 71 defined above, and det(<*_>nu<*/>) can be identified with the J19 72 determinant of the index bundle of this family.

J19 73 A flat connection A of formula determines a J19 74 holomorphic structure A' on formula, and the index of J19 75 the family DA can be identified with the index of J19 76 the family formula. (F is given the complex structure J19 77 compatible with its metric.) So, by Theorem 5.1 of [AS4] and the J19 78 Riemann-Roch theorem, c1(det(<*_>nu<*/>)) is represented J19 79 by the 2-dimensional part of

J19 80 formula

J19 81 where formula denotes integration along the fibers J19 82 (i.e. along F) and T(F) is a form representing the Todd J19 83 genus of F. Since formula, where formula, this J19 84 is just

J19 85 formula

J19 86 Note that, for formula and formula,

J19 87 formula.

J19 88 Hence c1(det(<*_>nu<*/>)) is represented by the J19 89 2-form <*_>omega<*/>, where

J19 90 formula

J19 91 (Here we have identified formula with R via J19 92 formula. Let formula be a symplectic 2-torus. J19 93 Then

J19 94 formula

J19 95 D. Results of Newstead.

J19 96 In this subsection we use results of Newstead ([N1,N2]) to show J19 97 that certain diffeomorphisms of representation spaces are J19 98 homologically trivial.

J19 99 Let formula be a separating simple closed curve. J19 100 Define

J19 101 formula

J19 102 Let F1 and F2 be the components J19 103 of F cut along <*_>gamma<*/>. Define

J19 104 formula

J19 105 (j = 1,2). Then we can make the identifications

J19 106 formula

J19 107 (In the first equation, some basing of the free loop J19 108 <*_>gamma<*/> is assumed.)

J19 109 Let formula be a diffeomorphism such that J19 110 formula and formula is the identity. <*_>tau<*/> J19 111 induces diffeomorphisms of F (which is the identity on J19 112 F2) and formula, which will also be denoted J19 113 <*_>tau<*/>. In the proofs of (3.10) and (3.12), we will need the J19 114 following lemma.

J19 115 (1.16) Lemma. formula is the J19 116 identity.

J19 117 Proof: Note that since [<*_>gamma<*/>] lies in the J19 118 commutator subgroup of formula, formula consists J19 119 entirely of irreducible representations. Hence there is a J19 120 fibration

J19 121 formula

J19 122 with fiber formula. Since SO(3) is a rational J19 123 homology 3-sphere, there is a Gysin sequence

J19 124 formula

J19 125 (All coefficients are in Q. formula is the Euler J19 126 class of the fibration.)

J19 127 I claim that formula is an isomorphism for i = 2 J19 128 or 3. This is obvious for i = 2, and injectivity is obvious J19 129 for i = 3. By Proposition 2.6 of [N2]), the natural map

J19 130 formula

J19 131 is an isomorphism (j = 1,2). Furthermore, formula J19 132 (Theorems 1 and 1' of [N2]). Therefore, by the Künneth formula, the J19 133 composite

J19 134 formula

J19 135 is an isomorphism. (The first arrow is induced by the fibration J19 136 formula.) Hence formula is onto.

J19 137 By Theorem 1' of [N2] and the Künneth formula, formula J19 138 is generated by classes of dimensions 2 and 3 for formula. J19 139 I claim that formula is generated by classes of dimensions J19 140 2 and 3 and formula, for formula. This is clear for J19 141 i<*_>unch<*/>3. Suppose it holds for formula. Let J19 142 formula. Then there exists formula lying in the J19 143 subring of formula generated by classes of dimensions 2 and J19 144 3 such that formula. Hence

J19 145 formula

J19 146 where, by inductive assumption, formula lies in the J19 147 subring of formula generated by classes of dimensions 2 and J19 148 3 and <*_>chi<*/>.

J19 149 So it suffices to show that <*_>tau<*/>* is the identity on J19 150 formulae and <*_>chi<*/>.

J19 151 Since <*_>tau<*/> acts on the fibration (1.17) in an J19 152 orientation preserving fashion, formula.

J19 153 Since formula is an isomorphism for i = 2 or 3, it J19 154 suffices to show that <*_>tau<*/> acts trivially on J19 155 formula. By the Künneth formula and the fact that J19 156 <*_>tau<*/> act trivially on formula, it suffices to show J19 157 that <*_>tau<*/> acts trivially on formula (i = 2,3). J19 158 This is done in [AM] (Lemma VI.2.1 and VI.2.2).

J19 159 Let x1, y1, ..., J19 160 xk, yk be a symplectic basis of J19 161 formula. This gives rise to identifications

J19 162 formulae

J19 163 Define formula by

J19 164 formula

J19 165 formula descends to a diffeomorphism of J19 166 formula, which will also be denoted by <*_>sigma<*/>. In J19 167 the proof of (3.59), we will need the following lemma.

J19 168 (1.18) Lemma. formula is the J19 169 identity.

J19 170 Proof: The proof of (1.16) can be adapted to this case J19 171 almost verbatim. It is left to the reader to check that the proofs J19 172 of Lemmas VI.2.1 and VI.2.2 of [AM] work with <*_>tau<*/> replaced J19 173 by <*_>sigma<*/>. (This amounts to observing that formula J19 174 is the identity and that <*_>sigma<*/> acts on various sphere J19 175 bundles in an orientation preserving fashion.)

J19 176 E. Special Isotopies.

J19 177 (1.19) Definition. An isotopy formula of R is J19 178 called special if

J19 179 1. formula is transverse to Q2 at J19 180 formula for all t.

J19 181 2. formula for all t.

J19 182 3. formula is symplectic, and hence (in view of 2 J19 183 above) preserves the fibers of the normal bundle <*_>nu<*/>.

J19 184 (1.20) Proposition. There exists a special isotopy J19 185 formula of R such that formula is transverse to J19 186 Q2 (i.e. their Zariski tangent spaces are transverse J19 187 at each point of formula).

J19 188 Proof: Since M is a QHS, Q1 J19 189 is transverse to Q2 at P and T1 J19 190 is transverse to T2 in S. Choose a compactly J19 191 supported isotopy of a tubular neighborhood of S<*_>unch<*/> J19 192 in R which moves formula transverse to formula J19 193 for each formula and is symplectic on formula. At J19 194 this stage Q1 is transverse to Q2 in J19 195 a neighborhood of S, and so we can find a compactly supported J19 196 isotopy of R<*_>unch<*/> which moves Q1 J19 197 transverse to Q2.

J19 198 F. Orientations.

J19 199 Orientations will be important in what follows, so in this J19 200 section we will establish orientation conventions. [Y] will J19 201 denote the orientation of the space Y. If Y is singular, J19 202 this means the orientation of the top stratum. If Y is a J19 203 bundle, it means an orientation of the fibers (not the total J19 204 space).

J19 205 First of all, orient the Heegaard surface F so that J19 206 [F] followed by a normal vector to F pointing into J19 207 Q2 gives [M]. This fixes an identification of J19 208 formula with R, and so fixes the sign of J19 209 <*_>omega<*/>.

J19 210 Complex vector spaces (and almost complex manifolds) have a J19 211 natural orientation: If a1, ..., an J19 212 is a basis over C, then a1, ia1, J19 213 ..., an, ian is an oriented basis J19 214 over R. Symplectic vector spaces and manifolds are oriented J19 215 according to a compatible almost complex structure. Thus J19 216 <*_>omega<*/> determines the orientations [R], J19 217 [S],[<*_>nu<*/>] and [<*_>xi<*/>]. <*_>eta<*/>j and J19 218 <*_>unch<*/>j, when lifted to formula, have J19 219 J-complex structures. This determines J19 220 [<*_>eta<*/>j] and <*_>unch<*/>j (as bundles J19 221 over formula).

J19 222 Choose orientations of <*_>unch<*/> and <*_>unch<*/> so that

J19 223 formula

J19 224 at points of formula. ([<*_>unch<*/>] is the J19 225 orientation lifted from [S].)

J19 226 In general, given a fibering formula, we choose J19 227 orientations so that [E] = [B][Y]. Thus orientations J19 228 on two of E, B or Y determine an orientation of the J19 229 third. If G acts on X, there is a fibering J19 230 formula (at least at points where G acts freely, which J19 231 is enough to determine orientations). We regard spaces of unit J19 232 vectors (e.g. <*_>unch<*/>j) as quotients of spaces of J19 233 non-unit vectors (e.g. <*_>theta<*/>j) by J19 234 R+, the positive reals. Let [SU(2)], J19 235 [S01] and [R+] be the J19 236 standard orientations. The following equations determine J19 237 orientations of spaces not yet oriented. (There are some cases of J19 238 over determination, and it is left to the reader to check that J19 239 these cases are consistent.)

J19 240 formulae

J19 241 (The third equation requires some comment. formula J19 242 determines an orientation of a double cover of a neighborhood of J19 243 formula in Qj, which induces an orientation J19 244 on Qj.)

J19 245 All boundaries will be oriented according to the 'inward normal J19 246 last' convention. (e.g. F is oriented as J19 247 <*_>delta<*/>W2.) In particular, this fixes an J19 248 orientation of <*_>delta<*/>F*. Hence the map

J19 249 formula

J19 250 is well-defined up to conjugation by elements of SU(2) J19 251 (which preserves orientation). Note that formula. Thus, J19 252 near regular points of R#, there is an J19 253 identification of the normal fiber (in R*# with J19 254 su(2), well-defined up to orientation preserving linear maps J19 255 (i.e. Ad (SU(2))). Choose [R*#] so that

J19 256 formula

J19 257 ([su(2)] is, of course, the standard orientation of J19 258 su(2).)

J19 259 Now we give an orientation convention for intersections of J19 260 manifolds. Let A and B be oriented properly embedded J19 261 submanifolds of an oriented manifold Y, intersecting J19 262 transversely in X. Let <*_>alpha<*/> be the normal bundle of J19 263 A in Y. Orient <*_>alpha<*/> so that

J19 264 formula

J19 265 formula is the normal bundle of X in B. J19 266 Orient X so that

J19 267 formula&figure

J19 268 Note that this orientation convention is compatible with the J19 269 convention given above for boundaries in the sense that

J19 270 formula

J19 271 Note also that the induced orientation of X depends on the J19 272 ordering of A and B.

J19 273 G. Dehn Twists and Dehn Surgery.

J19 274 Let formula be a smooth function such that f(r) = J19 275 0 for r near 1 and f(r) = 2<*_>pi<*/> for r near 2. J19 276 Let formula. Define formula by

J19 277 formula

J19 278 for formula and formula (see Figure 1.1). Note J19 279 that g is the identity near <*_>delta<*/>A and the J19 280 <}_><-|><+|>the <}/> isotopy class of g does not depend on the J19 281 choice of f.

J19 282 Let <*_>gamma<*/> be a simple closed curve in an oriented J19 283 surface F. Let formula be orientation preserving J19 284 embedding of A in F which maps the boundary components of J19 285 A to curves which are isotopic to <*_>gamma<*/>. Define J19 286 formula by

J19 287 formula

J19 288 The isotopy class of h<*_>gamma<*/> does not J19 289 depend on the choice of <*_>phi<*/>. Any map in the isotopy class J19 290 of h<*_>gammma<*/> is called a left-handed Dehn J19 291 twist along <*_>gamma<*/>. The inverses of such maps are called J19 292 right-handed Dehn twists.

J19 293 Let N be a 3-manifold and let formula be a J19 294 boundary component of N which is diffeomorphic to a torus. Let J19 295 formula be a primitive homology class (that is, one which J19 296 can be represented by a simple closed curve). Let formula J19 297 be a diffeomorphism which sends formula to a curve J19 298 representing a.

J19 299 formula

J19 300 is called the Dehn surgery of N along a. Note that J19 301 Na does not depend on the sign of a. If N J19 302 is oriented, we give Na the orientation induced from J19 303 formula.

J19 304 If formula is a knot in a 3-manifold M, then Dehn J19 305 surgery on K means Dehn surgery on M\U, where U J19 306 is an open tubular neighborhood of K. J19 307 J19 308 J20 1 <#FROWN:J20\>Let <*_>unch<*/> denote the neighborhood of J20 2 <*_>unch<*/>g parameterized by formula with J20 3 formula. In this paper, we will be interested in the J20 4 associated real analytic polar coordinates (Re <*_>unch<*/>, Im J20 5 <*_>unch<*/>, |t|, arg t): the coordinate function arg J20 6 t is S1-valued, but since we will soon lift to J20 7 a Z-cover and consider the lift of arg t to be J20 8 R-valued, here we will write arg t <*_>epsilon<*/> J20 9 [0, 2<*_>pi<*/>).

J20 10 Next, let <*_>unch<*/>g denote the Deligne-Mumford J20 11 compactified moduli space of stable curves and set formula. J20 12 Then <*_>pi<*/> extends to a nonsurjective map formula J20 13 which is a branched cover over its image. Moreover J20 14 <*_>unch<*/>g is a differentiable manifold ([BE], J20 15 [Ea-Ma], [Mas]).

J20 16 Now the neighborhood <*_>unch<*/> admits a Fenchel-Nielsen J20 17 parameterization based on uniformizing the surfaces J20 18 formula. On the surface Mt, choose curves J20 19 <*_>gamma<*/>1, ..., <*_>gamma<*/>3g-4 so that J20 20 Mt <*_>approximate-sign<*/> J20 21 {<*_>gamma<*/>0, <*_>gamma<*/>1, ..., J20 22 <*_>gamma<*/>3g-4} is a collection of 3-holed spheres. J20 23 Let <*_>sigma<*/><*_>unch<*/>, t denote the hyperbolic J20 24 metric constistent with the conformal structure on the Riemann J20 25 surface formula, and let formula denote the J20 26 <*_>sigma<*/><*_>unch<*/>, t-length of the unique J20 27 <*_>sigma<*/><*_>unch<*/>, t geodesic representing the J20 28 free homotopy class [<*_>gamma<*/>i] <*_>epsilon<*/> J20 29 <*_>pi<*/>1Mt. Let formula denote J20 30 the Fenchel-Nielsen twist angle (see [Ab] for details) of J20 31 [<*_>gamma<*/>i] of formula. Then formula J20 32 provide Fenchel-Nielsen polar co-ordinates for J20 33 <*_>unch<*/>; here we leave formula undefined. Also, J20 34 formula should be thought of as S1-valued, J20 35 but it will be notationally convenient to consider J20 36 formula.

J20 37 The local functions (Re <*_>unch<*/>, Im <*_>unch<*/>, J20 38 |t|, arg t) and (<*_>unch<*/>, <*_>unch<*/>) each provide J20 39 real analytic polar coordinates for <*_>unch<*/>, the former system J20 40 from a conformal geometry point of view, and the latter from a J20 41 hyperbolic geometry point of view. These two real analytic J20 42 structures are not equivalent: in this paper, we aim to relate the J20 43 two structures.

J20 44 Before we state the main theorems, we need a technical J20 45 definition of what we mean for a coordinate function to be real J20 46 analytic in a polar coordinate system. We say that a function J20 47 defined on a neighborhood of the origin in R2 J20 48 parametrized by the polar coordinate (r, <*_>theta<*/>) is J20 49 sector real analytic provided that for any ray from the J20 50 origin, there is a sector of the neighorhood at the origin J20 51 containing that ray in which the function has a convergent J20 52 expansion in the variables r and <*_>theta<*/> in that sector. J20 53 The definition extends immediately to a product of planes. We J20 54 prove

J20 55 THEOREM A. The coordinate functions formula, are J20 56 sector real analytic in (-log |t|)-1, arg t, and J20 57 <*_>unch<*/> near t = 0; the coordinate functions formula, J20 58 are also sector real analytic in (-log |t|)-1, arg t, and J20 59 <*_>unch<*/>.

J20 60 From our method, it follows

J20 61 THEOREM B. The hyperbolic metrics formula are J20 62 sector real analytic in (-log |t|)-1, arg t, and J20 63 <*_>unch<*/> near t = 0.

J20 64 We also conclude

J20 65 THEOREM C. If (l0, <*_>unch<*/>, J20 66 <*_>theta<*/>0, <*_>unch<*/> are Fenchel-Nielsen J20 67 coordinates for the neighborhood <*_>unch<*/>, then the coordinates J20 68 |t|, <*_>unch<*/>, arg t are sector real analytic in J20 69 formula, <*_>unch<*/>, <*_>theta<*/>0, and J20 70 <*_>unch<*/>.

J20 71 Remarks. (1) Additional notation extends the results to J20 72 surfaces with multiple nodes.

J20 73 (2) We would like to place the current work in the context of J20 74 [Wf] and [Wlpt].

J20 75 In [Wf], we showed that the hyperbolic metrics formula J20 76 were real analytic in formulae, and we provided a somewhat J20 77 explicit Taylor series. There only the hyperbolic structure of the J20 78 surfaces M<*_>unch<*/>, t was considered and the proof J20 79 relied on harmonic maps. We found that the harmonic maps were well J20 80 suited to the study of the single real analytic polar structure J20 81 (<*_>unch<*/>, <*_>unch<*/>): the maps were independent of the J20 82 twist variable formula, and so any sector real analytic J20 83 function of (<*_>unch<*/>, <*_>unch<*/>) was in fact real analytic J20 84 in a full polar neighborhood. In the present situation, we are J20 85 interested in comparing the conformal and hyperbolic geometry, and J20 86 we find that we must use different maps between surfaces; the J20 87 harmonic maps are unsuitable.

J20 88 In [Wlpt], we provided the initial expansion for the hyperbolic J20 89 metrics formula in (<*_>unch<*/>, t). The approach was J20 90 based on the plumbing construction, approximate hyperbolic metrics, J20 91 and solving the prescribed curvature equation. The method appears J20 92 not to provide that the hyperbolic metrics are sector real analytic J20 93 in (log |t|)-1, arg t, and <*_>unch<*/>. The motivation J20 94 for the current work was to treat both the real analyticity and the J20 95 complex parameterization.

J20 96 2. Maps between surfaces. In this section, we define J20 97 the map between surfaces that we will use, and we prove a lemma J20 98 about its relevant properties.

J20 99 2.1. We work on the augmented Teichmüller space J20 100 <*_>unch<*/>g (see [Ab]). Briefly, the Teichmüller space J20 101 Tg is an infinite cyclic cover of Pg J20 102 with covering group T(p) generated by isotopy classes of Dehn J20 103 twist about <*_>gamma<*/>0. Setting <*_>PI<*/> to be the J20 104 covering map formula, consider formula where J20 105 formula and let formula denote a point in J20 106 Tg lying over m(<*_>unch<*/>, t). Assign J20 107 <*_>unch<*/> the coordinates unch and lift the coordinate J20 108 functions on N near m(<*_>unch<*/>, t) to a neighborhood J20 109 of <*_>unch<*/>. When we continue the functions <*_>unch<*/>, |t|, J20 110 and arg t on Tg, we see that <*_>unch<*/> and |t| J20 111 are invariant under the action of the deck group T(p) while if J20 112 formula with formula, then the arg t coordinates J20 113 for <*_>unch<*/>1 and <*_>unch<*/>2 differ by J20 114 an integer multiple of 2<*_>pi<*/>.

J20 115 To obtain <*_>unch<*/>g we adjoin to J20 116 Tg classes representing the noded surfaces J20 117 M<*_>unch<*/>, 0 so that the coordinate functions J20 118 <*_>unch<*/> and |t| are continuous. The projection formula J20 119 extends to a map formula infinitely branched over the locus J20 120 of points t = 0. Let <*_>unch<*/>* be the preimage of <*_>unch<*/> J20 121 under <*_>unch<*/>; then <*_>unch<*/>* is sometimes called a J20 122 horocyclic neighborhood in <*_>unch<*/>g, and we have J20 123 provided coordinates (<*_>unch<*/>, |t|, arg t) on J20 124 <*_>unch<*/>*.

J20 125 2.2. In this section, we prove

J20 126 LEMMA 2.2. Fix (<*_>unch<*/>0, |t0|, J20 127 arg t0) with |t0| <*_>unch<*/> 0 and let J20 128 formula represent a point in <*_>unch<*/>g with J20 129 coordinates (<*_>unch<*/>, |t|, arg t) near J20 130 (<*_>unch<*/>0, |t0|, arg t0). Then J20 131 there exists maps formula so that formula is real J20 132 analytic in <*_>unch<*/>, (-log |t|)-1 and arg t and J20 133 the radius of convergence of the power series expansion of J20 134 formula about <*_>unch<*/>0, |t0|, J20 135 arg t0 is independent of <*_>unch<*/>0, J20 136 t0.

J20 137 Remark. Theorem B will follow from Lemma 2.2 by letting J20 138 |t0| tend to zero. We are forced to take this approach J20 139 rather than simply setting |t0| = 0 initially because we J20 140 do not know the expansion of formula near the node p, J20 141 and so we lack boundary conditions with which to simultaneously J20 142 open up the node and uniformize.

J20 143 In this approach, the size of the sector in which we are J20 144 guaranteed analyticity in the theorem depends on the radius of J20 145 convergence given by the lemma.

J20 146 Proof. We organize our argument into four steps.

J20 147 1. Constructing the maps. Important to our J20 148 construction of the map <*_>zeta<*/> will be a decomposition of J20 149 formula into three domains determined by its coordinates. J20 150 Our original surface M0 admitted a decomposition J20 151 as

J20 152 formula

J20 153 where the Beltrami differentials <*_>mu<*/>i were J20 154 supported on M* and where we now further assume that J20 155 M0 <*_>approximate-sign<*/> M* is given as J20 156 formula. Now define formula; here c is a small J20 157 number to be determined later. The cylinder C|t| J20 158 conformally embeds in formula and we will also use the J20 159 notation C|t| for the embedded cylinder. Now recall J20 160 that the map formula was conformal on M0 J20 161 <*_>approximate-sign<*/> M*, and define formula; we J20 162 notice from our constructions that the conformal type of J20 163 formula depends only on <*_>mu<*/>(<*_>unch<*/>), hence J20 164 only on <*_>unch<*/>. We conclude that formula can be J20 165 decomposed as

J20 166 formula

J20 167 Note that the middle component, formula, is a pair of J20 168 cylinders whose geometry depends only on |t|; moreover the J20 169 geometry of these cylinders is bounded independently of |t| J20 170 for formula. We will define the map <*_>zeta<*/> on each J20 171 domain and then patch the definitions together across the J20 172 frontiers.

J20 173 We begin by considering the cylinder J20 174 C|t0|, but before we define the map J20 175 <*_>zeta<*/> on C|t0|, we discuss some of J20 176 the geometry of a cylinder C|t|. First, it will be J20 177 more convenient for us to work with the conformally equivalent J20 178 domain formula where we identify formula with J20 179 formula. Second, we notice that A|t| admits J20 180 the (noncomplete) hyperbolic metrics formula; in this J20 181 metric the curve formula is a geodesic of length J20 182 formula, and the curves formula and formula J20 183 have constant geodesic curvature and are of length formula. J20 184 It is often enough to focus on the subdomain formula; the J20 185 domain formula is bounded by the geodesic J20 186 <*_>gamma<*/>|t| and is antiisometric to the interior of J20 187 formula.

J20 188 We now define the map formula by setting J20 189 formula and then defining u(|t0|; |t|)(z) and J20 190 v(|t0|; |t|)(z) in the coordinates on J20 191 A|to| by

J20 192 formula&caption

J20 193 Let formula denote the formula distance J20 194 function. Then the map formula has the important property J20 195 that

J20 196 formula

J20 197 Because u(|t0|; |t|)(x) is odd about J20 198 <*_>gamma<*/>|t|, we could also have defined <*_>zeta<*/> J20 199 to take formula onto formula and the interior of J20 200 formula onto the interior of formula; when we allow J20 201 |t| to tend to zero, this will be a useful characterization of the J20 202 map <*_>zeta<*/> between the domains formula and J20 203 formula.

J20 204 Later we will need to know formula and the Beltrami J20 205 differential formula. These we compute to be

J20 206 formula&caption

J20 207 and

J20 208 formula&caption

J20 209 where formula, and formula.

J20 210 It is perhaps easier to note that the first term in the J20 211 brackets of (2.2) is formula and that

J20 212 formula&caption

J20 213 and

J20 214 formula&caption

J20 215 The following properties of (2.1), (2.2), and (2.3) will be J20 216 crucial for our discussion:

J20 217 (i)formula,

J20 218 (ii)formula and formula,

J20 219 (iii) for formula is bounded independently of J20 220 |t0|, for |t0| small,

J20 221 for |t| > 0, the map formula and for |t| = 0, J20 222 the map formula on formula (also on the interior of J20 223 formula, with the natural conventions) where here we define J20 224 formula and most importantly of all,

J20 225 (v) for fixed formula, the families formula, J20 226 v(z), and Dv(z) are analytic in formula for J20 227 0<|t|<<*_>delta<*/> for some <*_>delta<*/> and can be J20 228 analytically continued to a neighborhood of formula.

J20 229 Next consider the components formula, which are J20 230 conformally a pair of cylinders of bounded geometry; we will assume J20 231 for convenience that formula does not depend on J20 232 (<*_>unch<*/>, t) and that B1 and J20 233 B2 are the two connected components. Divide each J20 234 cylinder Bi into thirds so that formula, so J20 235 that Bi1 share a frontier with J20 236 C|t|, Bi2 are the middle J20 237 thirds, and Bi3 share a frontier with J20 238 formula. Suppose that Bi2 is J20 239 parameterized by formula; using the anticonformal J20 240 equivalence of B1 with B2, we may J20 241 suppose that {|z| = <*_>beta<*/>} parameterizes the boundary J20 242 between B12 and J20 243 B11 and also the boundary between J20 244 B22 and B23. Then J20 245 we define formula to be a pair of twist diffeomorphisms J20 246 realizing the difference in arguments arg t - arg J20 247 t0, i.e. in the coordinated formula we J20 248 define

J20 249 formula&caption

J20 250 We will later define <*_>zeta<*/> on J20 251 Bi1 and Bi3 to J20 252 interpolate between the maps on C|t|, J20 253 Bi2, and formula.

J20 254 On the remaining subsurface formula, define J20 255 <*_>zeta<*/> to be a quasiconformal map formula depending J20 256 only upon <*_>unch<*/> and <*_>unch<*/>0 so that J20 257 formula is the identity and formula is analytic in J20 258 <*_>unch<*/>.

J20 259 We are left to construct <*_>zeta<*/> on the cylindrical J20 260 regions formula. Here we define <*_>zeta<*/> to interpolate J20 261 between the maps defined on formula so that (i) J20 262 <*_>zeta<*/> is analytic in (<*_>unch<*/>, |t|, arg t), with a J20 263 real analytic continuation possible to a neighborhood of J20 264 (<*_>unch<*/>0, t = 0), (ii) J20 265 <*_>zeta<*/>(<*_>unch<*/>0, |t0|, arg J20 266 t0; <*_>unch<*/>0, |t0|, arg J20 267 t0)(z) = z and (iii) formula. Such a family of J20 268 maps exists because we have defined <*_>zeta<*/> on formula J20 269 to have those properties and to also have derivatives which are J20 270 uniformly bounded on formula. This concludes our J20 271 construction of the map formula.

J20 272 2. Constructing the grafted metrics. Our next goal is J20 273 the construction of the hyperbolic metric formula on J20 274 formula in such a way that we can analyze its dependence on J20 275 <*_>unch<*/>, |t|, and arg t. Our approach is to use the J20 276 method of Wolpert in [Wlpt]: we first construct a family of smooth J20 277 metrics formula on formula which are hyperbolic on J20 278 C|t| and formula but which are generally not J20 279 hyperbolic on formula. We then consider the pullback metric J20 280 formula which will be hyperbolic on formula but J20 281 generally not hyperbolic in formula. J20 282 J21 1 <#FROWN:J21\>THE SELF-JOININGS OF RANK TWO MIXING J21 2 TRANSFORMATIONS

J21 3 DANIEL ULLMAN

J21 4 (Communicated by R. Daniel Mauldin)

J21 5 ABSTRACT. The class of rank one mixing transformations has J21 6 been known for some time to have certain so-called 'exotic' J21 7 properties. Specifically, any rank one mixing T has only the J21 8 two trivial factors, and nothing can commute with T but powers J21 9 of T itself. This much was known to Ornstein in 1969 [3]. In J21 10 1983 J. King showed that rank one mixing transformations possess an J21 11 even stronger property known as minimal self-joinings (MSJ). In J21 12 this note we investigate how these results can be extended to the J21 13 case of rank two mixing transformations. In this class, it is J21 14 possible for there to exist nontrivial factors and commuting J21 15 transformations: the square of a rank one mixing transformation and J21 16 certain two point extensions of a rank one mixing transformation J21 17 are rank two mixing [2]. What we prove is that, other than those J21 18 two kinds of rank two mixing transformations, this class also has J21 19 MSJ.

J21 20 Definitions. Assume throughout that T is a J21 21 measure-preserving automorphism of a Lebesgue probability space J21 22 (X, B, <*_>mu<*/>). By a tower for T of J21 23 height h we mean a set B <*_>unch<*/> B J21 24 together with its first h translates under T, provided J21 25 that these translates are disjoint. B is called the base J21 26 of the tower formula, and, for formula, J21 27 TiB is called a level of the tower. (Notice J21 28 that according to this terminology, a tower of height h has J21 29 h +1 levels.)

J21 30 The well-known Rohklin Lemma asserts that, given any J21 31 measure-preserving transformation T and any <*_>epsilon<*/>>0, J21 32 there is a tower M for T with formula.

J21 33 All partitions are presumed to consist of a finite number of J21 34 measurable sets. If Pj is a sequence of partitions J21 35 of X, we say Pj generates the J21 36 <*_>sigma<*/>-algebra B if B is the smallest J21 37 <*_>sigma<*/>-algebra containing all the Pj.

J21 38 A measure-preserving transformation T is called rank J21 39 n or less if there is an infinite sequence of sets of n J21 40 disjoint towers formula such that the partitions J21 41 Pj of X given by

J21 42 formula

J21 43 generate the full <*_>sigma<*/>-algebra B. Given such a J21 44 sequence, the towers with subscript j are called J21 45 j-towers. Obviously, we call a transformation rank n J21 46 (exactly) if it is rank n or less but not rank n - 1 or J21 47 less.

J21 48 It turns out that the class of rank n or less J21 49 transformations can be characterized as the family of automorphisms J21 50 isomorphic to a cutting and stacking transformation of the unit J21 51 interval with n or fewer towers at each stage. (The n =1 J21 52 version of this fact was first prove<&|>sic! by Baxter [1]; the J21 53 case of general n is not much harder.) I do not describe the J21 54 details of such a construction nor refer to cutting and stacking in J21 55 the sequel. Suffice it to say that this result allows us to assume J21 56 that the partitions Pj in (1) form a nested J21 57 sequence, each refining the previous.

J21 58 A TxT-invariant measure <*_>unch<*/> on the Cartesian J21 59 product space XxX, which projects to <*_>mu<*/> in both J21 60 directions, is called a self-joining of T. Any T has J21 61 certain obvious ergodic self-joinings; namely, product measure J21 62 <*_>mu<*/>x<*_>mu<*/> and the so-called off-diagonal J21 63 measures - measures defined on measurable rectangles by J21 64 formula for some integer n. If these are the only J21 65 ergodic self-joinings, then T is said to have J21 66 minimal self-joinings (MSJ). This definition is due to D. J21 67 Rudolph [4].

J21 68 The commutant of T is the group of measure-preserving J21 69 transformations that commute with T modulo the (necessarily J21 70 normal) subgroup of integral powers of T. A factor J21 71 <*_>sigma<*/>-algebra is a T-invariant J21 72 sub-<*_>sigma<*/>-algebra of B. B itself and J21 73 {<*_>phi<*/>, X} are factors of any T and are called J21 74 trivial. It is not hard to see that MSJ implies that T has J21 75 trivial commutant and factors. For if S commutes with T J21 76 and is not a power of T, formula defines a new ergodic J21 77 self-joining, and if F is a nontrivial T-invariant J21 78 (factor) <*_>sigma<*/>-algebra, then formula defines a J21 79 self-joining not all of whose ergodic components can be product J21 80 measure or off-diagonal measures.

J21 81 A map T is mixing (or satisfies the mixing property) J21 82 if, for all A and B <*_>unch<*/> B, formula J21 83 as n goes to infinity.

J21 84 THE CLASSIFICATION THEOREM

J21 85 Theorem. If T is a rank two mixing transformation, J21 86 then either

J21 87 (1) T is the square of a rank one mixing J21 88 transformation,

J21 89 (2) T is a two point extension of a rank one mixing J21 90 transformation, or

J21 91 (3) T has MSJ.

J21 92 The proof proceeds through a series of lemmas. Throughout, J21 93 T acts on a space (X, B, <*_>mu<*/>). For each J21 94 natural number j, X is partitioned into two towers J21 95 Mj(1) and Mj(2), called J21 96 j-towers, and a set formula. Mj J21 97 and Mj(2) have bases Bj(1) and J21 98 Bj(2) and heights hj(1) and J21 99 hj(2), respectively. Levels of j-towers are J21 100 naturally called j-levels. The partitions Pj of J21 101 X whose elements are j-levels successively refine each J21 102 other and generate B. We assume that the real numbers J21 103 <*_>mu<*/>(Mj(1)) and <*_>mu<*/>(Mj(2)), for J21 104 j = 1, 2, 3,..., are bounded away from zero, by <*_>gamma<*/>, J21 105 say. (Otherwise, T is rank one.) Finally, we assume that the J21 106 base of the j-towers are a union of at least twenty J21 107 (j+1)-levels; this can be arranged by passing to a subsequence J21 108 of towers jk.

J21 109 We begin with a lemma that guarantees that almost all points in J21 110 X lie in the middle 98% of their j-tower for at least 3/4 J21 111 of all j. For 0<<*_>alpha<*/><1/2 and i =1 or 2, let

J21 112 formula

J21 113 and set

J21 114 formula.

J21 115 This is the set of points on one of the j-towers that are J21 116 not too close to the top or bottom.

J21 117 Lemma 1. Let

J21 118 formula.

J21 119 Then <*_>mu<*/>S(1/100) =1.

J21 120 Proof. First notice that x<*_>unch<*/>S(1/99) J21 121 implies that formula for all integers n. Consequently, J21 122 formula, so

J21 123 formula.

J21 124 To show that <*_>mu<*/>S(1/100) =1, it is enough to show that J21 125 <*_>mu<*/>s(1/99) >0, since then the left-hand side of (2) is an J21 126 invariant set of positive measure.

J21 127 The idea is that the sets Dj(1/99), for j = J21 128 1, 2, 3,..., are close to being independent sets. We modify them J21 129 slightly (to produce independent sets) and then use the law of J21 130 large numbers.

J21 131 Find four natural numbers a(1), a(2), b(1), and J21 132 b(2) such that, if

J21 133 formula,

J21 134 then

J21 135 formula,

J21 136 formula,

J21 137 and

J21 138 formula.

J21 139 Condition (5) says simply that Ej must be a J21 140 union of j-levels the first of which is a subset of the base J21 141 of a (j - 1)-tower and the last of which is a subset of the J21 142 top level of a (j - 1)-tower. That a(1), a(2), J21 143 b(1), and b(2) can be chosen satisfying (3) and (4) J21 144 requires that the j-towers be many times higher than the J21 145 (j - 1)-towers, which is equivalent to the assumption made at J21 146 the end of the first paragraph after the statement of the J21 147 theorem.

J21 148 Let <*_>nu<*/> be the measure <*_>mu<*/> restricted to J21 149 formula and normalized; that is,

J21 150 formula

J21 151 for formula. Then the sets Ej are J21 152 independent with respect to nu, since no data of the form J21 153 formula or formula for k>j give any J21 154 information about which j-level the point x is on.

J21 155 Hence the strong law of large numbers tells us that, for J21 156 <*_>nu<*/>-almost every x,

J21 157 formula.

J21 158 But formula and so <*_>nu<*/>S(1/99) =1. This implies J21 159 that <*_>nu<*/>S(1/99) >0 and so <*_>nu<*/>S(1/100) =1 as J21 160 required.<*_>square<*/>

J21 161 Hereafter, we assume that the property that Lemma 1 ascribes to J21 162 almost all points in fact holds for all points. We may do this, J21 163 since the property is invariant.

J21 164 For any pair formula, we set rj equal to J21 165 the greatest nonpositive integer i such that either J21 166 Tix or Tiy is in formula. J21 167 Similarly, let sj be the smallest nonnegative J21 168 integer i such that either Tix or J21 169 Tiy is in the top level of a j-tower, that is, J21 170 in the set formula. We say that the interval of integers J21 171 between rj and sj is a frame J21 172 (the j-frame) for (x, y). Note that, when j is J21 173 sufficiently large, both x and y are on J21 174 j-towers, and for such j, the j-frame is J21 175 shorter than both hj(1) and hj(2).

J21 176 Lemma 2. Suppose <*_>unch<*/> is a self-joining J21 177 of T, (x, y)<*_>unch<*/>XxX, and x and y are not on the same orbit J21 178 of T. Suppose also that there is a subsequence J<*_>unch<*/>N of J21 179 natural numbers j such that (1) x and y lie in the middle J21 180 98% of the same j-tower, and (2) for all sets A and B in J21 181 X that are unions of levels of some tower,

J21 182 formula

J21 183 as j goes to infinity along the subsequence J. Then J21 184 formula, product measure.

J21 185 Comment. Condition (2) of the lemma is certainly satisfied J21 186 if (x, y) is generic for unch, which means J21 187 that for any sets A and B made out of levels of some J21 188 tower,

J21 189 formula.

J21 190 Proof. For j<*_>unch<*/>J, let Fj be J21 191 the phase shift between the j-block in which x sits J21 192 and the j-block in which y sits. By this I mean that J21 193 if n(x) and n(y) are the smallest nonnegative integers J21 194 such that formula and formula, then J21 195 formula. Since x and y are assumed to be on J21 196 different orbits of T, formula, as j goes to J21 197 infinity along J.

J21 198 Let A and B be subsets of X made up of J21 199 j0-levels. Fix <*_>epsilon<*/> >0. Because T is J21 200 mixing, there is an integer N so that

J21 201 formula.

J21 202 Pick j<*_>unch<*/>J so large that

J21 203 formulae

J21 204 and

J21 205 formula.

J21 206 By switching the roles of x and y if necessary, we J21 207 may assume that formula, so that Figure 1 is qualitatively J21 208 correct.

J21 209 figure&caption

J21 210 We now calculate:

J21 211 formula.

J21 212 The first equality here comes from (9) and the second holds J21 213 because formula if and only if formula and J21 214 formula, when formula. If formula, let

J21 215 formula,

J21 216 the set of points on the same j-tower as x, at a J21 217 height between 0 and sj-rj. Now (7) J21 218 implies that formula is made entirely of complete J21 219 j-levels, hence expression (10) above implies

J21 220 formula.

J21 221 Since C is at least one <}_><-|>hundreth<+|>hundredth<}/> J21 222 of Mj(i) (because of condition (1) of the lemma) we J21 223 see that <*_>mu<*/>C ><*_>gamma<*/>/100. (Remember that J21 224 formula.) Thus,

J21 225 formula,

J21 226 which, owing to (6) and (8), is formula. The outcome of J21 227 this calculation is that

J21 228 formula,

J21 229 or

J21 230 formula,

J21 231 where L is a constant independent of A and B. J21 232 Since this inequality holds for all A and B made up of J21 233 j0-levels, and since j0 is J21 234 arbitrary,

J21 235 formula.

J21 236 Finally, since <*_>unch<*/> is TxT-invariant (and J21 237 normalized), we conclude that formula and the lemma is J21 238 proved.<*_>square<*/>

J21 239 Lemma 3. Suppose <*_>unch<*/> is an ergodic J21 240 self-joining of T that is not product or off-diagonal measure. Then J21 241 for all <*_>epsilon<*/> > 0, there is a natural number N so that j J21 242 < N implies

J21 243 formula.

J21 244 Proof. Suppose that there are an <*_>epsilon<*/> > 0 and a J21 245 sequence of sets formula with formula, for J21 246 infinitely many j. Then formula, so lim J21 247 infjAj has a generic point (x, J21 248 y) for <*_>unch<*/>. If x and y were on the same J21 249 orbit, then <*_>unch<*/> would be an off-diagonal measure. If J21 250 x and y were not on the same orbit, then we could infer J21 251 from formula that condition (1) of lemma (2) is satisfied. J21 252 The genericity of (x, y) gives us condition (2) of that J21 253 lemma, which would imply that <*_>unch<*/> is product measure. The J21 254 lemma is proved.<*_>square<*/>

J21 255 Lemma 4. Suppose that <*_>unch<*/> is an ergodic J21 256 joining of T that is not product measure. Assume that both (x, J21 257 y1) and (x, y2) are generic for <*_>unch<*/> J21 258 and that x and yi are on opposite j-towers for all J21 259 sufficiently large j, for i = 1 and 2. Then y1 = J21 260 y2.

J21 261 Proof. y1 and y2 are on the same J21 262 j-tower for all sufficiently large j. Let J21 263 rj and sj be the beginning and the J21 264 end of the j-frame for (y1, J21 265 y2).

J21 266 Skip those j for which either y1 or J21 267 y2 is not in Dj(1/100). Lemma 1 J21 268 assures us that we are left with infinitely many j. Along the J21 269 subsequence that remains, we have formula.

J21 270 figure&caption

J21 271 There is a natural topology on X associated with the J21 272 towers Mj(i), and that is the one generated by J21 273 levels of towers. That is, a set is open (and closed) if it is a J21 274 countable union of elements of the partitions Pj. J21 275 J22 1 <#FROWN:J22\>Women between the ages of fifteen and fifty-nine now J22 2 normally have a job or are looking for one. In Denmark, 75.9 J22 3 percent of adult women were employed in 1987. Spain had the lowest J22 4 participation rate in the EC with 37.5 percent of adult women at J22 5 work (Jackson 1990: 5). Women most commonly work in the service J22 6 sector. Indeed, 76.4 percent of them find their jobs in that sector J22 7 (CEC 1989p: 124). Women are also most likely to work in feminized J22 8 work places. They are bank tellers, nurses, teachers, and cleaners J22 9 in hotels. They are unlikely to be bank managers, executives in J22 10 private business, well-paid technicians, or hotel managers. Women J22 11 who work in industry tend to be concentrated in a few sectors which J22 12 are less well paid and more labor intensive, such as the clothing J22 13 and textile industry. The situation in Germany illustrates the fact J22 14 of segregation. Ninety percent of working women find their jobs in J22 15 only twelve occupational categories (CEC 1989k: 87). Women also J22 16 fill a disproportionate number of part-time jobs; for J22 17 example, they provide 90 percent of the part-time work force in J22 18 Germany (CEC 1989v: 72). According to the latest report, 28.6 J22 19 percent of women's employment is part-time (CEC 1989p: 140). J22 20 Working women in the EC still have not achieved equality in pay or J22 21 opportunity despite laws requiring equality. Women's pay is J22 22 probably 31 percent less than men's (Jackson 1990: 50). However, J22 23 pay scales differ greatly among the member states. What women want J22 24 and what society expects for women also varies among the member J22 25 states. For example, a French woman employed in a bank has a much J22 26 greater chance of reaching a managerial position than does her J22 27 British counterpart. Also the French woman is much more likely to J22 28 find a place in a good public nursery for her child than her J22 29 British counterpart, and both the British and the French woman J22 30 would expect more social acceptance for their careers than would a J22 31 female bank employee in Portugal.

J22 32 NATIONAL POLICIES FOR WOMEN IN THE WORK FORCE

J22 33 The governments of all of the member states of the EC have laws J22 34 to protect and assist working women. In addition, the governments J22 35 have created high-level agencies or even ministries for women's J22 36 issues. The old paternalistic laws, such as those banning women J22 37 from night work, have gradually given way to more modern laws on J22 38 equal treatment. Discrimination is illegal in all countries, but J22 39 the definition of discrimination varies considerably, as does the J22 40 quality of enforcement (Landau 1985). A 1983 French law requires J22 41 employers to make an annual report about their personnel policies J22 42 for women. The law provides sanctions for transgressions as well as J22 43 protections for an employee making charges against an employer J22 44 (France 1984: 487). In general, most governments have been more J22 45 effective in banning overt discrimination than in devising policies J22 46 for affirmative action.

J22 47 Working mothers have more assistance from the law in EC J22 48 countries than they have in the United States. Every country has a J22 49 law which provides for maternity leave. Italian women are entitled J22 50 to twenty weeks of paid leave but British women to only six. In all J22 51 countries, employers may not fire a regular employee because of J22 52 pregnancy or refuse to allow a woman to return to work following J22 53 maternity leave. Parental leave is also beginning to appear in some J22 54 countries. Most countries have inadequate public provisions for J22 55 child care, but the French government provides a good system of J22 56 public child care centers. That system may be a factor in J22 57 explaining why French women are more likely to remain in the work J22 58 force during childbearing years than are women in most other EC J22 59 countries (OECD 1985: 34).

J22 60 New public policies for women are being devised in response to J22 61 unemployment and to changes in the work place. Schools are J22 62 encouraging girls to consider a variety of careers. Training J22 63 programs are being reformed so that women may participate more J22 64 easily. A great deal of research is being conducted in order to J22 65 ascertain what is needed in order to better use women in the work J22 66 force. The efforts are scattered and sporadic, but governments are J22 67 increasingly accepting responsibility to assure that women have the J22 68 preparation needed for modern job opportunities.

J22 69 THE DEVELOPMENT OF EC POLICIES FOR WOMEN

J22 70 The right of the EC to formulate a policy for working women J22 71 derives from the Treaty of Rome. Article 119 establishes the J22 72 principle of equal pay for men and women. The preamble to the J22 73 treaty as well as Articles 117 to 122 give the EC a general grant J22 74 of power for social policy.

J22 75 The 1970s was the decade when the EC began to address women's J22 76 issues. It was a period when both the Council of Ministers and the J22 77 Commission had leaders who were sympathetic to the social concerns J22 78 of the day. Equality, worker's rights, and social justice were J22 79 values which found their way onto the political agendas of the J22 80 countries of Western Europe and onto the agenda of the EC. The J22 81 Social Action Program of 1974 was the result. The program promised J22 82 action "to achieve equality between men and women as J22 83 regards access to employment and vocational training and J22 84 advancement and as regards working conditions including J22 85 pay" (CEC 1974b). The statement constituted the first J22 86 elaboration of the meaning of Article 119 and laid the foundation J22 87 for the EC to act over a broad range of job rights for women.

J22 88 The EC quickly started to fulfill its commitment by enacting J22 89 three directives on equal rights at work. They were the Equal Pay J22 90 Directive of 1975, the Equal Treatment Directive of 1976, and the J22 91 Social Security Directive of 1978. The meaning of each directive J22 92 has been broadened by subsequent rulings of the European Court, but J22 93 most member states have been remiss in enforcing them. Today the J22 94 definition of equal pay in the EC means equal pay for work of equal J22 95 value. Equal treatment now makes illegal all forms of sex J22 96 discrimination at work including hiring, training, and promotion. J22 97 Most importantly, it protects women against both direct and J22 98 indirect discrimination. The directive on social security applies J22 99 to both national social security systems and special occupational J22 100 and supplementary schemes. It does not require uniformity among the J22 101 national programs, and it allows those programs to contain special J22 102 benefits for women. It protects women against provisions which are J22 103 discriminatory even when the discrimination is indirect (CEC J22 104 1983).

J22 105 The organization of the Commission was changed in the 1970s in J22 106 response to the new interest in women's issues. The units added J22 107 continue to be responsible for EC policies for women. The equal J22 108 opportunities unit in the Directorate General for Employment, J22 109 Social Affairs, and Education (DG V) carries on the bulk of the J22 110 work. A handful of civil servants are responsible for the J22 111 information gathering, analysis, and consulting necessary for J22 112 preparing and overseeing policies for working women. DG V is under J22 113 the direction of the commissioner who holds the portfolio for J22 114 social affairs. (In 1991, the commissioner responsible for social J22 115 affairs was Vasso Papandreou. She was the first woman in that J22 116 position and one of the only two women ever to be a commissioner.) J22 117 A women's information service operates in the Directorate General J22 118 for Information, Communication and Culture (DG X). It is J22 119 responsible for disseminating information about women and publishes J22 120 a series called Women of Europe. The European Parliament J22 121 and the Economic and Social Committee have special committees to J22 122 deal with women's issues. Both institutions have advocated a strong J22 123 EC policy for women. The Court of Justice has also played a role in J22 124 developing the EC policy for women through a liberal interpretation J22 125 of EC law.

J22 126 The scope and ambition of the EC policy for working women J22 127 developed in the 1970s is quite remarkable. The policy contains J22 128 both legal measures to ban discrimination and nonlegal measures to J22 129 facilitate the social and psychological changes necessary for true J22 130 equality. Traditional family values intermingle with newer feminist J22 131 concerns in a broad range of initiatives. For example, the J22 132 Commission sponsored seminars to encourage bankers to be less J22 133 sexist in their personnel policies. The Commission studied J22 134 vocational education in order to ascertain why women remain in J22 135 feminized work places. The Commission delved into the question of J22 136 the relationship between family responsibilities and success in the J22 137 work place. The EC used the Social Fund to provide training for J22 138 women to enter jobs formerly inaccessible to them. During the J22 139 formative period of the 1970s, the EC chose an activist approach J22 140 which surpassed merely formulating measures essential to harmonize J22 141 national policies which might have inhibited competition in the J22 142 internal market.

J22 143 In 1981, the disparate EC activities for women were brought J22 144 together in the first action program to promote equal opportunities J22 145 for women. The opening sentence of the document states, J22 146 "The Community's longstanding commitment to the improvement J22 147 of the situation of women has established it as a pioneer and J22 148 innovator in this field" (CEC 1981). The rest of the J22 149 document does not match the bold opening sentence. Discussion in J22 150 the document is brief and focuses primarily on the problems working J22 151 women were facing because of the recession. Member governments were J22 152 given the primary responsibility of alleviating the problems.

J22 153 The annex of the document contains sixteen proposals for legal J22 154 and nonlegal measures to promote equality. In almost every case, J22 155 the responsibility for action is divided between the EC and the J22 156 member state. Frequently, the role of the EC is only to study the J22 157 situation and then consider action; however, six of the proposals J22 158 fit into the activist mold of the 1970s. They are:

J22 159 1. An EC law on equal treatment for women in occupational, J22 160 social security schemes.

J22 161 2. An EC law on equal treatment for self-employed women and J22 162 women in agriculture.

J22 163 3. An EC law on parental leave and leave for family reasons, J22 164 and on the building of public services and facilities to assist J22 165 working parents.

J22 166 4. Possible legislation on pregnancy and motherhood if the J22 167 Commission considers it necessary.

J22 168 5. Future legislation on steps needed for action to assist J22 169 women in achieving equal opportunity.

J22 170 6. Extension of EC action on vocational education so women can J22 171 participate in new technological sectors through the Social Fund J22 172 and the center for vocational training in Berlin.

J22 173 The dates for the action program, 1982-1985, coincided with the J22 174 period when the integration appeared stalled and economic problems J22 175 took precedence. Only the first two proposals listed above became J22 176 law according to the schedule given in the program (CEC 1984b); J22 177 however, the other proposals remained on the agenda of the J22 178 Commission.

J22 179 A second action program appeared in 1985, when Europessimism J22 180 was strongest. It was also the year when the EC was deeply involved J22 181 with two historic documents: the White Paper on Completing the J22 182 Internal Market and the Single European Act. Seen against that J22 183 time, the second action program is quite remarkable. Although it J22 184 contains no major new programs, it is a thoughtful and interesting J22 185 document. It shows the influence of research conducted in the J22 186 Commission over the past decade. Emphasis is given to the J22 187 psychological dimension of discrimination. The writers of the J22 188 document doubted the efficacy of laws to end discrimination. Ways J22 189 to change attitudes, and not just attitudes in the work place, were J22 190 needed. The basis of discrimination is in society and in the J22 191 family. The sharing of family responsibility is listed as the sine J22 192 qua non for true equality (CEC 1986b: 5). Many of the proposals in J22 193 the document reflect this orientation, such as the proposal for a J22 194 campaign to increase public awareness. Other proposals were a J22 195 reiteration of some in the first action program, which still await J22 196 acceptance. Four proposals were to become focal points of J22 197 controversy. They are:

J22 198 1. A legal instrument to facilitate action by women against J22 199 employers who discriminate. The instrument was to be based on the J22 200 principle of the reversal of the burden of proof.

J22 201 2. A code of practice on positive actions which should guide J22 202 employers and member states in order to facilitate providing equal J22 203 opportunity.

J22 204 3. A measure to protect working women during pregnancy and J22 205 motherhood.

J22 206 4. A directive on parental leave and leave for family J22 207 reasons.

J22 208 J23 1 <#FROWN:J23\>If any of these changes are substantial, they will J23 2 affect the relative values of the different ways of J23 3 institutionalizing property rights. If the result of these changes J23 4 is that a new property rights scheme can produce greater aggregate J23 5 benefits than the existing system does, the contracting parties J23 6 will consider the following cost-benefit calculation: Do these J23 7 additional benefits exceed the costs of changing the present J23 8 contract? If the answer is yes, the actors will enter into a new J23 9 contract institutionalizing new rights (North, 1990: 67). Here we J23 10 see the third important element of the transaction-costs theory: J23 11 Institutional change will occur only if the resulting outcome is J23 12 Pareto superior to the previous institutional arrangement.

J23 13 Note that at this point we are discussing only the motivations J23 14 of the actors in the individual exchange. The fact that low-cost J23 15 rules may be selected by marketlike competition has not yet entered J23 16 the analysis. North's reliance on transaction-costs minimization J23 17 reflects the approach of the new institutional economics. In J23 18 analyzing the contractual arrangements of social organization, the J23 19 theory proceeds from the "working rule that J23 20 low-cost organizations tend to supersede high-cost ones" J23 21 (Eggertsson, 1990: 213-14, emphasis in original). When exceptions J23 22 to this rule are observed, three factors should be considered.

J23 23 First, there may be hidden benefits that are not readily J23 24 apparent. This reduces to a claim that the apparent exception to J23 25 the working rule is in fact an instance of the more general rule of J23 26 maximizing benefits. Here a higher-cost institutional rule in fact J23 27 produces a more beneficial exchange than does a less costly one. J23 28 Second, the efforts of social actors to create low-cost J23 29 institutional forms may be constrained by the interests of the J23 30 state. This latter issue is closely related to issues of J23 31 intentional design and reform, but it is proposed as an explanation J23 32 of why the voluntary arrangements being analyzed here can be J23 33 restricted by some type of formal external constraint. The point J23 34 relevant to this analysis is the implicit claim that if it were not J23 35 for external constraints, rational social actors would voluntarily J23 36 create the least costly forms of organization. This is consistent J23 37 with the third exception to the minimization-of-costs rule: J23 38 uncertainty. According to this account, actors may not create the J23 39 least costly rules because they lack either the capacity or the J23 40 knowledge to establish them.

J23 41 This minimization-cost standard is a problematic one. As the J23 42 arguments in Chapter 2 demonstrate, the claim that the minimization J23 43 of aggregate transaction costs is the motivation for institutional J23 44 development and change is inconsistent with individual rationality. J23 45 The assumption of narrow rationality on which the transaction-costs J23 46 approach is based implies that strategic actors will prefer more J23 47 costly rules if these rules give them greater individual benefits J23 48 than those from less costly ones. Remember that in this case the J23 49 explanation is not inability or incapacity but, rather, J23 50 self-interest. The transaction-costs approach does not predict J23 51 costly rules produced by fully informed, self-interested behavior. J23 52 Here we return to some of the same explanatory problems we J23 53 encountered with the theory of social conventions. The standard J23 54 transaction-costs approach fails to explain either the J23 55 distributional features of informal social conventions or the J23 56 redistributive changes in these rules.

J23 57 I should clarify this criticism in the context of current J23 58 attempts by Libecap and North, among others, to incorporate J23 59 distributional considerations. The standard approach has J23 60 traditionally avoided the question of power asymmetries and instead J23 61 has focused on the problem of transaction-costs minimization among J23 62 symmetrically endowed actors (North, 1981). North (1990) J23 63 acknowledges the importance of asymmetries for explaining J23 64 distributional differences; Libecap (1989) explicitly addresses the J23 65 complications introduced by distributional concerns in the effort J23 66 to contract for socially efficient property rights. These attempts J23 67 are in the spirit of the sorts of arguments I am making. But what J23 68 these authors seem to have in mind are efforts at intentionally J23 69 creating formal property rights. The importance of power J23 70 asymmetries for explaining the emergence of informal rules J23 71 governing property rights and economic organization has not been J23 72 pursued; and this is the basis of my criticism. This failure can be J23 73 explained by the reliance on competition as the selection mechanism J23 74 for such decentralized emergence, competition thought to undercut J23 75 the importance of power in individual exchange.

J23 76 In regard to institutional change, the dubious relationship J23 77 between a criterion of Pareto-superior change and strategic J23 78 rationality has already been well exposed. Not only does the J23 79 transaction-costs approach predict a Pareto-improving institutional J23 80 change that might not be accepted by strategic actors (the J23 81 path-dependent argument), but it also fails to explain either J23 82 redistributive or Pareto-inferior change. This follows by J23 83 implication from the combination of a reliance on voluntary J23 84 agreement and a limitation to Pareto-superior change. If a change J23 85 entails a loss to some individuals as a cost of establishing new J23 86 property rights that produce greater aggregate benefits, narrowly J23 87 rational actors will not voluntarily agree to the change.

J23 88 The transaction-costs theory of exchange and competition has a J23 89 way to resolve this weakness: a reconciliation of transaction-costs J23 90 minimization (and the consequent maximization of aggregate J23 91 benefits) with the pursuit of individual gain by means of J23 92 compensation. We can see how the idea of compensation can be used J23 93 in this way by examining Posner's (1980) interpretation of property J23 94 rights and economic organization in primitive societies. He argues J23 95 for an analogy between primitive economic institutions and J23 96 insurance. That is, he explains the rules structuring economic J23 97 activity in these primitive societies as a means of distributing J23 98 risk in the community. The property rights scheme provides rules J23 99 governing the enjoyment of community resources. Distributional J23 100 advantages are granted to certain members of the community, and J23 101 they are explained as the product of an exchange with the less J23 102 advantaged, an exchange for the protection in the future if the J23 103 community is struck by hard times. Here Posner reconciles J23 104 distributional differences with the long-term efficiency of the J23 105 property rights scheme. This insurance analogy can be seen as a J23 106 form of compensation: The less advantaged are compensated for the J23 107 distributional bias in the economic rights scheme by the promise of J23 108 future insurance protection.

J23 109 It is easy to reduce this insurance explanation to a mere J23 110 functionalist assertion without some fairly rigorous empirical J23 111 evidence. To see this, consider the general logic of compensation. J23 112 Say that two actors currently share the benefits in the following J23 113 manner: $60,000 to A and $40,000 to B. Now assume that J23 114 circumstances change so that an alternative set of property rights J23 115 allows the actors to produce an additional $25,000 per year but J23 116 that this alternative scheme produces a new annual distribution: J23 117 $50,000 to A and $75,000 to B. The logic of the new J23 118 institutional economics predicts that as long as the aforementioned J23 119 limitations on change are not applicable, the actors will adopt the J23 120 new contract. But as long as the mechanism of change is one of J23 121 voluntary agreement, my argument is that the new rules will not be J23 122 adopted because they diminish A's benefits. If, however, J23 123 B agreed to compensate A by giving her a side payment of J23 124 at least $10,000 per year, then according to some accounts of J23 125 individual rationality, it would be rational for A to agree to J23 126 the changes.

J23 127 Here the logic is plausible, and in the case of the intentional J23 128 design of formal institutions, compensation is a possibility that J23 129 must be considered. But the utility of this concept for explaining J23 130 the spontaneous emergence of informal rights is highly J23 131 questionable. The empirical requirements necessary to satisfy the J23 132 theory are substantial, and there must be evidence of compensatory J23 133 payments between the individuals in the institutional change, J23 134 payments that are temporally related to these changes and are J23 135 anticipated before the change. In the case of spontaneous J23 136 emergence, an additional element must be shown: Either the form of J23 137 compensation must be a common practice in similar exchanges in a J23 138 society, or the compensation must be an element of the creation of J23 139 those contractual rules that are subsequently selected out by the J23 140 competitive process. Posner's insurance explanation exemplifies the J23 141 weakness of most accounts of compensatory-like mechanisms: The J23 142 evidence necessary to justify a compensation explanation is J23 143 lacking.

J23 144 Competition. But the transaction-costs approach invokes J23 145 more than the actors' intentions in an individual exchange to J23 146 explain the emergence of social institutions. Individual exchanges J23 147 merely produce a variety of possible institutional forms, but the J23 148 key selection mechanism is competition. Many explanations of J23 149 institutional development and change situate the decision to J23 150 establish social institutions in the context of a market or a J23 151 marketlike environment. The main influence of the market on the J23 152 choice of institutional form is in the competitive pressure it J23 153 supposedly exerts on the institutionalization process. There are J23 154 two related but conceptually distinct ways that competitive J23 155 pressure can enter into this analysis.

J23 156 First, as a dynamic effect, competition can be a selection J23 157 mechanism that determines the survival of various institutional J23 158 forms on grounds of survival and reproductive fitness. This is the J23 159 logic behind Alchian's (1950) model of evolutionary competition, J23 160 used in most economic analyses of institutional emergence. The J23 161 existence of a large number of firms seeking profits from a common J23 162 pool of consumers produces pressure for survival. Over time those J23 163 firms that employ less efficient techniques lose profits to those J23 164 that are more efficient. Losing profits eventually translates into J23 165 extinction. As the competitive process continues, only those firms J23 166 that use efficient techniques survive. Second, as a static effect, J23 167 competition can undermine the actors' bargaining power in a J23 168 particular interaction:

J23 169 The main curb on a person's bargaining power, and the J23 170 main pacifying influence on trade in general, is competition. A J23 171 person has competition if the party he wants to trade with has J23 172 alternative opportunities of exchange. The people who offer these J23 173 alternative opportunities to his opposite party are his J23 174 competitors. Competition restricts a person's bargaining power by J23 175 making the other less dependent and therefore less keen on striking J23 176 a bargain with him. (Scitovsky, 1971: 14)

J23 177 This latter effect enters into a theory of development and J23 178 change only to the extent that it establishes the environment in J23 179 which rational actors seek to produce institutions.

J23 180 The existence of competition raises questions for theories of J23 181 institutional emergence and change. The dynamic effect forms the J23 182 basis of the theory of exchange and competition. The relevant J23 183 question here is whether the competitive pressure is sufficient to J23 184 select out less efficient institutional rules. The static effect J23 185 relates mainly to arguments such as the one in the next chapter, J23 186 that would invoke the asymmetries of power in a society in order to J23 187 explain institutional emergence. The relevant question there is J23 188 whether the existence of competition prevents social actors from J23 189 using asymmetries in power to develop institutions that produce J23 190 systematic distributional consequences. This static effect is also J23 191 related to the theory of exchange and competition in that it J23 192 justifies ignoring power asymmetries in its own analysis. It is J23 193 important to remember that competition is not an either/or J23 194 phenomenon; there are degrees of competition and therefore degrees J23 195 of competitive effect. The best way to answer these questions is to J23 196 establish the empirical conditions under which competition affects J23 197 the emergence of social institutions. Because the issues are so J23 198 closely related, the analysis is best presented by addressing both J23 199 of these questions. I will clarify a few of the conclusions in my J23 200 later discussion of bargaining theory.

J23 201 What are the prerequisites for the existence of competition? J23 202 Lists of the necessary conditions for marketlike competition are J23 203 numerous. According to Scitovsky's analysis (1971) of competition, J23 204 the following conditions are necessary for the existence of J23 205 competition in social institutions: (1) a large number of J23 206 competitors in pursuit of a common pool of resources, (2) a set of J23 207 institutional alternatives differentiated only by their J23 208 distributional consequences, (3) full information about the J23 209 availability of alternatives, and (4) low transaction costs. If an J23 210 explanation of institutional change is to invoke competition as a J23 211 relevant factor, these conditions must be empirically satisfied.

J23 212 Discussions of competition in the market are numerous. What I J23 213 want to do is limit my analysis to those issues unique to the J23 214 question of competition in regard to social institutions, as this J23 215 will allow to me to emphasize the difficulties of satisfying these J23 216 conditions in the institutional case. Central to my argument is the J23 217 following distinction: Institutions are not goods. J23 218 J24 1 <#FROWN:J24\>Why Fewer Women Become Physicians: Explaining J24 2 the Premed Persistence Gap

J24 3 Robert Fiorentine and Stephen Cole

J24 4 Previous research indicates that the answer to the question of J24 5 why fewer women become physicians lies in the 'premed persistence J24 6 gap.' Women are no less likely than men to enter undergraduate J24 7 premed programs, but they are less likely to complete the program J24 8 and apply to medical school. This article presents data from a J24 9 study designed to test four plausible explanations of the J24 10 persistence gap that are consistent with the structural barriers, J24 11 normative barriers, and cognitive differences theories of gender J24 12 inequality. The findings do not support the 'perception of J24 13 discrimination' hypothesis, the 'discouragement' hypothesis, the J24 14 'self-derogation' hypothesis, and the 'anticipated role conflict' J24 15 hypothesis. Rather, the evidence suggests another explanation - the J24 16 normative alternatives approach, This approach holds that J24 17 contemporary gender norms offer women fewer disincentives to J24 18 changing or lowering their high-status career goals when J24 19 encountering hardship, self-doubt, and the possibility of J24 20 failure.

J24 21 INTRODUCTION J24 22 This article presents the latest results of a research program J24 23 designed to understand the contemporary causes of gender inequality J24 24 in the United States (Cole, 1986; Fiorentine, 1986, 1987, 1988a, J24 25 1988b). The specific area of research is medicine, one of the most J24 26 prestigious and lucrative occupations.

J24 27 As is true with other prestigious occupations, medicine is J24 28 characterized by extensive gender inequality. Men make up about J24 29 two-thirds of medical students and more than 80% of all practicing J24 30 physicians in the United States (Cole, 1986; U.S. Bureau of the J24 31 Census, 1987; Bickel, 1988). Men are overrepresented in the most J24 32 prestigious specialties, and they are more likely to hold positions J24 33 of authority in hospitals and clinics (Lorber, 1984).

J24 34 Three theoretical approaches have emerged in sociology and J24 35 social psychology to explain the underrepresentation of women in J24 36 high-status professional and executive positions. The J24 37 structural barriers approach holds that differences in J24 38 occupational achievement are a consequence of sex discrimination J24 39 that limits the opportunity of women. Because female gender may be J24 40 a 'discrepant status' that undermines trust and certainty so highly J24 41 desired by organizational players (Kanter, 1977) or because J24 42 prejudicial attitudes of employers force women into the less J24 43 desirable, secondary-sector jobs (cf. Bibb and Form, 1977; Bridges, J24 44 1982) or because capitalist production systematically exploits J24 45 female labor (Hartmann, 1976; Eisenstein, 1979), women may J24 46 encounter barriers to their mobility.

J24 47 The normative barriers approach assumes that gender J24 48 socialization brings young women to view the pursuit of success in J24 49 a high-status career as a transgression of norms (Angrist and J24 50 Almquist, 1975; Douvan, 1976). Anticipating social rejection J24 51 (Horner, 1972), and believing they would be unable to fulfill the J24 52 role expectations of wife and mother if they were committed to a J24 53 demanding career (Angrist and Almquist, 1975; O´Leary, 1974), young J24 54 women place limits on their ambitions, emphasize the primacy of J24 55 their domestic role, and select normatively appropriate, 'feminine' J24 56 occupations.

J24 57 The cognitive differences approach assumes that J24 58 because cultural stereotypes do not depict women in achievement J24 59 roles (cf. Weitzman et al., 1972), or because parents and J24 60 others are less encouraging of female achievement (Hoffman, 1972), J24 61 women have lower confidence in their ability to perform J24 62 successfully in a variety of achievement situations (Lenney, 1977; J24 63 Maccoby and Jacklin, 1974). With lower levels of confidence, women J24 64 are more likely to attribute their successes to 'external' or J24 65 'unstable' causes such as luck or effort, and their failures to J24 66 'internal' or 'stable' causes such as low ability or task J24 67 difficulty (cf. Weiner et al., 1971; Deaux, 1976; Frieze J24 68 et al., 1982). As successes are discounted while failures J24 69 are affirmed, women are less likely to enter into, persist with, or J24 70 perform well in a wide array of achievement tasks.

J24 71 In earlier research we looked closely at the structural J24 72 barriers explanation of the underrepresentation of female J24 73 physicians. Inasmuch as women comprise only about one-third of the J24 74 students currently admitted to medical school, it could be that J24 75 medical schools discriminate against female applicants. Medical J24 76 schools may employ restrictive quotas (Walsh, 1977), they could J24 77 subject female applicants to more rigorous admissions standards, of J24 78 they may systematically, if unconsciously, devalue their J24 79 credentials. But in an analysis of application and admission rates J24 80 of males and females to all American medical schools (Cole, 1986), J24 81 it was determined that female applicants have about the same J24 82 qualifications as male applicants, and they are just as likely to J24 83 be admitted to medical schools. Women account for one-third of the J24 84 admittees simply because they constitute one-third of the J24 85 pool of applicants.

J24 86 The evidence suggests that medical schools do not overtly J24 87 discriminate against female applicants. But what happens once they J24 88 enter medical school? Are women admitted only to be systematically J24 89 'cooled out' after they begin their training? The evidence suggests J24 90 not. Women graduate from medical school at almost the same rate as J24 91 men (Journal of the American Medical Association, 1984, J24 92 1986).

J24 93 If women are as likely as men to get into and through medical J24 94 school, then efforts to understand why fewer women become J24 95 physicians need to focus on gender differences in ambition rather J24 96 than differences in opportunity. It needs to be determined why J24 97 fewer women apply to medical school.

J24 98 Subsequent research has demonstrated that the answer lies in J24 99 the 'premed persistence gap': An equal ratio of women and men enter J24 100 undergraduate premed programs, but men are twice as likely to J24 101 complete the program and apply to medical school (Fiorentine, 1986, J24 102 1987). Further, it was determined that the persistence differential J24 103 cannot be explained entirely by differences in academic J24 104 performance, even though female premed students earn slightly lower J24 105 grades in the required premed courses. For while women with a J24 106 marginally competitive or a noncompetitive academic performance are J24 107 more likely to relinquish their medical career goals, women with a J24 108 competitive grade average of 3.50 or higher are no less likely J24 109 to persist in the program and apply to medical school.

J24 110 A similar persistence gap exists for male and female high J24 111 school students who aspire to high-status professional occupations J24 112 such as doctor, college teacher, scientist, and lawyer. Women with J24 113 an A average are just as likely as their male counterparts to J24 114 persist with these career aspirations, but at every lower academic J24 115 level, males are from one and one-half to two times more likely to J24 116 maintain their professional aspirations from the sophomore to the J24 117 senior years (National Center for Educational Statistics, 1983; J24 118 Fiorentine, 1986). This suggests that the premed persistence gap is J24 119 explained by general social or cultural processes rather than by J24 120 factors that are unique to premedical programs.

J24 121 Although that earlier study demonstrated that sex differences J24 122 in academic performance account for only a small portion of the J24 123 premed persistence differential, it was not designed to empirically J24 124 examine nonperformance explanations of the persistence gap. This J24 125 article presents data from a second study that tests four plausible J24 126 explanations that are consistent with the (informal) structural J24 127 barriers, normative barriers, and cognitive differences theories of J24 128 gender inequality.

J24 129 Perception of Discrimination

J24 130 Even though medical schools do not overtly discriminate against J24 131 female applicants, this 'objective reality' may not be the J24 132 subjective perception of many female premed students. It may be J24 133 that young women erroneously assume medical schools restrict J24 134 admission to all but the most academically distinguished female J24 135 students. If so, then those with a highly competitive performance J24 136 would be about as likely as young men to persist, but those with a J24 137 less competitive performance would underestimate their chances of J24 138 admission, and would be less likely to complete the premed program J24 139 and apply to medical school.

J24 140 Discouragement J24 141 Although there has been a great deal of change in gender J24 142 expectations over the last several decades, it could be that J24 143 professors, parents, peers, and lovers continue to treat the J24 144 medical career goals of young women with more indifference, less J24 145 encouragement, or even outright hostility. As premed programs are J24 146 usually competitive, even slight differences in encouragement may J24 147 lead women with marginally competitive and non-competitive J24 148 performances to believe the goal of becoming a physician is J24 149 unattainable or inappropriate. It would not be immediately J24 150 apparent, however, why women with a competitive performance are J24 151 either not discouraged by others or do not react to this J24 152 discouragement by relinquishing their medical career goals. One J24 153 possibility is that the negative reactions of others are not J24 154 sufficiently strong to sway the most determined, who eagerly and J24 155 realistically anticipate success.

J24 156 Self-Derogation J24 157 Consistent with the assumption of expectancy-value and J24 158 attribution theories (Atkinson, 1964; Weiner, 1986; Weiner et J24 159 al., 1971), it may be that young women enter the premed J24 160 program with lower expectancies of success, and attribute a J24 161 less-than-competitive performance to stable, internal causes (low J24 162 ability), or to stable, external causes (high task difficulty). If J24 163 so, then women would be less likely to believe they could turn a J24 164 marginally competitive or a noncompetitive performance into a J24 165 competitive one, and consequently, less likely to persist in the J24 166 premed program.

J24 167 Anticipated Role Conflict

J24 168 Finally, it may be that young women initially underestimate the J24 169 ease of negotiating family demands and a career in medicine. J24 170 Realizing they are forced to choose between their career goals and J24 171 their family plans, most relinquish their career goals, while some, J24 172 particularly those with a competitive academic performance, J24 173 relinquish their conventional family plans.

J24 174 METHOD J24 175 As in the original investigation, transcripts of all male and J24 176 female nontransfer premed students entering the State University of J24 177 New York at Stony Brook as freshmen between 1982 and 1985 were J24 178 acquired. During four weeks in April 1986, trained interviewers J24 179 attempted to locate and interview, via telephone, all persisting J24 180 and defecting students, including those who had withdrawn from the J24 181 university. Interviews were completed with 302 males and 240 J24 182 females. Only one male student refused to participate in the J24 183 survey. Interviews were completed with all persisting and defecting J24 184 premed students who were currently enrolled at the time of the J24 185 survey. There was no difference in the distribution of males and J24 186 females by year in college. Among the males in the sample, 26% were J24 187 freshmen, 26% sophomores, 25% juniors, and 22% seniors. Among J24 188 females, 27% were freshmen, 26% sophomores, 25% juniors, and 22% J24 189 seniors.

J24 190 As typical in attrition studies, some of the students who were J24 191 no longer enrolled at Stony Brook could not be located in a J24 192 national search, and were not interviewed. The sample represents J24 193 73% of all males and 60% of all females who began their studies as J24 194 premed students. Because females were more likely to drop out of J24 195 the premed program (and not leave a forwarding address or phone J24 196 number) they were more likely to be in the hard-to-find subsample. J24 197 Students who were not interviewed had very low GPAs (about a D+ J24 198 average), but there was no difference between males and females in J24 199 this group. Inability to locate these students had the following J24 200 effect on the results: (1) the sample has a higher grade average J24 201 than the population of all those who began as premed students and J24 202 (2) the proportion of persisters in the sample is higher than it J24 203 would be if there were data on all those who began as premeds. This J24 204 would not, as far as we can tell, have influenced any of the J24 205 conclusions about gender differences in persistence.

J24 206 Focused fact-to-fact interviews also were conducted with 23 J24 207 female and 13 male premed students during the fall 1985 and spring J24 208 1986 semesters. Students in the main sample (n = 542) were J24 209 categorized along lines of persistence, grades, and sex; and then a J24 210 small number within these broad categories were randomly selected J24 211 for interviewing.

J24 212 In order to assess initial expectancies of success prior to J24 213 entering the premed program, a list of high school students who J24 214 planned to begin as freshmen in the fall of 1986 and were J24 215 preregistered for any two of the required premed courses was J24 216 acquired from the registrar's office (n = 76). Sixty-eight of J24 217 these students could be located, and interviews were completed J24 218 during the same four-week period in April with the 62 students who J24 219 indicated they planned a 'premed' course of study.

J24 220 Premed persistence was assessed by self-report measures along J24 221 with actual records of medical school application acquired from the J24 222 Office of Undergraduate Studies. Respondents were determined to be J24 223 'persisters' if they either had applied to medical school (premed J24 224 students typically apply to medical school during the spring J24 225 semester of their junior year) or indicated they planned to J24 226 apply to medical school in the future. J24 227 J25 1 <#FROWN:J25\>DESIGN FOR VULNERABILITY: CUES AND REACTIONS J25 2 TO FEAR OF CRIME

J25 3 Jack L. Nasar

J25 4 Ohio State University

J25 5 Bonnie Fisher

J25 6 University of Cincinnati

J25 7 Fear of crime is a critical problem on university campuses. J25 8 This paper describes cues in the built environment that may J25 9 affect fear of crime. It develops and tests a theory about the J25 10 relationship between these cues and fear, and consequent reactions. J25 11 In our analysis of responses to open-ended questions, we found J25 12 that fear was heightened by several site-specific cues: poor J25 13 prospect for the passerby due to inadequate lighting, blocked J25 14 escape for the passerby, and concealment for the offender. J25 15 Respondents also reported avoidance, protective, and collective J25 16 actions in response to their site-specific fears. The results J25 17 suggest that reductions in fear (and actual crime) on campus may be J25 18 achieved through the design of micro-level physical J25 19 features.

J25 20 Introduction. Fear of criminal victimization threatens the J25 21 quality of life for many Americans (Gallup Poll, 1989). Almost half J25 22 of the U.S. population have reported feeling unsafe in areas within J25 23 a mile of their homes (National Opinion Research Center, 1987). J25 24 Many citizens have been found to feel unsafe in the neighborhoods J25 25 where they shop, work, go to school, and seek entertainment (Bureau J25 26 of Justice Statistics, 1984; Fisher, 1991).

J25 27 Fear of crime is also a significant problem on college J25 28 campuses, causing faculty, staff, students, and parents to demand J25 29 safer campuses (Gaines, 1989). Recent court rulings have held J25 30 universities liable for foreseeable victimizations (Raddatz, 1988), J25 31 and Congress has passed a law requiring colleges and universities J25 32 to publicly report their crime statistics (House Report 101-883, J25 33 Section 201-205, 1990). All of these factors pressure college J25 34 administrators to address the safety concerns and needs of the J25 35 university.

J25 36 To better understand causes of fear of crime, researchers have J25 37 examined demographics factors. They have found higher levels of J25 38 fear among socially or physically vulnerable individuals, such as J25 39 minorities, low-income people, women or the elderly, especially J25 40 after dark (Box, Hale, and Andrews, 1988; Skogan and Maxfield, J25 41 1981; Warr, 1984). Most young urban women, a group prominent on J25 42 many campuses, have been found to fear certain types of personal J25 43 crimes, such as rape (Warr, 1985); and when individuals are J25 44 fearful, they have been found to adopt different crime prevention J25 45 behaviors (Gates and Rohe, 1987; Lab, 1990).

J25 46 Researchers have also moved beyond demographics predictors to J25 47 examine contextual causes. In this regard, we see the occurrence of J25 48 victimization from two interrelated theoretical levels - one at the J25 49 macro-level and the other at the micro-level. From the macro-level J25 50 perspective, researchers have consistently found that J25 51 neighborhood-level problems, such as 'physical and social J25 52 disorders' or 'incivilities,' heighten fear of crime (Gates and J25 53 Rohe, 1987; Skogan and Maxfield, 1981; Wilson and Kelling, 1982). J25 54 While this macro-level perspective explains broad J25 55 characteristics of fear, it misses the micro-level cues in the J25 56 built environment that affect feelings of vulnerability and fear of J25 57 crime at a specific location. Yet, research has shown that J25 58 offenders rely on micro-level cues to select a suitable target (cf. J25 59 Taylor and Gottfredson, 1986). The cues indicate opportunities for J25 60 committing a crime. A sizable body of research has revealed links J25 61 between the built environment, crime, and crime prevention at the J25 62 street or block level (Jacobs, 1961; Jeffery, 1977), or the site J25 63 level, as in housing projects (Newman, 1972; Merry, 1981) or J25 64 schools (Hope, 1985).

J25 65 The role of the physical environment on university campuses has J25 66 been overlooked. Yet, campuses have high levels of crime and fear J25 67 of crime, and they often have macro-level conditions that lead to J25 68 these problems. Routine activity theory explains victimization as a J25 69 function of the routine activities of victims that place them at J25 70 risk (Cohen and Felson, 1979). Campuses exhibit all three risk J25 71 factors: the opportunity for victimization presented by potential J25 72 targets (students, faculty and staff), a supply of motivated J25 73 offenders nearby (neighborhoods with poor socio-economic J25 74 conditions), and poor guardianship (open access and varying levels J25 75 of security among students).

J25 76 With knowledge of how micro-level features affect crime and J25 77 fear of crime, university officials can institute design policies J25 78 to reduce the risk. They can plan, evaluate or alter campus J25 79 buildings and micro-level features to promote safety. Compared to J25 80 students and personnel, physical facilities are relatively J25 81 permanent. Thus, while educational programs only have short-term J25 82 value because they must be repeated for new students and personnel, J25 83 environmental design strategies can produce long-term positive J25 84 effects on crime and feelings of vulnerability.

J25 85 How can campuses be designed for safety? Unfortunately, there J25 86 has been little empirical evidence on the effects of the physical J25 87 arrangements on crime and fear of crime on campus. Most studies of J25 88 environmental strategies have concentrated on residential or J25 89 commercial areas (Kuskmuk and Whittemore, 1981). This paper draws J25 90 from both the macro- and micro-level perspectives to understand the J25 91 cues in the outdoor environment to feelings of vulnerability and J25 92 reactions to that fear. We present a theoretical model of how J25 93 micro-level site features affect fear of crime and we describe the J25 94 results of a study aimed at uncovering some of those cues in J25 95 relation to a campus setting. In particular, we analyzed written J25 96 responses to open-ended questions in relation to three buildings on J25 97 a university campus. The results revealed three micro-level cues as J25 98 important influences on site-specific fear of crime: refuge, escape J25 99 and prospect (depending on different light levels). Furthermore, we J25 100 found evidence of changes in behavior in response to the J25 101 site-specific fears.

J25 102 Cues and reactions to fear of crime. The built J25 103 environment may go relatively unnoticed until it creates a J25 104 substantial problem or pleasure. Nevertheless, the design of the J25 105 built environment does influence perceptions of safety (Fisher and J25 106 Nasar, 1992) which in turn may affect psychological well-being, and J25 107 spatial behaviors (Nasar, 1988; Russell, 1989).

J25 108 In areas where crime and fear are present, people regularly J25 109 evaluate their risk of victimization by scanning their immediate J25 110 environment for cues of danger (Appleton, 1975; Gates and Rohe, J25 111 1987; Goffman, 1971; Warr, 1990). Based upon their assessment of J25 112 the presence or absence of such cues, they adjust their behavior J25 113 accordingly. In discussing cues for alarm, Goffman (1971) J25 114 observed,

J25 115 Individuals seem to recognize that in some environments J25 116 wariness is particularly important, constant monitoring and J25 117 scanning must be sustained, and any untoward event calls forth a J25 118 quick and full reaction.(242)

J25 119 Individuals do not necessarily notice all signals as cues for J25 120 alarm. They only attend to those cues in their immediate J25 121 environment as signals for alarm. According to Goffman,

J25 122 The individual's surround ... (is) that region around J25 123 him within which signs for alarm ... can originate ... This is J25 124 likely to be measured by means of a radius that is only yards long. J25 125 His body is what he mainly can become immediately concerned about, J25 126 and it is principally vulnerable ...(253-254).

J25 127 As we move from place to place, our immediate surroundings J25 128 continually change, changing the position or distance of different J25 129 signals of alarm. Consequently, different types of situations give J25 130 rise to alarm while others do not. For example, Goffman noted,

J25 131 Walking along the street, the individual's surround J25 132 follows him; walking around a room, it does not, or does so only in J25 133 a small degree. Note, the individual is likely to concentrate his J25 134 scanning for signs for alarm at the moment and place of his J25 135 entering a bounded area. For often it will be at the doorway that J25 136 he will have to notice alarming things if he is to notice them in J25 137 time. (255-256)

J25 138 When a person's proximate surroundings have a cue for alarm, J25 139 the person feels vulnerable and reacts. As Goffman puts it,

J25 140 ... when the subject senses that something is up, his J25 141 attention and concern are mobilized; adaptive behavior occurs if J25 142 the alarms proves<&|>sic! 'real,' but if reassuring information is J25 143 acquired, the alarm proving false, his concentration will decay J25 144 quickly. (262)

J25 145 In contrast, when a cue for alarm is absent, "When the J25 146 world immediately around the individual portends nothing out of the J25 147 ordinary," Goffman writes that the person,

J25 148 will sense that ... he is safe and sound to continue on J25 149 with that activity at hand with only peripheral attention given to J25 150 checking up on the stability of the environment.(239)

J25 151 According to Goffman, then, individuals monitor their J25 152 environment for signals of danger. When they detect a danger J25 153 signal, they react accordingly. Otherwise, when appearances are J25 154 "natural" or normal (23), they go about their routine.

J25 155 It becomes important, then to identify what cues convey the J25 156 message of alarm or danger, how such cues provoke fear, and what J25 157 responses then follow. In discussing the design of vulnerability, J25 158 Goffman (1971) introduced the concept of 'lurk lines'. These lines J25 159 demarcate zones that lie beyond or behind the individual's line of J25 160 sight. These zones have also been called 'blind' spots (Warr, J25 161 1990). Lurk lines or blind spots occur in many places - the areas J25 162 behind an open door, inside an unlocked closet, or around a sharp J25 163 bend in hallway<&|>sic!. You do not necessarily have to walk J25 164 through a physical obstruction to discover these areas. You can see J25 165 potential blind spots, in the near distance and anticipate their J25 166 likely character.

J25 167 The concept of lurk lines and blind spots conforms with J25 168 Appleton's (1975) discussion of prospect and refuge. Appleton J25 169 argued that for evolutionary reasons humans favor places that J25 170 afford prospect (an open view) and refuge (protection). In J25 171 Goffman's (1971) and Warr's (1990) terms, humans dislike lurk lines J25 172 or blind spots. Places with prospect and refuge offer an J25 173 observation point from which humans can see, react, and if J25 174 necessary, defend themselves, as well as a protective space to keep J25 175 them from harm. Appleton (1975) went on to say that humans need not J25 176 enter an area to determine it's prospect and refuge. They J25 177 anticipate what those qualities would be if they were to enter the J25 178 place. For example, observers can look at a clear-cut J25 179 mountain-top in the distance and anticipate that it would J25 180 offer them prospect but little refuge should they venture there.

J25 181 We believe that potential offenders also value places which J25 182 offer prospect and refuge. As Lorenz's (1964: 181) said, such J25 183 places offer an advantage to "the hunter and hunted alike - J25 184 namely to see without being seen." This agrees with views J25 185 on the role of natural surveillance in reducing crime (cf. Jacobs, J25 186 1961, Jeffery, 1977; Newman, 1972); and several studies indicate J25 187 that robbers favor prospect and refuge. Bank robbers have been J25 188 found to target banks that have poor visibility of the inside from J25 189 the outside, and clear views out to escape routes and to see who J25 190 might enter (Camp, 1968). Tapes of bank robberies show that robbers J25 191 select banks and follow a path that let them see as much as J25 192 possible while remaining unseen (Wise, 1983). In a study of 78 J25 193 tapes of sting operations, Archea (1985) found that robbers operate J25 194 from locations with prospect ("enough visual access to gain J25 195 and maintain control") and refuge ("low visual J25 196 exposure from areas not in direct control") (p. 249). Stoks J25 197 (1983) found that rape outdoors tend to occur in areas with refuge J25 198 ("physically confined" spaces) and limited prospect J25 199 ("barriers") nearby areas of pedestrian movement (p. 334). J25 200 In sum, the research confirms prospect and refuge as relevant J25 201 constructs in relation to site-specific opportunities for crime. It J25 202 was our hypothesis that prospect and refuge would also affect fear J25 203 of crime among passersby in outdoor settings.

J25 204 In theory, humans would feel vulnerable passing by places such J25 205 as alcoves and tall, dense shrubs that afford potential offenders J25 206 refuge (concealment) and potential victims limited prospect. J25 207 Offenders would have hiding places, into which passersby could not J25 208 see in time to escape or avoid an attack. Conversely, where J25 209 pedestrians have wide and deep prospect onto an area which has no J25 210 places of refuge for an offender, they would feel safer because J25 211 they could anticipate and avoid an attack.

J25 212 Expanding the notion of prospect, we argue that insufficient J25 213 lighting at night also provokes feelings of vulnerability because J25 214 it limits prospect. The idea that the night may be scary is not new J25 215 (Fisher and Nasar, 1992; Warr, 1985). Previous research has found J25 216 that darkness represents a potent sign of danger, and combined with J25 217 novelty and uncertainty can produce considerable fear (Warr, 1990). J25 218 We believe, however, that it is not only darkness per se, but also J25 219 its effects on prospect that effect site-specific fear of crime. J25 220 J26 1 <#FROWN:J26\>Population Decline from Two Epidemics on the J26 2 Northwest Coast

J26 3 ROBERT BOYD

J26 4 Direct contact by Indians of the Northwest Coast culture area J26 5 with Euro-Americans began very late, in 1774. Therefore, J26 6 documentation of the contact process is relatively good. It appears J26 7 that the introduction of Old World diseases was similarly late and J26 8 coincided with contact. Although Campbell (1989), using J26 9 archeological data, has made a case for penetration of the J26 10 Northwest Coast by the hemispheric smallpox epidemic of the 1520s, J26 11 all other evidence (archeological, historical, Indian oral J26 12 tradition) for Northwest disease introduction postdates 1774. That J26 13 evidence is considerable, with nearly the full list of high J26 14 mortality infectious diseases introduced in rapid fire sequence, J26 15 with accompanying heavy mortalities strongly suggesting that these J26 16 diseases were new to the area, or at least present again after a J26 17 very long absence.

J26 18 Smallpox epidemics appeared roughly every generation - the J26 19 first, in the late 1770s following direct contact, over the entire J26 20 coast, and others in 1801, 1836-1838, 1853, and 1862-1863 in more J26 21 limited regions (Boyd 1990:137-143). This interesting sequence J26 22 appears to have been created by the dual factors of introduction J26 23 from an outside source plus presence of a large enough pool of J26 24 non-immune susceptibles in the Northwest born since the J26 25 last epidemic. Evidence for depopulation from the initial outbreaks J26 26 is limited to descriptions of abandoned villages and native J26 27 recollections, but high mortalities from the later epidemics are J26 28 well documented in the Indian censuses and estimates compiled by J26 29 the Hudson's Bay Company and by early government officials. Besides J26 30 smallpox, malaria was introduced to and became endemic in a sizable J26 31 portion of the southern coast, and records indicate considerable J26 32 population decline concurrent with its introduction. Other 'new' J26 33 diseases that contributed to Northwest Coast Native American J26 34 population decline included measles, influenza, dysentery, whooping J26 35 cough, and (in a different way) tuberculosis and venereal diseases J26 36 (Boyd 1985, 1990).

J26 37 This is the situation for the region as a whole. This chapter J26 38 will discuss the records on population decline associated with two J26 39 localized outbreaks - on the lower Columbia in the decade following J26 40 the introduction of malaria in 1830, and in the Queen Charlotte J26 41 Islands during the years of the last great smallpox epidemic, J26 42 1862-1863. The two epidemics appear to qualify as virgin soil or J26 43 near virgin soil; malaria had not been known in the Northwest prior J26 44 to 1830, and smallpox had been absent from the Charlottes since the J26 45 late 1770s. The data on depopulation in these two instances are J26 46 remarkably good and may provide clues to patterns of depopulation J26 47 in similar virgin soil situations in other places and at earlier J26 48 dates in the Americas.

J26 49 MALARIA ON THE LOWER COLUMBIA, AFTER 1830

J26 50 The Lower Columbia drainage, here defined as the area from the J26 51 river mouth to near The Dalles and including the basins of large J26 52 tributaries such as the Willamette and Cowlitz, was home to several J26 53 Indian groups. The two largest, Chinookans on the banks of the J26 54 Columbia and Kalapuyans in the Willamette Valley, are also the most J26 55 completely documented as far as population decline is concerned; J26 56 and they are used as the base of discussion here.

J26 57 In the late 1820s the population of these two peoples J26 58 approximated 18,580; within a decade it had dropped to 2,433, or 13 J26 59 percent of the initial number. Detailed discussions of the sources J26 60 for these numbers have been given elsewhere (Boyd 1985, 1990, J26 61 1991a; Boyd and Hajda 1987); suffice it to say that the preepidemic J26 62 estimates are from Meriweather Lewis and William Clark and the J26 63 Hudson's Bay Company, and the postepidemic numbers are from the J26 64 1841 U.S. Exploring Expedition.

J26 65 What caused this monumental decline in Lower Columbia J26 66 populations? The contemporary sources (a large body of data) J26 67 uniformly ascribe it to the disease called 'fever and ague' by J26 68 Americans and 'intermittent fever' by the British. Despite some J26 69 controversy over the identity of this disease (e.g., Cook 1955; J26 70 Taylor and Hoaglin 1962), epidemiological evidence strongly J26 71 suggests that it was malaria (Boyd 1975), described most often in J26 72 later accounts (e.g., Townsend 1839:197; Brackenridge 1931:141, J26 73 221; J.R. Dunn 1846) as the tertian form (Plasmodium J26 74 vivax and P. falciparum). The sources are unanimous J26 75 in stating that this disease was new to the area, and unknown J26 76 before 1830 (e.g., McLoughlin 1941:88). This points to a virgin J26 77 soil situation. Malaria, as has been fully established by research J26 78 (F. Dunn 1965; Wood 1975), is an Old World disease, which was J26 79 introduced in sequence to different regions of the Americas, where J26 80 suitable vectors, mosquitoes of the genus Anopheles, were J26 81 native (Harrison 1978). Once introduced, it became endemic. In the J26 82 unbroken tropical and subtropical regions of the Amazon, J26 83 circum-Caribbean, and Southeast United States this introduction was J26 84 early and apparently has gone unrecorded (Ashburn 1947:112-123; J26 85 Friedlander 1969). In the Lower Columbia and adjacent California J26 86 central valley (Cook 1955) malaria arrived late and was both J26 87 observed and recorded.

J26 88 The statistics from the area are unambiguous in documenting a J26 89 dramatic population decline following the introduction of 'fever J26 90 and ague.' What remains unclear are the dynamics of this decline. J26 91 Such high mortalities are not characteristic of endemic malarial J26 92 regions today, nor of those occasional instances where a J26 93 combination of factors inflate mortality rates in endemic areas to J26 94 numbers that merit the term 'epidemic' (Wood 1979:260). It is J26 95 essential to remember that on the Lower Columbia the outbreak was a J26 96 virgin soil epidemic, and virgin soil outbreaks of whatever disease J26 97 are almost by definition unusually virulent (Crosby 1976). This may J26 98 be part of the answer to the large decline. Three other J26 99 contributing factors are: the Indian means of dealing with the J26 100 alternating spells of chills and fever characteristic of the J26 101 disease, secondary diseases that precipitated mortality, and J26 102 fertility decline directly or indirectly associated with J26 103 malaria.

J26 104 A small, consistent, yet independently written body of sources J26 105 from the earliest years of the epidemic ascribe the high mortality J26 106 to inadequate treatment of the disease. During the fever spells, J26 107 Indians would plunge into cold water; when chills came they J26 108 retreated to sweat lodges. Sudden death followed. The phenomenon is J26 109 securely documented; the actual dynamics of death are not (Boyd J26 110 1979). Quinine sulfate, the medicine used by Whites, was in short J26 111 supply in the early 1830s (Allan 1882:79; McLoughlin 1941). Factor J26 112 two: Modern malaria is a disease of complications; death, when it J26 113 comes, is usually from a secondary illness superimposed on a J26 114 weakened malarial constitution (Brown 1983:65). Children, in J26 115 endemic regions, form the bulk of casualties (Wood 1979:258-259). J26 116 Though evidence is sparse, deaths listed in the record book of an J26 117 1830s Willamette Valley mission school, for instance are ascribed J26 118 to such secondary diseases (including influenza and tuberculosis) J26 119 introduced to Indian children chronically ill from malaria (Shepard J26 120 1922). It seems likely that much of the mortality in later years J26 121 may have been due to secondary illnesses.

J26 122 These two factors, in combination, appear to have contributed J26 123 to the mortality side of the population equation in the Lower J26 124 Columbia valley. But it appears that there may have been a decline J26 125 in fertility as well in this newly malarial population. J26 126 Post-malaria Chinookan and Kalapuya censuses (Hudson's Bay J26 127 Company 1838; Spalding 1851; Boyd 1985: chart 26) incorporate J26 128 information on population structure, specifically age and sex. J26 129 Censuses of groups within the epidemic focal area show low - below J26 130 30 - percentages of children. Elsewhere in the Northwest, among J26 131 Indian populations counted by the Hudson's Bay Company but not J26 132 subjected to disease, the percentages are much higher, generally in J26 133 the 30-40 percent range. In the case of the 1851 counts, mortality J26 134 from the 1848 measles epidemic (Boyd 1991b) decreased the J26 135 percentage of children even more. But in 1838 and 1851, two J26 136 underlying explanations for fertility decline also seem likely. J26 137 First, in immediate postepidemic years (MacArthur 1961:8), there J26 138 was a sudden drop in fertility due to disrupted marital units. J26 139 Second, there was considerable malaria-caused anemia, which, among J26 140 women of child-bearing age, led to spontaneous abortions and J26 141 stillbirths (Wood 1979:258).

J26 142 The approximate 87 percent decline in Lower Columbia Indian J26 143 populations in the decade following the introduction of malaria J26 144 therefore appears not to have been a simple process. Inappropriate J26 145 Indian treatments precipitated deaths in cases where proper J26 146 treatment would have resulted only in temporary debilitation; J26 147 secondary ailments moved in and pushed weakened malarial cases on J26 148 to death, particularly among the very young; and maternal anemia is J26 149 likely to have caused spontaneous abortions and miscarriages, J26 150 resulting in a drop in fertility. The decline was regular and J26 151 cumulatively dramatic: in 1841 an American physician, resident J26 152 since 1835, estimated that the population of the Willamette Valley J26 153 Indians diminished by one-quarter annually (Bailey in Wilkes J26 154 1926:57), a proportion fully in keeping with before (late 1820s) J26 155 and after (1841) Kalapuyan population figures.

J26 156 Socially, after 1841, most Chinookan and Kalapuyan populations J26 157 were approaching extinction as viable and ethnically distinct J26 158 entities. Inhabitants of Chinookan winter villages, tied to their J26 159 local land base, remained in place with minimal regrouping (Dart J26 160 1851) until they were reduced to a few survivors or overwhelmed by J26 161 migrants from more robust populations (Sahaptin or Salishan) from J26 162 the peripheries of the endemic malaria zone (Tappan 1854). The 1851 J26 163 Kalapuya census showed that these people likewise remained in their J26 164 local band territories, with negligible regrouping. The average J26 165 band size in 1851 was a mere 53 people, and the bands were evenly J26 166 spread over the entire valley. As on the Columbian, more robust J26 167 outsiders (Sahaptin Klikitat) had begun to move in, and by 1851 J26 168 they were equivalent in numbers to the sum total of the Willamette J26 169 Valley indigenes. The Kalapuyan population nadir was reached in the J26 170 1850s, and numbers (in what was a highly mixed population) began to J26 171 increase again only after enforced concentration on the Grand Ronde J26 172 Reservation and regular access to Western medicines.

J26 173 SMALLPOX IN THE QUEEN CHARLOTTE ISLANDS, 1862-1863

J26 174 The steady population decline of the Lower Columbia Indians J26 175 following the introduction of malaria was very different in nature J26 176 from the sudden catastrophic drop of the Haida population of the J26 177 Queen Charlottes caused by the 1860s smallpox epidemic, even though J26 178 the percentage losses were comparable.

J26 179 The 1862-1863 British Columbia smallpox epidemic has been J26 180 documented (Yarmie 1968; Pethick 1978). Over 20,000 Indians, nearly J26 181 60 percent of the pre-epidemic total, died in British Columbia and J26 182 the Alaska Panhandle (Duff 1964; Boyd 1985, 1990; 142-144). In J26 183 classic fashion, the disease arrived with an infected individual on J26 184 a ship from San Francisco. The ship docked at Victoria, and the J26 185 disease spread rapidly to the crowded Indian encampment on the J26 186 city's outskirts. Instead of quarantining the Indians, the J26 187 authorities evicted them, and fleets of Haida, Tlingit, Tsimshina, J26 188 and Kwakiutl traders sailed back to their homelands, taking the J26 189 epidemic with them.

J26 190 On the densely populated (relatively speaking) and isolated J26 191 Queen Charlotte Islands, the effect was rapid and devastating. A J26 192 Hudson's Bay census dating from 1839-1842 showed a total of 8,428 J26 193 Haidas (table 1); there is no evidence that this number changed J26 194 significantly in the next 20 years. The people were concentrated in J26 195 several winter villages spread along the coastline in areas of J26 196 heavy resource concentration (especially halibut, the local J26 197 staple). Each village was associated with a particular J26 198 matriclan.

J26 199 table&caption

J26 200 Census of the Haida taken 1882-1884 (Chittenden 1884; Petroff J26 201 1882) show 1,598 survivors, or a population drop of over 80 J26 202 percent, nearly all of which must on present evidence be assigned J26 203 to the smallpox epidemic. This exceedingly large loss was due to J26 204 several factors. First, the strain of smallpox virus in the 1860s J26 205 epidemic was particularly virulent; second, Haida settlement was J26 206 (relatively) dense and continuous, facilitating transmission; J26 207 third, the Queen Charlotte population was particularly vulnerable, J26 208 as there was no segment with acquired immunity, as was the case J26 209 with most other British Columbia Indian populations. Whereas J26 210 Tsimshian, Tlingit, and others had experienced an outbreak 24 years J26 211 prior, the Haida had not known the disease for more than 90 years J26 212 (Boyd 1985; Blackman 1981:23; cf. MacDonald 1983:17). It was, in J26 213 essence, a virgin soil experience for them. A fourth factor was J26 214 vaccine. While sizable numbers of Tlingit and some Kwakiutl had J26 215 been vaccinated in the 1830s, and the Metlakatla Tsimshian in 1862, J26 216 none of the Haida had been vaccinated (Boyd 1985, chapter 4 and J26 217 appendix).

J26 218 J27 1 <#FROWN:J27\>The Division of Household Labor

J27 2 The most obvious predictors of wives' perceptions of fairness J27 3 might seem likely to be the outcomes related to the quantity of J27 4 labor, such as the extent of the husband's contribution to J27 5 household labor and the total amount of labor performed by the J27 6 wife. Benin and Agostinelli (1988) have shown that women are J27 7 relatively unaffected by increased household labor participation on J27 8 the part of husbands unless the labor is specifically spent in J27 9 those tasks traditionally defined as 'feminine' chores, such as J27 10 cooking or cleaning. What might explain this pattern? First, the J27 11 traditional male tasks involve little investment of time relative J27 12 to the female tasks. Second, male involvement in female tasks is J27 13 likely to have direct and visible consequences for the wife's J27 14 workload. Third, since husbands' contributions to such tasks cross J27 15 traditional gender lines, they will be more salient and more likely J27 16 to be perceived as involving a special effort by husbands to be J27 17 'fair.' We thus expect perceptions of fairness to be affected by J27 18 husbands' participation in 'female' household tasks, but not by J27 19 their participation in 'male' or gender neutral tasks. Since our J27 20 arguments for this effect hinge in part on the reduced need for J27 21 wives' labor, we also predict a direct effect of wives' total labor J27 22 on perceptions of fairness.

J27 23 However, we do not expect even this aspect of husbands' J27 24 household labor contribution to be related to perceptions of J27 25 fairness in all families. As Thompson (1991) rightly emphasizes, J27 26 women's orientations to various household outcomes vary J27 27 considerably. It is still the case that many families allocate J27 28 labor market participation exclusively to husbands (for the sample J27 29 utilized here, about a third of the families have nonemployed J27 30 wives), with household labor presumably becoming the wife's J27 31 responsibility. For such families, it is unlikely that husbands' J27 32 participation in housework would have much impact on perceptions of J27 33 fairness. Hence, all of our analyses will be carried out separately J27 34 for families in which the wife is employed and those in which she J27 35 is not. We expect that the quantity of labor variables will have J27 36 their clearest impact for families in which the wife is J27 37 employed.

J27 38 Qualitative Aspects of Household Labor

J27 39 Thompson has reasonably argued that quantitative aspects of the J27 40 division of labor or "the distribution of time and J27 41 task" do not tap the full range of possibly important J27 42 outcomes. For example, Oakley (1974) found that wives who were J27 43 dissatisfied with the division of labor in their home were likely J27 44 to complain about the monotony and fragmentation of their chores. J27 45 Dissatisfied wives were also likely to complain about the J27 46 loneliness of household labor; that is, they expressed concern over J27 47 the lack of companionship during the performance of household J27 48 labor. Thus, although the relationship between 'satisfaction' and J27 49 perceptions of fairness may be complicated, it is reasonable to J27 50 hypothesize, on the basis of Oakley's findings, that wives' J27 51 perceptions of the extent to which their household labor is lonely J27 52 and complicated will be related to their perceptions of the J27 53 fairness of the division of labor.

J27 54 Appreciation J27 55 Kessler and McCrae (1982) conclude that husbands' willingness J27 56 to participate in housework is important to wives' in part because J27 57 it carries a 'symbolic meaning' for wives, such that their work is J27 58 recognized and appreciated by the husbands. Thompson (1991) also J27 59 argues that among the most important outcomes of the division of J27 60 household labor are symbolic outcomes, particularly the J27 61 significance of caring. We would argue that the 'satisfaction' J27 62 involved in such a demonstration will be greater when wives J27 63 perceive that their work is, indeed, appreciated. We expect, J27 64 therefore, that women who do not perceive their household labor to J27 65 be appreciated will be more likely to feel that the division of J27 66 labor is unfair, and that this effect should be evidenced whether J27 67 or not she is employed.

J27 68 Ideological Factors

J27 69 What role might we expect gender ideology to play in women's J27 70 reactions to division of house-hold labor? A number of J27 71 authors (e.g., Huber and Spitze, 1983) assume that gender J27 72 ideologies are an important determinant of the allocation of chores J27 73 within the home, and we will be able to test this assumption. J27 74 However, since task allocation certainly takes place through a J27 75 negotiation process (Atkinson & Huston, 1984), and one that might J27 76 be somewhat protracted (Hochschild, 1989) there will undoubtedly be J27 77 a far from perfect relationship between gender ideology and the J27 78 division of labor. Given the general finding that men, on average, J27 79 contribute very little to house-hold labor, it is J27 80 reasonable to assume that there will be some direct relationship J27 81 between women's gender ideology and their assessment of fairness in J27 82 the division of labor.

J27 83 Additionally, however, we would expect ideology to be related J27 84 to all three of the factors in Thompson's (1991) model. First, J27 85 husbands' contributions to the division of household labor should J27 86 be a particularly salient outcome for less traditional wives. J27 87 Second, less traditional wives should be more inclined to make the J27 88 cross-gender comparisons that would produce dissatisfaction with J27 89 husbands' lack of participation. Third, less traditional wives will J27 90 presumably be less likely to accept the traditional justification J27 91 for low levels of husbands' contributions to household work. Thus, J27 92 we would expect an interaction effect involving ideology and J27 93 husband's contributions, such that women with egalitarian J27 94 ideologies will react strongly to the extent of their husband's J27 95 participation, while those with more traditional ideologies will J27 96 not. And we would expect this effect to be particularly clear for J27 97 those families in which both partners participate in the paid labor J27 98 force.

J27 99 DATA AND VARIABLES

J27 100 Data for this study are taken from the 1988 National Survey of J27 101 Families and Households (Sweet, Bumpass, & Call, 1988). The NSFH J27 102 provides a cross-sectional national sample of 13,017 respondents J27 103 aged 19 and older. The sample here is limited to 778 married white J27 104 women for whom data were complete across the variables required J27 105 here-in. Our smaller sample size results from the J27 106 restriction of the study to only white, currently married wives for J27 107 whom data on both attitudinal and household labor measures are J27 108 complete. The primary limitations on sample size result from the J27 109 use of only married households (68% of the total sample), women J27 110 (50% of the total sample), whites (85% of the total sample), and J27 111 individuals less than 65 years of age (86% of the total sample). J27 112 The sample was also restricted to include only respondents who J27 113 reported a weekly household labor performance of 120 hours or J27 114 less.

J27 115 Household labor. For each couple, the NSFH includes J27 116 reports from each partner/spouse regarding their own household J27 117 labor contributions and those of their partner/spouse. Each partner J27 118 provides an estimate of the hours spent per week on eight specific J27 119 household tasks. The tasks are: (a) preparing meals, (b) washing J27 120 dishes, (c) cleaning house, (d) outdoor tasks, (e) shopping, (f) J27 121 washing and ironing clothes, (g) paying bills, and (h) auto J27 122 maintenance. Given the focus of this investigation on wives' J27 123 reactions to the division of labor, measures of both husbands' and J27 124 wives' household labor are based on reports from wives.

J27 125 Two primary aspects of the quantitative division of household J27 126 labor are of central interest here: how much work is done by J27 127 husbands and how much by wives. First, with regard to husbands' J27 128 household labor, our major focus will be on male labor in J27 129 female-dominated tasks, the number of hours spent by husbands J27 130 in those chores typically envisioned as 'female' tasks (meal J27 131 preparation, dishes, cleaning house, and ironing and washing J27 132 clothes), as measured by wives' estimations. We will also begin our J27 133 analysis of the effects of male labor on wives' perceptions of J27 134 fairness with a demonstration that male labor in J27 135 male-dominated tasks (outdoor tasks and auto maintenance) and J27 136 male labor in gender neutral tasks (shopping and paying J27 137 bill) have little to do with perceptions of fairness. Second, J27 138 total female labor is the total number of hours spent in J27 139 all household labor per week by wives.

J27 140 Qualities of household labor. Three measures of J27 141 wives' assessment of the nature of their household labor were J27 142 available in the NSFH. Responses were taken from the question: J27 143 "How would you describe the work you do around the house? J27 144 Would you say it is: (1) boring-interesting, (2) J27 145 complicated-simple, (3) lonely-sociable?" Responses were J27 146 taken on a 7-point scale, with the responses shown above J27 147 representing the respective end points of each scale.

J27 148 Appreciation. The extent to which women perceive their J27 149 household labor to be appreciated is based on the same question J27 150 used to assess the qualities of household labor. One of the 7-point J27 151 scales presented in the NSFH was J27 152 "unappreciated-appreciated."

J27 153 Gender ideology. An indexed measure of wives' J27 154 sex-role ideology was created from responses to the J27 155 following questions: (a) It is much better for everyone if the man J27 156 earns the main living and the woman takes care of the home and the J27 157 family, (b) Preschool children are likely to suffer if their mother J27 158 is employed, (c) Parents should encourage just as much independence J27 159 in their daughters as in their sons, and (d) In a successful J27 160 marriage, each partner must have the freedom to do what they want J27 161 individually. Wives answered each question on the basis of a J27 162 5-point scale, ranging from strongly agree to strongly disagree. J27 163 Each item was coded appropriately, with a high score indicating J27 164 egalitarian sex-role orientations. Coefficient alpha for the scale J27 165 was .55.

J27 166 A measure of ideology focused more directly on the division of J27 167 labor, family-labor ideology, was taken from responses to J27 168 the question: "If a man and a wife both work full-time, J27 169 they should share household tasks equally." Here, responses J27 170 ranged across a 5-point scale, with a higher score indicating J27 171 egalitarian family-role orientations.

J27 172 Employment. The employment variable is based on a question J27 173 regarding number of hours per week spent in the paid labor force. J27 174 Our first intention was to separate the wives into three groups J27 175 (employed full time, employed part time, and nonemployed), but for J27 176 these data less than 10% of the women were employed fewer than 35 J27 177 hours per week. We decided, therefore, to combine all 'working' J27 178 women, and the employed group includes all women who report any J27 179 hours of paid employment.

J27 180 Perceptions of fairness. Wives' perception of J27 181 fairness of the division of labor was measured with the J27 182 following question: "How do you feel about the fairness in J27 183 your relationship in each of the following areas? (household J27 184 chores)." Responses were on a 5-point scale, ranging from J27 185 "very unfair to me" to "very unfair to J27 186 him." A code of 5 indicates "very unfair to my J27 187 partner," 4 is "somewhat unfair to my J27 188 partner," 3 corresponds to a response of "fair," 2 J27 189 represents "somewhat unfair to me," and 1 is J27 190 "very unfair to me." Since very few wives J27 191 responding perceived the division of labor to be unfair to their J27 192 partner (see below), we will refer to higher scores on this J27 193 variable as indicating "fairness."

J27 194 table&caption

J27 195 Among the 496 employed wives in this sample, 61% perceive the J27 196 division of labor to be fair, 28% "somewhat unfair to J27 197 me," 6% "very unfair to me," and about 4% J27 198 as either "somewhat" or "very" unfair to him; the J27 199 mean is 2.64. Among the 282 nonemployed wives, 71% report a J27 200 "fair" division of labor, 24% see it as "somewhat J27 201 unfair to me," 2% as "very unfair to me," J27 202 and 3% as somewhat or very unfair to him; the mean is 2.76.

J27 203 RESULTS

J27 204 Division of Labor

J27 205 Table 1 presents the mean hours per week spent in household J27 206 labor, by each spouse, across the eight household chores, as well J27 207 as their percentage distribution. As shown, wives and husbands J27 208 perform significantly different total amounts of labor and J27 209 different types of labor. In those households in which the wife is J27 210 employed, wives average 31.06 hours per week of total labor, while J27 211 their husbands average only about 15.28. The disparity between male J27 212 and female total labor is, of course, even greater within those J27 213 households in which the wife is nonemployed. Here, wives perform J27 214 approximately 42.04 hours per week of total labor, while their J27 215 husbands average 12.23 hours per week. Obviously, the employment J27 216 status of wives has a significant impact on their total labor in J27 217 the home; their employment status has a substantially weaker J27 218 association with husbands' household labor. J27 219 J27 220 J28 1 <#FROWN:J28\>Rather than seek integration of the best elements of J28 2 each model, model-based professional organizations place behavioral J28 3 scientist-practitioners in a multiple-choice stance in which they J28 4 are encouraged to affiliate with one model.

J28 5 Furthermore, many literature reviews, such as those published J28 6 in Psychological Bulletin and Clinical Psychology J28 7 Review organize their presentations around separate causal J28 8 models and evaluate their comparative efficacy. This strategy also J28 9 places causal models in a competitive stance and fails to J28 10 acknowledge that behavior disorders may be a function of several J28 11 classes and levels of causal variables, either concurrently or J28 12 under different conditions, and that some variables and causal J28 13 relationships in different causal models may be compatible and J28 14 equally valid. As Blalock (1964) noted, this leads to J28 15 "jurisdictional disputes" among proponents of J28 16 models that are not necessarily incompatible.

J28 17 Causal models can also be artificially contrasted by focusing J28 18 on only one parameter of a behavior disorder (e.g., focusing only J28 19 on the 'occurrence' of depression). This narrow focus is limiting J28 20 in that it fails to recognize that different parameters of a J28 21 disorder are affected by different variables. The cognitive model J28 22 of depression, presented in chapters 1 and 2, provides a good J28 23 example. An examination of the published longitudinal research J28 24 might suggest that "beliefs of helplessness" should J28 25 be minimized as an explanatory concept for the occurrence of J28 26 depressive symptoms because the evidence that it reliably precedes J28 27 the onset of depressive episodes is weak. However, if we focus on J28 28 the magnitude or duration of depressive symptoms, the J28 29 research is more supportive of a causal role for "learned J28 30 helplessness" (Barnett & Gotlib, 1988).

J28 31 'Alternative' causal models are sometimes simply different J28 32 levels of explanation for the same phenomena. We can base J28 33 a useful causal model for lamp illumination on either initiating J28 34 electron flow through a resistor or on the behavior of throwing a J28 35 light switch. Both can be useful, are not incompatible, and are J28 36 simply different levels of explanation. Similarly, 'learned J28 37 helplessness beliefs,', 'neurotransmitter deficits,' and 'social J28 38 rejection experiences' need not be incompatible causal concepts for J28 39 depression (Beck & Young, 1985; Boyd & Levis, 1980; Coyne, Kahn, & J28 40 Gotlib, 1987; Lewinsohn & Hoberman, 1982). They may reflect only J28 41 different labels, different measurement procedures, and different J28 42 levels of foci for the same phenomena.

J28 43 The relative validity or power of causal models may vary across J28 44 domains or conditions. For example, between-spouse communication J28 45 difficulties may be an important factor in marital distress, but J28 46 significantly less so in the context of a supportive extended J28 47 family. Similarly, the relative importance of diet and life J28 48 stressors as determinants of hypertension risk may differ between J28 49 European-Americans and African-Americans, or between younger and J28 50 older adults. The various models, however, should not be viewed as J28 51 necessarily competing but as potentially complimentary; elements of J28 52 each may be necessary to account for causal relationships in J28 53 different domains.

J28 54 Unspecified Causal Mechanisms

J28 55 Because causal inference demands a logical connection between J28 56 variables, and because causal mechanisms are important in treatment J28 57 decisions, the mechanisms of action of a causal variable (i.e., how J28 58 or in what manner the causal variable exerts its influence) must be J28 59 identified or at least amenable to reasonable hypotheses (Hyland, J28 60 1981). For example, we can hypothesize that aerobic conditioning J28 61 can reduce premenstrual distress (Gannon, 1985), hypertension J28 62 (Danforth, et al., 1990), depression (Doyne, Chambless, & Beutler, J28 63 1983), heart attacks (Dubbert, Rappaport, & Martin, 1987), and J28 64 obesity (Brownell & Foreyt, 1985). But, as noted in chapter 1, it J28 65 is important that the mechanisms of these hypothesized causal J28 66 relationships be identified. 'How does exercise reduce J28 67 premenstrual distress or hypertension?'

J28 68 Causal relationships with unspecified mechanisms may have J28 69 predictive and clinical efficacy. Consider the clinical utility of J28 70 a reliably identified causal relationship between aerobic J28 71 conditioning and hypertension, even if the causal mechanism is J28 72 unknown. However, the utility of any hypothesized causal J28 73 relationship is limited if the mechanism that underlies that J28 74 relationship is not articulated. Knowledge of a causal mechanism J28 75 helps us to identify the 'active components' of a causal J28 76 relationship, to refine our intervention procedures, and to develop J28 77 more effective and efficient interventions. Going back to our J28 78 exercise example, it is useful to know that moderate levels of J28 79 aerobic exercise can reduce premenstrual distress for some women J28 80 (Gannon, 1985). However, if we also identify a possible causal J28 81 mechanism for that relationship (e.g., that aerobic conditioning J28 82 increases receptor-site sensitivity to endogenous opiates that may J28 83 be released during premenstrual phases or affects prostaglandin J28 84 release), we can possibly develop other interventions that operate J28 85 through the same mechanism.

J28 86 However, the question of 'how' an hypothesized causal variable J28 87 affects the target behavior is too infrequently addressed in causal J28 88 models of behavior disorders. How does childbirth sometimes lead to J28 89 postpartum depression (Atkinson & Rickel, 1984)? How does a severe J28 90 threat to self-esteem sometimes lead to paranoid ideation (Haynes, J28 91 1986b)? How does sexual abuse of a child sometimes lead to J28 92 disturbed interpersonal relationships as an adult (Harter, J28 93 Alexander, & Neimeyer, 1988)? How do self-efficacy beliefs J28 94 sometimes influence a person's social interactions (Bandura, J28 95 1977a)? How does an 'accepting' and 'positive' client-therapist J28 96 relationship sometimes lead to a reduction in interpersonal J28 97 anxiety?

J28 98 In summary, many causal models of behavior disorders have J28 99 unspecified causal mechanisms. As a result, these models will (1) J28 100 be less likely to evolve into more powerful models, (2) have a J28 101 limited impact on our ability to explain behavior, and (3) have J28 102 limited utility for the development of more effective J28 103 interventions.

J28 104 Unacknowledged Domains

J28 105 As indicated in chapter 2, causal relationships are never J28 106 unconditional. They have domains (i.e., boundaries or necessary J28 107 conditions) outside of which the causal relationships are no longer J28 108 valid. In psychopathology, these domains include population J28 109 characteristics, internal states, developmental stages, variable J28 110 values, temporal factors, and environmental contexts. Domains may J28 111 apply to the types of causal variables related to the behavior J28 112 disorder as well as the strength and form of causal J28 113 relationships.

J28 114 Many causal models of behavior disorders fail to specify their J28 115 domains adequately. For example a causal model that suggests that J28 116 self-efficacy beliefs affect the probability of posttreatment J28 117 relapse (Marlatt, Baer, Donovan, & Kivlahan, 1988) should carefully J28 118 delimit the social context in which that proposed relationship is J28 119 applicable. Similarly, hormonal models of gender behavior should J28 120 specify the environmental conditions in which hormonal factors will J28 121 influence sexual orientation (Ellis and Ames, 1987). Also, it is J28 122 important to stipulate the conditions under which stimulus pairings J28 123 will produce a classically conditioned response (Rescorla, 1988; J28 124 Papini & Bitterman, 1990), defensiveness and withdrawal will lead J28 125 to long-term marital distress (Gottman & Krorkoff, 1989), a single J28 126 exposure to a stimulus will produce a phobic response (McNally, J28 127 1987), and social support will mediate the impact of environmental J28 128 stressors (Alloway & Bebbington, 1987).

J28 129 Unacknowledged domains can lead to several conceptual and J28 130 methodological errors. First, a causal relationship may erroneously J28 131 be presumed to be absent. This can occur when the causal model is J28 132 tested outside of its domain of operation. Second, domains can J28 133 provide information about causal mechanisms and an unspecified J28 134 domain restricts access to this information. For example, J28 135 long-duration and short-duration stressors may have J28 136 opposite effects on the immune system. Some studies have suggested J28 137 a strengthening of the immune system during brief stress and a J28 138 weakening during protracted stress (Miller, 1983), illustrating a J28 139 'chronicity' domain for the effects of the stressors. Examining J28 140 causal relationships within these chronicity domains may lead to a J28 141 better understanding of causal mechanisms involved in immune system J28 142 deficiencies. Therefore, unacknowledged domains restrict the J28 143 evolution of causal models.

J28 144 Third, undefined domains for causal relationships can lead to J28 145 inappropriate clinical applications. If negative outcome J28 146 expectancies contribute to the maintenance of depression, but not J28 147 its onset (a behavior disorder parameter domain), cognitive J28 148 intervention efforts may be more effective if used during J28 149 depressive episodes to reduce its duration than if used as a J28 150 prevention strategy between depressive episodes.

J28 151 Excessively High Level

J28 152 As noted in chapters 1 and 2, causal models of a behavior J28 153 disorder can be expressed at different levels. Furthermore, various J28 154 levels of causal models can be clinically and empirically useful, J28 155 depending on their intended application. Consider the utility of J28 156 both political-sociological and biological models of J28 157 post-traumatic stress disorders of Vietnam veterans. J28 158 However, there is probably a hyperbolic functional relationship J28 159 between the level of a causal model and its utility in that J28 160 excessively low and excessively high levels often have diminished J28 161 clinical utility for treating and preventing behavior disorders as J28 162 well as diminished predictive and explanatory utility. However, J28 163 behavioral scientists-practitioners err more frequently by J28 164 proposing excessively high-level models.

J28 165 As pointed out earlier the difficulty with a high-level causal J28 166 variable is that it includes many lower-level variables. Therefore, J28 167 higher-level variables and models do not permit distinction among J28 168 the multiple possible causal variables and paths that they J28 169 represent. For example, finding that 'marital distress' has a J28 170 strong causal relationship to a behavior disorder, such as J28 171 depression or alcoholism, has limited clinical utility because this J28 172 variable does not identify the specific forms, mechanisms, or J28 173 parameters responsible for the causal relationship. The active J28 174 elements in 'marital distress' can include hostile verbal J28 175 exchanges, physical abuse, sarcastic comments, defensiveness, J28 176 withdrawal, infrequent presence in the home, feeling unloved, J28 177 anxiety and emotional arousal, or lack of positive statements to J28 178 the spouse. Therefore, attributing a behavior disorder to 'marital J28 179 distress' provides only an array of possible causal variables J28 180 and paths.

J28 181 High-level behavior-disorder constructs are also problematic. A J28 182 causal model for 'anxiety' is insufficiently specified because we J28 183 do not know if the model applies to physiological, subjective, J28 184 cognitive, or behavioral components of anxiety or to its onset, J28 185 magnitude, or duration (Bernstein, Borkovec, & Coles, 1986).

J28 186 Personality variables have played a prominent role in the J28 187 behavioral sciences but are particularly vulnerable to the J28 188 criticism of being dysfunctionally high level. Constructs such as J28 189 'locus of control,' 'hardiness,' 'authoritarianism,' 'need for J28 190 dependency,' 'assertiveness,' 'self-esteem,' 'Type A behavior J28 191 pattern,' 'emotional adjustment,' 'self-respect,' 'sexuality,' and J28 192 'need for achievement,' include so many possible lower-level J28 193 variables that they are rendered scientifically and clinically J28 194 debilitated. Personality variables suffer from an added J28 195 disadvantage as causal variables because they usually imply J28 196 stability across time and conditions and, therefore, are less J28 197 useful as explanations of variance in the parameters of behavior J28 198 disorders (Epstein, 1979, 1980).

J28 199 Less molar causal variables are also open to this criticism. J28 200 For example, it has been reliably demonstrated that there is a J28 201 causal association between 'exercise' and a variety of behavior J28 202 problems (e.g., response to stressors, hyper-tension). J28 203 However, at a lower level of analysis, it is apparent that some J28 204 types of exercise (i.e., weight lifting) can increase blood J28 205 pressure while others (e.g., jogging) can decrease it; exercise J28 206 early in the day can facilitate sleep onset while exercise later in J28 207 the day can inhibit it, and the mediating effects of exercise on J28 208 responses to psychosocial stressors may be related to the length of J28 209 time a person has been exercising. Again, the concept of 'exercise' J28 210 can be suggestive of causal relationships but requires lower-level J28 211 specification to facilitate the design of intervention programs.

J28 212 As with other elements of limited causal models, excessively J28 213 high-level causal variables also promote inferential and J28 214 measurement errors. Proposing a causal relationship between J28 215 'exercise' and response to psychological stressors, without more J28 216 precisely describing the relationship, increases the chance that it J28 217 will be measured at the wrong time of day, with the wrong sampling J28 218 rate, or with the wrong assessment instruments.

J28 219 Identifying higher-level causal variables can be a useful first J28 220 step in psycho-pathology research and treatment programs. J28 221 However, higher-level variables should be viewed as preliminary J28 222 'markers' for lower-level and more heuristic causal relationships. J28 223 Subsequent inquiry can increase the chance of identifying causal J28 224 variables and paths with greater clinical and empirical utility.

J28 225 Linearity J28 226 Most causal models do not specify the mathematical form of J28 227 their functional relationships. Consequently, most applications and J28 228 tests of the models presume that the relationships are linear in J28 229 form. However, as noted by <}_><-|>umerous<+|>numerous<}/> scholars J28 230 and clinicians (e.g., Asher, 1976; Biddle & Marlin, 1987; Bishop, J28 231 Fienberg, & Holland, 1975; Blalock, 1964; Bridgman, 1931; Grove & J28 232 Andreasen, 1986, James et al., 1982; Miller, 1983), many functional J28 233 relationships adhere more closely to parabolic, sine-wave, log, J28 234 exponential, or other nonlinear forms.

J28 235 The main drawback to an erroneous presumption that a causal J28 236 relationship is linear in form is, again, an increased chance of an J28 237 inferential error - underestimating the strength of a causal J28 238 relationship. This can happen when statistical techniques based on J28 239 a presumed linear relationship are applied to date that are J28 240 nonlinear. J28 241 J29 1 <#FROWN:J29\>Conflict Talk:

J29 2 Sociolinguistic Challenges to Self-Assertion and How Young J29 3 Girls Meet Them

J29 4 Amy Sheldon

J29 5 University of Minnesota

J29 6 Cultural stereotypes which interpret girls as less forceful or J29 7 less assertive than boys in pursuing their own agendas, J29 8 particularly during conflict episodes, are questioned. A theory of J29 9 double-voice discourse is proposed to characterize a type of J29 10 conflict talk that has a dual orientation, in which speakers J29 11 negotiate their own agenda while simultaneously orienting toward J29 12 the viewpoint of their partner. In double-voice discourse, J29 13 self-assertion is enmeshed with addressee-oriented mitigation. J29 14 Examples of such discourse in 3- and 4-year-old middle-class white J29 15 girls' conflict talk are analyzed. Gender differences in conflict J29 16 talk are seen to be contextualized variations that reflect J29 17 differences in the organization of same-sex groups. Girls' use of J29 18 the pretend frame in negotiating is discussed. Also highlighted is J29 19 the importance of analyzing children's language in the context of J29 20 conversational turns in order to develop a fuller interpretation of J29 21 utterances.

J29 22 This paper is an investigation of how young girls argue during J29 23 social play, the kinds of tactics they use to further their own J29 24 interests in disputes. As Maccoby (1986) pointed out, "we J29 25 have a clearer picture of what girls' groups do not do than J29 26 what they do do" (p.271). She called for "a J29 27 more clearly delineated account of interaction in female social J29 28 groups" (p. 271). Such an account requires description of J29 29 the ways in which language creates and maintains social J29 30 relationships among girls in everyday conversation.

J29 31 A theory of double-voice discourse (name taken from J29 32 Bahktin, 1929) is proposed to describe a linguistic style in white, J29 33 middle-class, preschool girls' social interactions. The term points J29 34 to a principle of dual orientation that shapes the agenda and style J29 35 of their conflict talk. One of the speaker's orientations is toward J29 36 her own agenda, toward the self. It asserts the speaker's own J29 37 wishes and proposes activities that are in the speaker's interest. J29 38 The other orientation is toward the other members of the group. As J29 39 a result, self-assertion is enmeshed in an orientation toward the J29 40 other. Self-assertion is thus regulated and contextualized J29 41 by the speaker's relationship-centered orientation.

J29 42 A theory of double-voice discourse captures the linguistic J29 43 complexity and creativity with which young girls manage disputes. J29 44 It also raises questions for traditional beliefs about femininity J29 45 and masculinity, which treat gender as a polarity or a comparison J29 46 of opposites. Such traditional views of gender are ill-conceived J29 47 and inadequate to develop an account of girls' sociolinguistic J29 48 interaction. Gender and context are confounded (Goodwin, 1980; J29 49 Sheldon, 1990; Thorne, 1990). We can best see how talk is gendered J29 50 if we take into consideration the context in which it emerges J29 51 (e.g., the sex of the speakers and what the speakers are trying to J29 52 accomplish). Considering children's talk from the perspective of J29 53 gender has the advantage of raising methodological issues that J29 54 focus attention on how we study children's language.

J29 55 First, I will raise concerns about androcentric interpretations J29 56 in some recent research on gender differences in children's talk. J29 57 Next, I will describe the theory of double-voice discourse and J29 58 relate it to children's solidarity-based task orientation. I will J29 59 then discuss methodological advantages of a qualitative approach to J29 60 the study of children's discourse and analyze examples of preschool J29 61 girls' conflict talk to show how double-voice discourse can J29 62 be read from them. I will conclude with remarks on the study of J29 63 language and gender.

J29 64 Gender and conflict. In most parts of American J29 65 society women's conflict talk is constrained by the expectation J29 66 that they will be 'nice.' 'Tough' talk, hard bargaining, and J29 67 'confrontational' talk is taboo for women. Men, however, have the J29 68 license to argue in directly demanding ways. They can engage in J29 69 unmitigated rivalry. Women are criticized if they speak as men do J29 70 (Campbell, 1988, 1989; Coates, 1987b; Lakoff, 1975; Thorne, J29 71 Kramarae, & Henley, 1983). Consequently, social talk which is J29 72 considered ordinary 'assertive' talk for men is likely to be J29 73 perceived as 'confrontational,' 'bossy' (or worse) when uttered by J29 74 women. Does the talk of very young girls show evidence of the J29 75 cultural taboo on tough talk? How do young girls express themselves J29 76 in assertive ways while they keep to the cultural mandate that they J29 77 not be 'too assertive'?

J29 78 It is difficult to describe the full range of girls' and J29 79 <}_><-|>womens'<+|>women's<}/> talk without echoing stereotypical J29 80 thinking about feminine and masculine behavior. Our thinking about J29 81 conflict has an androcentric bias. As a society, we view aggression J29 82 as the conflict norm. Equating conflict with aggression, however, J29 83 does not fully capture how girls argue. The dispute management norm J29 84 for girls is different from that for boys (Goodwin & Goodwin, J29 85 1987). Expecting conflict to have an aggressive component can J29 86 prevent us from even noticing conflict in girls' groups in which a J29 87 more subdues, yet nonetheless assertive, conflict style is J29 88 common.

J29 89 If we reconsider verbal conflict by focusing on feminine J29 90 conflict styles, there are several interesting consequences. First, J29 91 we must ask what agency, self-assertion, and power mean for J29 92 females, because looking at conflict through an androcentric lens J29 93 hides feminine agentic behavior and obscures the dynamics of J29 94 feminine power. Second, we see the constructive and facilitative J29 95 aspects of double-voice conflict, which is attentive to the social J29 96 cost of self-assertion, in contrast to the constrictive and even J29 97 destructive aspects of single-voice, baldly aggressive conflict, J29 98 which is enacted with little regard to social cost. Finally, we J29 99 gain a much clearer understanding of interaction in female J29 100 groups.

J29 101 Some problems with research on gender differences in J29 102 children's conflict talk. In a review of the literature of J29 103 gender differences in children's language and language development, J29 104 Klann-Delius (1981) concluded that this area is in "dire J29 105 need of being developed." Two recent quantitative studies J29 106 have observed the way that young children's talk is gendered. J29 107 Miller, Danaher, and Forbes (1986) studied over 1,000 quarrels by J29 108 24 racially and socioeconomically mixed 5- to 7-year-old children. J29 109 They concluded that "boys are more concerned with and more J29 110 forceful in pursuing their own agendae, and girls are more J29 111 concerned with maintaining interpersonal harmony" (p. 543). J29 112 What led the authors to the interpretation that boys were more J29 113 "forceful" in trying to get their own way is that their J29 114 dispute style was more heavy-handed than girls'. Boys used more J29 115 threats and physical force. By contrast, girls were interpreted as J29 116 more concerned with maintaining group harmony because they used J29 117 more conflict-mitigating strategies, such as compromise, evasion, J29 118 acquiescence, and clarification of intent.

J29 119 This conclusion by Miller et al. (1986) is influenced by J29 120 traditional views of gender hierarchy. It implies that girls are J29 121 not as self-assertive as boys because (a) when measured against the J29 122 masculine norm for conflict, which is coercive, girls do not use as J29 123 much verbal brute force to get what they want; and (b) part of J29 124 girls' agenda is to be empathic, and empathy presumably limits J29 125 self-assertion. To describe the boys as "more J29 126 forceful" in pursuing their own agenda is ambiguous as well J29 127 as politically loaded. It is ambiguous because forceful means J29 128 both 'effective' as well as "overpowering ... using J29 129 force." It is political because it implies that girls are J29 130 not as effective as boys in conflict situations; hence, girls are J29 131 not good at verbally managing conflict, at furthering their own J29 132 interests when opposed by someone. It is also a political J29 133 conclusion because it values and emphasizes a masculine mode of J29 134 brute force over a feminine mode of conflict mitigation. A J29 135 different conclusion about feminine conflict style is possible, J29 136 however; one that values the feminine conflict process and does not J29 137 interpret it as weakness, as something 'less' than the masculine J29 138 mode. From such a perspective, mitigation can be understood as J29 139 functioning to tone down coercion and domination, to bring about J29 140 adjustment and accord, and to restore group function. Girls and J29 141 women are skillful at negotiating constructive conflicts. This view J29 142 permits us to ask what important effects a constructive conflict J29 143 process has on how girls' groups function.

J29 144 To say that boys are "more forceful" J29 145 persuaders, or "more assertive" (Sachs, 1987), J29 146 overlooks the very important work that mitigation does to further J29 147 self-assertion in the conflict process and reinforces cultural J29 148 stereo-types that portray girls and women as submissive, J29 149 ineffective, and weak. It equates effectiveness in conflicts with J29 150 aggression. It measures self-assertion and independence too J29 151 narrowly. To remedy this, I propose that the term coercive be J29 152 used to describe only the heavy-handed conflict style. This frees J29 153 the word forceful, meaning effective, to describe girls' (and J29 154 boys') moderate conflict styles.

J29 155 In another study of gender differences in children's talk, J29 156 Leaper (1991) analyzed the conversations of 138 middle to J29 157 upper-middle class 4- to 9-year-old children in either same- or J29 158 mixed-sex dyads. Leaper's findings are consistent with those of J29 159 Miller et al. (1986). Girls used more collaborative speech J29 160 acts, defined as "direct" and "affiliative" (e.g., J29 161 invitations to play, constructive offers, mutual affirmations). J29 162 Boys used more controlling speech acts, defined as J29 163 "direct" and "distancing" (e.g., insults, orders, J29 164 refutations, and nonacceptance). Leaper (1991) hypothesizes,

J29 165 Given that girls have been found to demonstrate more J29 166 mutual coordination, responsivity, and elaboration in their J29 167 conversations ... the female pairs were expected to use more J29 168 affiliative speech acts ... and demonstrate more cooperative J29 169 exchanges compared to male pairs. In contrast, since boys have been J29 170 observed to be more demanding and domineering in their interactions J29 171 than girls, ... male dyads were hypothesized to display more J29 172 controlling speech acts and more domineering exchanges. (p. J29 173 800)

J29 174 The girls and boys in these two studies actually used both J29 175 self-assertive and supportive speech. But because there was a J29 176 significant difference in the comparisons of interest, it is easy J29 177 to lose sight of the similarities between girls and boys and to be J29 178 left with the differences, despite the efforts of the authors to J29 179 stress similarities. To address this problem in the study of J29 180 conflict talk, disputes can be described, instead, in a way which J29 181 better captures the complexity of girls' talk, thus revealing the J29 182 imaginative and elaborate ways in which girls are self-assertive J29 183 and powerful within the constraint of being relationship J29 184 oriented.

J29 185 Conflict is a contest of wills. Feminine conflict, because it J29 186 requires the overlay of mitigation to avoid jeopardizing J29 187 interpersonal harmony, asks for more sociolinguistic sensitivity J29 188 than the more direct masculine conflict style. In fact, Sachs J29 189 (1987) found that preschool girls have learned already how to J29 190 assert themselves "with a smile." Camras (1984) J29 191 studied what she identified as dominant and subordinate J29 192 middle-class children in pre-school, kindergarten, and J29 193 second grade. Dominant boys were much less polite than subordinate J29 194 girls or boys, but they were also much less polite than dominant J29 195 girls. Camras (1984) interprets these results as showing that J29 196 dominant girls "are gradually socialized to mask their J29 197 exercise of power during conflicts with use of polite J29 198 language" (p. 263).

J29 199 That girls and women are prescribed to assert themselves in a J29 200 way that is responsive to others has been noted by Carol Gilligan J29 201 (1987) and Jean Baker Miller (1986). This means that feminine J29 202 agency functions in a different way than masculine agency, not that J29 203 females are less agentic than males. Feminine self-assertion J29 204 requires responsiveness to others, whereas masculine agency does J29 205 not necessarily do so. The self-in-relation models of feminine J29 206 groups proposed by Miller and by Gilligan also predict less J29 207 hierarchical power relationships in female groups in contrast to J29 208 the masculine model in which power is a relation of domination over J29 209 others.

J29 210 Double-voice discourse. Double-voice discourse is a J29 211 talk style that is predicted by the self-in-relation model of J29 212 feminine development (Gilligan, 1987; Miller, 1986). The term J29 213 double refers to the perspective-taking stance of this style J29 214 in which the speaker expresses a double orientation or double J29 215 alignment. The primary orientation is to the self, to one's own J29 216 agenda. The other orientation is to the members of the group. The J29 217 orientation to others does not mean that the speaker necessarily J29 218 acts in an altruistic, accommodating, or even self-sacrificing J29 219 manner. It means, rather, that the speaker pays attention to the J29 220 companion's point of view, even while pursuing her own agenda. As a J29 221 result, the voice of the self is enmeshed with and regulated by the J29 222 voice of the other.

J29 223 Double-voice discourse is the norm in groups that are J29 224 solidarity based. The best example of such groups are girls' (or J29 225 women's) groups. Their social orientation is more often or more J29 226 consistently relationship centered. J29 227 J30 1 <#FROWN:J30\>Recovery of traumatic memories leads to J30 2 reintegration of the split and isolated multiples.

J30 3 There is another aspect to Freud's early belief that the cure J30 4 of neurosis lies in the remembrance of traumatic experiences. There J30 5 is no other treatment that so stresses remembering, and there is no J30 6 religion other than Judaism that makes a religious duty of J30 7 remembrance of traumatic events. "You shall not forget that J30 8 your forefathers were slaves in Egypt and you shall teach it to J30 9 your children and to your children's children" is one of J30 10 the cardinal commandments of Judaism. The Pass-over Seder is a J30 11 dramatization of that traumatic event and the redemption from it, J30 12 so that it will not be forgotten. The Jew must remember that his J30 13 forefathers were slaves. Freud repudiated Judaism as a religion and J30 14 consciously was an atheist who followed no religious practices or J30 15 ceremonies; however, he never repudiated his identity as a Jew or J30 16 his cultural adherence to Judaism. On the contrary, he was proud of J30 17 it. I would suggest that the psychoanalytic emphasis on remembering J30 18 as the essence of the cure was a return of the repressed or perhaps J30 19 a return of the disavowed that was in part determined by the J30 20 unconscious part of Freud's identity as a Jew. This, of course, J30 21 does not affect the theoretical validity or the degree of practical J30 22 utility of the cure through remembering, nor does it deny the J30 23 clinical inspiration for the theory. Theories, like all J30 24 psychological states and products, are, to use another Freudian J30 25 concept, overdetermined; that is, they have many causes. The source J30 26 of an idea has nothing to do with its value; to think so is to J30 27 commit a genetic fallacy. After I wrote this, I came across Yosef J30 28 Hyman Yerushalmi's brilliant and moving Freud's Moses: Judaism J30 29 Terminable and Interminable (1991), in which he expresses a J30 30 similar understanding of the origin of some of Freud's J30 31 psychoanalytic theorizing.

J30 32 Breuer, the third of Freud's spiritual fathers broke with him J30 33 over the issue of sexuality. Love turned to hate, or, more J30 34 accurately, the flip side of Freud's ambivalence toward fathers J30 35 came to the fore, and Freud found it necessary to cross the street J30 36 when he saw Breuer, his presence being so distasteful to Freud. J30 37 There followed a period of lonely isolation during which Freud met J30 38 and fell in love with Wilheim Fleiss, a charismatic Berlin J30 39 internist to whom he was related by marriage. Freud was neither the J30 40 first nor the last to be fascinated by Fleiss. Confident, J30 41 successful, and uncritically admired by many, Fleiss was just what J30 42 Freud needed. Brilliant, if erratic and eccentric in his ideas, J30 43 Fleiss was receptive to Freud's otherwise and otherwhere unwelcome J30 44 theorizing. Fleiss had a mesmerizing charm and was probably more J30 45 than a little crazy. His theory that all illnesses were caused by J30 46 nasal disorders, the nose being a sexual organ, has found little J30 47 scientific support, nor has his belief that all natural phenomena J30 48 could be accounted for by combinations and permutations of the J30 49 female (28-day) and male (23-day) cycles. Fliess's pseudoscientific J30 50 numerology probably owes an unconscious debt to cabalistic number J30 51 mysticism - altogether, an unlikely consort for the Helmholtzian, J30 52 scientifically rigorous Freud, but the heart has its reason, and a J30 53 passionate relationship developed between the two men. Their J30 54 contact was mostly through their correspondence, occasionally J30 55 punctuated by congresses, Freud's term for their J30 56 anxiously anticipated meetings, a term that suggests both J30 57 grandiosity and sexuality. Reading Freud's side of their J30 58 correspondence, which is all that has survived (Freud, 1985), we J30 59 get a sense of intense intellectual excitement: here are two men J30 60 approaching 40 who sound like adolescents who have just discovered J30 61 the world of ideas, with all the passion and excitement that goes J30 62 with that discovery. Of course, Fliess's excitement is an inference J30 63 from Freud's letters, but it certainly appears to be there. Freud's J30 64 letters to Fliess are a depiction of life of the educated Jewish J30 65 middle class of late 19th-century Vienna that have all the J30 66 vividness and richness of a great novel. Sentences filled with J30 67 Freud's deep love of children alternate with sarcastic comments on J30 68 his academic rivals, discussion of current political events, and J30 69 theoretical 'drafts'. The overall effect is exhilarating. Freud's J30 70 early theories about neurosis, anxiety, and the role of sexuality J30 71 are all given trial balloons in the drafts he sent to Fliess. The J30 72 most extensive of the drafts is Freud's 'Project for a Scientific J30 73 Psychology' (1895/1950), which he abandoned and never published. It J30 74 is a brilliant attempt to give a quantitative neurological J30 75 explanation of psychological states and of psychopathology. It was J30 76 Freud's last attempt to reduce psychology to physiology. Although J30 77 he never abandoned the belief that a neurochemical explanation of J30 78 mental events was possible, he himself turned to purely J30 79 psychological explanations to account for both normal and J30 80 pathological events. It is true that his psychological models and J30 81 accounts retain a physicalistic basis, and much of Freud's J30 82 theorizing is based on a "hydraulic model" of J30 83 forces, pressures, flows, and blockages. It is a model based on J30 84 19th-century physics. It is also true that his theorizing becomes J30 85 more and more a theory about meaning, and about relationships, and J30 86 becomes truly psychological rather than pseudopsychological J30 87 physics.

J30 88 During Freud's almost two-decade-long relationship with Fliess, J30 89 he suffered a "considerable psycho-neurosis" J30 90 (Jones, 1961, p. 198) himself. Freud's emotional pain drove him to J30 91 undertake his self-analysis, in which Fliess served as a sort of J30 92 analyst by mail, and more important, was a transference figure J30 93 eliciting all of Freud's intense feelings of love and hate for his J30 94 father. Although it is unlikely that the two men were actually J30 95 lovers, there is no question that Wilheim Fliess was the great love J30 96 of Freud's life.

J30 97 In the course of his self-analysis and his relationship with J30 98 Fliess, Freud 'discovered' the Oedipus complex and wrote what is J30 99 usually considered his most important work, Interpretation of J30 100 Dreams (1900/1953a). In analyzing his dreams, Freud came to J30 101 see that dreams have the same structure as symptoms. They too are J30 102 disguised expressions of forbidden wishes. He concluded that all J30 103 dreams are wish fulfillments. In the course of his self-analysis, J30 104 he discovered much about himself: about his intense rivalry with J30 105 and ambivalence toward his father; about his murderous feelings J30 106 toward his infant brother, Julius; about his drivenness; and about J30 107 his narcissistic vulnerability.

J30 108 The dreams reported in Interpretation of Dreams make J30 109 a unique contribution to the autobiographical literature of the J30 110 West. They expand the account of self to include a new dimension. J30 111 The self asleep - at least while dreaming - now becomes an integral J30 112 part of self. Descartes's questions about distinguishing dreams and J30 113 waking reality as a vital component of reality testing become J30 114 irrelevant, and Locke's concern about the continuity of self during J30 115 sleep is seen in a new light: dream consciousness is just as much J30 116 consciousness, just as integral to the self, as waking J30 117 consciousness. The injunction "Know Thyself" J30 118 changes in meaning as the locus of self shifts to that which is not J30 119 known, to the unconscious as represented in disguised and distorted J30 120 forms in the dream. The self is now more unknown and unknowable, J30 121 apart from undergoing the rigors of analysis, than hitherto J30 122 believed. Freud's technique of dream analysis is double-edged: on J30 123 the one hand, it gives us a tool for knowing the self; on the other J30 124 hand, it reveals a new, unknown territory that must be reclaimed J30 125 before the self can be either known or integral.

J30 126 Having gone public in a unique, if partial and selective, way, J30 127 Freud put an important part, by his lights the most important part, J30 128 of himself up for scrutiny by any and all; and indeed his dreams J30 129 have been interpreted and reinterpreted in a bewildering variety of J30 130 ways, both from within and from without the psychoanalytic J30 131 movement. One of the most fascinating perspectives on Freud's J30 132 dreams is that of Carl Schorske (1980), who looks at their J30 133 political meaning and significance and sees Freud as J30 134 "regressing" from the political (adult's) to the familial J30 135 (child's) world, from external reality to internal reality, because J30 136 of the disintegration of the Austrian-Hungarian empire, its series J30 137 of defeats in war, and growing dissension, corruption, and J30 138 decadence; also, increasingly virulent anti-Semitism (Karl Lueger J30 139 was installed as the anti-Semitic mayor of Vienna just as J30 140 Interpretation was published) made action in the outer J30 141 world increasingly futile and hopeless. Freud's dreams do indeed J30 142 have many political references, and Freud like Plato before him J30 143 takes the relation between social classes as representative of, or J30 144 isomorphic to, the relationships of the parts of the psyche. J30 145 Additionally, Freud's metaphors of self and mind are consistently J30 146 political, and even sometimes military: defense, resistance, J30 147 occupation, and drive.

J30 148 Schorske interprets what Freud calls the manifest J30 149 dream, the dream as dreamt, which Freud distinguishes from the J30 150 latent dream , which is where his interest lies. In J30 151 Freud's theory of the mechanism of dreams (which serves as a J30 152 paradigm for his theory of mind in the sense of self) the dream J30 153 thoughts that are forbidden childhood wishes, derivative of drives J30 154 (instinctual energies) striving for discharge, are J30 155 "converted" by the dream work into the manifest J30 156 dream through the mechanisms of displacement, condensation, J30 157 symbolization, visualization, and secondary revision. The task of J30 158 dream interpretation is to work backwards from the manifest dream J30 159 to the latent dream thoughts by listening to the dreamer's J30 160 association to each dream element. Secondary revision is the mind's J30 161 reworking of the dream material to give it more apparent sense and J30 162 continuity than it possesses, that is, to give the dream a better J30 163 story line. Dreams make use of current materials (the "day J30 164 residue") but always equally, or more than equally, J30 165 represent in distorted form the events and desires of childhood. J30 166 Dreams are always egoistic. The censor imposes the dreamwork on the J30 167 latent dream thoughts so they do not arouse so much anxiety as to J30 168 wake the dreamer.

J30 169 Freud has now moved from psychopathology to a normal J30 170 psychological phenomenon, dreaming, and found that dreams are J30 171 compromise formations in just the same way as symptoms. He is now J30 172 in a position to expound a general psychology, an omni-applicable J30 173 account of human nature. In the years following the J30 174 Interpretation of Dreams, Freud went on to apply his J30 175 paradigm to jokes, art, hallucinations, religion, and culture in J30 176 general, finding each to have the same basic structure as J30 177 compromises and disguised wish fulfillments.

J30 178 In the famous "specimen dream of J30 179 psychoanalysis", the dream of Irma's injection, Freud for J30 180 the first time subjects a dream of his own to analysis. In the J30 181 dream, the dreamer is in a large reception hall receiving guests, J30 182 including Irma, who is a former patient who is still ill. By the J30 183 time the dream ends, Irma's continued illness is blamed on at least J30 184 three other persons, including one who represents Breuer, Freud J30 185 interprets the dream wish as the desire to be blameless as well as J30 186 to pay back some old scores. Irma in real life was Emme Eckstein, J30 187 whom Fliess had operated on for "nasal neurosis" J30 188 (which was plain madness), an intrusive application of his wild J30 189 theory to a human being. To make matters worse, he left the packing J30 190 in, which infected (long before antibiotics) and almost killed the J30 191 patient, who suffered the torments of the damned and was given J30 192 psychological interpretation of her difficulties by Freud. Freud J30 193 told her that her symptoms were a holding onto her illness, which J30 194 was a manifestation of her negative transference to him. Freud's J30 195 dream was certainly an attempt to find himself guiltless by J30 196 projecting blame for Irma's difficulties onto others, but Freud J30 197 missed the main thrust, the deepest wish, behind the dream: to find J30 198 Fliess blameless in order to protect his (Freud's) idealized love J30 199 object from contamination and devaluation. Freud missed the motive J30 200 power of our need for ideal objects, for perfect lovers with whom J30 201 we can identify and perhaps merge. Fliess was such an ideal object J30 202 for him. If Fliess was a transference object, as according to J30 203 Freud's theory he had to be, then it was his father who was to be J30 204 protected from the charge of injuring a woman. The childhood wish J30 205 represented in distorted form in the dream was his wish that Father J30 206 be perfect and blameless. J30 207 J31 1 <#FROWN:J31\>THE TIMING OF AN INTERPRETATION

J31 2 A Comparative Review of an Aspect of the Theory of Therapeutic J31 3 Technique

J31 4 Lawrence Josephs

J31 5 Interpretation in psychoanalysis is often considered to be as J31 6 much an art as a science. And perhaps the most intuitive aspect of J31 7 interpretive work, an aspect that is crucial to therapeutic J31 8 outcome, is the timing of an interpretation. When do we interpret J31 9 the transference? When do we interpret resistance? How deeply do we J31 10 interpret? (And by 'deep' do we mean highly defended-against J31 11 material, archaic fantasy, or early life historical events?) When J31 12 would we simply empathize with the phenomenology of conscious J31 13 self-experience and when should we confront and challenge J31 14 resistance to discussing anxiety-laden unconscious J31 15 conflict? By what criteria do we define an interpretation as J31 16 premature or perhaps as too late? To what degree is the timing of J31 17 an interpretation dependent on the presence of a prerequisite J31 18 atmosphere of safety? How do we assess whether early interpretation J31 19 of latent negative transference will either strengthen or impair J31 20 the formation of a working alliance? The answers to these questions J31 21 are partially based on the analyst's evaluation of the particular J31 22 dynamics of the individual patient, yet to a large degree the J31 23 answers to such questions derive from the analyst's theory of J31 24 therapeutic technique. Although timing is in many respects J31 25 dependent on the analyst's intuitive grasp of a unique analytic J31 26 moment and is informed by past experience conducting analyses, the J31 27 analyst's theory of therapeutic technique provides a preconscious J31 28 sensibility that quietly guides the analyst's intuition as to the J31 29 proper timing of an interpretation. In other words, intuition is J31 30 partially derivative of a preconscious theory-derived rationale. It J31 31 is preconscious in the sense that one's theory of therapeutic J31 32 technique is not, for the most part, subject to repression, denial, J31 33 or disavowal but nevertheless operates prereflectively, often J31 34 without explicit conscious formulation.

J31 35 For the purpose of this paper interpretation will be defined in J31 36 the broadest sense as the provision of verbal feedback which aims J31 37 at increasing the patient's self-understanding. Thus, J31 38 clarification, empathic reflection, and confrontation will be J31 39 considered as forms of interpretive activity in addition to the J31 40 more narrow definition of interpretation that involves linking J31 41 transference manifestations in the here and now to genetic J31 42 constellations deriving from the there and then. The timing of an J31 43 interpretation refers to the sequencing and pacing of one's J31 44 interpretive activity and as such reflects a pre-conscious J31 45 interpretive strategy; continually modified as it will be by the J31 46 vicissitudes of the ongoing therapeutic interaction. The purpose of J31 47 this paper will be to review, compare, and contrast recommendations J31 48 concerning the timing of an interpretation that derive from diverse J31 49 theories of therapeutic technique and offer an integrative overview J31 50 in the service of resolving some of the disputes under J31 51 discussion.

J31 52 FOUNDATIONAL MODELS

J31 53 The Topographic and Psychodynamic Approaches

J31 54 The original theory of how to time an interpretation is to be J31 55 found in Freud's (1911, 1912a, 1912b, 1913, 1914, 1915) seminal J31 56 papers on technique which were based primarily on his topographic J31 57 and psychodynamic models, having been written prior to the J31 58 introduction of the structural model in 1923. Well-known J31 59 psychoanalytic aphorisms informing the timing of an interpretation J31 60 derive from these papers, including the principles that one should J31 61 always work from surface to depth and that one should always J31 62 interpret resistance before content. In these papers, resistance J31 63 emerged as the key concept on which the assessment of the proper J31 64 timing of an interpretation must be based. Resistance to free J31 65 association was conceived of as a surface phenomenon that blocked J31 66 access to greater depth. According to Freud (1914), the analyst J31 67 "contents himself with studying whatever is present for the J31 68 time being on the surface of the patient's mind, and he employs the J31 69 art of interpretation mainly for the purpose of recognizing the J31 70 resistances which appear there, and making them conscious to the J31 71 patient" (p. 147). Resistance analysis was seen as the J31 72 primary technique through which the analyst helped the unconscious J31 73 to become conscious.

J31 74 Freud discovered in his psychoanalytic work that resistance to J31 75 the uncovering of unconscious contents was a ubiquitous phenomenon. J31 76 Patients blocked introspective understanding of unconscious J31 77 conflicts in inhibiting the flow of free association and they would J31 78 not accept the analyst's interpretation of unconscious contents, be J31 79 it forbidden wishes or painful memories, until the analyst had J31 80 interpreted the resistance to the awareness of such warded-off J31 81 contents. If one interpreted prematurely (i.e., too deeply) or too J31 82 late (i.e., too superficially) resistance was exacerbated rather J31 83 than ameliorated.

J31 84 Expecially pertinent to the timing of an interpretation was a J31 85 consideration of the dynamics of the transference. The art of J31 86 correct timing largely focuses on the problem of overcoming the J31 87 resistance to free association, and the major form of resistance is J31 88 invariably to be found in the transference:

J31 89 Over and over again, when we come near to a pathogenic J31 90 complex, the portion of that complex which is capable of J31 91 transference is first pushed forward into consciousness and J31 92 defended with the greatest obstinacy .... These circumstances tend J31 93 towards a situation in which finally every conflict has to be J31 94 fought out in the sphere of transference. (Freud, 1912b, p. J31 95 104)

J31 96 In terms of the relationship of the timing of an interpretation J31 97 to the nature of the transference, Freud (1913) gave this J31 98 recommendation:

J31 99 So long as the patient's communications and ideas J31 100 run on without any obstruction, the theme of transference should be J31 101 left untouched. One must wait until the transference, which is J31 102 the most delicate of all procedures, has become a J31 103 resistance. (p. 139)

J31 104 Yet, before transference can be interpreted as a resistance, J31 105 there is a crucial preparatory phase that must occur if J31 106 interpretation is to be effective at all:

J31 107 When are we to begin making our communications to the J31 108 patient? ... The answer can only be: Not until an effective J31 109 transference has been established in the patient, a proper J31 110 rapport with him. It remains the first aim of the treatment to J31 111 attach him to it and to the person of the doctor. To ensure this, J31 112 nothing need be done but to give him time. If one exhibits a J31 113 serious interest in him, carefully clears away the resistances that J31 114 crop up at the beginning and avoids making certain mistakes, he J31 115 will of himself form such an attachment. (Freud, 1913, p. J31 116 139)

J31 117 Part of what makes for an effective interpretation is not only J31 118 the correct content of the interpretation but the nature of the J31 119 transference at the time the interpretation is made: "The J31 120 patient, however, only makes use of the instruction in so far as he J31 121 is induced to do so by the transference; and it is for this reason J31 122 that our first communication should be withheld until a strong J31 123 transference has been established" (Freud, 1913, p. J31 124 144).

J31 125 The Structural Viewpoint

J31 126 Freud discovered that resistance analysis was no easy task. J31 127 Although resistance is closer to the surface than the wishful J31 128 impulses and painful memories whose expression resistance serves to J31 129 inhibit, resistance itself is nevertheless a deeply unconscious J31 130 phenomenon for which there is no ready access to awareness. In J31 131 other words, there is always resistance to the awareness of J31 132 resistance. In the structural model, resistance was traced to the J31 133 unconscious defensive activities of the ego:

J31 134 There can be no question but that this resistance J31 135 emanates from his ego .... We have come upon something in the ego J31 136 itself which is also unconscious, which behaves exactly like the J31 137 repressed - that is, which produces powerful affects without itself J31 138 being conscious and which requires special work before it can be J31 139 made conscious. (Freud, 1923, p. 17)

J31 140 The technical rule of interpreting resistance before content J31 141 could be amended to read as a recommendation to analyze ego before J31 142 id, ego defined as the unconscious mechanisms of defense.

J31 143 Deviations in Technique

J31 144 By the end of his career, Freud (1937) began to assess the J31 145 limitations of psychoanalysis as a therapeutic modality. He had J31 146 discovered that psychopathology and character structure were J31 147 massively resistant to change, that patients were massively J31 148 resistant to the awareness of their resistance to change, that J31 149 whatever changes did transpire occurred very slowly and only with J31 150 painstaking working-through of anxiety-laden conflictual J31 151 issues, and that, even when change occurred, regressive J31 152 developments could always undo it. Despite this tragic vision of J31 153 human development Freud believed that his technical recommendations J31 154 provided the optimum conditions for a 'talking cure.' Naturally, J31 155 succeeding generations of analysts would attempt to improve results J31 156 with technical innovations given the modest therapeutic J31 157 expectations that Freud suggested.

J31 158 If Freud's technical recommendations are treated as a baseline, J31 159 it becomes apparent that technical innovations have deviated in J31 160 basically two directions. On the one side, the central idea has J31 161 been that classical technique is too conservative in addressing J31 162 anxiety-laden unconscious conflicts and that resistance to the J31 163 awareness of unconscious processes needs to be addressed more J31 164 directly, vigorously, and systematically. On the other side, the J31 165 central idea has been that classical technique has laid excessive J31 166 emphasis on the analysis of defense, resistance, and latent J31 167 transference in a context of abstinence, in neglect of establishing J31 168 a therapeutic relationship and building psychic structure, J31 169 activities that are seen as preceding and laying down the J31 170 foundation for meaningful analysis of unconscious conflict. The J31 171 debate has centered on matters of relative emphasis and priority, J31 172 with the issues resting on a continuum from conservative to radical J31 173 approaches. When classical technique has been seen as too J31 174 conservative, the innovation has been to address latent J31 175 transference more directly, actively, and systematically. Reich J31 176 (1933) was one of the first innovators in that regard, chastising J31 177 his peers for insufficient attention to latent negative J31 178 transference and suggesting that nonverbal characterological J31 179 resistances to treatment be consistently interpreted as J31 180 manifestations of latent negative transference. Klein (1952) shared J31 181 Reich's focus on systematically interpreting the latent negative J31 182 transference, but felt that links to archaic unconscious fantasy J31 183 could be established much more quickly than was the case utilizing J31 184 classical technique. Searles (1979), Langs (1976), and Gill (1982) J31 185 also advocated the primacy of interpretation of the latent negative J31 186 transference, but felt that transference interpretation needed to J31 187 be linked to the triggering stimuli in the here-and-now therapeutic J31 188 interaction. The overall innovation in technique in all of the J31 189 aforementioned suggestings is that unconscious conflict is most J31 190 quickly brought to the fore through a primary and active J31 191 interpretation of the latent negative transference. The difference J31 192 in these approaches resides in whether transference interpretation J31 193 is most efficaciously linked to character analysis, archaic J31 194 fantasy, or current interpersonal transactions. The analyst's J31 195 resistance to employing such innovations is usually traced to his J31 196 or her fear of the patient's anger in confronting latent negative J31 197 transference, fear of psychosis in interpreting archaic fantasy, J31 198 and fear of admitting interpersonal involvement interpreting the J31 199 here-and-now interaction.

J31 200 When the critique of the classical approach is for insufficient J31 201 attention to relationship building, technical recommendations J31 202 center on interpretive strategies that are thought to facilitate a J31 203 therapeutic relationship as well as greater emphasis on nonverbal J31 204 affective attitudes that are thought to facilitate the proper J31 205 ambience for treatment. Be it the ego psychologist's fostering of a J31 206 working alliance and an observing ego, the object relations J31 207 theorist's provision of containment or a holding environment, or J31 208 the self psychologist's facilitation of a selfobject transference J31 209 through empathy, the essential idea is the same: A therapeutic J31 210 climate, which is partially curative in and of itself, must be J31 211 established before conflict defense analysis can be fruitfully J31 212 conducted. Verbal interpretation from this perspective tends to J31 213 focus on the articulation of preconscious processes that are J31 214 thought to be crucial in ego development and in the establishment J31 215 of a cohesive sense of self.

J31 216 Each approach claims to achieve maximal therapeutic benefit and J31 217 possesses a critique of the inadequacies of other approaches. When J31 218 the technical approach is seen as unnecessarily avoidant of J31 219 addressing unconscious conflict and therefore superficial, the J31 220 treatment is devalued as re-educative supportive psychotherapy that J31 221 bolsters defenses rather than analyzes them. When the approach is J31 222 seen as going too deep too quickly, the assumption is that J31 223 treatment will result in excessive negative therapeutic reaction, J31 224 intellectualized insight, false self-compliance, and a moralistic J31 225 maturity ethic. The basic question in regard to interpretation J31 226 boils down to one of timing: how deep to go how quickly? In working J31 227 from surface to depth, must painstaking elucidation of the J31 228 phenomenological surface precede deeper interpretation or is such J31 229 an approach only working with the manifest content and avoiding a J31 230 more active, direct, and systematic analysis of latent negative J31 231 transference?

J31 232 J32 1 <#FROWN:J32\>Chapter 4

J32 2 The Frequency of Dirty Word Usage

J32 3 How frequently are dirty words used in American culture? How J32 4 often does one person use them throughout the day? Although many J32 5 people in broadcasting, education, law, and social science make J32 6 judgments about the frequency of dirty word usage, there are no J32 7 sound data on which to make those decisions. If one examines the J32 8 literature on general word frequency, one finds little information J32 9 on dirty word usage. Entries in standard dictionaries are standard J32 10 speech. Studies of word frequency tend to be studies of standard J32 11 word frequency. One can examine colloquial dictionaries or those J32 12 dedicated to argot, slang, or jargon. These provide no indication J32 13 of how frequently such terms are used. A few minutes on a busy J32 14 street corner convinces the listener that Americans use dirty words J32 15 frequently in public but how frequently? This chapter is about J32 16 determining the relative frequencies of dirty words in American J32 17 English using standard empirical techniques. We will look at J32 18 traditional means of establishing word frequency and some of the J32 19 problems with establishing how often people swear. The conclusion J32 20 from the available data is that these words are used relatively J32 21 frequently.

J32 22 Why Word Frequency?

J32 23 For many years those interested in human communication and J32 24 verbal behavior have known that words are used with different J32 25 frequencies in communication. The notion of variable frequency has J32 26 led investigators to study how frequency differences affect a wide J32 27 variety of language-related activities (reading, creativity, J32 28 language learning, problem solving, or comprehension). Given that J32 29 dirty word usage is a behavior that most social scientists have J32 30 constantly ignored, how can we tell how often dirty words are used? J32 31 What impact does differential usage have?

J32 32 One of the early concerns was the role of word frequency in J32 33 both face-to-face communication and through electronic media like J32 34 radio, short wave, or encoded channels. It was pointed out by Zipf J32 35 (1949), for example, that relatively few words were used extremely J32 36 often in communication and that the higher the frequency of J32 37 occurrence a word had, the more likely it was to be used in J32 38 general. Shannon and Weaver (1949) incorporated 'Zipf's Law' into J32 39 their grand theory of communication, using word frequency to make J32 40 predictions about the probability of various sayings or messages J32 41 during communication. These predictions were important to produce J32 42 efficient and quick communication and to eliminate errors by using J32 43 predictable words. These findings were applied in military and J32 44 government communication systems following World War II. However, J32 45 the grand theory building that dominated psychology in the 1950's J32 46 has died out and contemporary psychologists now focus on specific J32 47 issues and problems involving human communication. Interest has J32 48 shifted to how word frequency influences a specific type of J32 49 language processing. The frequency question remains present in J32 50 almost every aspect involving information processing abilities, J32 51 such as attention, problem solving, pattern recognition, learning, J32 52 rehabilitation, and memorization. Below are some of the recent J32 53 applications of word frequency in contemporary investigations to J32 54 show why the dirty word frequency question has remained J32 55 unanswered.

J32 56 Word frequency has been a fairly consistent predictor of J32 57 response time differences in language processing tasks. High J32 58 frequency words are processed faster and more easily than words at J32 59 low frequencies. The result has been called 'the word frequency J32 60 effect' in whatever process is under investigation. In a lexical J32 61 decision task, for example, subjects are shown a string of letters J32 62 and asked to tell whether the string forms a word. One of the best J32 63 predictors of response time is how frequently the word is used; the J32 64 higher the frequency, the lower the reaction time.

J32 65 Word frequency effects have been reported in other language and J32 66 memory tasks. In an object-naming task, where subjects are asked to J32 67 name objects as fast as possible, results indicate that word J32 68 frequency is related to the ease of naming the object. In word J32 69 legibility tasks, frequent words were found to be as legible as J32 70 single letters, although infrequent words are less legible than J32 71 either. In lexical access tasks, word frequency effects the time it J32 72 takes to decide if a target word has occurred in a sentence. Word J32 73 frequency effects have been found in many memory tests of recall, J32 74 recognition, and others (see Jay, 1980a). People who study reading J32 75 and learning to read know that frequent or familiar words make J32 76 information processing tasks easier and faster than obscure or low J32 77 frequency words.

J32 78 It should be clear that in each of these studies dirty words J32 79 are never used. What is known about linguistic processes has J32 80 had little or nothing to do with dirty words. If all science on J32 81 language stopped now, we would know little about dirty word usage J32 82 or how dirty word usage relates to more normal language use.

J32 83 A recent question has been whether word frequency is related to J32 84 the age of acquisition of a language by children. Carroll and White J32 85 (1973 a,b) found that frequent words tend to be those that are J32 86 acquired early in life. Another variable related to acquisition is J32 87 the number of different meanings that a word has. One point here is J32 88 that frequent words have more interpretations than low frequency J32 89 words. So, age of acquisition depends both on meaning and frequency J32 90 aspects of word usage. It is important to know what children learn J32 91 early in life when examining patients with brain damage. Lesser J32 92 (1978, pp 110-112), for example, shows that word frequency is J32 93 needed to study aphasic (language loss) patients. Patients may use J32 94 childish or highly frequent language when recovering from brain J32 95 damage. Some of these aphasics may use only dirty words when J32 96 recovering because the words were learned early and well, and may J32 97 be stored in parts of the brain untouched by the trauma.

J32 98 There should be no doubt that word frequency has an important J32 99 influence on communication and communication problems. But, have we J32 100 stopped to ask the question, where did all these frequency data J32 101 come from in the first place?

J32 102 The Frequency Estimation Problem: Why There Are No Dirty J32 103 Words

J32 104 There are hundreds of millions of speakers of English using J32 105 tens of thousands of words on thousands of different occasions. To J32 106 accurately count actual word usage is impossible. Therefore, J32 107 estimates must be used. Throughout the history of frequency J32 108 research, the question of proper frequency estimation has received J32 109 minimal attention, and one wonders whether any of these previous J32 110 estimates were accurate. While the tasks mentioned above have been J32 111 conducted with appropriate methods, those interested in taboo J32 112 language estimation would find them inadequate and inaccurate to J32 113 estimate taboo language frequency. Future interest in controlling J32 114 taboo language frequency or offensiveness requires an alternative J32 115 to traditional methods and reports.

J32 116 One major problem is the use of inappropriate normative samples J32 117 of word frequency. Since its publication, psychologists and others J32 118 interested in language frequency have been using the Thorndike and J32 119 Lorge (1944) norms to estimate word frequency. However, the count J32 120 (or others like it) are inadequate for three reasons: (a) the J32 121 sample is restricted only to written, not oral usage, (b) the J32 122 written sample is restricted to a limited domain of reading J32 123 materials (mostly children's and popular adult literature), and (c) J32 124 it is outdated (language changes with time). Eriksen (1963) has J32 125 demonstrated quite clearly that the Thorndike-Lorge count is J32 126 inadequate to estimate oral usage and that it in fact J32 127 underestimates the frequency of many colloquially used words. Two J32 128 collections of written word frequency norms were published by J32 129 Kucera and Francis (1967) and more recently by Carroll, Davies, and J32 130 Richman (1971). These are updated and less restricted than J32 131 Thorndike-Lorge, but they are still limited to written samples.

J32 132 The problem with written norms is not that they are inaccurate, J32 133 but that they are used incorrectly by researchers. The written J32 134 norms generally are appropriate when applied to tasks involving J32 135 textual material; that is, reading processes. The written norms are J32 136 not appropriate for research concerning oral usage, however. J32 137 Written norms do not apply directly to colloquial, conversational J32 138 situations such as parent-child interaction, discourse processes, J32 139 or language influenced by sociolinguistic variables (social or J32 140 physical setting). The point is that sociolinguistic variables are J32 141 crucial to understanding dirty word usage. It is context that J32 142 controls dirty word usage and these factors must be accounted J32 143 for.

J32 144 The bottom line on studies of written samples is that one is J32 145 rarely going to find dirty words used in the sample because they J32 146 are collected from biased material, even though that material may J32 147 be appropriate to design children's reading texts, for example.

J32 148 Counting Oral Frequency: Almost Good Enough

J32 149 Written and oral speech are two different forms of language. In J32 150 addition to different rules (e.g., the future perfect does not seem J32 151 to exist in oral speech) and different distribution of rules they J32 152 have in common (e.g., the perfect form of the verb appears much J32 153 more in written speech), written language has the benefit of more J32 154 polish. Oral speech is marked by hesitations, interruptions, J32 155 incomplete expressions, is more prone to imprecise or incorrect J32 156 definitions - all of which can be corrected with a little J32 157 proofreading in the written form.

J32 158 Oral speech is also at the mercy of a number of sociolinguistic J32 159 influences. Spoken language, particularly in its more colloquial J32 160 form, is more sensitive to the relation between speaker and J32 161 listener, the degree of social relaxation, and the topic of J32 162 discussion (see Jay, 1978c). When these variables J32 163 <{_><-|>flucuate<+|>fluctuate<}/>, so does the kind of language J32 164 selected and used in conversation. In light of these contextual J32 165 constraints, then, the estimation of oral frequency requires J32 166 looking at oral - not written - samples of language for taboo J32 167 speech.

J32 168 Several attempts have been made to examine the frequency of J32 169 word usage in conversational English (Jay, 1980a). One widely J32 170 mentioned study is French, Carter, and Koenig's (1930) collection J32 171 of words and sounds from telephone conversations. This study is J32 172 cited because it provides information about vowel, consonant, and J32 173 word frequency data. The study has one major flaw. They omitted J32 174 some 25% of the data representing utterances such as exclamations, J32 175 interjections, proper names, titles, letters, numbers, and J32 176 profanity. In fact profanity was 40% of the material omitted! J32 177 While the study could give a relatively accurate picture of J32 178 conversations, it compromised a true picture of dirty word usage. J32 179 Fairbanks (1944) updated the French et al. study and compared the J32 180 spoken language samples of college students with diagnosed J32 181 schizophrenics. More recently, Black, Stratton, Nichols, and Chavez J32 182 (1985) published a word count based on college student classroom J32 183 language, as did Berger in 1968.

J32 184 Several studies concentrate on children's oral language. The J32 185 need for these data in reading, learning, and comprehension J32 186 applications should be clear. These data serve the purpose of J32 187 designing age-appropriate textbooks but do not indicate anything J32 188 about how children use dirty words or how often. One of the best J32 189 collections and most current counts is Spoken Words J32 190 (1984) by Hall, Nagy, and Linn. These investigators report the data J32 191 as a function of situation and social status of the children's J32 192 family. However, when speech was recorded in classes or at the J32 193 dinner table, very few dirty words were spoken. Dirty words are J32 194 highly context dependent and even young children have learned not J32 195 to use them most of the time at school or at the dinner table. Thus J32 196 few of them appear in the corpus. While these norms provide a J32 197 useful foundation for writing or evaluating children's reading J32 198 texts, as intended, they give the impression that children do not J32 199 produce dirty words. Consequently, there is a risk of thinking J32 200 children are much more naive about matters involving sex and J32 201 aggression than they are, if the conclusion is based on biased word J32 202 counts. Cameron (1969) made one of the more natural attempts to J32 203 collect adult language data as a function of social setting. His J32 204 sample, though restricted to a few college settings, is useful to J32 205 indicate how speech changes from relaxed to formal social J32 206 environments. The major problem with the Cameron norms stems from J32 207 an inadequate sampling procedure. He asked his 'overhearer' to J32 208 record by hand "the first three words they heard during the J32 209 conversation at 15 second intervals" (p 102). Such J32 210 recording is subject to constraints from attention, perception, or J32 211 recording bias, especially when overhearers knew he was interested J32 212 in certain types of words. J32 213 J32 214 J33 1 <#FROWN:J33\>The most successful representations make fairly J33 2 specific assumptions about the way in which the text should be J33 3 interpreted (e.g., as a story about one of a small number of J33 4 topics) and about the kinds of questions that might be asked about J33 5 the text (e.g., questions about actor's intentions). Some work has J33 6 been done to adapt the natural language understanding techniques to J33 7 an information retrieval setting, but there is little near-term J33 8 hope that these techniques could be used to represent large J33 9 document collections and arbitrary queries [Sparck Jones and Tait, J33 10 1984].

J33 11 Within the information retrieval community, a number of J33 12 techniques have been developed that can represent the content of J33 13 documents and information needs. These representations have a much J33 14 different flavor than NLP representations. They are generally based J33 15 on simple, very general, features of documents (e.g., words, J33 16 citations) and represent simple relationships between features J33 17 (e.g., phrases) and between documents (e.g., two documents cite the J33 18 same document). The focus here is on simple, but general, J33 19 representations that can be applied to most texts rather than on J33 20 specialized techniques which capture more information but are J33 21 applicable only in narrow contexts. Information retrieval J33 22 representations also make extensive use of the statistical J33 23 properties of representations features and attempt to make use of J33 24 information produced by human analysis (e.g., manual indexing) when J33 25 available.

J33 26 Over the last decade there has been considerable interaction J33 27 between the AI and information retrieval communities; AI techniques J33 28 have been adapted to an IR setting and the IR focus on 'real' J33 29 document collections and on thorough experimental evaluation has J33 30 helped to expand the focus of AI research.

J33 31 Given the availability of a number of representation techniques J33 32 that capture some of the meaning of a document or information need, J33 33 our basic premise is that decisions about which documents match an J33 34 information need should make use of as many of the representation J33 35 forms as practical. The remainder of this paper develops a J33 36 theoretical framework for retrieval that allows multiple J33 37 representations to be combined.

J33 38 In the next section, we describe the major types of retrieval J33 39 models. Section 7.3 presents the motivation for a retrieval model J33 40 based on inference. In Section 7.4, we review related research on J33 41 inference and network models. Sections 7.5 and 7.6 describe the J33 42 basic inference network model and how it is used. Section 7.7 J33 43 addresses the issue of causality in a network model. The final J33 44 section discusses recent results and research directions.

J33 45 7.2 Current Retrieval Models

J33 46 A retrieval model fixes the details of the representations used J33 47 for documents and information needs, describes how these are J33 48 generated from available descriptions, and how they are compared. J33 49 If the model has a clear theoretical basis we call it a formal J33 50 retrieval model; if the model makes little or no appeal to an J33 51 underlying theory we call it ad hoc. We use the terms J33 52 theory and model here in the mathematical or logical sense in which J33 53 a theory refers to a set of axioms and inference rules that allow J33 54 derivation of new theorems. A model is an embodiment of the theory J33 55 in which we define the set of objects about which assertions can be J33 56 made and restrict the ways in which classes of objects can J33 57 interact.

J33 58 Four retrieval models are particularly important in IR J33 59 research: the Boolean, cluster-based, probabilistic, and J33 60 vector-space models. Most of the commercial and prototype IR J33 61 systems currently available are based on some variation of these J33 62 models, and some understanding of them is necessary for our J33 63 discussion of the inference net model in the next sections.

J33 64 Boolean. Boolean retrieval forms the basis of most major J33 65 commercial retrieval services, but is generally believed to be J33 66 difficult to use and has poor recall and precision performance J33 67 since the model does not rank documents. In the Boolean model we J33 68 have a finite set of representation concepts or features R = J33 69 {r1, ..., rk} that can be assigned to J33 70 documents. A document is simply an assignment of representation J33 71 concepts and this assignment is often represented by a J33 72 binary-valued vector of length k. The assignment of a J33 73 representation concept ri to a document is J33 74 represented by setting the ith element of the vector J33 75 to true. All elements corresponding to features not assigned J33 76 to a document are set to false.

J33 77 An information need is described by a Boolean expression in J33 78 which operands are representation concepts. Any document whose set J33 79 of representation concepts represents an assignment that satisfies J33 80 the Boolean expression is deemed to match the information need, all J33 81 other documents fail to match the information need. This evaluation J33 82 partitions the set of documents, but provides no information about J33 83 the relative likelihood that documents within the same partition J33 84 will match the information need.

J33 85 Relevance in Boolean retrieval, then, is defined in terms of J33 86 satisfiability of a first-order logic expression given a set of J33 87 document representations as axioms. Several attempts have been made J33 88 to extend the basic Boolean model to provide document ranking.

J33 89 Cluster-based retrieval. Cluster-based retrieval is J33 90 based on the Cluster Hypothesis which asserts that similar J33 91 documents will match the same information needs [van Rijsbergen, J33 92 1979]. Rather than comparing representations of individual J33 93 documents to the representation of the information need, we first J33 94 form clusters of documents using any of several clustering J33 95 algorithms and similarity measures. For each cluster, we then J33 96 create an 'average' or representative document and compare this J33 97 cluster representative to the information need to determine which J33 98 clusters best match. We then retrieve the clusters that are most J33 99 likely to match the information need rather than the individual J33 100 documents. There are several ways to identify the clusters to be J33 101 retrieved, particularly when using hierarchical clustering J33 102 techniques that allow navigation of the cluster hierarchy.

J33 103 Since many techniques are used to compare the query with the J33 104 cluster representative, there is no single definition of relevance J33 105 for cluster-based retrieval. Rather, relevance is partially defined J33 106 by the model that forms the basis of the comparison. The similarity J33 107 measures used to define clusters and the method used to create the J33 108 cluster representatives also play a part in defining relevance J33 109 since they determine which documents will be judged similar to a J33 110 cluster representative that matches the information need.

J33 111 Vector-space retrieval. In the vector-space model, we J33 112 have a set of representation concepts or features R = J33 113 r1,..., rk}. Documents and queries J33 114 are represented as vectors of length k in which each element J33 115 corresponds to a real-valued weight assigned to an element of the J33 116 representation set. Several techniques have been used to compute J33 117 these weights, the most common being tf.idf weights which are J33 118 based on the frequency of a term in a single document (tf) and J33 119 its frequency in the entire collection (idf). These J33 120 tf.idf weights are discussed in more detail in [Salton and J33 121 McGill, 1983].

J33 122 Documents and queries are compared using any of several J33 123 similarity functions, the most common function being the cosine of J33 124 the angle between their representation vectors.

J33 125 Since several techniques have been used to compute weights for J33 126 the vector elements, the vector-space model has no single form of J33 127 document or query representation, although all representations have J33 128 a common form. Similarly, since several similarity functions have J33 129 been used, relevance has no single definition.

J33 130 The vector-space model is historically important since it forms J33 131 the basis for a large body of retrieval research that can be traced J33 132 back to the 1960's. The vector-space model has been criticized as J33 133 an ad hoc model since there is relatively little J33 134 theoretical justification for many of its variations.

J33 135 Probabilistic retrieval. Probabilistic retrieval is J33 136 based on the Probability Ranking Principal which asserts that the J33 137 best overall retrieval effectiveness will be achieved when J33 138 documents are ranked in decreasing order of probability of J33 139 relevance [Robertson, 1977].

J33 140 There are several different probabilistic formulations which J33 141 differ mainly in the way in which they estimate the probability of J33 142 relevance. Using a representative model, a document J33 143 di and an information need fj are J33 144 represented as the now familiar vectors of length k in which J33 145 each element is true if the corresponding representation concept is J33 146 assigned to the document or query. If we let F represent the J33 147 set of representations for information needs and D represent J33 148 the set of document representations, then we can define an event J33 149 space F x D and our task becomes one of determining which J33 150 of these document-request pairs would be judged relevant, that is, J33 151 estimating P(R|d<*_>i<*/>,fj). We then use Bayes' J33 152 theorem and a set of independence assumptions about the J33 153 distribution of representation concepts in the documents and J33 154 queries to derive a ranking function that computes J33 155 P(R|d<*_>i<*/>,fj) in terms of the probabilities J33 156 that individual representation concepts will be assigned to J33 157 relevant and non-relevant documents. Different independence J33 158 assumptions lead to different forms of the model. Given estimates J33 159 for these two probabilities (say, from a sample of documents judged J33 160 relevant and from the entire collection), we can compute J33 161 P(R|d<*_>i<*/>,fj).

J33 162 Probabilistic models are in many ways similar to the J33 163 vector-space model [Bookstein, 1982; Turtle and Croft, 1992]. Both J33 164 can be considered to be generalizations of the Boolean model in J33 165 that they can support partial matching using Boolean queries. J33 166 Probabilistic models, however, provide a sounder theoretical base J33 167 for the design of IR systems, and have significantly contributed to J33 168 our understanding of some aspects of IR, such as term weighting, J33 169 ranking, and relevance feedback.

J33 170 7.3 Retrieval Based on Inference and Networks

J33 171 Recent retrieval research has suggested that significant J33 172 improvements in retrieval performance will require techniques that, J33 173 in some sense, 'understand' the content of documents and queries J33 174 [van Rijsbergen, 1986; Croft, 1987] and can be used to infer J33 175 probable relationships between documents and queries. In this view, J33 176 information retrieval is an inference or evidential reasoning J33 177 process in which we estimate the probability that a user's J33 178 information need, expressed as one or more queries, is met given a J33 179 document as 'evidence.'

J33 180 The idea that retrieval is an inference or evidential reasoning J33 181 process is not new. Cooper's logical relevance [Cooper, 1971] is J33 182 based on deductive relationships between representations of J33 183 documents and information needs. Wilson's situational relevance J33 184 [Wilson, 1973] extends this notion to incorporate inductive or J33 185 uncertain inference based on the degree to which documents support J33 186 information needs. The techniques required to support these kinds J33 187 of inference are similar to those used in expert systems that must J33 188 reason with uncertain information. A number of competing inference J33 189 models have been developed for these kinds of expert systems and J33 190 several of these models can be adapted to the document retrieval J33 191 task.

J33 192 In this paper, we describe a retrieval model based on inference J33 193 networks. This model is intended to

J33 194 <*_>bullet<*/>Support the use of multiple document J33 195 representation schemes. Research has shown that a given query will J33 196 retrieve different documents when applied to different J33 197 representations, even when the average retrieval performance J33 198 achieved with each representation is the same. Katzer, for example, J33 199 found little overlap in documents retrieved using seven different J33 200 representations, but found that documents retrieved by multiple J33 201 representations were likely to be relevant [Katzer et J33 202 al., 1982]. Similar results have been obtained when comparing J33 203 term- with cluster-based representations [Croft and Harper, 1979] J33 204 and term- with citation-based representations [Fox et J33 205 al., 1988].

J33 206 <*_>bullet<*/>Allow results from different queries and query J33 207 types to be combined. Given a single natural language description J33 208 of an information need, different searchers will formulate J33 209 different queries to represent that need and will retrieve J33 210 different documents, even when average performance is the same for J33 211 each searcher [McGill et al., 1979; Katzer et J33 212 al., 1982]. Again, documents retrieved by multiple searchers J33 213 are more likely to be relevant. A description of an information J33 214 need can be used to generate several query representations (e.g., J33 215 probabilistic, Boolean), each using a different query strategy and J33 216 each capturing different aspects of the information need. These J33 217 different search strategies are known to retrieve different J33 218 documents for the same underlying information need [Croft, J33 219 1987].

J33 220 <*_>bullet<*/>Facilitate flexible matching between the terms or J33 221 concepts mentioned in queries and those assigned to documents. The J33 222 poor match between the vocabulary used to express queries and the J33 223 vocabulary used to represent documents appears to be a major cause J33 224 of poor recall [Furnas et al., 1987]. Recall can be J33 225 improved using domain knowledge to match query and representation J33 226 concepts without significantly degrading precision.

J33 227 The resulting formal retrieval model integrates several J33 228 previous models (probabilistic, Boolean, and cluster-based) in a J33 229 single theoretical framework [Turtle, 1990]. J33 230 J33 231 J34 1 <#FROWN:J34\>Manipulating the +/- UG distinction in terms of J34 2 Coppieters' percentage figures turns out to be something of a J34 3 vacuous exercise, however, since many of the percentages are based J34 4 on small numbers of exemplars, thus lowering their face validity. J34 5 Recall that the percentages for NNS-NS deviance are derived from J34 6 varying numbers of tokens of each linguistic variable. At one J34 7 extreme of NNS-NS divergences are the variables Imparfait vs. J34 8 Pass<*_>e-acute<*/> Compos<*_>e-acute<*/> (39.5% divergence) and J34 9 <*_>a-acute<*/> vs. de + Infinitive (34.7%), which are J34 10 exemplified by 5 tokens and 2 tokens, respectively. At the other J34 11 extreme is the A-over-A variable (14.4%), exemplified by 6 tokens. J34 12 As numbers of tokens decrease, percentage figures become less valid J34 13 indices of response patterns. It would be a simple matter, for J34 14 example, to deflate or inflate the deviance for the J34 15 <*_>a-acute<*/> vs. de + Infinitive category by changing J34 16 (or adding, or subtracting) a single well-chosen item. J34 17 Clark 1973 has demonstrated the dangers of inference from few J34 18 tokens of a given category: there is no evidence that the findings J34 19 will generalize beyond the items sampled and apply to the category J34 20 as a whole.

J34 21 A contingency in data elicitation further clouds the issue of J34 22 locus of competence differences. For some 41 of the 107 items, J34 23 subjects were given a choice between paired forms (e.g. J34 24 Est-ce que tu {as su/savais} conduire dans la J34 25 neige? 'Did you {manage/know how} to drive in the J34 26 snow?'). They were then asked by the experimenter to decide: (1) J34 27 whether both forms were acceptable; (2) if so, whether a difference J34 28 in meaning was entailed in the differing forms; (3) if so, J34 29 "what that difference consisted of in the framework of J34 30 sentence interpretation" (550). These 41 items were tokens J34 31 of five linguistic variables: the Imparfait/Pass<*_>e-acute<*/> J34 32 Compos<*_>e-acute<*/> distinction, il/elle vs. ce, J34 33 prepositions <*_>a-acute<*/> and de + Infinitive, J34 34 prenominal vs. postnominal adjectives, and article use. The other J34 35 elicitation procedure required subjects merely "to provide J34 36 straightforward well-formedness judgments" (550) J34 37 for single (unpaired) sentences. This was done for "most of J34 38 the other 66 sentences" (550), representing four linguistic J34 39 categories: the Object + Predicate construction, Causatives and J34 40 Clitics, A-over-A constraint, and the de + Adjective J34 41 construction. Reviewing the data in Table 1, above, one may discern J34 42 a striking pattern. For the top five variables - just those J34 43 variables tested by the first technique - NNS diverged dramatically J34 44 from the NS (NNS-NS average = 29.2%) and the range of NNS-NS J34 45 divergence is disparate (19.9% - 39.5%). For those variables tested J34 46 under the second procedure, on the other hand, the NNS-NS J34 47 divergence is relatively small (average = 17.3%), with the range J34 48 clustering tightly (14.4% - 18.9%). Comparing these figures with J34 49 the deviance figures for the +/- UG distinction (average for J34 50 putatively - UG items = 27.2%, range = 16.9% - 39.5%; average for J34 51 putatively + UG items = 17.4%, range = 14.4%-18.9%), one notes that J34 52 the magnitude and heterogeneity of NNS-NS divergence are somewhat J34 53 better predicted by the procedural contingency than by Coppieters' J34 54 version of the +/- UG distinction.

J34 55 The issue of elicitation procedure has special significance for J34 56 de + Adjective structure. Coppieters registers his surprise J34 57 that NNS should diverge so minimally from NS (16.9%) on such a J34 58 "comparatively obscure" construction (561). It is J34 59 an ad hoc move to invoke frequency as a predictor of divergence for J34 60 this structure alone. Left unmentioned is the fact that minimal J34 61 divergences for this structure are an embarrassment for the +/- UG J34 62 distinction as a general predictor of NNS-NS differences. As it J34 63 happens, the de + Adjective construction was tested under the J34 64 second procedure; for it and the other structures tested in this J34 65 manner, NNS-NS divergences were small. There is no anomaly to J34 66 account for if one allows for the possibility that the asymmetry in J34 67 NNS-NS differences is an artifact of a procedural asymmetry. J34 68 Coppieters is apparently aware of the procedural asymmetry, but J34 69 does not attribute differential results to it (553-54).

J34 70 Though one can only speculate about whether differences in J34 71 procedure are in fact causally related to differences in results, J34 72 it is nevertheless easy to see how there might be a relationship. J34 73 The first procedure involved a three-part, fine-grained decision J34 74 task in the context of a face-to-face interview with the J34 75 experimenter. In the second procedure, subjects were asked merely J34 76 to give nominal statements of well-formedness. In the latter J34 77 instance, it is unlikely that subtle cues could be communicated by J34 78 the experimenter to the subjects. However, that likelihood J34 79 necessarily increases in an elaborated discussion of form, meaning, J34 80 and interpretation, as was the case in the first procedure. I offer J34 81 this observation about the relative possibility of experimenter J34 82 effects as a statement of objective fact about experimental J34 83 procedure (see Heringer 1970, Labov 1975, Rosenthal 1966), not to J34 84 dispute Coppieters' assertion that he reinforced all subjects in J34 85 their judgments and gave no indication when judgments diverged from J34 86 native norms (552-53). Experimenter effects can be unintentional J34 87 and unnoticed; measures should be taken to reduce their J34 88 likelihood.

J34 89 More to the point, one must consider the nature of the data J34 90 resulting from the two different elicitation procedures. Recall J34 91 that the results displayed in Fig. 1 and the two left columns of J34 92 Table 1 represent cumulative departures from NS norms, the norms J34 93 being in the form of nominal statements of acceptability or J34 94 grammaticality. The second procedure generates just such nominal J34 95 statements for each item, and departures from these norms are J34 96 straightforwardly tallied. The first procedure, by contrast, J34 97 involves direct comparisons of two sentences. In this case, the J34 98 subjects declare that the sentences are equally acceptable, or that J34 99 they prefer one sentence over the other. Thus in one procedure J34 100 judgments of acceptability reflect assessments against some J34 101 external criterion of acceptability, while in the other procedure J34 102 the judgments reflect comparisons of one sentence with another. A J34 103 further difference between the procedures should also be noted. The J34 104 first procedure, in eliciting three-part answers, generates complex J34 105 commentaries on subtle interactions of syntax and semantics. Anyone J34 106 who has ever elicited (or offered) such comments recognizes that J34 107 they may contain references to imagined contexts, statements J34 108 reflecting judgments based on degrees of well-formedness, J34 109 expressions of ambivalence, etc. This type of data may be J34 110 contrasted with the simple nominal statements ('correct or J34 111 good'/'incorrect or bad') generated by the second procedure. J34 112 (Moreover, what is meant by 'good' and 'bad' is not spelled out for J34 113 participants, creating the possibility of intersubject variability J34 114 as to the basis of judgment.) In light of multiple differences J34 115 between the two procedures, one must recognize that their results J34 116 are inherently dissimilar and therefore not comparable. The two J34 117 types of data should not be lumped together as if they were J34 118 identical.

J34 119 However, Coppieters seems to do just that. The results of the J34 120 first procedure, in order to be considered relevant input for the J34 121 NNS-NS comparisons displayed in Fig. 1 and Table 1, have apparently J34 122 been converted into nominal data. The details of this conversion J34 123 operation are not spelled out. How does one establish a unique NS J34 124 norm, and nominal departures from it, when the task generates a J34 125 complex response? By what algorithm is a comparison of two related J34 126 items, with an accompanying three-part, fine-grained discussion, J34 127 transformed into an independent statement of acceptability for each J34 128 of the items? These questions speak to the reproducibility of the J34 129 results, as well as to their validity.

J34 130 With respect to the more general question of whether there are J34 131 NNS-NS competence differences, Coppieters' methodology is again J34 132 problematic. Recall that the differences he attests are derived J34 133 from comparisons of NS and NNS deviations from a prototypical J34 134 native norm. The norms represent the majority of NS nominal J34 135 judgments. For 90 of the 107 items, there was 80% or more agreement J34 136 in judgment among NS. For the other 17 items, "the majority J34 137 opinion of NS's" (553) must be understood as any level of J34 138 agreement above 50%. Clearly, it would have been desirable to J34 139 establish norms at a high level of NS agreement for all items. Even J34 140 if this had been the case, one must question the use of nominal J34 141 data to establish norms. Nominal judgments are insensitive to J34 142 degrees of grammaticality or acceptability. Chaudron 1983 has J34 143 argued that scaled judgments of grammaticality (e.g. 1 = totally J34 144 unacceptable/ungrammatical; 5 = perfectly acceptable/grammatical) J34 145 are more informative than nominal data, inasmuch as they permit J34 146 intergroup comparisons (e.g. t-tests and ANOVAs) on the J34 147 basis of means and variance from means. The psychological validity J34 148 of scalar grammaticality statements is supported by Barsalou 1987 J34 149 and others working in category/concept theory. The category of J34 150 well-formedness is susceptible to gradedness effects in J34 151 experimental performance, just like other categories such as BIRD J34 152 (robins and sparrows are judged more birdlike than emus, ostriches, J34 153 penguins, and apteryxes). Moreover, Chomsky 1986 has argued for J34 154 relative grammaticality in syntax as a function of the number of J34 155 barriers crossed in extractions. In sum, there are numerous reasons J34 156 for preferring scalar responses to nominal statements in J34 157 determining differences between NS and NNS.

J34 158 The composition of the subject groups in the Coppieters study J34 159 presents additional problems. To qualify for participation, J34 160 nonnative subjects had to perform at a near-native level of speech, J34 161 as determined by native-speaker friends of the subjects and by the J34 162 experimenter himself. In addition, by Coppieters' reckoning, all J34 163 performed orally at a level of 'superior' by ACTFL criteria. (For a J34 164 description of the ACTFL oral proficiency interview, see ACTFL J34 165 1986.) However, both of these screening procedures are imprecise J34 166 and nondeterministic. Informal near-native ratings cannot assure J34 167 native-like performance across a variety of linguistic structures. J34 168 It is quite possible to sound 'native-like' by avoiding production J34 169 of complex structures which are not fully mastered. It is also J34 170 possible to sound 'native-like' in restricted contexts, such as J34 171 bank transactions or philosophy lectures. Indeed, for restricted J34 172 and practiced contexts, there may be apparent pragmatic mastery, J34 173 while for other contexts there may be manifest inadequacies. The J34 174 second screening measure leaves a great deal of latitude for J34 175 composing the near-native group, inasmuch as the ACTFL category of J34 176 'superior' ranges between the 3rd and 5th levels of a 5-point J34 177 scale. By virtue of these screening procedures, it is quite J34 178 possible that the near-native group was not particularly J34 179 proficient. It is not clear that Coppieters satisfies the J34 180 desideratum of Long 1990 for investigating the general question of J34 181 competence differences, viz., that at least some subjects sampled J34 182 be among "the very best" L2 learners.

J34 183 The biographical profiles of NNS and NS subjects pose further J34 184 problems. As has been documented by Ross 1979 and others, one may J34 185 expect differences in grammaticality judgments - even among native J34 186 speakers - if sociological and educational variables are not J34 187 controlled. Coppieters' 21 NNS were "highly J34 188 educated" (551) and included 19 professors or researchers, J34 189 eleven of whom were professors of linguistics or literature. In J34 190 contrast, his 20 NS had merely "had some education" J34 191 (551). Alone, such background differences could account for a good J34 192 deal of the NNS-NS deviance. That is, failure to control for J34 193 background differences between the NNS and NS increases the chances J34 194 that the two groups will not coincide on judgments for J34 195 "subtle areas" (Long 1990:281) of the grammar, just J34 196 as NS or NNS with heterogeneous backgrounds would be likely to J34 197 disagree among themselves.

J34 198 The risks of heterogeneity are also evident in the native J34 199 language backgrounds of the NNS. Recall that Coppieters' subjects J34 200 are from a variety of disparate linguistic backgrounds. In J34 201 principle, this design feature would allow one to determine whether J34 202 results generalize beyond speakers of a single native language. J34 203 However, with only 2 subjects representing Farsi, only 5 subjects J34 204 representing 'Oriental' languages, and 6 different native languages J34 205 among the 14 native speakers of Germanic and Romance, one cannot J34 206 assume validity of results for the individual native language J34 207 groups, much less generalize across groups.

J34 208 Because of these shortcomings, and others that will come to J34 209 light in later sections of this paper, the Coppieters study must be J34 210 viewed with some skepticism. At the very least, it should be J34 211 evident that the two main questions of maturational effects in L2A J34 212 - whether there are competence differences between NNS and NS, and J34 213 if there are, which linguistic domains are affected - remain open. J34 214 Further examination of the issues is therefore warranted. As a J34 215 first step, it is important to determine whether the principal J34 216 findings of Coppieters can be replicated with methodological J34 217 modifications and under conditions of tighter procedural J34 218 control.

J34 219 J34 220 J35 1 <#FROWN:J35\>The other sources for the Eastern dialect give forms J35 2 containing [ts] or [<*_>c-hacek<*/>] rather than [ks]. These J35 3 include Lean and Mulia, the two archaic sources cited by Lehmann J35 4 (1920:656), who give //lochac//, presumably [lo<*_>c-hacek<*/>ak], J35 5 and all of the modern sources, e.g., Dennis and Royce (1983:14), J35 6 who give lots'akh. Campbell and Oltrogge (1980:219) J35 7 reconstruct *lots'ak. The [ks] reported by some sources is J35 8 likely an erroneous transcription, due to lack of familiarity with J35 9 ejectives. Moreover, none of the sources contains any word of the J35 10 form sak or sax that might be what Greenberg J35 11 considers to be the second member of the form, or anything like J35 12 ala-k 'glowing' or lak 'glow'. Nor do these J35 13 forms appear in the notebook. The notebook contains no headword J35 14 'glow', and no entry for Jicaque under the headword 'shine', nor J35 15 has inspection of all of the Jicaque entries yielded any plausible J35 16 sources for these forms. The only entries under 'sun' are J35 17 loksak, latsak, and loksaki. It thus J35 18 appears that there is no evidence whatsoever internal to Jicaque J35 19 for analyzing laksak into lak-sak. In other J35 20 words, the comparison is between two words, each containing the J35 21 sequence lak (if the Jicaque contains a [k] at all), in J35 22 neither of which there is the slightest evidence for an analysis in J35 23 which lak is a morpheme. In contrast, Greenberg fails to J35 24 mention the etymology proposed by Mason (1918:19), namely, that J35 25 ma<*_>unch<*/>alak is derived from J35 26 ma<*_>unch<*/>al 'burn, blaze' (cf. J35 27 ma<*_>unch<*/>ale 'flames'). The weakness in Mason's J35 28 etymology is that no suffix -ak is known, but, there being J35 29 evidence for one of the two morphemes, it is nonetheless more J35 30 plausible than Greenberg's.

J35 31 246 STONE2

J35 32 LIA cites M. i<*_>unch<*/>ak and J35 33 i<*_>unch<*/>ik 'knife' and asak'a 'flint'. The J35 34 last is attested but is indicated by Mason (132) to be J35 35 Antonia<*_>n-tilde<*/>o. Neither of the forms for 'knife' appears J35 36 in the notebook, which gives M. <*_>c-hacek<*/>a:k, A. J35 37 <*_>c-hacek<*/>ikh, formula, and J35 38 formula. Nor do they appear in Mason (130), who gives M. J35 39 <*_>c-hacek<*/>ak, A. formula. The form J35 40 i<*_>unch<*/>ak is cited by Greenberg and Swadesh J35 41 (1953:219). The forms in LIA as well as the form in Greenberg J35 42 and Swadesh (1953) appear to derive from Sapir (1917:8), who gives J35 43 the forms (i)<*_>unch<*/>a:k and J35 44 (i)<*_>unch<*/>ik. In the footnote to this entry, Sapir J35 45 cites the form as <*_>unch<*/>ak, suggesting that the J35 46 forms with the initial i may be typographical errors, with J35 47 i in place of the articular prefix t, the parentheses J35 48 intended to segregate it from the stem.

J35 49 255 THROAT

J35 50 LIA lists A.p-e:nik'a. The notebook has J35 51 pe:nik'a:i, which is the form given by Mason (127). The J35 52 basis for segregating the /p/ is unknown.

J35 53 4.2. Grammatical evidence. Chapter 5, 'Grammatical J35 54 Evidence for Amerind,' cites Salinan data in eleven places, J35 55 sections 6, 13, 16, 19, 45, 74, 80, 84, 88, 90, 100. Of these, the J35 56 following five call for comment.

J35 57 In section 19 (p. 288) LIA gives the J35 58 Miguele<*_>n-tilde<*/>o first-person plural pronoun as J35 59 ka. The correct form, according to Mason (32) is J35 60 formula.

J35 61 Section 80 (p. 311) reads in its entirety:

J35 62 The following etymology represents a past-tense marker J35 63 in HOKAN: Yuru-mangu<*_>i-acute<*/> iba, Coahuilteco J35 64 pa-, Tequistlatec -pa, Salinan be, Pomo (hi)ba, J35 65 Karok -<*_>unch<*/>ipa<*_>approximate-sign<*/>pa- (near J35 66 past), and Shasta p'- (distant past, habitual past). The J35 67 marker probably also occurs in Salinan iwa-<*_>unch<*/>, which J35 68 when suffixed to nouns means 'that which was formerly', e.g. J35 69 noq<*_>unch<*/> 'head', formula 'skull'. The last J35 70 element, -<*_>unch<*/>, is a common noun formant in Chumash. The J35 71 agreement in a form *ipa among Yurumangu<*_>i-acute<*/> in the J35 72 extreme south, Coahuilteco in the middle, and Karok in the far J35 73 north is striking.

J35 74 This passage presents a number of difficulties. First, Salinan J35 75 be is not a simple past tense morpheme as Greenberg J35 76 glosses it. Mason (1918:35) glosses be as 'when, definite J35 77 past time', as in such examples as be:-ya 'when I went'. Sapir J35 78 (1920:307) suggests that be:ya is really "an J35 79 indicative e:ya 'I went' subordinated by the J35 80 demonstrative stem pe, pa 'the, that'." J35 81 Sapir's view is supported by Mason's statement that "Pure J35 82 sonant b has been found only in the case of the demonstrative J35 83 article pe..." (1918:11). If Sapir is right, be J35 84 is not a tense morpheme at all.

J35 85 Second, the agreement among the forms is perhaps less striking J35 86 than Greenberg suggests. Of the eight forms cited, only four J35 87 (Karok, Pomo, Yurumangu<*_>i-acute<*/>, Salinan J35 88 iwa<*_>unch<*/>) have the initial i, as we shall J35 89 see, the initial i of iwa<*_>unch<*/> is probably J35 90 not original. Moreover, some of these morphemes are prefixes while J35 91 others are suffixes. Finally, as discussed below in 5.3, it is J35 92 unclear whether the Yurumangu<*_>i-acute<*/> past tense morpheme is J35 93 the infix -iba- or the suffix -bai.

J35 94 Third, the discussion of iwa<*_>unch<*/> is seriously J35 95 flawed. To begin with, this suffix and the examples cited are not J35 96 Salinan. The suffix iwa<*_>unch<*/> is not mentioned in J35 97 any source on Salinan that I have consulted, nor are the examples J35 98 cited as illustrating the use of this suffix J35 99 (noq<*_>unch<*/> 'head', formula 'skull'). Nor do J35 100 they appear under the headings 'head' and 'skull' for Salinan in J35 101 Greenberg's Hokan notebook. Indeed, Salinan has no [q] at all.

J35 102 Rather, the suffix iwa<*_>unch<*/> and the examples J35 103 of its use are Chumash. noq<*_>unch<*/> and J35 104 formula are among the examples of the use of the suffix J35 105 iwa<*_>unch<*/> in Barbare<*_>n-tilde<*/>o Chumash given J35 106 by Beeler (1976:259), one of the sources listed in the notebook. J35 107 The notebook lists noks under 'head' for J35 108 Inese<*_>n-tilde<*/>o and Barbare<*_>n-tilde<*/>o Chumash.

J35 109 Nonetheless, the issue of the cognation of Chumash J35 110 iwa<*_>unch<*/> with the other forms cited still arises. J35 111 Two facts militate against Greenberg's analysis. First, the initial J35 112 i is very likely not original. There are two related suffixes J35 113 in Chumash, the suffix iwa<*_>unch<*/> which derives J35 114 nouns meaning 'dead, defunct, former', and the verbal past tense J35 115 wa<*_>unch<*/>. The nominal suffix is invariant, as is J35 116 the verbal suffix in Barbare<*_>n-tilde<*/>o (Beeler 1976). In J35 117 Inese<*_>n-tilde<*/>, however, according to Applegate (1972:102-3), J35 118 an epenthetic copy of the last vowel of the stem is inserted before J35 119 the past tense marker when the stem ends in a sonorant. Although J35 120 the history is not known, it seems very likely that the nominal J35 121 suffix is historically derived from the verbal past tense, and that J35 122 the initial i represents a frozen epenthetic vowel.

J35 123 Second, there is little basis for segmenting out the final J35 124 <*_>unch<*/>. If iwa<*_>unch<*/> is derived from the J35 125 verbal suffix wa<*_>unch<*/>, it is unlikely that the J35 126 final <*_>unch<*/> is a noun-forming suffix. Noun-forming J35 127 suffix <*_>unch<*/> is not described in any source on Chumash that J35 128 I have consulted, including the two most detailed grammatical J35 129 descriptions, Applegate (1972) for Inese<*_>n-tilde<*/>o and Beeler J35 130 (1976) for Barbare<*_>n-tilde<*/>o. There is a suffix J35 131 -V<*_>unch<*/> described by both Beeler (1976:258) and J35 132 Applegate (1972:213) as "resultative," and it is to this J35 133 suffix that Greenberg refers. The V means that the suffix J35 134 takes the form of a vowel followed by <*_>unch<*/>. According to J35 135 Beeler (1976) this vowel is unpredictable, but Applegate (1972:93) J35 136 gives rules for predicting the vowel. In any case, the derivation J35 137 envisioned by Greenberg is not clear. If we start from a verbal J35 138 past tense suffix *ipa, which in Chumash comes to be added to J35 139 nouns as well as to verbs, it would seem unnecessary to add a J35 140 resultative suffix, and one has to wonder why the basic past tense J35 141 suffix would acquire the resultative suffix attached to it in J35 142 nouns.

J35 143 In section 84 (p. 311), LIA cites a suffix J35 144 -<*_>unch<*/>e 'desiderative' for Antonia<*_>n-tilde<*/>o. J35 145 Mason discusses such a suffix, but with final glottal stop J35 146 formula, on page 49, and although he glosses it J35 147 'desiderative', he makes it clear that he is far from confident of J35 148 this interpretation, a hesitation fully justified by the varied J35 149 meanings of the examples adduced.

J35 150 In section 88 (p. 313), LIA refers to a Salinan imperative J35 151 morpheme -i-. This suffix is not described by Mason in his J35 152 discussion of the verbal morphology (1918:34-54), nor in Turner J35 153 (1987), so that it appears to be quite spurious, and given the lack J35 154 of documentation, one despairs of tracking it down. The key turns J35 155 out to be Rivet (1942:33), which presents the same equation as in J35 156 LIA, minus the Karok form, which was evidently added by J35 157 Greenberg. Rivet cites Sapir (1921:71), in which we find, as entry J35 158 number 28, the following:

J35 159 Sal. -i-, imperative suffix with third person J35 160 pronominal object (e.g. m-alel-i-k ASK HIM!): Yana J35 161 -'i', imperative suffix.

J35 162 This constitutes the entire discussion of imperative -i- J35 163 in the literature. At the very least Greenberg is to be taken to J35 164 task for using as evidence a morpheme for which the evidence is so J35 165 skimpy, especially when it does not appear in Mason (1918), the J35 166 only published grammatical description of the language and the J35 167 source of Sapir's data.

J35 168 Whether Sapir's analysis should be accepted is unclear. Salinan J35 169 has a number of third-person singular objective suffixes, which J35 170 according to Mason (1918:46-47) take the form -o, -ko, J35 171 -xo, and -k. The different suffixes are associated with J35 172 different classes of verb: "... the -p prefix nearly J35 173 invariably takes the suffix -o or -ko as its third J35 174 personal objective form while the objective form in -k occurs J35 175 exclusively with the -k prefix" (Mason 1918:39). That J35 176 is, the forms in -k are associated with what Mason considered J35 177 to be the intransitive verbs, later argued by Sapir (1920:307-8; J35 178 1921:69-70) to be stative. In his discussion of the imperative, J35 179 Mason (1918:41) states that : "The imperative takes its J35 180 third person pronominal object in -ik, never in -o or J35 181 -ko."

J35 182 Sapir's reasoning appears to have been that the i of this J35 183 suffix could be isolated as an imperative since the suffix appears J35 184 only in imperatives and since other third-person singular objective J35 185 suffixes contain k. Against this we may consider the fact that J35 186 imperative i occurs nowhere else and that it is not possible J35 187 to analyze the third-person singular objective morpheme simply as J35 188 k; even if we could extract k from ko, we would be J35 189 left with the forms in o. Moreover, judging from Mason's J35 190 examples of the use of -ik, namely J35 191 k-<*_>a-acute<*/>mamp-ik 'take it out!', J35 192 <*_>a-acute<*/>mes-ik 'shout to him!', and m-alel-ik J35 193 'ask him!', it appears that -ik occurs even with active verbs, J35 194 which according to Mason's generalization do not take the objective J35 195 prefixes containing k. On balance, it seems that -ik must J35 196 be treated as a unit, and that Sapir's analysis is overly J35 197 aggressive.

J35 198 In section 90 (p. 313) LIA cites k- as the imperative J35 199 prefix. Actually, according to Mason (1918:41) this prefix appears J35 200 only in the plural, and not in all cases.

J35 201 5. Yurumangu<*_>i-acute<*/>. Yurumangu<*_>i-acute<*/> J35 202 forms occur in a total of twenty-six entries in LIA, to wit: J35 203 H15, H32, H35, H44, H61, H69, H72, H87, H142, H158, H161, A9, A79, J35 204 A86, A102, A104, A191, A243, A269, G19, G61, G80, G88, G90, J35 205 G102.

J35 206 5.1. Segmentation. Morphological analysis is often J35 207 difficult even in languages for which we have unlimited data; in a J35 208 case like Yurumangu<*_>i-acute<*/>, where we have only a tiny J35 209 corpus containing few related forms, we must proceed with the J35 210 utmost caution and must expect to remain unsure on many points.

J35 211 Rivet made a valiant effort at analysis on the basis of the J35 212 data available to him, using both what little language-internal J35 213 evidence there was, and what was suggested to him by similar forms J35 214 in Hokan languages. Many of these proposals are interesting and J35 215 would be worthy of pursuit if we were convinced of the affiliation J35 216 of Yurumangu<*_>i-acute<*/> with Hokan, but until such an J35 217 affiliation is proved they remain the merest speculation.

J35 218 The result of the dearth of Yurumangu<*_>i-acute<*/> data J35 219 combined with the fecundity of Rivet's etymological imagination is J35 220 that most of the morphological analyses he proposed rest on J35 221 hypothetical affiliations with Hokan, and so remain undemonstrated. J35 222 Rather than exercising caution and utilizing only justifiable J35 223 analyses, Greenberg simply accepts Rivet's analyses and presents J35 224 them as if they were clearly justified internal to J35 225 Yurumangu<*_>i-acute<*/>, as we shall now see in detail.

J35 226 Rivet (1942:28-29) posits a prefix a-, appearing on both J35 227 verbs and nouns. Other than the fact that a considerable number of J35 228 words begin with a, he cites no language-internal evidence for J35 229 the existence of such a prefix. In many cases he gives no evidence J35 230 of morphological complexity of any kind. Where he does give J35 231 evidence, it consists of comparisons with other languages. J35 232 J35 233 J36 1 <#FROWN:J36\>One

J36 2 Palestine and the

J36 3 Coming of the Cold War,

J36 4 January 1947 - March 1947

J36 5 During the winter of 1947, the future of Palestine became J36 6 immersed in East-West policy considerations extending far beyond J36 7 the Eastern Mediterranean. Britain's sudden retreat from empire J36 8 enabled American Zionists and the U.S. State Department to exploit J36 9 rising war jitters as they competed with one another in the mass J36 10 media to create a new conventional wisdom on Palestine. The J36 11 American Zionist Emergency Council (AZEC) insisted American J36 12 involvement in Greece and Turkey should be followed by support for J36 13 a Jewish state. The State Department's Near East desk argued that J36 14 policy would only lead to Communist gains throughout the oil-rich J36 15 region. Both sides attempted to use the American mass media, J36 16 particularly the New York Times, to win converts to their J36 17 cause. They competed in describing a vision of reality that would J36 18 serve as the context for policymaking. The beginning of this J36 19 struggle to shape conventional wisdom at the outset of the Cold War J36 20 and the role the mass media played in that conflict is the subject J36 21 of this chapter.

J36 22 Americans reading the daily press and listening to the nation's J36 23 network commentators in the opening days of 1947 had reason to J36 24 believe that the world was entering an era of peace. Although a J36 25 state of war still technically existed, President Harry S. Truman J36 26 on December 31, 1946 declared the hostilities of World War II J36 27 officially terminated. Emergency laws exercised by the Executive J36 28 Branch since 1941 were ended. A presidential proclamation pledged J36 29 the United States would work with other nations in building a world J36 30 in which justice would replace force.

J36 31 The New York Times reported that American Secretary J36 32 of State James F. Byrnes and Soviet Foreign Minister Vyacheslav M. J36 33 Molotov had just completed negotiations in New York that amicably J36 34 resolved differences on more than forty issues. United Nations J36 35 Secretary General Trygve Lie indicated in his New Year's message J36 36 that a "sound foundation" had been laid and he J36 37 looked to the future with "sober confidence." The J36 38 chief United States delegate to the United Nations, Warren R. J36 39 Austin, told a radio audience that he was convinced the United J36 40 States and the Soviet Union "had come to a better J36 41 understanding of each other" and that he had "great J36 42 faith" that a start had been made on the "long road J36 43 away from war." The nation's foremost radio commentator, J36 44 Hans V. Kaltenborn of the National Broadcasting Company (NBC), J36 45 caught this spirit of optimism when he told listeners, "the J36 46 days ahead will confirm mankind's steady march to better J36 47 things." The United States, Kaltenborn was convinced, had J36 48 "seized the torch of freedom and held it aloft for others J36 49 to see and follow."

J36 50 The New Year's cheer proved short-lived. The British, J36 51 bankrupted by the war and wracked by labor problems, abruptly J36 52 announced in February a pullback from their commitments in the J36 53 Eastern Mediterranean. The collapse of the British economy and the J36 54 Labour government's sudden fallback from empire jolted New J36 55 York Times publisher and board chairman Arthur Hays J36 56 Sulzberger, who remained convinced that the "common J36 57 good" of the free world required a strong Britain. The J36 58 prospect of a disintegrating British empire had Times J36 59 editorial writers fearing the emergence of a world in which the J36 60 United States stood alone in defense of democratic freedoms J36 61 everywhere under attack. Kaltenborn shared this sense of sudden J36 62 foreboding. Gone was the "new age of peace" that J36 63 seemed to have been dawning only weeks before. The British J36 64 cabinet's decision to reduce Britain's armed forces by 340.000 men, J36 65 plus the acute shortages of food, fuel and clothing throughout the J36 66 United Kingdom, created a "dangerous situation," J36 67 Kaltenborn believed, in the postwar world.

J36 68 Britain's pullback from the Eastern Mediterranean exposed J36 69 misgivings long held by members of the American media and the State J36 70 Department over the Soviet Union's postwar aims. In the fifteen J36 71 months that followed, the world was rhetorically split into two J36 72 hostile camps, one Soviet, the other American, separated by J36 73 "deep distrust and mutual suspicion." Palestine, a J36 74 country with a violent history of its own, became absorbed within J36 75 the East-West war of words when Britain, eager to trim a costly J36 76 100,000 man peacekeeping force, announced in late February it was J36 77 turning the Palestine problem over to the United Nations. It was J36 78 these events that set the stage for a mass-mediated battle for J36 79 public opinion at the beginning of the Cold War.

J36 80 JOURNALISM'S FIRST DRAFT ON HISTORY

J36 81 The early postwar rhetoric of Kaltenborn and Sulzberger J36 82 reflects a journalism of self-promotion and self-conscious public J36 83 service, if not self-abnegation. Each man intended to play J36 84 a bigger part in policymaking than either was permitted during the J36 85 war years. Born in Milwaukee, Kaltenborn joined the Fourth J36 86 Wisconsin Voluntary Infantry to fight in the Spanish-American War. J36 87 In the two decades that followed he hustled stories for the German J36 88 language press in Wisconsin before joining the staff of the J36 89 Milwaukee Journal, and later Herbert F. Gunnison's J36 90 Brooklin Daily Eagle. On April 4, 1922 Kaltenborn J36 91 presented a radio address at the Army Signal Corps station in New J36 92 York City, which years later he billed as "the first J36 93 editorial opinion over the air." He made news when he J36 94 climbed Mt. Fuji in 1927 and visited the Soviet Union two years J36 95 later. In 1932 he covered the Democratic National Convention for J36 96 the Columbia Broadcasting System (CBS), later interviewed Hitler J36 97 and Mussolini, reported on the Spanish Civil War and the J36 98 appeasement at Munich.

J36 99 Kaltenborn considered himself a "contemporary J36 100 historian" and saw his fifteen-minute newscasts for NBC J36 101 during the war as first drafts of history. NBC nourished this image J36 102 and on October 16, 1941 organized a town meeting in which J36 103 Kaltenborn answered questions about the war from two hundred of San J36 104 Francisco's "outstanding citizens." NBC also J36 105 arranged for a newsreel camaraman to be present. Subsequently, J36 106 Kaltenborn appeared weekly in newsreels shown across the country. J36 107 During the first six months of the war, more than one million J36 108 theatre goers asked for Kaltenborn's views. While he admitted that J36 109 he "didn't know all the answers," he believed it J36 110 was his responsibility to "guide public opinion.

J36 111 Both Kaltenborn and Sulzberger saw the press as a major player J36 112 in postwar policymaking. Kaltenborn observed that the J36 113 responsibility to lead opinion belonged to the press as much as to J36 114 those "who place their hands over the Bible and swear J36 115 to do their duty in public office." Each made democracy J36 116 possible by their public service. Sulzberger told the nation's J36 117 leading editors that how the press handled its "sacred and J36 118 special mission" might well "determine the destiny J36 119 of the world." The war had shown too clearly, Sulzberger J36 120 claimed, that only "an informed democracy" was J36 121 strong enough to survive in a world where totalitarianism was on J36 122 the march.

J36 123 During the war the press had been an agreeable, if at times, J36 124 reluctant partner in reporting the Roosevelt administration's war J36 125 news and views. Part of the reason was that the press generally J36 126 supported the war effort. In April, 1942, Kaltenborn endorsed the J36 127 idea that total war required a cooperative press in April 1942. J36 128 "We want freedom of speech and press for patriotic J36 129 Americans whose one concern is to win the war," he noted. J36 130 "We want silence from all others." But as the war J36 131 went on, many in the press wearied in the role of messenger service J36 132 for the Roosevelt administration. Radio newsmen in May, 1943 went J36 133 public with a complain that government censors often left J36 134 "the Japs better informed than Americans." Wire J36 135 service reporters were convinced a false image of the war was being J36 136 communicated through the government's determination "to J36 137 hold back and play down American casualties." The American J36 138 Society of Newspaper Editors, representing the nation's largest J36 139 dailies, resented the growth of "pernicious J36 140 propaganda" disseminated by government bureaucrats. It J36 141 charged in April, 1944 that news management had confused the J36 142 American people.

J36 143 Arthur Krock, the Times veteran Washington correspondent, J36 144 believed the war and the controversy growing out of government J36 145 efforts to manage the news would alter the policymaking environment J36 146 in the immediate postwar period. Krock thought the press would be J36 147 less likely to accept the administration's call for a non-partisan J36 148 foreign policy. He wrote that the State Department would be J36 149 incapable of managing postwar foreign policy without significant J36 150 participation by the Congress and the American people. Times J36 151 management shared Krock's conviction. Sulzberger conceived of the J36 152 Times as an "American institution" called upon J36 153 to preserve the country's fragile freedoms through vigorous J36 154 editorial crusading. Charles Merz, the paper's editorial page J36 155 chief, determined to "stir the American people and the J36 156 Congress to their responsibilities." Sunday editor Lester J36 157 Markel saw that responsibility as "educating public J36 158 opinion" to what diplomat correspondent James Reston J36 159 described as "the changes and convulsions in the world in J36 160 which America must operate."

J36 161 The determination of Times management to participate in J36 162 postwar policymaking was a significant departure from the days when J36 163 Adolph S. Ochs seriously considered scrapping the paper's editorial J36 164 page. The Ohio country boy turned grocer's clerk and druggist's J36 165 apprentice told an interviewer that he could get a larger J36 166 circulation by printing a newspaper with all advertisements and no J36 167 news rather than a paper with all news and no advertisements. His J36 168 work as an assistant in the composing room of the Knoxville J36 169 (Tenn.) Tribune and as staff writer and editor of the J36 170 Chattanooga Times convinced him that "news which J36 171 told the exact truth so far as possible" also made good J36 172 economic sense. In 1896 he became publisher and controlling owner J36 173 of the New York Times, then a struggling enterprise in J36 174 the era of the yellow press dominated by Joseph Pulitzer and J36 175 William Randolph Hearst. Ochs' determination to cultivate an elite, J36 176 monied readership resulted in a twenty fold increase in Times J36 177 earnings during the first generation of his stewardship, and it led J36 178 to the Times' dominance "within its own sphere of J36 179 usefulness." Ochs reasoned that only one in 100 readers J36 180 read all the news that Times' editors saw fit to print. But J36 181 that one "would tell the other ninety-nine" and the J36 182 Times would get the reputation of being the "complete J36 183 newspaper."

J36 184 Sulzberger saw interpretation and leading elite opinion as J36 185 central to the postwar role of the Times. Sulzberger had been J36 186 a member of a noteworthy Jewish family that featured J36 187 philanthropists, scholars and jurists, when he married Ochs' only J36 188 daughter in 1917, positioning himself as the heir apparent within J36 189 the Times hierarchy. On Ochs' death in 1935, Sulzberger became J36 190 publisher, president, and chairman of the board at the Times J36 191 and wielded his power to pick associates who shared his commitment J36 192 to a journalism of "public service." Sulzberger saw J36 193 protecting the nation's freedom from totalitarian intrusions at the J36 194 center of that service. He warned fellow editors that the press was J36 195 like the canary miners took down into the shafts with them. J36 196 "It fell over at the least sign of poisonous gas", J36 197 Sulzberger pointed out, and this warning gave others a chance to J36 198 escape. While the Times primarily considered Soviet foreign J36 199 policy before and during the World War II in terms of that nation's J36 200 security interests, by the winter of 1947 it was beginning to smell J36 201 poison gas. "Communists and fellow travelers" J36 202 represented a threat to the American way of life, Sulzberger told J36 203 the nation's editors in April, 1947; the publisher who J36 204 "knowingly employed a communist or any other type of J36 205 totalitarian" or gave him "any place of J36 206 influence" in news or editorial departments J36 207 "threatened the United States itself." The defeat J36 208 of Nazi Germany did not mean fascism was dead, Times J36 209 diplomatic correspondent Herbert L. Matthews wrote, expressing this J36 210 new orthodoxy. Communist authoritarianism was a "Red J36 211 Fascism."

J36 212 Sulzberger and the Times were seen as an important target J36 213 for those within the Truman administration who were pressing for a J36 214 get tough policy toward the Soviet Union. In December, 1945 Navy J36 215 Secretary James V. Forrestal told Sulzberger and Brooks Atkinson, J36 216 the paper's Moscow correspondent, that "the only thing /the J36 217 Russians) recognize is stark force." Forrestal's case to J36 218 Sulzberger was that the Russians had "no respect for the J36 219 normal human weaknesses, such as justice, kindliness and J36 220 affection." At a cabinet meeting three weeks later J36 221 Forrestal urged President Truman to call an emergency meeting of J36 222 Sulzberger and other leading newspapermen, as well Kaltenborn and J36 223 the nation's leading radio commentators, impressing on them J36 224 "the seriousness of the present situation."

J36 225 J37 1 <#FROWN:J37\>During these sessions, the PDRY was consistently J37 2 reported in Sanaa as arguing that the transitional period between J37 3 the announcement of unity and the holding of national elections J37 4 should be extended beyond the six months stipulated in the draft J37 5 constitution. PDRY leaders reportedly feared that six months was J37 6 not enough time to overcome their regime's poor image in the eyes J37 7 of the electorate. There was also speculation that both the YAR and J37 8 PDRY leaders feared that opposition to the draft constitution was J37 9 growing, so they wanted to accelerate the unification process to J37 10 deny opponents time to organize.

J37 11 Predictably, the issue of assigning senior posts in the new J37 12 government proved difficult. Throughout the government structure, J37 13 the principle of a 50-50 split of positions, and of assigning J37 14 northern deputies to southern department heads and vice versa, was J37 15 maintained. The bargaining sessions were long and at times J37 16 rancorous as officials jockeyed endlessly to ensure themselves J37 17 favorable spots in the new hierarchy. Salih and Bidh were named J37 18 president and vice president, respectively, of the five-man J37 19 Presidential Council provided for in the draft unity constitution; J37 20 the PDRY head of state, Haydar Abu Bakr al-Attas, became prime J37 21 minister of a 39-member cabinet, and PDRY prime minister Said Yasin J37 22 Numan was installed as speaker of the parliament. The unity J37 23 agreement said that sufficient time should be allowed for J37 24 transition to permanent unity and specified a two and one-half J37 25 year, rather than a six-month, transition period. The J37 26 constitutional referendum called for at Aden and stipulated in the J37 27 draft constitution was not mentioned.

J37 28 The new state was something of a Potemkin facade. The situation J37 29 was symbolized by the new flags seen around Sanaa, from which it J37 30 was obvious that the green star of the YAR had simply been removed J37 31 to leave the unadorned red, white, and black tricolor of the new J37 32 state. The 'united' ministries tended to be divided into two camps J37 33 of rival northern and southern officials who were more inclined to J37 34 keep to themselves than to integrate with their new colleagues. J37 35 These divisions were particularly sharp in bodies such as the J37 36 Education Ministry, where northern and southern ideas differed J37 37 sharply on fundamental substantive issues such as curriculum. As J37 38 more than one senior official admitted, the complex problems of J37 39 unifying the PDRY's socialist economy with the free market system J37 40 of the YAR were at first swept under the rug with the two economies J37 41 left to function largely as they had done in the past. In the field J37 42 of defense, the general staffs were successfully integrated into a J37 43 unified defense ministry, and a few units were moved from south to J37 44 north and vice versa. Unification at the rank-and-file level, J37 45 however, was left to take place over a protracted period of time, J37 46 in order not to sacrifice military efficiency.

J37 47 The haste with which the two sides came together had its costs J37 48 in the early months of unity. The low salaries that had been paid J37 49 to southern officials were at first not raised quickly enough to J37 50 keep pace with the cost of living, which rose rapidly toward J37 51 northern levels despite the maintenance of some consumer goods J37 52 price subsidies. This problem, combined with administrative failure J37 53 to meet payrolls in Aden, created severe hardship and short-term J37 54 discontent in the south. Another cause of friction was a hasty J37 55 attempt to roll back the land reform program under which the PDRY J37 56 government had distributed agricultural land in the years following J37 57 independence from Great Britain. The dislocations caused by this J37 58 effort led to demonstrations in at least one southern province. At J37 59 the same time, the government was slow to deliver on promises to J37 60 restore real estate confiscated from businessmen who had fled north J37 61 from Aden following southern independence. This failure hurt J37 62 efforts to induce these businessmen to reinvest in the stagnant J37 63 southern economy.

J37 64 Despite these difficulties, however, the merger took place J37 65 without the major disruptions some had predicted, and the small J37 66 size of demonstrations mounted in Sanaa on 22 May against the union J37 67 was a fair measure of the weakness of anti-unity sentiment in the J37 68 country. Officials noted at that time that their strategy was to J37 69 capitalize on the enthusiasm for unity that existed in 1990 and not J37 70 allow the formidable practical problems of unification to cause J37 71 them to miss this major opportunity. In retrospect, and J37 72 notwithstanding the severe problems the Republic of Yemen has faced J37 73 in its first two years, their calculation appears sound.

J37 74 THE POLITICS OF UNIFICATION

J37 75 There were several reasons why the leaders in Sanaa and Aden J37 76 were able to overcome a legacy of mutual suspicion and fear created J37 77 by 22 years of politico-military combat and to unify Yemen J37 78 in less than six months. The first and foremost was the sea change J37 79 in the international political climate brought about by the Soviet J37 80 Union's acceptance of change in Eastern Europe in the late 1980s J37 81 and the implications for the Aden regime of Moscow's new attitude J37 82 toward its friends elsewhere in the world. The southern leadership J37 83 doubtless<&|>sic! felt compelled to alter radically its political J37 84 course and to strike the best deal possible with Sanaa as quickly J37 85 as possible.

J37 86 The other reasons for the success of the unification effort are J37 87 interrelated. In the 1980s, the consolidation of the authority of J37 88 the YAR government in tribal areas left Salih, whose personal stake J37 89 in a successful unity initiative was high, greater room for J37 90 maneuver in unity negotiations than he had previously enjoyed. J37 91 Salih's greater maneuverability vis-<*_>a-acute<*/>-vis the tribes J37 92 grew out of the evolution of the complex Yemeni-Saudi relationship, J37 93 which was an important element influencing the success of the J37 94 initiative. To understand the radical transformation in the J37 95 attitude of the two sides toward one another, the political and J37 96 economic positions of Aden and Sanaa in November 1989 must be J37 97 considered.

J37 98 Aden: Fall 1989, Nowhere to Turn

J37 99 By the fall of 1989, the regime in Aden had very limited J37 100 options. The bloodletting of January 1986 was a repudiation of Ali J37 101 Nasir Muhammad and his effort to seek an opening to the West, seen J37 102 by many as the key to the PDRY's economic salvation. Also gone from J37 103 the scene, however, were the most prominent of Ali Nasir Muhammad's J37 104 hard-line opponents. Although they were eulogized as 'martyred' J37 105 heroes, the country's new leaders showed little enthusiasm for J37 106 returning the country to a radical communist course.

J37 107 PDRY politics following the January events were essentially J37 108 stagnant. The regime was divided into factions opposed to one J37 109 another along regional as well as ideological lines, with hard-line J37 110 North Yemeni members of the former NDF pitted against moderates, J37 111 many of whom came from Hadhramaut Province. These splits sharply J37 112 limited the regime's ability to develop a clear policy direction. J37 113 The only success the new PDRY leadership could claim was the J37 114 rapprochement with the YAR that produced the joint investment zone, J37 115 the proliferation of other less significant agreements with Sanaa, J37 116 and the gradual diminution of the status of Ali Nasir Muhammad and J37 117 his exiles in the YAR.

J37 118 In social terms at least, the regime had some reason for J37 119 self-congratulation. Between 1967 and 1985, illiteracy was reduced J37 120 from 97 to 59 percent. In the YAR, the illiteracy rate stood at 80 J37 121 percent in 1985. In achieving this progress, the PDRY relied J37 122 largely on native, well-motivated teachers whereas the YAR depended J37 123 heavily on Egyptians. The PDRY could also point with pride to its J37 124 liberal family law that gave women rights unparalleled in most J37 125 Islamic countries.

J37 126 The economy, by contrast, was largely dysfunctional. Land and J37 127 fisheries reforms had produced declines in agricultural production, J37 128 and the port of Aden, once one of the busiest in the world, played J37 129 almost no role in international commerce. Industrial production had J37 130 declined, and, despite 20 years of socialism, more than 50 percent J37 131 of the gross national product still came from the private sector, J37 132 while hard currency remittances from workers abroad accounted for J37 133 half the government's annual budget.

J37 134 Nowhere was the bankruptcy of the PDRY's system more apparent J37 135 than in the petroleum sector. The discovery of oil in the YAR by J37 136 the US-based Hunt Oil company in 1984 had led to exports of 200,000 J37 137 barrels of oil a day five years later. A more or less J37 138 contemporaneous Soviet discovery just across the border in the PDRY J37 139 had put Aden $500 million in debt with nothing to show for it other J37 140 than 60 or so mostly non-functioning wells, a pipeline suspected of J37 141 leaking, an outmoded oil processing complex, and a few barrels of J37 142 crude oil trucked sporadically to Aden's decrepit refinery. Ali J37 143 Nasir Muhammad's opening to the West had brought several Western J37 144 companies into the PDRY, but none had made a discovery.

J37 145 Against this somber backdrop came the changes in Moscow's J37 146 policy toward Eastern Europe. The political message that Moscow was J37 147 unwilling to stand behind the regimes there was complemented by J37 148 widespread reports in the latter part of 1989 that the Soviet J37 149 military and economic aid that had been vital to the Aden J37 150 government's survival was to be drastically reduced. Exactly what J37 151 the Soviets said to the Aden leaders about aid levels, and when J37 152 they said it, is not well known. There is little doubt, however, J37 153 that such cuts were an imminent prospect.

J37 154 The PDRY leaders' strategy for improving their bleak situation J37 155 was to move cautiously in the direction of political and economic J37 156 reform. In the latter part of 1989, the government organized local J37 157 council elections in which candidates outside the ruling Yemen J37 158 Socialist Party were allowed to run; a high percentage of non-party J37 159 members were elected. The government also appeared to reach a tacit J37 160 understanding with newspaper editors permitting freer journalistic J37 161 expression. On the economic front, discussions began with the J37 162 Soviet Union in 1989 aimed at making available to Western companies J37 163 a substantial portion of the area in Shabwa Province where the J37 164 Soviets had been looking for oil. The government also adopted a law J37 165 intended to encourage domestic and foreign private investment and J37 166 to permit investors to take a percentage of their profits out of J37 167 the PDRY. Finally, wealthy Saudis of South Yemeni origin, as well J37 168 as southern businessmen who had moved to the YAR, were invited to J37 169 make investments in the PDRY economy.

J37 170 Aden's limited economic liberalization efforts, however, proved J37 171 a failure. Whatever negotiations there had been with the Soviets J37 172 about the release of oil acreage seemed to be going nowhere. Soviet J37 173 exploration in Shabwa largely ceased, and the date for the opening J37 174 of the pipeline from Shabwa to the coast was continually postponed. J37 175 Efforts to attract private investment languished as businessmen J37 176 waited until the fate of the regime became clear.

J37 177 The response of South Yemenis, like that of investors, made it J37 178 plain that the regime's political strategy was too little too late. J37 179 The spring of 1990 saw considerable ferment in the media - in J37 180 contrast to the silence of the still highly 'disciplined' fourth J37 181 estate in the YAR - as well as the first stirrings of political J37 182 demonstrations and strikes. South Yemenis also made clear their J37 183 strong pro-unity feelings. When President Salih went to Aden in J37 184 November 1989, he was greeted by crowds calling for unity whose J37 185 enthusiasm was obviously not staged. At the same time, some of the J37 186 political parties that had been active in both the YAR and the PDRY J37 187 in the early 1960s began to resurface in the PDRY. While these J37 188 parties were insignificant as political forces, their appearance J37 189 reflected the popular interest in genuine liberalization.

J37 190 Pressure for unification from South Yemenis continued in the J37 191 spring of 1990. By then, the leadership in Aden, short of other J37 192 alternatives, appeared to have reached the conclusion that J37 193 unification was its best option. In theory, it could have continued J37 194 to resist pressures for unification, opting instead for more J37 195 radical political change, for a renewed effort to open the oil J37 196 fields to Western companies, and for strategms designed to avoid J37 197 the need both to play second fiddle to the YAR leadership and to J37 198 share their oil wealth with Sanaa. Their decision against such J37 199 actions and instead to respond favorably to Salih's unity J37 200 initiative almost certainly stemmed from a calculation that popular J37 201 feeling in the PDRY both against them and in favor of unity was too J37 202 strong to permit them to stay in power.

J37 203 J38 1 <#FROWN:J38\>10.2 Connections between City Growth and J38 2 Economic Development

J38 3 Compared with Third World averages, India did not develop J38 4 rapidly between 1960 and 1981. Much of the difference can be J38 5 attributed to far lower total factor productivity growth rates J38 6 (chap. 7). Indeed, while poor productivity growth can account for J38 7 only a portion of India's slow city growth experience after 1964 J38 8 (chap. 6), it can account for a great deal of the poor economic J38 9 performance overall.

J38 10 Underlying India's dismal productivity growth performance are J38 11 structural and intrasectoral distortions. To some degree, these J38 12 could be reduced by appropriate rural policies (notably, providing J38 13 insurance incentives to encourage peasant farmers to switch to J38 14 higher yield but riskier crop mixes: see Singh, 1979) and J38 15 industry-oriented packages (in particular, by removing location and J38 16 input pricing distortions created by current policies: see Sekhar, J38 17 1983). In the presence of such distortions, traditional policies J38 18 that are oriented toward capital formation will have only a modest J38 19 impact, mainly because returns are low. This was the key flaw of J38 20 the Mahalanobis strategy (chap. 9). Policies that favored urban J38 21 manufacturing at the expense of rural production ultimately J38 22 resulted in increasing input and urban food costs, as well as in J38 23 sharply declining returns to investment.

J38 24 Many public sector policies needlessly reduce real incomes of J38 25 the poor as well. Excessive modern sector investment is one such J38 26 practice. As chapter 9 shows, the Mahalanobis strategy channels J38 27 investment to the most capital-intensive sectors, thereby J38 28 minimizing the impact of new investment on labor demand, and hence J38 29 on unskilled wages. The neglect of agriculture also results in food J38 30 price hikes, again hurting the poor. Migration restrictions and J38 31 incomes policies tend to diminish the earnings of most unskilled J38 32 workers still further.

J38 33 Furthermore, highly 'prorural' development strategies reduce J38 34 city growth rates only slightly, largely because urban-rural J38 35 linkages have become so strong in India. Manufacturing has always J38 36 been a major user of raw materials, so an augmented supply of those J38 37 raw materials encourages urban-based manufacturing growth. Because J38 38 city workers are heavy consumers of foodstuffs, any improvement in J38 39 food production that lowers food prices also lowers urban wages, J38 40 thus stimulating city growth. In addition, because Indian J38 41 agriculture has become increasingly dependent on manufactured J38 42 intermediates, agricultural output growth stimulates the demand for J38 43 manufacturing output, creating urban employment. As the Indian J38 44 economy is relatively closed to trade, all of these urban-rural J38 45 linkages tend to be stronger than in the SOE in which such linkages J38 46 spill over into changes in foreign trade volumes. India's poor J38 47 export performance also implies that the major outlet for increased J38 48 output must be the domestic market. Thus, if rural demands lag J38 49 behind, urban output will suffer. Furthermore, the growth of Indian J38 50 towns and cities appears to have spurred agricultural productivity J38 51 in nearby areas by generating increased demand and improving the J38 52 supply of critical intermediate inputs (Dasgupta and Basu, J38 53 1985).

J38 54 In summary, while Indian economic growth may well be greater in J38 55 the 1990s than it was in the 1960s and 1970s, strong forces limit J38 56 the extent to which it can be carried by highly unbalanced sectoral J38 57 productivity advance favoring urban sectors. Development is likely J38 58 to be most dramatic in the coming decade if urban and rural J38 59 productivity advance<&|>sic! are both enhanced simultaneously.

J38 60 10.3 Modeling India and Other Developing Countries

J38 61 Let us turn to a third set of findings coming from the previous J38 62 chapters and consider what we have learned about computable general J38 63 equilibrium models. Our general assessment in chapter 3 was the J38 64 BMW's simulation of a recent portion of Indian economic history J38 65 provided a surprisingly close fit to reality (even though our J38 66 vision of that reality was often blurred by imperfect data). In J38 67 addition, counterfactual simulation with BMW has offered detailed J38 68 insight into the determinants of city growth. Nevertheless, some of J38 69 our tales about Indian city growth could have been told without the J38 70 help of an elaborate CGE model. But this statement can only be made J38 71 ex post facto: the analyst cannot start by assuming which J38 72 urbanization and growth forces are most important, which J38 73 interactions matter most, and which assumptions are most J38 74 critical.

J38 75 Although BMW is a complex computer program, it is not a J38 76 black box. We were surprised at the number of times the model J38 77 yielded results inconsistent with our economic intuition. But J38 78 unanticipated results were never accepted (nor hidden) out of J38 79 deference to the program's inscrutable omniscience. Instead, J38 80 virtually all results were subjected to scrutiny that enabled us to J38 81 determine exactly which forces were, and which were not, driving J38 82 the results. In other words. BMW is not a black box; nor should J38 83 other CGE models be viewed as black boxes. The model is a tool, not J38 84 a Ouija board.

J38 85 Like all models, a CGE can only provide plausible answers to J38 86 the questions it was designed to address. And like all models, it J38 87 makes assumptions. The BMW model of India assumes that prices and J38 88 quantities are, for the most part, determined by the simultaneous J38 89 clearing of all markets. As such, it is designed to explain J38 90 medium-term and long-run phenomena. It was not designed to J38 91 assess short-run responses to shocks, for here disequilibrium J38 92 rather than equilibrium assumptions seem appropriate. Similarly, J38 93 BMW has been developed so that we can address questions relating to J38 94 spatial movements and sectoral growth. But it cannot be expected to J38 95 provide great insights into policies and shocks impinging on only a J38 96 very narrow part of the economy.

J38 97 One of the most valuable contributions of BMW (and something J38 98 that can in principle be obtained from other CGE models) is a clear J38 99 understanding of the model's errors. Just as one would examine J38 100 residuals in econometric work, we devote considerable attention to J38 101 those episodes and those parts of the economy in which the model J38 102 did not seem to fit Indian history well. By analyzing the errors, J38 103 we have been able to identify periods and sectors for which a J38 104 neoclassical model does poorly and to highlight the need for J38 105 further research. It turns out that the main errors all have fairly J38 106 obvious explanations.

J38 107 One also may ask what can be learned from BMW that could not be J38 108 learned from a different framework. Relative to other CGEs, the BMW J38 109 India model richly details government behavior and focuses on J38 110 infrastructure, capital market fragmentation, and the rural-urban J38 111 spatial division (as do Kelley and Williamson, 1984). But CGE J38 112 models do not offer the only means by which to analyze J38 113 urbanization. It would have been possible to develop a small J38 114 econometric model of Indian city growth. Such models are obviously J38 115 useful. Furthermore, they avoid 'one-point estimation' of many J38 116 parameters, which is inherent in most CGE models. But such models J38 117 are highly aggregative; they typically contain few explanatory J38 118 variables and cannot be used to analyze urbanization with the J38 119 richness offered here. Rather than designating one method as J38 120 inferior to the other, it makes more sense to regard them as J38 121 complementary, each with different strengths and weaknesses.

J38 122 Demographic forecasting offers yet another way to analyze J38 123 urbanization (see, for example, Rogers, 1984). These models J38 124 typically have extensive detail with respect to demographic J38 125 variables; in particular, they usually contain many regions, J38 126 population groups, migration probabilities, fertility rates, and J38 127 age- and sex-specific death rates. What they do not have are J38 128 behavioral specifications or endogenous economic variables. But it J38 129 is imporper to assert smugly the superiority of economic models: J38 130 ours, after all, aggregates highly over demographic variables and J38 131 treats all of them, but not migration, as exogenous.

J38 132 10.4 Can We Generalize from India?

J38 133 BMW has been designed to capture critical features of the J38 134 Indian economy. Our assessment is that it does so rather well, J38 135 enabling us to draw a large set of conclusions concerning the J38 136 country's urbanization and growth process. But to what extent can J38 137 these conclusions be generalized to other countries? The answer J38 138 hinges on the similarity between India's demo-economic structure J38 139 and that of other developing countries. Naturally, authors like J38 140 their findings to be as general as possible. Alas, there are some J38 141 major obstacles in the way.

J38 142 A leading obstacle is the greater openness of most developing J38 143 countries relative to India. In India's semiclosed economy, a J38 144 sectoral demand boom can fizzle if the costs of traded goods or J38 145 nontraded inputs are bid up rapidly. In an SOE, traded goods' J38 146 supply curves are flatter, since traded goods can be imported at a J38 147 fixed price. Thus, a booming sector is not held back so much by the J38 148 presence of a stagnant, low-productivity sector in an SOE; indeed, J38 149 the more unbalanced the productivity advance, the more unbalanced J38 150 the output growth. But India, with its small, moribund export J38 151 sector and pervasive import controls, cannot accommodate truly J38 152 dramatic unbalanced growth quite so easily. In consequence, J38 153 unbalanced urban-rural sectoral productivity advance, a driving J38 154 force behind rapid city growth, will have a much more pronounced J38 155 impact in an SOE than in India.

J38 156 Two other key features of India's economic structure are its J38 157 relatively large and developed rural nonagricultural sector and J38 158 urban manufacturing's dependence on domestically produced raw J38 159 materials rather than on imported intermediates. The latter feature J38 160 ensures that rural and urban areas are far more closely integrated J38 161 than, say, in many African countries. This linkage makes 'runaway J38 162 city growth' difficult if not impossible and ensures reasonably J38 163 balanced growth, but it also tends to put the brakes on the overall J38 164 rate of economic progress. The large rural service sector J38 165 essentially competes with the urban sectors for released peasant J38 166 labor, but it can also release labor to agriculture and to the J38 167 cities, thus reducing competition between them. If Indian J38 168 agriculture continues to modernize rapidly -and one should not be J38 169 excessively optimistic here -then the presence of a large rural J38 170 service sector will provide something of a surplus labor pool that J38 171 will continue to permit rapid growth of output (but, less happily, J38 172 continue to depress unskilled wages) both in farming and in the J38 173 cities.

J38 174 In short, India has many unique features. Thus, the question J38 175 remains: Are there any generalizable findings? The most important J38 176 ones probably center on the role of government. The inegalitarian J38 177 nature of public sector employment (bidding up skill differentials) J38 178 and demands (consuming urban goods) is unlikely to be unique to J38 179 India. So is the need to follow proagricultural policies if large J38 180 welfare gains for the poor are a target. As a whole, the J38 181 inefficient government policies outlined above are unlikely to be J38 182 dramatically different in most other Third World economies. This J38 183 comment applies most strongly to urban incomes policies and J38 184 migration restrictions. Furthermore, while IMF-style policies are J38 185 likely to produce much more growth in SOEs, they are unlikely to be J38 186 more egalitarian than in India. The distinction between policies J38 187 that generate large real GDP gains and those that increase living J38 188 standards of the poor is also generally valid.

J38 189 The forces driving India's city growth slowdown are to some J38 190 extent also general, although they operate most strongly in a J38 191 semiclosed economy. The relative unimportance of rural push factors J38 192 is probably wide-spread; it was certainly true of J38 193 nineteenth-century industrial revolutions as well (Williamson, J38 194 1990). Similarly, the importance of unbalanced rates of J38 195 productivity advance, rather than their average level, in driving J38 196 urbanization is likely to be even stronger in the SOEs, as Kelley J38 197 and Williamson (1984) have shown. The importance of urban J38 198 nonmanufacturing sectors and rural nonagricultural sectors in J38 199 understanding growth and urbanization in India is undoubtedly J38 200 matched elsewhere as well.

J38 201 One of the striking findings from our counterfactual J38 202 simulations was that the misallocation of investment has never been J38 203 terribly important in India. In a sense, economic inefficiencies J38 204 are more likely to matter at the firm level: returns to investment J38 205 are, after all, fairly low everywhere. Since SOEs have more rigidly J38 206 determined spheres of comparative advantage, sectoral investment J38 207 choices are probably more important than in India.

J38 208 In extending the model to other countries -or to India of the J38 209 1990s -several modifications need to be considered. In particular, J38 210 the choice of sectors and the degree of openness must vary from J38 211 country to country. So, too, must the assumed degree of factor J38 212 market integration and assumptions concerning domestic restrictions J38 213 on the tradeability of goods across space. Public sector behavior J38 214 is also largely country specific. Finally, countries such as India J38 215 that have experienced significant modernization in portions of J38 216 their agricultural economies probably should have more than one J38 217 agricultural sector in a CGE model. J38 218 J39 1 <#FROWN:J39\>Building Trades

J39 2 The building and construction trades have made the most direct J39 3 gains in using pension funds offensively. Funds from building J39 4 trades union pensions formed three labor banks in Colorado, New J39 5 Mexico, and Arizona. The building trades have also channeled funds J39 6 into union-built construction. Over eleven projects across the J39 7 country with varying arrangements and goals have pooled money from J39 8 regional pension funds. One notable example is the Bricklayers and J39 9 Laborers Non-Profit Housing Company, Inc. The Bricklayer Union and J39 10 Laborer Union pension fund buys federally insured certificates of J39 11 deposit (with small subsidies from participating unions) from the J39 12 U.S. Trust Bank in Boston, which, in turn, finances construction J39 13 loans for union-built, low-price housing for South Boston J39 14 residents.

J39 15 Other building trade investment projects operate like the Union J39 16 Labor Life Insurance Company (ULLICO) 'J for Jobs' project, which J39 17 invests union pension funds in an open-ended trust that invests in J39 18 union-built construction. The Housing Investment Trust (HIT) has, J39 19 since 1964, financed union-only housing construction projects with J39 20 union pension money. In 1987, the AFL-CIO has expanded the HIT J39 21 scope and established the Building Investment Trust (BIT), which J39 22 will finance commercial and industrial real estate investments. In J39 23 May 1988, the Bakery and Confectionery Pension Plan and the J39 24 Bricklayers National Pension Fund committed $15 million to the BIT; J39 25 its first project was a $5.8 million hotel complex in Taos, New J39 26 Mexico. BIT's advantages to pension funds are its relatively low J39 27 administrative costs and the assurance that union labor will be J39 28 used in the construction and maintenance of the project.

J39 29 The Sheet Metal Workers are using their national pension funds J39 30 in a fashion that could be interpreted as a type of industrial J39 31 policy. They seek to increase the demand for their members by J39 32 closing gaps in financial markets and in the surety industry. In J39 33 order to help contractors in the asbestos removal industry obtain J39 34 bonds with lower premiums, the union bought a bonding company. The J39 35 union pension fund already owns a stake in an asbestos removal J39 36 contractor (U.S. Bureau of National Affairs 1988c). Edward J39 37 Carlough, President of the Sheet Metal Workers, calls pension funds J39 38 the labor movement's 'Star Wars Weapon' (U.S. Bureau of National J39 39 Affairs 1988c). Approximately $500 million has been directed into J39 40 real estate projects by building trades investment financing J39 41 foundations (Westerbeck 1985).

J39 42 The building trades have come closest to systematically J39 43 challenging the appropriateness of using only the market rate of J39 44 return in assessing the quality of an investment. Instead, they use J39 45 a Keynesian multiplier argument and argue that pension funds, J39 46 unlike other trust funds where all the income comes from financial J39 47 investments, have two sources of income - employer and employee J39 48 contributions and investment earnings - and the rates of return J39 49 from both sources, not just from the more visible financial J39 50 markets, must be considered. The multiplier argument is primary J39 51 rhetorical and has not been quantified.

J39 52 Most of the DOL challenges to building trade investment J39 53 projects have failed because ERISA's prudent expert rule allows J39 54 each investment vehicle to be evaluated within the context of the J39 55 portfolio and the interests of the participants (Leibeg 1980). The J39 56 unsuccessful 1981 DOL challenge of the Florida Operating Engineers J39 57 low-interest mortgage program (for plan participants) showed that J39 58 legal pension-fund investments must have a reasonable rate of J39 59 return and must serve the needs of the participants (U.S. Bureau of J39 60 National Affairs 1986a, 1986c). The Reagan administration had sent J39 61 mixed signals about these union initiatives. The DOL continues to J39 62 issue warnings about union social investment projects, but then J39 63 President Reagan, speaking to a building trades meeting in 1982, J39 64 applauded the unions' efforts in setting an example for local J39 65 initiatives (Smith 1984). In fact, the building trades' activities J39 66 have not threatened the traditional control of capital.

J39 67 Each project has different effects in terms of reallocating J39 68 capital. Some projects clearly correct a market failure, or J39 69 barrier, by providing funds to projects that would have otherwise J39 70 faced a credit problem. The Boston Bricklayer and Laborer project J39 71 is an example of reallocating capital. Other projects can be viewed J39 72 as substitutes for organizing new members. The enemy in the J39 73 building trades is not always perceived as the nonunion worker or J39 74 the aggressive employer; and in periods of high unemployment the J39 75 nonunion contractor is not blamed as much as insufficient demand. J39 76 The building trades have sidestepped organization in favor of using J39 77 pension-fund monies as a primary way to increase the market J39 78 share of unionized construction (U.S. Bureau of National Affairs J39 79 1985; Freeman 1985).

J39 80 State and Local Governments

J39 81 State and local social investing, practiced by public funds in J39 82 about twelve states, is confined to mortgages or loans to small J39 83 businesses. The main purpose is to increase employment in the J39 84 relevant region by filling a capital gap and not by providing loans J39 85 below 'market rates.' The state of Michigan's project to coordinate J39 86 investments of private and public funds to provide small business J39 87 loans and the Pennsylvania MILRITE project are just two examples of J39 88 the many affirmative investment strategies of governments. These J39 89 trends can be tracked through the BNA's Pension Reporter, J39 90 Labor and Investments, and other pension-industry J39 91 publications.

J39 92 From a distance, these programs look like models of socially J39 93 responsible investment because the government, endowed with the J39 94 duty to maximize social goals, invests. Only some projects J39 95 reallocate capital to worthy projects that otherwise would have J39 96 faced a credit shortage. Other projects involve granting some J39 97 privilege to the private sector and usually do not involve the J39 98 public sector labor unions. Moreover, I have not found cases where J39 99 the state and local governments have opted for owner control; they J39 100 have not operated the projects they finance through pension funds. J39 101 Another, more political, state and local use of pension plans is J39 102 represented by the efforts of the late California State Treasurer, J39 103 Jesse Unruh, who, at the end of 1984, organized the Council of J39 104 Institutional Investors (CII), which now has representatives from J39 105 twenty-two state and local pension funds, six multiemployer funds, J39 106 and one single-employer plan administrator (Pensions and J39 107 Investment Age 1989). The CII seeks to monitor and influence J39 108 corporate policies that affect its state and local pension-fund J39 109 investments, such as green mail, golden parachutes, and poison J39 110 pills, because these practices threaten dividend income. CII was J39 111 very vocal in its criticism of GM management's buyout of its J39 112 largest individual shareholder, H. Ross Perot. Despite GM's J39 113 assertion that a $720 million price tag on $350 million worth of J39 114 shares represented a good investment, CII, as well as most of the J39 115 financial press, cited H. Ross Perot's increasingly shrill J39 116 criticism of GM management, for example, his public questions, such J39 117 as, "Why does it take longer to design a GM car than it did J39 118 to fight World War II?" as a leading motive for the buyout J39 119 (Kraw 1989). CII forced a meeting with senior members of GM J39 120 management. A few years later, in January 1990, two CII members, J39 121 the New York and California state funds, in reportedly J39 122 uncoordinated efforts wrote to GM asking for a say in the J39 123 replacement of Roger Smith as GM chief. Most of the CII's tools are J39 124 public embarrassment, shareholder resolutions, proxy votes, and J39 125 capital boycotts. CII actions often seek to protect shareholders J39 126 from entrenched management resisting a takeover. In many cases, J39 127 labor would be directly opposed to the interests of the J39 128 shareholders and in agreement with managements' efforts to save the J39 129 plant from a hostile takeover and possible plant closing.

J39 130 Under criticism from Edward V. Regan, New York State J39 131 comptroller and sole trustee of the $44 billion New York State and J39 132 Local Retirement Fund, Governor Marie Cuomo's Commission on 'The J39 133 New York State Pension Investment Task Force' recommended that J39 134 pension funds obtain guarantees from federal agencies to invest in J39 135 projects to improve the state's infrastructure. Regan argued J39 136 against the Commission's report and financier Felix Rohatyn's J39 137 similar suggestions that the federal government was overburdened by J39 138 guarantees and the first and foremost duty of the fund was to earn J39 139 the highest return.

J39 140 This sort of reasoning was further displayed in 1990 when J39 141 several state and local pension funds threatened to divest J39 142 themselves of the stock of all Pennsylvania-based companies to J39 143 protest Pennsylvania's restrictions on hostile takeovers. The J39 144 pension funds' position is that hostile takeovers bid up the price J39 145 of the stock and any restriction on hostile bids constrains the J39 146 potential profit from holding that equity. Lawmakers in J39 147 Pennsylvania want to halt hostile takeovers in order to stabilize J39 148 communities and the business environment. What the Pennsylvania J39 149 pension plans do will reveal how sharp the horns of the dilemma J39 150 actually are when it comes to defining socially responsible J39 151 pension-fund strategies.

J39 152 Churches, Wealthy Individuals, and Endowments

J39 153 The Interfaith Center for Corporate Responsibility and the J39 154 Investor Responsibility Research Council are research and advocacy J39 155 institutes with strong church connections. Their publications J39 156 provide advice on share-holder rights and strategies. Most J39 157 of the alternative investments of university endowments have been J39 158 in South African divestment and clean fund investments. In addition J39 159 to specific actions and research functions, the financial industry J39 160 has responded to the concern of churches, unions, endowments, and J39 161 wealthy individuals who fear their investments are encouraging J39 162 antisocial activities by the offering of 'clean funds' (Domini and J39 163 Kinder 1984). Currently, in the U.S., eight mutual funds and three J39 164 money market funds advertise holdings that are screened according J39 165 to various criteria. Two Canadian funds, the Summa Fund and the J39 166 Ethical Growth Fund, and one British fund, The Stewardship Unit J39 167 Trust, use social screening criteria in selecting investments. Each J39 168 fund seems specifically tailored to one or more groups. For J39 169 instance, the PAX World fund does not contain 'sin' stocks, such as J39 170 holdings in tobacco or liquor companies or defense contractors. The J39 171 performance of these socially screened funds beats the market J39 172 average. As a 1986 Wall Street Journal headline noted, J39 173 "Investors Can Do All Right By Doing Good". So do J39 174 the brokerage houses and money managers who have developed this J39 175 market niche.

J39 176 From the point of view of labor and churches, however, the J39 177 structure of capital markets limits the effectiveness of the clean J39 178 fund strategy. Multinationals and conglomerates issue debt and J39 179 equity. Targeting money into a particular region, or to a J39 180 particular subsidiary of a conglomerate, is not possible, because J39 181 capital allocation decisions are made internally. Managers make J39 182 investment decisions. Unions face the same problems a shareholders J39 183 if they attempt to transform companies through the ownership of J39 184 corporate equity. The advantage the labor movement has over most J39 185 shareholders - its national presence - can either create J39 186 enormous clout, or unresolvable conflicts of interest among unions. J39 187 (How successful would the Shell boycott have been if the Oil, J39 188 Chemical and Atomic Workers Union represented Shell employees?)

J39 189 But, trustees of funds of many corporations and entities with J39 190 trust funds are asking the same question Harbrecht (1959) and J39 191 Barber and Rifkin (1978) exhorted labor unions themselves to ask. J39 192 In the words of a pension attorney, "What is it that says J39 193 trustees of the American Cancer Society can't say (its pension J39 194 fund) won't invest in tobacco stocks" (Gerald Feder, quoted J39 195 in Pensions and Investment Ages 1989).

J39 196 Corporations

J39 197 Corporations have discovered in the 1980s that their pension J39 198 finds can be convenient sources of cash and can serve other J39 199 corporate needs. A combination of legal, political, and economical J39 200 factors explains why single-employer pension funds are now J39 201 used as a corporate financing tool.

J39 202 The interpretation, and Reagan administration's enforcement of J39 203 ERISA, all but encouraged corporations to put their funds to J39 204 innovative and, indeed, alternative uses. To fend off hostile J39 205 takeovers, or to court 'white knights', a corporation can often J39 206 obtain DOL permission to direct its employee pension fund to buy J39 207 the company's stock (as long as it is not 'overpriced'). A J39 208 corporation can terminate a plan that happens to have more assets J39 209 than legal liabilities, pay the liabilities, and use the extra J39 210 cash. The corporation can also manipulate the rate of return the J39 211 fund is assumed to earn in order to alter the cash contributions J39 212 required to fund the liability. Raising the assumed rate assumes J39 213 the fund will earn more and, therefore, require less from the J39 214 firm.

J39 215 Between 1980 and 1987 about $9 billion had been recaptured by J39 216 corporations in defined-benefit pension plan terminations J39 217 (VanDerhei and Harrington 1989: 189). J39 218 J40 1 <#FROWN:J40\>Straight talk from Wall Street

J40 2 A highly regarded stock analyst points out what's right - and J40 3 wrong - with bank investor relations

J40 4 By Thomas K. Brown

J40 5 What makes an effective investor relations program? The answer J40 6 depends to an extent on your position. I will provide the J40 7 perspective of a Wall Street bank stock analyst.

J40 8 We'll start with the person in charge of investor relations. J40 9 This person must have a thorough understanding of the company and J40 10 its financial statements, as well as access to senior J40 11 management.

J40 12 He or she must be able to perform the delicate task of J40 13 providing information to analysts and investors without crossing J40 14 the line of differential disclosure.

J40 15 This person cannot be purely a cheerleader, nor a sugar-coater. J40 16 The goal is to present a fair picture of current conditions.

J40 17 There is no one spot in the organization from which the head of J40 18 investor relations should come.

J40 19 For example, at Wachovia Corp. and Mellon Bank Corp. the heads J40 20 of investor relations are individuals located in the finance part J40 21 of the organization. Both are excellent. Bank of America Corp. has J40 22 one individual whose sole responsibility is investor relations and J40 23 he is the best in the business. Fleet/Norstar Financial Group J40 24 operates with a team involving the chief financial officer and two J40 25 individuals with responsibilities in addition to investor J40 26 relations. Yet the result is quite effective.

J40 27 The form may vary, but in all these cases the end result is an J40 28 effective program.

J40 29 Unfortunately, in too many companies, including some of the J40 30 largest banking firms, the head of investor relations is either not J40 31 knowledgeable about the company and its financials or lacks timely J40 32 access to senior management. These companies are known for their J40 33 mishandling of important information, for differential disclosure, J40 34 and for the dreaded 'surprise' announcement.

J40 35 Responsive voice. As important as the quality of J40 36 information disseminated is its timeliness. Institutional analysts J40 37 and investors are under ever-increasing pressure to gain access to J40 38 important information and draw investment conclusions.

J40 39 At First Union Corp. and Wells Fargo & Co., investor relations J40 40 staffs are briefed about the company's quarterly earnings just J40 41 before they are released. In this way, several individuals at the J40 42 company are then able to answer questions from analysts and J40 43 investors once the earnings release is made public.

J40 44 At other companies there is only one individual available to J40 45 answer questions. This inevitably leads to long delays in receiving J40 46 answers and effectively puts some analysts at a significant J40 47 competitive disadvantage.

J40 48 However, because there are more investors and analysts than J40 49 staffers, at times priorities will have to be made as to who is J40 50 called first.

J40 51 Don't just return calls in the order they are received. In my J40 52 opinion, the top priorities should be large shareholders and those J40 53 Wall Street analysts who have been particularly visible (both J40 54 positive and negative) in providing research on the company. This J40 55 may seem unfair, but it will lead to quickest dissemination of J40 56 information.

J40 57 Good, timely disclosure. Analysts are never satisfied J40 58 with the level of disclosure. So it's only natural that one of the J40 59 traits that I find in the best investor relations programs is J40 60 excellent disclosure of timely information.

J40 61 Recent examples of breakthroughs in the disclosure of important J40 62 new information include:

J40 63 <*_>bullet<*/>A reconciliation of flows in nonperforming assets J40 64 provided by Valley National Corp., Bank of Boston Corp., and J40 65 Shawmut National Corp.

J40 66 <*_>bullet<*/>First Tennessee National Corp. provides a J40 67 breakdown of its loan portfolio by risk rating.

J40 68 <*_>bullet<*/>Wells Fargo provides excellent disclosure of the J40 69 cash payments received on nonperforming loans.

J40 70 The goal of increased disclosure should be to put meat on the J40 71 bones of the company's required financial statements. The enhanced J40 72 disclosure of asset quality trends previously described, as well as J40 73 line-of-business results, are quite helpful in this regard. A J40 74 better understanding of what is occurring at the company can only J40 75 help to reduce investor uncertainty. Over time, this will lead to a J40 76 higher relative stock price valuation.

J40 77 One of the principal sources of disclosure is the annual J40 78 report. I believe this is the single most important communication a J40 79 company makes to its share-holders and potential investors. J40 80 It should provide the reader with a thorough understanding of the J40 81 company's products, markets, strategy, and recent financial J40 82 performance.

J40 83 Frankly, I am amazed that the investor relations departments at J40 84 numerous companies do not provide any input into the company's J40 85 annual report - a mistake.

J40 86 There are numerous examples of companies that publish J40 87 outstanding annual reports. Among them are Fleet/Norstar, First J40 88 Union, Barnett Banks Inc., and Norwest Corp.

J40 89 Unfortunately, some of the highest-quality companies J40 90 publish some of the worst annual reports. As long as these J40 91 companies continue to deliver superior results, this won't be a J40 92 major problem. But if they stumble, the lack of disclosure will J40 93 lead to investor uncertainty and a sharply lower relative J40 94 valuation.

J40 95 No differential disclosure. It is difficult to be J40 96 responsive and provide timely disclosure of information without J40 97 crossing the line and providing differential disclosure to one J40 98 analysts or investor. The issue can only be managed. It cannot be J40 99 avoided completely.

J40 100 However, there are some obvious steps that can be taken to J40 101 reduce differential disclosure.

J40 102 For example, don't allow analysts and investors to visit the J40 103 company after the quarter is over and before earnings are reported. J40 104 Try as they might, senior management seldom is able to conduct J40 105 these meetings without providing an important insight into the J40 106 quarterly results.

J40 107 In addition, when analysts and investors call after the quarter J40 108 is over and before the results are announced, the investor J40 109 relations department should discuss only the results from the first J40 110 two months of the quarter and make this clear to the caller.

J40 111 Finally, senior management must spend more time telling J40 112 employees what constitutes inside information. I am surprised at J40 113 the number of company executives who know they must not disclose J40 114 certain information to analysts and professional investors. Yet J40 115 they will freely discuss such information with friends in social J40 116 settings.

J40 117 Leaks of information destroy an investor relations program and J40 118 raise doubts about management's credibility - and internal J40 119 controls.

J40 120 Educate the Street. Effective investor relations J40 121 isn't simply being responsive to Wall Street's ever-changing J40 122 desires. Sometimes investor relations programs need to take a stand J40 123 on issues.

J40 124 My personal peeve is the widespread use of the J40 125 reserve-to-nonperforming loan ratio as the sole measure of reserve J40 126 adequacy. Any good banker knows that this ratio is a poor indicator J40 127 of reserve adequacy. Yet management let Wall Street adopt this J40 128 measure over the past few years as its principal tool in reserve J40 129 adequacy.

J40 130 Educate the brass. An effective investor relations J40 131 program must educate the company's senior management, and J40 132 particularly its chief executive officer, in the workings of Wall J40 133 Street, for example:

J40 134 <*_>bullet<*/>The difference between good companies and J40 135 good stocks. A good or even great company does not necessarily J40 136 make a good investment over the short or intermediate term. And J40 137 those are the time frames of most investors, whether they say so or J40 138 not.

J40 139 As of mid-November in 1991, J.P. Morgan had had an outstanding J40 140 year and it is widely regarded as an excellent company. Yet it has J40 141 been one of the market's worst-performing bank stocks. Analysts can J40 142 recognize J.P. Morgan as an outstanding company and yet not J40 143 recommend its stock - and be helping their clients in the J40 144 process.

J40 145 <*_>bullet<*/>Analysts' varying perspectives. Most J40 146 Wall Street analysts are not good stock pickers.

J40 147 Many are really investment bankers hiding out in the research J40 148 department; others are just good at understanding industry trends. J40 149 A good investor relations program will explain to senior management J40 150 the strengths and weaknesses of the analysts who are making public J40 151 comments or writing reports.

J40 152 For example, when the analysts at the company's principal J40 153 investment banking firm write a glowing report about the company, J40 154 it's up to the investor relations department to remind senior J40 155 management of the analysts' bias. Or, when an analyst who doesn't J40 156 know the company well publishes a report that is somewhat off base, J40 157 the investor relations department should explain this to senior J40 158 management and then contact the analyst.

J40 159 <*_>bullet<*/>Don't let the CEO blow up over negative J40 160 comments. When a negative report is written or negative J40 161 comments are made, it is up to the investor relations department to J40 162 encourage the CEO not to act irrationally by cutting the analyst J40 163 off from the information flow or by preventing the banking company J40 164 from doing any business with the analyst's firm.

J40 165 Such steps are only understandable when careless analysis has J40 166 damaged the company or if the analyst is deliberately spreading J40 167 misinformation.

J40 168 An effective investor relations program can significantly J40 169 influence how investors view the company, which will impact how the J40 170 company's earnings stream is capitalized over the long J40 171 run. Too often, managements judge the effectiveness of their J40 172 investor relations programs by short-term movements in their stock J40 173 price - a critical mistake.<*_>square<*/>

J40 174 J40 175 ADA compliance is uncharted territory

J40 176 With just a short time remaining before some provisions of the J40 177 Americans with Disabilities Act (ADA) take effect, community J40 178 bankers are generally less concerned about compliance with the law J40 179 than their counterparts in large metropolitan-area banks.

J40 180 Nevertheless, they have some real concerns, not the least of J40 181 which is how to finance structural and other changes that may be J40 182 required. "From an employment standpoint, we have a second J40 183 floor without elevator access, which pretty well leaves the J40 184 handicapped out," notes Gerald Hansen, vice-president and J40 185 controller of Itasca (Ill.) Bank &Trust Co., with $172 million in J40 186 assets. "We can make arrangements that will alleviate the J40 187 problem in terms of interviews, but if someone were qualified to J40 188 work in that area, we would have to consider the cost of installing J40 189 an elevator in the building." The law states that J40 190 "readily achievable" measures to remove physical J40 191 and communications barriers must be taken, but such language has J40 192 yet to be tested in the courts, and the expense of installing an J40 193 elevator probably would not be found readily achievable.

J40 194 But the problem is not an isolated one. Michael Derr, assistant J40 195 vice-president, operations, at the $144 million-assets Bank J40 196 of Glen Burnie (Md.), faces a similar situation at his bank's main J40 197 office. That facility was built in the early 1950s, notes Derr, J40 198 with very limited access to the second floor.

J40 199 The bank has four remote branches as well, "and the J40 200 bank is surveying those facilities and identifying areas that could J40 201 be considered barriers," says Derr. Like most banks, J40 202 accessibility to automated teller machines ranks high on the list J40 203 of potential trouble spots, but access to night depositories and J40 204 rest rooms can't be overlooked either, notes Derr. "We'll J40 205 have to make a judgement as to what is practical," he J40 206 says.

J40 207 Derr and dozens of other bankers attended an ABA-sponsored J40 208 symposium on ADA in late October. Speakers, including those J40 209 representing various disabled groups, encouraged banks of all sizes J40 210 to have a committee in place whose task is to formulate a strategy J40 211 for complying with the law. Such a measure will help persuade a J40 212 court of law that a bank is attempting to meet the requirements of J40 213 ADA, if and when litigation occurs.

J40 214 Disabled neighbors. "We're not anticipating J40 215 litigation," says Paul Sciacchitano, executive J40 216 vice-president and cashier of Old Point National Bank, Hampton, J40 217 Va., which has $263 million in assets. "A lot of disabled J40 218 people deal with us already, since we're one street away from a J40 219 Veteran's Administration center," he adds.

J40 220 A year ago, Sciacchitano's bank surveyed its locations and J40 221 identified areas where access for the disabled could be improved. J40 222 Since then, Sciacchitano has become more sensitive to the issue of J40 223 accessibility for the disabled. He's observed that a local J40 224 fast-food restaurant, for example, put a wheelchair ramp in a J40 225 virtually inaccessible location leading to a doorway that is only J40 226 about 20 inches wide. "And the ramp was placed in the J40 227 middle of a parking space with no clear access to it," says J40 228 <}_><-|>Sciaccitano<+|>Sciacchitano<}/>. "It was done as an J40 229 afterthought and showed no real consideration."

J40 230 Most disconcerting to Sciacchitano and other bankers is the J40 231 fact that the only real enforcement of the law regarding facility J40 232 accessibility lies with the Department of Justice. The Equal J40 233 Employment Opportunity Commission (EEOC) will oversee compliance of J40 234 the law's Title I provisions, which deal with human resources J40 235 issues.

J40 236 "Compliance is more subjective than objective," J40 237 says Sciacchitano, "but the monetary penalties involved are J40 238 rather objective. J40 239 J41 1 <#FROWN:J41\>Motor Carrier Deregulation and Highway Safety: J41 2 An Empirical Analysis

J41 3 DONALD L. ALEXANDER

J41 4 I. Introduction

J41 5 The passage of the Motor Carrier Reform and Modernization Act J41 6 of 1980 (MCA 1980) effectively marked the end of rate and entry J41 7 regulation in the interstate trucking industry, and most economists J41 8 agree that this legislation has created many important economic J41 9 benefits during the 1980s. For example, shipping rates fell in real J41 10 terms, service expanded and improved in quality, and generally J41 11 resources were used more efficiently. There is increasing concern, J41 12 however, that deregulation has led firms to reduce safety J41 13 expenditures in order to remain competitive, and that trucking J41 14 accidents and fatalities have increased as a consequence [4; 9]. J41 15 Indeed, Brock Adams (former U.S. Secretary of Transportation) J41 16 claims that trucking accidents have increased since 1980 for this J41 17 very reason.

J41 18 The limited empirical evidence reported in the literature J41 19 suggests otherwise. Moore [17], for example, shows that accident, J41 20 fatality, and injury rates have fallen since 1980 despite the rapid J41 21 increase in the number of truck-miles traveled [13]. His analysis, J41 22 however, is based on a comparison of rates across time, and does J41 23 not reveal any of the potential economic or institutional forces J41 24 that may be affecting the evolution of these data. In a more J41 25 systematic analysis, Traynor [22] finds that deregulation has J41 26 reduced accident rates in California, although it may be difficult J41 27 to extend his results to the national experience as a whole.

J41 28 This paper has two major objectives. The first is to estimate J41 29 an empirical model using a pooled, cross section of state data to J41 30 determine whether the evidence Traynor [22] reports for California J41 31 holds, in general, for all other states. The second objective is to J41 32 determine empirically those factors that may be driving the results J41 33 in Moore [17]. Pooling these data allow<&|>sic! us to test for the J41 34 impact of deregulation on accident, fatality, and injury rates J41 35 while holding state-specific factors constant; an empirical J41 36 approach that is not possible at a higher level of aggregation.

J41 37 The paper is organized in the following manner. In the second J41 38 section I discuss three possible ways deregulation may have J41 39 affected accident rates in interstate trucking. The third section J41 40 presents the empirical model and regression results, and the final J41 41 section summarizes the major conclusions drawn from this J41 42 investigation.

J41 43 II. Motor Carrier Deregulation and Highway Safety

J41 44 The current theoretical literature provides a framework for J41 45 discussing the potential link between changes in deregulation and J41 46 highway safety. I will use this framework to focus on three aspects J41 47 of deregulation that are likely to affect highway safety in the J41 48 trucking industry.

J41 49 Before deregulation, rate bureaus acted as cartels and set J41 50 shipping rates at supracompetitive levels. Since the Interstate J41 51 Commerce Commission (ICC) restricted the entry of new trucking J41 52 firms, incumbent firms were able to earn economic rents that were J41 53 shared with organized labor. After deregulation, however, the ICC J41 54 essentially permitted free entry and naturally there was an influx J41 55 of new firms. Winston et al. [31], for example, report that the J41 56 number of truckers with ICC operating authority increased from J41 57 18,045 in 1980 to 36,948 in 1986. One implication for highway J41 58 safety is that trucking accidents may have increased because of the J41 59 additional traffic congestion. Moreover, if the new entrants were J41 60 inexperienced in handling trucks on congested highways, it is J41 61 likely accidents would have increased until the new drivers gained J41 62 the necessary driving experience.

J41 63 The implicit assumption underlying this argument is that a J41 64 driver's behavior towards safety remains unchanged in response to a J41 65 change in the regulatory environment, which Peltzman [18] has J41 66 questioned on theoretical grounds. Although deregulation in the J41 67 trucking industry did not involve a specific change in safety J41 68 regulation, Peltzman's model provides a useful framework to examine J41 69 the effect of deregulation on the behavior of owner-operators.

J41 70 The basic model is that drivers face a choice between driving J41 71 intensity (e.g., faster speeds, longer hours) and the probability J41 72 of an accident, which is affected by several economic factors: the J41 73 price of an accident, income, and a regulatory parameter which we J41 74 will associate with the driver's promotion of safety. In a J41 75 cross-section model, it is difficult to argue that interstate J41 76 drivers respond to differential 'accident prices' across each J41 77 state. Moreover, it is equally difficult to argue that secular J41 78 changes in income that Peltzman discusses are at work to the same J41 79 extent in this analysis since we are using data for a shorter time J41 80 span. Therefore, I will ignore the first and second factors and J41 81 focus on the third.

J41 82 We can think of deregulation shifting the driver's demand for J41 83 driving intensity in several ways. The widely-held view is that J41 84 deregulation has led owner-operators to reduce safety expenditures J41 85 to remain competitive, which raises the probability of an accident J41 86 for a given level of driving intensity. It is not quite clear, J41 87 however, why owners would reduce safety expenditures when their J41 88 discounted future profits depend on providing timely and safe J41 89 deliveries today.

J41 90 On the other hand, deregulation may have induced firms, which J41 91 employ drivers for hire, to increase their safety expenditures. J41 92 Suppose a trucking firm uses two complementary inputs, labor J41 93 (L) and safety (S), per truck to deliver a given quantity J41 94 of goods. We can think of S broadly as the resources the firm J41 95 uses to maintain some level of safe operation, which may include J41 96 any investment in safety training for drivers or the purchase of J41 97 truck-related safety equipment. In competitive markets, the level J41 98 of safety the firm provides determines the wages the firm must pay J41 99 to compensate drivers for any relative risks in trucking. If, for J41 100 example, the firm offers better training or installs more (or J41 101 better) safety equipment, then wages would be commensurately lower. J41 102 As wages fell after deregulation, this may have created an J41 103 incentive for firms to hire more drivers and to provide additional J41 104 safety to compensate drivers for any apparent risks in trucking. In J41 105 addition, it is quite possible that firms provided the drivers with J41 106 greater safety resources (i.e., equipment or training) because the J41 107 new drivers were relatively inexperienced. The implication is that J41 108 if firms increased safety expenditures because wages fell, then J41 109 accidents would have probably fallen as well. Moreover, if the J41 110 additional expenditures were used to purchase truck-related safety J41 111 equipment, then fatalities and injuries would have fallen too.

J41 112 And finally, deregulation may have affected driving intensity J41 113 directly if drivers were induced to make more deliveries by driving J41 114 longer hours and at faster speeds. Therefore, it would be important J41 115 to control for differences in speed limit enforcement and vehicle J41 116 inspections when attempting to explain differences in accident J41 117 rates across states.

J41 118 The above discussion indicates there are reasons to expect that J41 119 accidents may have increased or decreased as the result of J41 120 deregulation. The next section discusses the empirical model which J41 121 attempts to determine the net impact of deregulation, while holding J41 122 various other factors constant.

J41 123 III. The Empirical Model and Regression Results

J41 124 The Empirical Model

J41 125 The sample consists of a pooled, cross-section of state data J41 126 for 1977, 1982, and 1987. Since the MCA was passed in 1980, the J41 127 sample can be partitioned into two periods: 1977 represents a J41 128 regulation year, whereas 1982 and 1987 represent deregulation J41 129 years. The variable descriptions and sources are discussed in the J41 130 appendix.

J41 131 The dependent variable is the number of accidents per J41 132 truck-miles traveled per state. The numerator represents accidents J41 133 that occurred in a particular state, which were reported by truck J41 134 drivers engaged in interstate transportation. The denominator is an J41 135 estimate of the truck-miles traveled per state, and includes all J41 136 trucks that traveled more than 200 miles from their base of J41 137 operation. To the extent that the denominator includes some J41 138 intrastate travel, the denominator is likely to overstate the miles J41 139 logged by interstate truckers. Therefore, the dependent variable is J41 140 likely to understate the actual accident rate per truck-miles J41 141 traveled for interstate carriers, and any effect of deregulation J41 142 that we uncover would be a conservative estimate of the actual J41 143 impact.

J41 144 The set of independent variables include factors which have J41 145 some theoretical or institutional basis for explaining the J41 146 variation in accidents per truck-miles traveled across states. The J41 147 first is the number of highway police officers per highway mileage J41 148 (POLICE). This measure is intended to control for differences J41 149 in enforcement resources used to detect speed and weight J41 150 violations. More officers per highway mile should lead to more J41 151 careful driving and, consequently, less accidents. Thus, we expect J41 152 a negative sign for this variable.

J41 153 The second, third, and fourth variables relate to traffic J41 154 conditions on interstate highways which could arguably affect J41 155 accident rates across states. These conditions are: the average J41 156 speed of interstate traffic per state (AVESPEED); the variance J41 157 of speed on interstate highways; and a density variable J41 158 (CONGESTION) to proxy the level of road congestion. Lave [14] J41 159 argues that it is the variance of highway speed that causes J41 160 accidents and not the mean speed. The intuition is that if all J41 161 vehicles were traveling at the same speed (i.e., variance is 0), J41 162 then chances of an accident occurring are almost zero at any mean J41 163 speed. Recently, Levy and Asch [15] challenge this view and report J41 164 some evidence which shows that both the mean and variance of J41 165 highway speed affect accident rates. Since this issue appears to be J41 166 unresolved, I have included both measures in the model.

J41 167 AVESPEED is the estimated statewide average highway speed J41 168 for all vehicles. The variance is calculated as the difference J41 169 between 'the speed at or below which 95 percent of the vehicles are J41 170 traveling' (85TH) and the statewide average. In the regression J41 171 model, one could write the expression as

J41 172 formula.

J41 173 Lave, however, suggests that the expression be rewritten as

J41 174 formula.

J41 175 The rationale is that the coefficient for AVESPEED in the J41 176 second expression reflects the relative effect of both the average J41 177 and variance of speed (i.e., <*_>beta<*/>-<*_>THETA<*/>). J41 178 Thus, if the variance has a larger impact than the average (i.e., J41 179 <*_>beta<*/><<*_>THETA<*/>), then the coefficient on AVESPEED J41 180 would be negative while the coefficient on 85TH would be J41 181 positive. This explains why a negative sign for AVESPEED is J41 182 counter intuitive, since the coefficient measures the relative size J41 183 of each effect.

J41 184 CONGESTION is simply the total number of automobile J41 185 registrations normalized by the estimated highway mileage for each J41 186 state. This factor is intended to control for differences in road J41 187 congestion across states, since greater congestion is likely to J41 188 increase the probability of a collision between two vehicles. Thus, J41 189 we expect a positive sign for this variable.

J41 190 I included two additional variables to control for differences J41 191 in weather conditions across states. The first is the average J41 192 number of rain days per state (RAINDAYS). I anticipate a J41 193 positive coefficient for this variable since rain is likely to J41 194 impair vision and road conditions. The second is the average J41 195 snowfall (SNOW) per state, and the expected sign for this J41 196 variable is uncertain. On the one hand, one may argue that more J41 197 snow leads to more accidents because of slippery roads. On the J41 198 other hand, more snow might reduce travel if drivers wait until the J41 199 roads are cleared, which makes a negative sign plausible.

J41 200 Finally, I included a dummy variable (MCA80) in the model J41 201 to control for differences attributable to deregulation; 1977 is a J41 202 regulation year while 1982 and 1987 are deregulation years. Thus, J41 203 MCA80 equals 1 for 1982 and 1987 and 0 otherwise. It is J41 204 possible, however, that any empirical difference between the time J41 205 periods is unrelated to the shift from regulation to deregulation, J41 206 and I acknowledge this potential interpretation. Nonetheless, any J41 207 difference that is uncovered in the analysis will be discussed in J41 208 the context of the existing empirical literature [3; 17].

J41 209 The Empirical Evidence: Rates per Truck-Miles J41 210 Traveled

J41 211 Table 1 reports the estimates from four regression equations, J41 212 and several interesting findings emerge from these results. First, J41 213 the MCA80 variable has a negative sign in each of the four J41 214 equations, but is only significant (using a two-tailed test) at J41 215 conventional levels in the fatality, injury, and property-damage J41 216 equations. These results suggest that drivers experience the same J41 217 accident rate that they did before deregulation, but that the J41 218 accidents involved fewer fatalities and injuries.

J41 219 The insignificance of MCA80 in the accident equation J41 220 presents a puzzle; that is, how and why would deregulation affect J41 221 the fatality and injury rates, but not the accident rate? J41 222 J42 1 <#FROWN:J42\>Part 1. Jurisprudence is Adjudication

J42 2 a. Critique of legal positivism

J42 3 Dworkin constructs his legal theory largely in response to J42 4 legal positivism and utilitarianism. Legal positivism, Dworkin J42 5 claims, defines law merely as a set of rules or 'social facts,' and J42 6 allows judges to legislate when rules run out in hard cases. As J42 7 such, Dworkin argues legal positivism contains two main defects: J42 8 (1) an inaccurate description of adjudication in American legal J42 9 practice, and (2) an inadequate theoretical account of law and J42 10 legal obligation. Utilitarianism, Dworkin claims, enforces the J42 11 preferences of the majority over the preferences of the minority to J42 12 improve social welfare. Dworkin considers utilitarianism a J42 13 deficient political theory because individual rights depend upon J42 14 the shifting sands of political compromise and majority will. J42 15 Consequently, Dworkin's legal theory establishes certain individual J42 16 rights beyond the control of the majority, and links positive law J42 17 to political theory and thereby to virtue and justice. Dworkin thus J42 18 rejects, partially at least, both legal positivism and J42 19 utilitarianism.

J42 20 Dworkin starts his examination of law with a critique of Hart's J42 21 version of legal positivism. Hart's positivism has three central J42 22 tenets: (1) the law of a community consists of special rules J42 23 identifiable by the manner in which they were adopted, (2) the set J42 24 of legal rules is exhaustive of the law, and (3) a legal obligation J42 25 derives only from a legal rule. Hart further divides rules into two J42 26 types: primary and secondary rules. Primary rules grant rights or J42 27 impose legal obligations upon members of the community. For J42 28 example, the criminal law consists of primary rules. Secondary J42 29 rules stipulate how primary rules are formed and validated. Hart J42 30 calls a fundamental secondary rule a 'rule of recognition.' The J42 31 latter is legitimate because it is accepted by the community. In J42 32 the United States, the rule of recognition is the federal J42 33 constitution since the legitimacy of any particular law can be J42 34 traced through a complicated chain of validity back to the federal J42 35 constitution.

J42 36 According to Hart, judges decide cases by applying rules of J42 37 law. When a case is not governed by any existing rule of law, a J42 38 hard case, the judge decides the case by exercising his discretion. J42 39 The new rule the judge forms becomes part of the legal order and is J42 40 valid because, under Hart's system, the judge has power to create a J42 41 new rule when existing rules do not provide guidance in a J42 42 particular case, Judges hence "may be said to make J42 43 'choices' among possible alternatives or to exercise a 'legislative J42 44 discretion.' Since most cases involve the simple J42 45 application of rules, the legislative powers of the judge are J42 46 limited.

J42 47 Dworkin charges that Hart's theory of judicial discretion J42 48 inaccurately describes what judges in the United States do when J42 49 they decide a hard case. Judges use principles, in addition to J42 50 rules, to decide cases. Principles differ from rules because J42 51 principles are abstract, general and flexible. In support of his J42 52 claim, Dworkin cites Riggs v. Palmer, denying a murderer J42 53 his inheritance on the ground that "no man shall profit by J42 54 his own crime," and Henningsen v Bloomfield Motors, J42 55 Inc., voiding a limitation of warranty provision in a consumer J42 56 contract principally on the ground that unfair bargains are J42 57 unenforceable. Because the courts in Riggs and Henningsen J42 58 invoked principles not rules to decide the case, Dworkin concludes, J42 59 Hart's theory of judicial discretion fails to describe judicial J42 60 decision making and therefore is wrong.

J42 61 The origin of legal principles "lies not in a J42 62 particular decision of some legislature or court, but in a sense of J42 63 appropriateness developed in the profession and the public over J42 64 time. Their continued power depends upon this sense of J42 65 appropriateness being sustained." When a judge decides a J42 66 hard case, the judge does not simply create a decision in a vacuum; J42 67 rather the judge invokes the applicable principles of law and J42 68 applies them to the case. The judge does not create law and apply J42 69 it retroactively to the parties; the judge enforces moral and legal J42 70 rights preexisting the case although not captured by any single J42 71 rule of law. Principles exist independently of legal institutions J42 72 enacting rules of law because they are part of the community's J42 73 moral and political culture.

J42 74 Dworkin draws two broad conclusions from the putative failure J42 75 of Hart's theory of discretion to reflect actual judicial practice. J42 76 First, he claims Hart's theory of law does not identify all laws in J42 77 the society because it fails to account for the existence of J42 78 principles that judges commonly use to decide cases. Second, the J42 79 master rule of recognition, to the extent that it ignores rules, is J42 80 not a master rule defining all laws. If the master rule were J42 81 redefined to capture principles, Dworkin maintains, it would become J42 82 so broad as to be meaningless. Dworkin concludes "if we J42 83 treat principles as law we must reject the positivists' first J42 84 tenet, that the law of a community is distinguished from other J42 85 social standards by some test in the form of a master J42 86 rule." This raises the possibility, Dworkin contends, that J42 87 legal obligation rests on constellations of principle, as well as J42 88 rules of law. More important, the critique of legal positivism J42 89 provides the primary material for Dworkin to create the rights J42 90 thesis.

J42 91 Scholars claim Dworkin reduces positivism to a theory no one J42 92 actually holds since positivists recognize restraints upon judicial J42 93 discretion. Sullivan correctly notes that "judicial J42 94 discretion is more tightly circumscribed than Dworkin's caricature J42 95 indicates." Though the judge is free to weigh various J42 96 considerations, "this does not entail that decisions J42 97 resulting from this process are arbitrary, or that the judge's J42 98 discretionary power is therefore completely unconstrained." J42 99 Nevertheless, genuine differences differentiate Dworkin's legal J42 100 theory from positivism. For positivism "law is J42 101 fundamentally characterized by the notion of a rule," J42 102 whereas for Dworkin it is a process of discovering the political J42 103 morality implicit in positive law. Positivism distinguishes law J42 104 from morality by identifying a master rule of recognition; Dworkin J42 105 denies the existence of a master rule of recognition and locates J42 106 law in the practice of interpretation. While some scholars argue J42 107 that Dworkin's legal theory merely amends positivism, his theory J42 108 nevertheless investigates the origin of law beyond the mere fact of J42 109 its enactment by a legitimately constituted legal institution.

J42 110 b. The rights thesis

J42 111 The rights thesis corrects two flaws in the positivist's J42 112 account of judicial discretion: (1) treating the judge as deputy to J42 113 the appropriate legislature, and (2) claiming judges decide cases J42 114 in two stages, first reviewing the law books to locate pertinent J42 115 rules, and second setting aside the law books when pertinent rules J42 116 are not found. Dworkin says judges are not and should not be J42 117 legislators for two reasons. First, judges are not elected and J42 118 therefore, under democratic theory, are not entitled to make law, J42 119 and second, the judicial creation of ex post facto J42 120 legislation punishes the losing party. Dworkin also denies that J42 121 judges decide cases in two stages as positivism maintains. Rather J42 122 judges enforce the preexisting rights of parties grounded in legal J42 123 principles. According to Dworkin, adjudication should be as J42 124 unoriginal as possible.

J42 125 Central to the rights thesis is the distinction between J42 126 arguments of policy and arguments of principle. Arguments of J42 127 prinicple justify a political decision by showing that the decision J42 128 respects or secures some individual or group right. Dworkin states J42 129 the "argument in favor of anti-discrimination statutes, J42 130 that a minority has a right to equal respect and concern, is an J42 131 argument of principle." Arguments of policy justify a J42 132 political decision by showing the decision advances or protects J42 133 some collective goal of the community as a whole. Dworkin states J42 134 the "argument in favor of a subsidy for aircraft J42 135 manufacturers, that the subsidy will protect national defense, is J42 136 an argument of policy." While Dworkin realizes principles J42 137 and policies mix, and therefore recognizes the distinction between J42 138 them is more subtle and complex than his examples suggest, he J42 139 nevertheless advances the claim that a principle cannot be J42 140 outweighed by every social policy.

J42 141 Dworkin rejects arguments of policy as a legitimate basis to J42 142 decide cases because they fail to recognize the existence of rights J42 143 and require the judge to legislate. Arguments of policy do not J42 144 provide a stable vehicle to secure rights since they depend upon J42 145 variable factors designed to promote the social welfare. If J42 146 arguments of policy determined rights, the latter would fluctuate J42 147 according to whatever factor advanced the social welfare at a J42 148 particular historical moment. The law and economic analysis theory J42 149 illustrates the problem of basing rights on utility, since the J42 150 efficient decision may deviate from prior law, and hence frustrate J42 151 the expectations of the parties. Dworkin claims citizens are J42 152 entitled to rely on rights and duties flowing from the law, and are J42 153 entitled to request the court to enforce them. If a plaintiff is J42 154 entitled to win a lawsuit, Dworkin maintains, the plaintiff always J42 155 had the right to win and the defendant always had a duty to act. J42 156 Economic analysis runs afoul of this conception of adjudication J42 157 because economic analysis defines rights ex post facto on J42 158 grounds of efficiency.

J42 159 Dworkin divides rights into four categories; (1) background (2) J42 160 institutional (3) abstract and (4) concrete. Background rights are J42 161 rooted in political theory and are not necessarily recognized as J42 162 rights by legal institutions. For example, a political theory may J42 163 demand "to each according to his needs, from each according J42 164 to his ability," although no legal institution in the J42 165 United States yet recognizes that claim. On the other hand, an J42 166 institutional right "provides a justification for a J42 167 decision by some particular and specified political J42 168 institution" and therefore, unlike certain background J42 169 rights, has the force of law. An abstract right is "a J42 170 general political aim" such as "Congress shall J42 171 enact no law abridging the freedom of speech," while a J42 172 concrete right gives practical content to its corresponding J42 173 abstract right. For example, the right of newspapers to publish J42 174 defense plans classified as secret provided the publication will J42 175 not create an immediate physical danger to troops is a concrete J42 176 expression of the abstract right contained in the first J42 177 amendment.

J42 178 The rights thesis enforces existing concrete and legal rights J42 179 of an institutional type. Dworkin draws an analogy between the J42 180 institution of chess and the institution of law to explain what he J42 181 means by the institutional character of legal rights. He considers J42 182 how a chess referee would interpret a rule of chess which provides J42 183 that "the referee shall declare a game forfeit if one J42 184 player 'unreasonably' annoys the other in the course of J42 185 play" when one player smiles continually at his opponent to J42 186 unnerve him. Since the rule does not define the term J42 187 'unreasonably,' Dworkin says, the referee must construct a theory J42 188 of the game of chess to interpret the rule. The theory of the game J42 189 of chess is derived from the rules constituting the game.

J42 190 Dworkin first observes that the chess referee cannot interpret J42 191 the rule by imposing personal convictions. For example, the chess J42 192 referee may believe that individuals have a right to equal welfare J42 193 without regard to intellectual abilities and rely upon this J42 194 conviction to find that annoying behavior is reasonable so long as J42 195 it reduces the importance of intellectual ability in deciding who J42 196 will win the game because chess is a game of intellect. However, J42 197 "(s)ince chess is an intellectual game, (the chess referee) J42 198 must apply the forfeiture rule in such a way to protect, rather J42 199 than jeopardize, the role of intellect in the context." J42 200 Therefore, the discretion of the referee is fettered by the nature J42 201 of the game of chess which disqualifies personal convictions of the J42 202 referee contrary to the game's point.

J42 203 The chess referee must determine the abstract concept of the J42 204 game of chess and interpret the rule to implement that concept. The J42 205 abstract concept of the game of chess is identified by analyzing J42 206 its institutional rules and by posing a series of questions J42 207 designed to identify the game's character. Since chess is a game of J42 208 intellect, Dworkin suggests, the referee may need to construct not J42 209 only the concept of chess, but also the concept of intellect itself J42 210 to interpret the rule forbidding annoying behavior. 'Intellect' is J42 211 the point of the game. While the referee exercises judgment to J42 212 define the concrete right of the players, the exercise of judgment J42 213 does not reflect the referee's personal convictions. J42 214 J43 1 <#FROWN:J43\>Conceptualizing Anti-Gay Violence

J43 2 JOSEPH HARRY

J43 3 This chapter attempts a conceptualization of the motivations J43 4 and situations surrounding the hate crime of violence against gay J43 5 males and lesbians. (The term gay will henceforth be used to J43 6 refer both to gay males and to lesbians. Gay males and J43 7 lesbians will be used to discriminate between the groups.) J43 8 Violence is anti-gay when its victims are chosen because they are J43 9 believed to be homosexual. This definition excludes common crimes J43 10 committed against gay males or lesbians when the homosexuality of J43 11 the victim is unknown or irrelevant to the choice of victim. J43 12 Although some research has been done on the victims of anti-gay J43 13 violence (Committee on the Judiciary, 1986; Harry, 1982; Miller & J43 14 Humphreys, 1980), there is little knowledge about the perpetrators. J43 15 In this chapter, I attempt to enlarge on this scarce data.

J43 16 MOTIVATIONS FOR ANTI-GAY VIOLENCE

J43 17 As Berk and his colleagues suggest in Chapter 8, the J43 18 perpetrators of anti-gay violence are very largely male, in their J43 19 late teens or early twenties, strangers to the victim(s), in J43 20 groups, and not engaged in victimization for profit. Anti-gay J43 21 violence seems to be committed during the peak years of J43 22 delinquency/criminality (Hindelang, Gottfredson, & Garofalo, 1978). J43 23 Anti-gay violence may thus be but one element of the J43 24 general delinquency complex in which correlations are found among J43 25 most kinds of illegal behaviors. If so, it may require little J43 26 special explanation beyond those usually offered for delinquency J43 27 and crime; that is, no special psychological propensities on the J43 28 part of the offender need be assumed.

J43 29 Even if, however, the typical anti-gay offender is a generic J43 30 criminal disengaged from the conventional moral order, some closer J43 31 examination is required to explain why or when he may engage in a J43 32 particular type of offense (male pronouns are used throughout to J43 33 highlight the likelihood that perpetrators are male). Whereas J43 34 disengaged delinquents are free to commit a variety of illegal J43 35 activities, such freedom does not mean they will engage in any one J43 36 particular activity. Motivations and situational circumstances are J43 37 needed to focus their attention on a particular illegal J43 38 possibility. Why commit anti-gay violence versus rape or armed J43 39 robbery or burglary? What is there about beating homosexuals that J43 40 appeals to offenders?

J43 41 I suggest that most anti-gay violence arises out of the J43 42 interactions of male groups in their late adolescence or early J43 43 twenties. For many persons, the period of adolescence constitutes J43 44 an extended 'moral holiday' during which bonds to the adult moral J43 45 order are attenuated by involvement in an adolescent subculture, J43 46 the principal emphases of which are hedonism and autonomy from J43 47 adult control. Such adolescents find themselves most at home not in J43 48 school or in the family but in the company of same-age peers. Such J43 49 company is unstructured, informal, and largely devoted to J43 50 recreational pursuits, both legal (e.g., sports) and illegal (e.g., J43 51 drugs). Although the social groups of the adolescent and J43 52 immediately postadolescent worlds consist of both same-sex and J43 53 mixed-sex groups, groups of gay-bashers seem to be almost J43 54 exclusively male. In a few accounts of gay-bashing incidents, a J43 55 female consort of the offenders served as appreciative audience. J43 56 This acknowledged, it remains that the offenders are overwhelmingly J43 57 male and usually act in groups.

J43 58 One depiction of such male adolescent groups has been provided J43 59 by Matza (1964, pp. 49-64) in what he calls the "situation J43 60 of company." In this situation, adolescents are constantly J43 61 mutually pressured to prove their commitment to the male gender J43 62 role. Engaging in a variety of illegal or deviant acts is one way J43 63 to prove their daring, their maleness, their adulthood. In the J43 64 rather primitive eyes of the adolescent male, sexual and violent J43 65 acts are the two main means through which they can prove their male J43 66 commitment. For example, adolescent males have been found much more J43 67 likely than females (68% versus 44%) to tell their friends about J43 68 their first experience with sexual intercourse (Carns, 1973), J43 69 apparently because reporting such intercourse has status value in J43 70 the eyes of peers.

J43 71 Although violence can also validate one's commitment to being a J43 72 male, it has risks. In the legitimate forms of sports, one can J43 73 lose. Also, many forms of available sports are supervised by adults J43 74 and hence do not fit well with the emphases of the adolescent J43 75 subculture. Most illegal forms of violence, such as fighting, offer J43 76 the possibilities of losing, being injured, possibly being J43 77 arrested, and having one's status considerably deflated. Hence, J43 78 although it is important for the adolescent male to be able to talk J43 79 a good fight, actually engaging in one is risky business.

J43 80 The option of gay-bashing offers a nearly ideal solution to the J43 81 status needs of the immature male. When done in groups, it offers J43 82 little risk of injury. It provides immediate status rewards in the J43 83 eyes of one's peers because, unlike verbal reports of sexual J43 84 conquest, it provides direct and corroborated evidence of one's J43 85 virility. It offers only minimal likelihood of arrest both because J43 86 the offenders are rarely known to the victim and because the victim J43 87 is unlikely to report the incident to the police. Gay-bashing J43 88 serves to validate one's maleness in the areas of both violence and J43 89 sexuality. It is a sexual, but not homosexual, act because it J43 90 reaffirms one's commitment to sexuality exclusively in its J43 91 heterosexual form. Occasionally, gay-bashing incidents include J43 92 forcible rape, either oral or anal. Given the context of coercion, J43 93 however, such technically homosexual acts seem to imply no J43 94 homosexuality on the part of the offenders. The victim serves, both J43 95 physically and symbolically, as a vehicle for the sexual status J43 96 needs of the offenders in the course of recreational violence.

J43 97 The offenders' choice of victim is made appropriate by the J43 98 institution of gender. Although young males living in the situation J43 99 of company and morally adrift may find anti-gay violence appealing, J43 100 such behavior requires that the laws and norms of civil society be J43 101 morally neutralized. In cases of gay-bashing, the offender is not J43 102 simply on a moral holiday, as he may be when committing common J43 103 property offenses, nor is he simply grabbing excuses out of thin J43 104 air to justify seriously criminal behavior. He is resorting to an J43 105 alternative set of norms based upon the institution of gender: that J43 106 set of norms, imbibed mostly unconsciously from birth, that J43 107 prescribes our sense of what is 'masculine' and what is 'feminine' J43 108 in thought, affect, and behavior.

J43 109 The gender institution often operates as a set of subterranean J43 110 values justifying illegal conduct when more acceptable J43 111 justifications (e.g., self-defense) cannot be found within the law. J43 112 Our dominant institution of gender contributes to the view that J43 113 male-female rape is justified if the victim behaved in a J43 114 'provocative' or 'unladylike' manner. It also allows the J43 115 perpetuation of wife-beating. Gay-bashing seems similarly to be J43 116 based on a popularly accepted belief, in this case that the only J43 117 justifiable forms of sex are those between males and females. In J43 118 the case of gay-bashing at least, moral neutralization is based J43 119 upon "denial of the victim" (Sykes & Matza, 1957) J43 120 and of her or his moral worth as a human being. By viewing the J43 121 victim as worthy of punishment for having violated gender norms, J43 122 the offender not only excuses himself from opprobrium but sees J43 123 himself as rendering gender justice and reaffirming the natural J43 124 order of gender-appropriate behavior.

J43 125 The above arguments may seem to predict too much J43 126 gay-bashing, just as Matza (1964, pp. 25-26) argued that J43 127 cultural theories of delinquency predict too much delinquency J43 128 because they imply a continuing commitment by the juvenile to J43 129 delinquent behaviors. Matza's point was that, if juveniles are so J43 130 committed to delinquency, they would engage in it almost on a J43 131 full-time basis. Similarly, if anti-gay violence is an ideal means J43 132 for the attainment of sexual status by young males and is based on J43 133 such a basic institution as gender, it would seem that gay-bashing J43 134 should be a daily occurrence involving significant percentages of J43 135 both the homosexual and the heterosexual populations. To deal with J43 136 this issue, we need some idea of the extent of anti-gay violence. J43 137 Because relevant statistics are few, we divide gay-bashing J43 138 incidents into three types based on the age of the victim.

J43 139 First are serious physical assaults and homicides J43 140 committed against adult lesbians and gay males such as those J43 141 reported in the House Criminal Justice Subcommittee hearings on J43 142 Anti-Gay Violence (Committee on the Judiciary, 1986). These J43 143 reported assaults are clearly the most serious ones and do not J43 144 include the common, random beatings of homosexuals that occur in J43 145 the streets, parks, and parking lots of America. Most assaults go J43 146 unreported either because the victim fears being discredited by J43 147 family, the law, or employers or because the assault was less J43 148 serious, although still criminal.

J43 149 Second are assaults and related harassments of lesbian and J43 150 gay male adolescents by their peers, such as those that gave J43 151 rise to the Harvey Milk School in New York City for homosexual J43 152 adolescents. The existence of such a school implies that J43 153 mistreatment of homosexual adolescents is pervasive in the J43 154 adolescent world.

J43 155 Finally, probably far more common than either of the other J43 156 forms of assault and harassment are the beatings of effeminate J43 157 boys, both future homosexuals and heterosexuals (Saghir & J43 158 Robins, 1973, pp. 18-23) that occur during childhood. These J43 159 beatings occur because the boys do not confirm to the extremely J43 160 rigid rules of the male gender role. They also reaffirm the J43 161 offender's commitment to that role before his peers. J43 162 Psychologically, they serve the same function as the more serious J43 163 gay-bashing of adulthood. They differ from the latter in two ways, J43 164 however. First, they are more accepted in conventional adult norms. J43 165 Second, they do not suggest as much criminality and probable moral J43 166 disengagement from the norms of civil society on the offenders' J43 167 part as does adult gay-bashing. Culturally, however, the childhood J43 168 and adult incidents are the same. Whether persons who engage in J43 169 adult gay-bashing have also engaged in childhood 'sissy-bashing' is J43 170 unknown.

J43 171 If we view the above three age-based types of incidents as J43 172 gay-bashings that differ only in the ages of the participants J43 173 involved, the ideas offered to explain anti-gay violence may not J43 174 predict too much. Gay-bashing may be endemic during childhood and J43 175 decline in frequency with age while at the same time it increases J43 176 in seriousness and leathality. As males approach adulthood, most J43 177 become more secure in their gender roles, so that proving their J43 178 gender adequacy becomes less obsessive and gender deviance in J43 179 others becomes less salient. Hence the motivations for gay-bashing J43 180 may decline with the advent of an adulthood that is not defined in J43 181 the stark imagery of the immature male.

J43 182 THE SITUATIONS OF ANTI-GAY VIOLENCE

J43 183 The views of gay-bashers are clearly in agreement with those of J43 184 the large majority of the population who disapprove of J43 185 homosexuality (see Chapter 5). Reporting date from the General J43 186 Social Survey (National Opinion Research Center, 1988) in 1987, 82% J43 187 of the population found homosexuality "always J43 188 wrong" or "almost always wrong." This J43 189 percentage has changed little since 1973 and may have increased J43 190 slightly since the 1970s. For purposes of analysis, we divide this J43 191 82% into three categories. Most strongly opposing homosexuality are J43 192 a small number of activists who go out of their way to find J43 193 homosexuals to assault. Such strongly motivated persons would J43 194 typically go to a place where homosexuals are known to gather such J43 195 as a gay ghetto (Levine, 1980) or to the environs of a gay bar. J43 196 Somewhat less opposed to homosexuality would be the larger number J43 197 of opportunists who are not sufficiently motivated to seek out J43 198 homosexuals to victimize but will assault them as occasions arise. J43 199 Such situations would typically arise in non-gay-defined settings J43 200 when persons who are visibly homosexual appear. The remainder of J43 201 the 82% are those who disapprove of homosexuality but not strongly J43 202 enough to engage in gay-bashing. This group is theoretically J43 203 important because it is by far the largest of the three and it J43 204 consists of those who might normally be expected to serve as J43 205 guardian citizens in cases of assault (Cohen & Felson, 1979). In J43 206 the case of common crimes among heterosexual participants, such J43 207 guardians serve the function of being interveners or of calling the J43 208 police. In cases of anti-gay violence, however, it is doubtful that J43 209 many of this large group who disapprove of homosexuality would be J43 210 willing to actively assist the victim. J43 211 J44 1 <#FROWN:J44\>AMERICAN LEGAL THOUGHT AND LEGAL REFORM

J44 2 A. Introduction

J44 3 The Federal Rules of Civil Procedure, implemented in 1938, and J44 4 the Federal Rules of Evidence, enacted in 1975, are designed, we J44 5 are told, to promote the "just, speedy, and inexpensive J44 6 determination of every action." They are to be J44 7 "construed to secure fairness in administration, J44 8 elimination of unjustifiable expense and delay, and promotion of J44 9 growth and development of the law of evidence to the end that the J44 10 truth may be ascertained and proceedings justly J44 11 determined." The stated goals of these transsubstantive J44 12 rules, then, are that the truth be determined and disputes be J44 13 justly resolved.

J44 14 The foundational assumptions underlying the claim that the J44 15 Federal Rules of Civil Procedure and the Federal Rules of Evidence J44 16 are instruments that permit the discovery of truth and the 'just' J44 17 resolution of disputes are three related phenomena: first, the J44 18 general 'optimistic rationalism' pervading most of Western legal J44 19 and intellectual thought from the Enlightenment; second, the legal J44 20 'progressivism' of influential early to mid-20th-century American J44 21 reformers, who acted as catalysts for both procedural and J44 22 evidentiary reform; and third, the jurisprudential reaction to J44 23 American legal realism, which coalesced after World War II into J44 24 legal process or reasoned elaboration.

J44 25 The Federal Rules of Procedure and the Federal Rules of J44 26 Evidence were explicitly presented as means to the goals of 'truth' J44 27 and 'justice' in part due to this broader Western and narrower J44 28 American intellectual milieu. These goals were also channeled by a J44 29 deep public and professional reverence for both Law and the Rule of J44 30 Law. Finally, the legal profession was dedicated to the beauty and J44 31 utility of the adversary system, the hallmark of the J44 32 'Anglo-American' system of adjudication. These explicit statements J44 33 were not part of the wellspring of the Federal Rules of Evidence, J44 34 the American Law Institute's 1942 Model Code of Evidence. Part of J44 35 the failure of the Model Code of Evidence was due to its apparent J44 36 disavowal of these goals.

J44 37 The abiding belief of early 20th-century legal progressive J44 38 thought was that legal reform could rationally aid in the progress J44 39 of a legal system toward consensual notions of 'truth' and J44 40 'justice'. Legal realism, while having little contemporary impact J44 41 on the legal profession, shattered a jurisprudential faith in legal J44 42 progress toward truth and justice. The restructuring of legal J44 43 progressive thought into reasoned elaboration or legal process J44 44 after World War II required a fundamentally different justification J44 45 for a 'rational' and progressive administration of justice. This J44 46 justification, however, was unacceptable to a legal profession then J44 47 essentially unaffected by legal realism. While legal academics J44 48 could not longer faithfully argue that the goal of the trial was J44 49 truth, nor that the administration of justice was concerned with J44 50 substantive rather than procedural justice, the legal profession J44 51 and the public continued to believe in both goals. Invoking the J44 52 goals of truth and justice to garner public and professional J44 53 support was necessary to the passage of the Federal Rules of J44 54 Evidence; the structure of the Federal Rules, because it is based J44 55 on the Model Code of Evidence, undermines those goals.

J44 56 B. Optimistic Rationalism

J44 57 William Twining describes the tenets of 'optimistic J44 58 rationalism' as a congery of beliefs in truth, reason, and justice J44 59 under law. Events occur independently of human observation, and J44 60 past events can be truthfully reconstructed in the present, J44 61 although "establishing the truth about alleged past events J44 62 is typically a matter of probabilities or likelihoods falling short J44 63 of <}_><-|>compete<+|>complete<}/> certainty." Ascertaining J44 64 the truth is accomplished by listening to experts explain and J44 65 interpret relevant data and through the 'common-sense' J44 66 generalizations of society. In adjudicating disputes, establishing J44 67 the truth must be based on relevant evidence and justice can be J44 68 accomplished only if the truth is established on the basis of J44 69 relevant evidence. Further, justice can be accomplished only if the J44 70 method of fact finding is 'rational.' Rational decision making J44 71 means making decisions based on inferences from relevant evidence. J44 72 Rational decision making based on relevant evidence will thus lead J44 73 the fact-finder to the truth and to 'correctness' in decision J44 74 making. The search for truth, then, is at the core of a system of J44 75 justice. Since, however, decisions about the truth of factual J44 76 allegations occur in an imperfect, human setting, the concern for J44 77 justice is not a concern for an idealized justice but a justice J44 78 under [positive] law, which means that truth will not always be J44 79 discovered or a correct decision rendered and further means that J44 80 the goal of 'correctness' may be matched or superseded by other J44 81 social goals.

J44 82 The 'Anglo-American' system of adjudication - the adversary J44 83 system - structures and channels these tenets of optimistic J44 84 rationalism. Unlike trial by compurgation or trial by ordeal, the J44 85 adversary system was perceived as a rational system for the J44 86 discovery of truth and the pursuit of justice. In the adversary J44 87 system, each participant, with the notable exception of the J44 88 parties, plays a significant role in fulfilling the requirements of J44 89 optimistic rationalism. The attorneys for the parties investigate J44 90 and sift the facts pertinent to their (opposing) cases and offer J44 91 and object to the introduction of evidence; the judge impartially J44 92 decides disputed issues of law, including the admissibility of J44 93 evidence; and the jury, given the conflicting evidence presented by J44 94 both parties and instructions on the applicable law by the judge, J44 95 decides the disputed issues of fact and renders a verdict for a J44 96 party. This system provides checks on abuses by counsel (by the J44 97 judge), by the judge (by counsel on appeal), and by the jury J44 98 (through jury instructions, limiting their purview to issues of J44 99 'fact' and, in egregious cases, permitting the court to render a J44 100 judgment notwithstanding the verdict or to inquire into the J44 101 validity of the verdict), and so limits any departures from J44 102 rationality.

J44 103 C. Legal Progressivism and Procedural Reform

J44 104 The story of the codification of the rules of evidence is J44 105 further linked to the story of legal progressivism, for the J44 106 interest in a code of evidence rules is based on the legal J44 107 progressives' spirit of legal reform. In 1904-5, Wigmore's J44 108 Treatise was published. This four-volume first edition was an J44 109 immediate critical and commercial success. Dean Wigmore became the J44 110 unchallenged authority on the law of evidence in America.

J44 111 The publication of Wigmore's Treatise was "the J44 112 most important event in the history of the law of evidence in this J44 113 century." Wigmore's Treatise was not simply a J44 114 compendium of cases and a rationalization of inconsistencies in the J44 115 law of evidence but also a call for reform. If the legal system was J44 116 to be a rational system for the discovery of truth, as Wigmore J44 117 believed, the rules of evidence needed to be applied consistently J44 118 with those goals and to be workable in practice, that is, in J44 119 trials. Wigmore's ideas for reforming the law of evidence were part J44 120 of the emergence of legal progressivism, or sociological J44 121 jurisprudence, led by Roscoe Pound.

J44 122 In 1906 Pound spoke at the annual meeting of the American Bar J44 123 Association in St. Paul, Minnesota, about the reasons for public J44 124 dissatisfaction with the administration of justice in American J44 125 courts. Among the reasons for public dissatisfaction with the J44 126 American legal system was contentious procedure, which turned J44 127 litigation from a search "for truth and justice" J44 128 into a game or sport "that the parties should fight out ... J44 129 in their own way without interference." Decrying the J44 130 sporting theory of justice, Pound cited Wigmore for the proposition J44 131 that this view inaccurately depicted the adversary system. The J44 132 sporting theory disfigured the administration of justice and J44 133 mistakenly led even the "most conscientious" judge J44 134 to believe that he was "not to search independently for J44 135 truth and justice" and to assume that "errors in J44 136 the admission or rejection of evidence are presumed to be J44 137 prejudicial and hence demand a new trial." This gave the J44 138 community "a false notion of the purpose and end of J44 139 law."

J44 140 Pound's call was for a true "scientific J44 141 jurisprudence" based on the use of experts to make the J44 142 legal system more efficient. Making judges 'scientists' would J44 143 instill in judges an expertise which would create a greater J44 144 efficiency in the administration of justice. It would also alter J44 145 the administration of justice by creating an emphasis on J44 146 substantive justice in the courts. Two years later, Pound fleshed J44 147 out both these themes in a Columbia Law Review article. J44 148 The science of law was a means to the end of "reason, J44 149 uniformity, and certainty." A scientific jurisprudence was J44 150 a search for full justice, for "solutions that go to the J44 151 root of the controversies," for equal justice, and for J44 152 exact justice. Law was scientific in order to eliminate J44 153 "the personal equation in judicial administration, to J44 154 preclude corruption and to limit the dangerous possibilities of J44 155 magisterial ignorance." The scientific administration of J44 156 justice, however, was not to be confused with a mechanistic J44 157 jurisprudence, although a degeneration of legal science could lead J44 158 to stagnation and "petrification" in the legal system.

J44 159 The antidote to the problem of "petrification" was J44 160 "a pragmatic, a sociological legal science." J44 161 "The sociological movement in jurisprudence is a movement J44 162 for pragmatism as a philosophy of law; for the adjustment of J44 163 principles and doctrines to the human conditions they are to govern J44 164 rather than to assumed first principles; for putting the human J44 165 factor in the central place and relegating logic to its true J44 166 position as an instrument."

J44 167 Pound then noted that the law of procedure and evidence J44 168 suffered "especially from mechanical J44 169 jurisprudence." An insistence on perceiving procedure and J44 170 evidence in conceptual terms led judges to view them as ends rather J44 171 than means, and Pound gave examples of this error. He concluded by J44 172 suggesting the enactment of "a common-sense and J44 173 business-like procedure."

J44 174 The advent at the beginning of the 20th century of sociological J44 175 jurisprudence, also known as legal progressivism, progressive J44 176 proceduralism, and progressive-pragmatism, was part of the general J44 177 progressive movement and specifically part of the intellectual J44 178 departure from formalism. Pound, the progenitor of sociological J44 179 jurisprudence, relied, like all good progressives, on the J44 180 "ideology of bureaucracy" to support his efforts at J44 181 reforming the legal system. In general, "[p]rogressivism J44 182 believed in the management of government by experts and advocated J44 183 the expansion of the executive branch, primarily in the form of J44 184 administrative regulatory agencies, at the expense of the Congress J44 185 and the courts." Specifically, formalist jurisprudential J44 186 theory employed a priori reasoning rather than reasoning based on J44 187 actual economic and social conditions. The use of disinterested J44 188 experts in adjudication would make the administration of justice J44 189 more rational and just. Such reform was a gradual reform, J44 190 conservative in the sense of taking the best from the American past J44 191 and molding it to the present. Political and legal progressives, as J44 192 their name suggests, believed in the evolution of human progress, a J44 193 gradual but continued movement toward greater enlightenment about J44 194 the human condition. As advocates for efficiency, expertise, and J44 195 progress, progressives claimed that their movement was J44 196 nonideological. All bureaucrats, including judges, if correctly J44 197 trained and learned as 'scientists,' could act disinterestedly in J44 198 support of progress. Finally, some legal progressives, including J44 199 Pound and Wigmore, believed in moral absolutes. While society's J44 200 values were often inchoate and in flux, there was some consensus J44 201 about values.

J44 202 Pound's ideas for legal reform gradually captured the attention J44 203 of influential academics and 'elite' members of the legal J44 204 profession. In a 1937 article looking back at the early proposals J44 205 for legal reform, Wigmore called Pound's 1906 speech "the J44 206 spark that kindled the white flame of progress." Wigmore J44 207 noted that on the morning after Pound's speech was given, he met J44 208 with William Draper Lewis, then of the University of Pennsylvania, J44 209 and they, along with others, "resolved to do something J44 210 about it in our own limited spheres." In 1936, a writer J44 211 discussing the third draft of the proposed Federal Rules of Civil J44 212 Procedure in the American Bar Association Journal traced J44 213 the movement for reform of the rules of civil procedure to Pound's J44 214 1906 speech.

J44 215 When Pound spoke to the American Bar Association, he was dean J44 216 of the University of Nebraska School of Law, a "hitherto J44 217 obscure Nebraska jurist." Two years later, Wigmore J44 218 recruited Pound to Northwestern, and shortly after that, he was J44 219 named Story Professor of Law at Harvard Law School. By 1916, Pound J44 220 was dean at Harvard, and during his 20-year term he consolidated J44 221 Harvard's preeminence in legal education. The preeminence of Pound J44 222 at Harvard and Wigmore at Northwestern eased the transition of the J44 223 legal academy from formalist to 'progressive-pragmatist' notions of J44 224 jurisprudence.

J44 225 J45 1 <#FROWN:J45\>Chapter 5

J45 2 Rebuilding the American City

J45 3 There is a potential cycle for change. It begins with the local J45 4 problem of urban poverty and central-city decay, then moves to J45 5 local public recognition, which generates a local response. That J45 6 response is severely constrained and confounded by lack of J45 7 resources and power. In the best of circumstances - and we will J45 8 argue the case for this - the conflict between attempts to deal J45 9 locally with the problems of poverty, on the one hand, and lack of J45 10 resources, on the other, will lead to coalitions and pressures on J45 11 Congress, the federal judiciary, the White House, and federal J45 12 agencies. In the face of these pressures, Congress will pass better J45 13 federal laws and offer more generous budgets, the executive branch J45 14 will better regulate the national economy, and industry will J45 15 develop a more progressive response to competition in the global J45 16 economy. These changes, in turn, will lead not only to better J45 17 conditions, such as stronger labor demand, more attention to J45 18 education, and broad health care coverage, but will also provide J45 19 the funds municipalities need to become better places in which to J45 20 work and live.

J45 21 Changes of this sort will not happen automatically or easily. J45 22 Even when reforms begin, desirable as they would be, major changes J45 23 in either markets or policy are unlikely in the short run. Neither J45 24 is any set of partial reforms likely to 'solve the poverty J45 25 problem.' Cognizant of these severe limitations, in this chapter we J45 26 aim to be practical, to search for means by which - at the least - J45 27 the serious problems of urban poverty will get written prominently J45 28 into the political agenda.

J45 29 It is not enough to call for a return to generous, liberal J45 30 federal policy. Neither our analysis and the recommendations we J45 31 make nor the excellent and more detailed proposals of others will J45 32 stimulate governmental generosity. The authors of such proposals J45 33 have no access to the White House basement, where they might push J45 34 good legislation through Congress, to remake the country in their J45 35 (and our) better image. Instead, we believe, better policy to J45 36 minimize poverty will result only from new political forces, which J45 37 are most likely to be rooted in the poverty of the central city. We J45 38 believe, that is, that an urban political strategy is the most J45 39 practical approach for attacking America's poverty problems.

J45 40 The time is ripe for this plan. City governments are poor and J45 41 weak, and although they would like to solve the poverty problem, J45 42 they are unable. The federal government, so distant from urban J45 43 poverty, is preoccupied with international economic and political J45 44 affairs. But as the problems mount, city officials and J45 45 community-based organizations will increase their pressures and try J45 46 to form new political coalitions. As these problems threaten J45 47 national productivity, new solutions will become more attractive to J45 48 various national groups, such as industrial leaders who fear for J45 49 their international competitive advantage. If these city-based J45 50 coalitions can be formed, then inroads can be made to improve J45 51 federal policies and transfer some real power to the cities, and a J45 52 cycle of positive feedback can begin.

J45 53 This argument will proceed, section by section, through this J45 54 chapter. First, we review the history of federal-local relations in J45 55 fighting poverty. We begin by pointing out that federal aid has J45 56 drastically declined. Cities are short of resources and nearly J45 57 powerless in the face of suburban disparities and economic J45 58 pressures from big business. The situation has been made worse by J45 59 the rivalries forced on cities by federal programs and their J45 60 antineighborhood bias.

J45 61 In the second section, we provide a selection of proposals for J45 62 sensible, efficient, and efficacious federal programs to solve the J45 63 urban poverty crisis. We observe the various options for public J45 64 policy. The major portion of this section is devoted to a review of J45 65 proposals for better federal policy. It is well for the reader to J45 66 recall that the national response to global economic pressures can J45 67 vary: Japan, France, Germany, Italy, and the Scandinavian J45 68 countries, for example, have adopted policies considerably J45 69 different from those adopted in the United States. Even in Great J45 70 Britain, intercity rivalry is less destructive, because national J45 71 laws and budgets provide a common base for family and urban J45 72 services. In particular, countries make political choices among J45 73 technical options to help guide capitalist development. The United J45 74 States has chosen, partly by lack of plan, regressive policies that J45 75 guide choice of technology and work arrangements in J45 76 counterproductive directions. The country could, however, plan more J45 77 progressively. Reforms could encourage the educating and J45 78 strengthening of the workforce, from the bottom up. This would be J45 79 in contrast to the current practice of dividing and further J45 80 separating labor, destroying opportunity for those at the J45 81 bottom.

J45 82 In the third section of this chapter, we examine the potential J45 83 for political support to implement governmental programs. We raise J45 84 a troubling question: from where will political support come for J45 85 these reforms? We briefly review four possibilities, but feel J45 86 compelled to judge three of them unlikely. The fourth, which J45 87 stresses the latent strength of grass-roots politics in cities, J45 88 leads us to the last section, where we focus on strengthening the J45 89 urban role in the quest for better policy. There we will turn J45 90 briefly to the heart of the matter - how we may work collectively J45 91 inside cities to gather political support to fight poverty.

J45 92 We examine the possibilities for a renewed and revived J45 93 municipal politics. We first observe that one way to attack the set J45 94 of problems treated in this book (poverty, low productivity, social J45 95 division, and urban decay) is through local, progressive J45 96 experiments. Their success has been documented in several cities. J45 97 Chicago, Hartford, and Cleveland are among the examples, along with J45 98 the more widely discussed but smaller city experiments in J45 99 Burlington, Santa Monica, and Berkeley. If these experiments were J45 100 to be multiplied and extended, they could show the way to the J45 101 needed reconstruction of urban America. The evidence suggests there J45 102 is room for municipal maneuvering in spite of the dismal prospect J45 103 of a continued negative federal policy toward global competition, J45 104 and it also suggests what kinds of programs are most effective.

J45 105 We are more optimistic, still. When enough local change takes J45 106 place, and when more experiments arise from the economic demands J45 107 and political pressure of impoverished ghetto populations of J45 108 African Americans, Latinos, and recent immigrants, they will J45 109 provide the stimulus for coalitions to fight for better national J45 110 policies for raising productivity and improving the U.S. response J45 111 to global challenges. Once there are better national policies, they J45 112 will stimulate still better local reforms, and the cycle may J45 113 reinforce positive change.

J45 114 Federal Aid, Municipal Expectations, and Antipoverty J45 115 Programs

J45 116 We open this section on federal-urban relations with a brief J45 117 response to conservative pronouncements on the problems of the poor J45 118 and the central city.

J45 119 A Note on Neoconservatism

J45 120 The American city shows a pressing need for more adequate J45 121 national-level policies. The core of the metropolis is failing. J45 122 Central cities are falling apart physically, economically, and J45 123 socially. Whole neighborhoods are decaying, the people in them are J45 124 suffering, and social disorganization threatens entire cities. Far J45 125 too many people are poor across the nation, not just in the cities J45 126 and not only when out of work. Their numbers are not declining, J45 127 even during what the indicators say are economic good times. A J45 128 generation has reached adulthood in poverty, and the children of J45 129 that generation are threatened with worse. The gulf between haves J45 130 and have-nots in this country has never been greater, and political J45 131 communication never worse.

J45 132 Few can doubt that the United States needs a new approach to J45 133 problems of poverty, nor can they doubt the needs of the central J45 134 city. It is difficult, therefore, to accept conservatives' J45 135 arguments that we should leave well enough alone. It is hard to J45 136 believe their theories that the situation will get better by J45 137 itself. The evidence of the 1980s casts great doubt that problems J45 138 of poverty will be resolved or even seriously reduced by benefits J45 139 trickling down from general prosperity.

J45 140 The conservative argument has been much popularized, but it is J45 141 false. Most troublesome for our work at this juncture is a tendency J45 142 in much contemporary discussion to use rhetoric that at once J45 143 trivializes systematic causes of poverty and magnifies the problems J45 144 thought to derive from improper individual behavior. To put this J45 145 bias in context, we borrow ideas from political economist Albert J45 146 Hirschman, who has examined the problem of rhetoric in a broader J45 147 but closely related context.

J45 148 The rise of the welfare state in the twentieth century, J45 149 Hirschman asserts, can be seen as the third stage in a protracted J45 150 zig-zag struggle over centuries for the "development of J45 151 true economic and social citizenship." The first stage was J45 152 the back-and-forth struggle for civil rights of speech, thought, J45 153 religion, and justice. The second stage involved the effort to win J45 154 political rights by extending the vote; and the third stage was the J45 155 broader struggle to expand social and economic rights, J45 156 "recognizing that minimum standards of education, health, J45 157 economic well-being, and security are basic to the life of a J45 158 civilized person." Arguments for and against these J45 159 developments of modern society have used greatly exaggerated J45 160 rhetoric: progressives extoll the advantages of expanded rights, J45 161 while conservatives warn of dangers. At each stage there may be J45 162 progress, followed by proposals that attempt to undo the most J45 163 recent gains. We are now in a period when 'reactionary rhetoric' is J45 164 particularly prominent.

J45 165 Rhetorical and ideological backlashes stem not simply from J45 166 gloomy estimates of human capacity (as by Edmund Burke on the J45 167 French revolution or Thomas Malthus on the utility of starvation J45 168 for checking the growth of the English working class), or from fear J45 169 by the privileged classes that derives from their being outnumbered J45 170 by the common people. Support for reaction is also provided by J45 171 theoretical predispositions of the social sciences, especially the J45 172 myth of self-regulating economics, which allows free-market J45 173 enthusiasts to denounce as strongly 'perverse' any effects from J45 174 progressive interferences with the 'natural' laws of supply and J45 175 demand. The argument that welfare is the cause of poverty is a J45 176 prominent example of this sort of reactionary argument, neatly J45 177 echoing centuries of similar reaction to various stages of J45 178 progress.

J45 179 The ideological onslaught of the last twenty years against J45 180 redistributive policies has been widely justified in terms of J45 181 national economic policy. Although the most negative and racist J45 182 accompaniments of this policy have been kept usually out of sight, J45 183 the agendas of those who abuse the theories of free-market J45 184 economists and other archconservative social scientists have J45 185 sometimes been transparent. The theories lend themselves to this J45 186 abuse, as is suggested by the quantity of 'counterintuitive' J45 187 reasoning to which we have been subjected. Simulation models are J45 188 designed to show that "at times programs cause exactly the J45 189 reverse of desired results," as would be the case, for J45 190 example, if by providing good housing for the poor the City of J45 191 Boston would attract impoverished migrants and therefore worsen its J45 192 average housing conditions. It is claimed that "our efforts J45 193 to deal with distress themselves increase distress." J45 194 Conservatives argue that "we tried to provide more for the J45 195 poor and produced more poor instead. We tried to remove the J45 196 barriers to escape from poverty, and inadvertently built a J45 197 trap."

J45 198 These expectations of counterintuitive, reversed, and J45 199 inadvertent consequences of progressive social policy exist more in J45 200 the flawed reasoning of the right-wing critics than in reality. J45 201 Although unanticipated consequences do often result from public J45 202 (and private) actions, it is important to recognize that, as J45 203 Hirschman points out, "there is actually nothing certain J45 204 about such perverse effects." It is claimed by J45 205 conservatives, to take but one example, that minimum-wage J45 206 legislation dries up jobs for the poor by making labor too J45 207 expensive. But there is in fact little such evidence, and it could J45 208 be in theory that a higher legal floor to wages would have J45 209 precisely the intended salutory influence, that is, higher minimums J45 210 would have "a positive effect on labor productivity and J45 211 consequently on employment." As the terms of public debate J45 212 have shifted so as to frame a more conservative and less J45 213 compassionate view, reformers have more and more difficulty J45 214 defending in public perfectly reasonable attempts (such as the J45 215 legislation of a higher minimum wage) to improve basic conditions J45 216 for the poor.

J45 217 J46 1 <#FROWN:J46\>International Evidence on the Historical J46 2 Properties of Business Cycles

J46 3 By DAVID K. BACKUS AND PATRICK J. KEHOE

J46 4 We contrast properties of real quantities with those of J46 5 price levels and stocks of money for ten countries over the last J46 6 century. Although the magnitude of output fluctuations has varied J46 7 across countries and periods, relations among real quantities have J46 8 been remarkably uniform. Properties of price levels, however, J46 9 exhibit striking differences between periods. Inflation rates are J46 10 more persistent after World War II than before, and price-level J46 11 fluctuations are typically procyclical before World War II and J46 12 countercyclical afterward. Fluctuations in money are less highly J46 13 correlated with output in the postwar period but are no more J46 14 persistent than in earlier periods. (JEL E32, E31)

J46 15 We study fluctuations in output, prices, and money in ten J46 16 countries for which at least a century of annual data are J46 17 available. We also examine the cyclical behavior of components of J46 18 national output: private consumption expenditures, fixed J46 19 investment, government purchases of goods and services, and net J46 20 exports. Our objective is to document some of the salient features J46 21 of business cycles. We know that in many respects these countries J46 22 and time periods have been markedly different. The ten countries J46 23 differ in their institutions, their monetary and fiscal policies, J46 24 their industrial compositions and structures, and their average J46 25 aggregate growth rates. The question is whether they share, despite J46 26 these differences, similar features of business cycles.

J46 27 We find a great deal of regularity in the cyclical behavior of J46 28 real quantities. Although the magnitude of output fluctuations J46 29 varies across countries and over time, relations among variables J46 30 are remarkably stable. Investment is consistently 2-4 times as J46 31 variable as output; consumption is about as variable as output; and J46 32 both investment and consumption are strongly procyclical. The trade J46 33 balance is generally countercyclical, exhibiting larger deficits J46 34 during booms than during recessions. The exception to this J46 35 regularity in quantities is government purchases, which exhibit no J46 36 systematic cyclical tendency. Patterns of price-level fluctuations, J46 37 however, have changed markedly. Before World War II, prices were J46 38 predominantly procyclical; since then, they have been consistently J46 39 countercyclical. They have also been, in most countries, J46 40 substantially more persistent since World War II than in earlier J46 41 periods. We also find for the post-World War II period that J46 42 fluctuations in the stock of money have been less highly correlated J46 43 with output. There is no general tendency across countries, J46 44 however, toward greater persistence of money growth rates.

J46 45 Our study is an outgrowth of business-cylce research by J46 46 Robert Lucas (1977), Finn Kydland and Edward Prescott (1990), and J46 47 others that, in turn, retains some of the flavor of the tradition J46 48 of Arthur Burns and Wesley Mitchell (1946) at the National Bureau J46 49 of Economic Research (NBER). The goal of this work is, for the most J46 50 part, to summarize the properties of macroeconomic data without J46 51 imposing much theoretical structure. The resulting empirical J46 52 regularities can then serve as a guide to a variety of future J46 53 theoretical developments. A common theme in this line of research J46 54 is that the business-cycle phenomenon consists not simply of J46 55 fluctuations in aggregate output, but also of common patterns of J46 56 correlation between different aggregate time series. We report J46 57 properties of international fluctuations in a manner that conforms J46 58 with some recent work on American business cycles and thus extends J46 59 this work to a much wider range of countries and time periods. Our J46 60 motivation is international in another respect: our own research J46 61 (Backus and Kehoe, 1987; Backus et al., 1992) concerns the dynamics J46 62 of inernational trade and the relationships among business cycles J46 63 in different countries. A useful by-product is additional evidence J46 64 on the question of whether output fluctuations since World War II J46 65 have been smaller than those prior to World War I. This question J46 66 has been the subject of active debate in the United States, J46 67 including papers by Christina Romer (1986, 1989), Steven Sheffrin J46 68 (1988), and Nathan Blake and Robert J. Gordon (1989). Like J46 69 Sheffrin's study, ours puts this debate in an international J46 70 context. We include several countries not studied by Sheffrin, J46 71 notably Australia, Canada, and Japan, and introduce new data for J46 72 Sweden.

J46 73 Our data set covers ten countries with at least a century of J46 74 annual data on national output: Australia, Canada, Denmark, J46 75 Germany, Italy, Japan, Norway, Sweden, the United Kingdom, and the J46 76 United States. For the most part, countries with national income J46 77 accounts for such a long period are also those with the highest per J46 78 capita output today. Several others, including India, report J46 79 partial time series, but we doubt that these series are J46 80 sufficiently accurate for the study of short-term fluctuations. J46 81 Estimates of national output in the ten countries vary in quality, J46 82 but in some cases we think they are superior to the U.S. data.

J46 83 We begin, in Section I, by describing the data. While data for J46 84 earlier periods are unquestionably less reliable than modern data, J46 85 in some countries they appear to be good enough to provide an J46 86 accurate picture of business cycles prior to World War II. The data J46 87 for several countries seem to be significantly more accurate than J46 88 the Kuznets-based estimates for the United States, primarily J46 89 because raw-data sources are better in these countries. In Section J46 90 II, we compare output volatility before World War I (the J46 91 prewar period), after World War II (the postwar period), J46 92 and between the wars (the interwar period). Until recently, J46 93 the presumption has been that prewar U.S. output fluctuations were J46 94 two or three times larger than those of the postwar period. Romer J46 95 (1989), however, suggests that at least part of this difference is J46 96 the result of systematic measurement error in prewar GNP that J46 97 overstates its cyclical variability. Our international data set J46 98 provides additional evidence on this question.

J46 99 For the ten countries, we find that interwar fluctuations in J46 100 real output are uniformly larger than those of the postwar period. J46 101 With the single exception of Japan, the standard deviations of J46 102 output fluctuations are from two to four times larger in each of J46 103 the ten countries. We find, however, no consistent pattern for the J46 104 prewar-postwar comparison. In six of the ten countries studied by J46 105 Sheffrin (1988), prewar fluctuations are no more than 60-percent J46 106 larger than those of the postwar period. However, in the other four J46 107 (Australia, Canada, Sweden, and the United States) the fluctuations J46 108 are considerably larger in the prewar period. The U.S. case has J46 109 been discussed extensively, and it appears that part of the excess J46 110 volatility of the prewar period can be attributed to measurement J46 111 error (Romer, 1989). Romer's preferred estimate of prewar J46 112 volatility is only 30-percent higher than for the postwar period, J46 113 but Balke and Gordon (1989) argue for a number closer to 100 J46 114 percent. Sheffrin (1988) considers a similar case for Sweden and J46 115 concludes that the excess volatility in the prewar era is not J46 116 primarily the result of measurement error. We find, as do Michael J46 117 Bergman and Lars Jonung (1989) with different methods, that about J46 118 half of the excess volatility Sheffrin finds in the prewar period J46 119 disappears when revised as estimates of prewar output are used. J46 120 Australia and Canada have the most extreme differences between J46 121 periods, with output three and two times more volatile, J46 122 respectively, in the prewar period. The data for both countries are J46 123 reasonably good, so the greater volatility of measure output J46 124 probably indicates a change in the variability of real economic J46 125 activity.

J46 126 In Section III, we examine the behavior of components of the J46 127 national product: consumption, gross investment, government J46 128 spending, and net exports. We find that many of the properties of J46 129 postwar business cycles in the United States are evident in other J46 130 countries and periods. Consumption expenditures have been J46 131 procyclical and have approximately the same standard deviation as J46 132 output. Investment has also been uniformly procyclical and J46 133 generally varies, in percentage terms, from two to four times more J46 134 than output. Government spending has generally been more variable J46 135 than output, but it has been countercyclical almost as often as J46 136 procyclical. Net exports have been, for the most part, J46 137 countercyclical. We also find that correlations between measured J46 138 output movements in different countries are typically positive and J46 139 more pronounced in the postwar period than in the prewar period.

J46 140 In Section IV, we examine movements in price levels and money J46 141 stocks. Here we find, in contrast to the regularity of real J46 142 quantities, two significant changes in the cyclical behavior of J46 143 prices. We find, first, that price changes in most countries have J46 144 been more persistent in the postwar period than in the prewar J46 145 period. This finding extends related work by Jeffrey Sachs (1980), J46 146 Charles Schultze (1986), and John Taylor (1986) on the United J46 147 States and work by Gordon (1983) on the United States, the United J46 148 Kingdom, and Japan, to a larger set of countries. We also find, in J46 149 the prewar and interwar periods, that output and price-level J46 150 fluctuations are positively correlated in most of the ten J46 151 countries. However, in the postwar period, price fluctuations have J46 152 been consistently countercyclical. We find a slight decline in the J46 153 correlation of money and output in the postwar period, but no J46 154 general tendency for greater persistence of money growth rates. We J46 155 conclude this section with some speculative remarks on potential J46 156 explanations for the observed changes in price behavior.

J46 157 I. The Data

J46 158 We start with a description and evaluation of the data, J46 159 emphasizing in particular the methods used to construct prewar J46 160 national income accounts; sources and definitions are described in J46 161 Appendix A. Although national accounts are based to a large extent J46 162 on a common framework, sources of raw data differ across countries, J46 163 especially in the prewar period. Countries with the best source J46 164 material tend to have the most reliable estimates of national J46 165 income. The United Kingdom, for example, has had an annual income J46 166 tax in effect continuously since 1842, while in the United States J46 167 the federal personal income tax was only made possible in 1913 by J46 168 the Sixteenth Amendment to the Constitution. As a result, the J46 169 United Kingdom has much better data on the income side of the J46 170 national accounts for the prewar period than the United States. In J46 171 other countries, the establishment of statistical bureaus to J46 172 measure production and employment, frequently on an industry basis, J46 173 makes production-based accounts feasible. In the United States such J46 174 sources of annual data are extremely limited. For this reason, and J46 175 because accounting methods have improved in the decades since Simon J46 176 Kuznets's (1961) work on prewar U.S. GNP, estimates for several of J46 177 the countries we study are likely to be better than the American J46 178 data examined by Romer (1989) and Balke and Gordon (1989).

J46 179 Problems with prewar U.S. data have been well-documented by J46 180 Kuznets (1961) and Romer (1986, 1989). Kuznets and his coworkers J46 181 constructed national income accounts for the United States from J46 182 1869. The cornerstone of this work, and most later work as well, is J46 183 William Shaw's (1947) commodity output series: estimates of value J46 184 added in manufacturing, mining, and farming. Shaw's estimates, and J46 185 therefore those of national income, were severely constrained in J46 186 the prewar period by the absence of comprehensive annual data J46 187 sources. The most informative sources (see Shaw, 1947 part II) were J46 188 periodic federal censuses, including especially the Census of J46 189 Manufactures, available every ten years from 1869 to 1899 and every J46 190 five years from 1899 to 1919. One source of annual data is reports J46 191 on industry published by eight states. These states accounted for J46 192 between 10 and 39 percent of total manufacturing in census years, J46 193 and the reports typically covered only part of each state's J46 194 manufacturing output (see Shaw, 1947 table II:4). The state reports J46 195 were supplemented with occasional government reports and industry J46 196 publications. Kuznets (1961) interpolated further between the J46 197 census benchmarks of 1869, 1879, and 1889 by using a variety of J46 198 industry-output indicators (see the notes to tables II:1-5 in J46 199 Kuznets [1961]), since neither he nor Shaw was able to measure J46 200 commodity output directly for the 1869-1889 period. Finally, both J46 201 Shaw and Kuznets estimated nominal value added, which was converted J46 202 to real terms at a disaggregated level using producer price J46 203 indexes.

J46 204 Romer (1989), however, bases her criticism of prewar U.S. data J46 205 not on the fragmentary source material used to produce estimates of J46 206 commodity output, but on the method Kuznets (1961) used to J46 207 extrapolate from commodity output to GNP. Kuznets's problem was to J46 208 estimate GNP from information on commodity output alone, since J46 209 direct measures of other components were not consistently available J46 210 even for census years. J46 211 J47 1 <#FROWN:J47\>AFRICAN-AMERICAN SOCIETY AND EDUCATION

J47 2 Letha A. (Lee) See

J47 3 The study of social inequality in the United States has J47 4 properly focussed on the fate of African-Americans. Although other J47 5 minorities have endured privations based on their language or J47 6 religion, their identity is in their own hands to some extent: J47 7 there are no physical barriers to a change of an individual's J47 8 tongue or faith. This is not so with race. Gender would similarly J47 9 define identity by nature, but even though women are still treated J47 10 unequally across almost all racial and ethnic groups, civilized J47 11 societies increasingly denounce sexual discrimination.

J47 12 The African-Americans, at twenty-eight million, considerably J47 13 outnumber the Native Americans (about one million), the Asians, or J47 14 the mixed race Hispanic community (sixteen million). Their J47 15 situation still constitutes the 'American problem' that has been J47 16 identified for generations. It should be clear that the problem is J47 17 only partly theirs to solve, for the society as a whole must change J47 18 too, and would benefit immensely from its solution.

J47 19 On the other hand, there is evidence that a minority of J47 20 African-Americans have succeeded, despite the inferior J47 21 opportunities available to most of their members. Of course, J47 22 without their handicap of widespread victimization, these notables J47 23 would likely have won achievements which were more substantial, J47 24 earlier, easier, and achievements might have been recognized over a J47 25 broader range of ventures.

J47 26 This chapter deals briefly with four aspects of inequality in J47 27 society to set the context for a discussion of education: income, J47 28 housing, criminal justice, and health care. Other data on J47 29 inequality (such as social class) in the United States are drawn J47 30 upon where appropriate, but this chapter argues that J47 31 African-Americans exist as a statistically significant sub-group J47 32 within most of the other categories of disadvantage found in J47 33 American society - in fact a statistically larger share of J47 34 disadvantaged categories than would be expected. In short, race J47 35 does not explain everything, but if you are black and in the United J47 36 States, it has a pervasive inhibition on opportunities of every J47 37 kind. Educational programs of schooling and teacher education are J47 38 then addressed to see where intervention is most promising. J47 39 Self-help programs are identified, recognizing the difficulties of J47 40 securing broad public support.

J47 41 Background to Inequality for African-Americans

J47 42 African-Americans came to the United States as slaves in most J47 43 cases. Although they sometimes came via the British colonies of the J47 44 Caribbean, most came directly from Africa. Although other peoples J47 45 were sometimes enslaved and the Africans mixed with various other J47 46 races, slavery remained the dominant experience of their group J47 47 (more or less exclusively) until the Civil War. From the time that J47 48 slave trading was abolished during the 1830s (with legislation from J47 49 several states reinforced by the effective blockade of slaving by J47 50 the British navy) the numbers of African-Americans have grown by J47 51 natural causes rather than by continued migration. Only a few J47 52 thousand have left the U.S. for other nations such as Liberia or J47 53 Canada. In short, for over 150 years the United States has been the J47 54 only home of African-Americans, and for 125 years they have been J47 55 citizens. But not equal ones.

J47 56 Deprived of the vote initially, threatened by lynch mobs until J47 57 the present generation, denied equal access to many public services J47 58 in both government and private institutions, the African-American J47 59 is still not able to enjoy equal status. This inequality remains J47 60 institutionalized although no longer formalized in law. Most of the J47 61 social functions of American life create separate categories for J47 62 white and non-white, and black is both the largest and probably the J47 63 most disadvantaged group among the latter. At the personal level, J47 64 racism may not be evident, but almost any set of statistics can be J47 65 broken into categories that reflect the racial exploitation. Of J47 66 course, current data do not describe a society that is inevitable J47 67 or desirable. Since the systems they describe are capable of being J47 68 changed if there is sufficient social and political will, education J47 69 has an important role in improving the deplorable conditions J47 70 revealed in studies of other aspects of society: income, housing, J47 71 health, and crime.

J47 72 Income J47 73 Economic developments for African-Americans in the United J47 74 States reflect the continuum of possibilities. Blackwell (1985) J47 75 asserts that "many segments of the black community J47 76 experienced major economic progress" between the Civil J47 77 Rights Act of 1964 and the election of Ronald Reagan in 1980. J47 78 U.S. News and World Report (1986) claimed that the J47 79 majority of African-Americans are prospering and that they had J47 80 doubled their proportion in the middle class. But these two J47 81 articles still portray these gains as insignificant in the context J47 82 of economic inequality. The Reagan period brought appalling J47 83 reverses in economic equality for the majority of J47 84 African-Americans. An Urban League report entitled The State J47 85 of Black America (Swinton 1989) shows per capita income for J47 86 African-Americans remained steadily at 51 percent of that for J47 87 whites. The aggregate incomes of both groups continued to grow, but J47 88 the income gap widened by about $2000. Black family income was only J47 89 42 percent that of whites and spelled disaster for large families J47 90 in the urban setting.

J47 91 U.S. News and World Report (1986) reported that 1.1 J47 92 million black males were unemployed in 1985, compared to .5 million J47 93 in 1970. Half of all black teenagers who had started a job are now J47 94 unemployed. The explanation offered by the Department of Commerce J47 95 (1986) is that there are no longer well-paid jobs in manufacturing J47 96 where unskilled blacks would be employed. In the last fifteen J47 97 years, of twenty-three million jobs created in the private sector, J47 98 more than 90 percent were in the service sector.

J47 99 High unemployment rates among African-Americans are related to J47 100 the entry of new immigrant groups into the employment force, and to J47 101 the structural shift from low-skill jobs to jobs requiring J47 102 technical skills (See 1986). The internationalization of the J47 103 world's labor force (partly because American businesses are opening J47 104 factories offshore) points to an increasing need for American youth J47 105 to have job training and general education. Indirect evidence J47 106 indicates that investment in public schooling can be partly offset J47 107 against added costs of long-term unemployment, or can provide a J47 108 partial solution to the high rate of unemployment among blacks, J47 109 thereby providing a respectable cost/benefit argument for offering J47 110 training services.

J47 111 Black unemployment and poverty are both high, and high relative J47 112 to the figures for white Americans. The 1986 Census Bureau reports J47 113 that 33.7 million people (14.4 percent of the U.S. total) are poor. J47 114 Among African-Americans, the rate is 33.8 percent, for black J47 115 children 51.1 percent, and for the black elderly 31.7 percent. J47 116 Blacks are three times as likely as whites to find themselves in J47 117 poverty. Black families headed by women are twice as likely to be J47 118 poor. Swinton (1989) contends that not only did African-Americans J47 119 not share in the recent economic revolution, but the black-white J47 120 gap is growing. The black poverty rate doubled from 1969 to 1988 J47 121 (12 percent instead of 6 percent) and unemployed tripled (1.7 J47 122 million instead of 570,000).

J47 123 This economic inequality in America reflects upon the marginal J47 124 participation by African-Americans in the economic community. Not J47 125 only are there smaller numbers but the nature of the jobs makes J47 126 workers vulnerable to displacement from automation, technological J47 127 changes, and shifts to off-shore operations. Unskilled and J47 128 semi-skilled jobs are disappearing at the rate of 35,000 per week, J47 129 nearly two million per year. This pressure on the African-American J47 130 community is curiously functional, for Gans (1974) noted that any J47 131 social system can ensure that its 'dirty work' is done at low wages J47 132 if there are no alternatives for part of the work force.

J47 133 African-Americans are losing ground, giving rise to a nation of J47 134 the truly disadvantaged. Evidently America is growing into two J47 135 nations, one black, one white; separate and unequal.

J47 136 Housing J47 137 Blackwell (1985) argued that African-American housing should be J47 138 judged by the standards created and used by the empowered Americans J47 139 - the whites. By these standards, African-American housing is J47 140 grossly overcrowded, substandard, and expensive. It contributes to J47 141 homelessness even as it provides a limited form of housing.

J47 142 This situation evolved as a succession of Republican J47 143 administrations shifted the focus from construction of public J47 144 housing to permitting private landlords to build or convert rental J47 145 units for eligible families. These changes were followed by a J47 146 series of rental subsidies (Bell 1970). To reach provisional J47 147 agreements with landlords, landlords were allowed to subdivide J47 148 existing apartments into exceedingly small units, resulting in J47 149 overcrowding becoming commonplace (Forman 1978).

J47 150 The population density in public housing was three times as J47 151 high for African-Americans as for whites. In 1980, black families J47 152 were larger than those of whites, but their apartments were J47 153 smaller. The 1986 Census also indicated that a substantially larger J47 154 number of ancillary <}_><-|>indviduals<+|>individuals<}/> resided J47 155 with black families.

J47 156 There are not enough housing units for the African-American J47 157 population, and the existing units cost too much. Today's problems J47 158 arose from the urban renewal programs of the late 1950s and the J47 159 1960s, when the goal was to demolish blighted areas and construct J47 160 new dwellings, office buildings, and highways. African-Americans J47 161 were forced from their homes and traditional neighborhoods and the J47 162 reduced number of available housing units became available for the J47 163 poor in new ghettos (Gilderbloom 1989). The Reagan administration J47 164 added to existing problems by massive cuts in domestic spending, J47 165 including a large reduction in housing aid. Spending for low income J47 166 housing fell from $32 billion in 1980 to $7 billion in 1988. This J47 167 national agenda halted construction of low income housing despite J47 168 residential density being at its highest levels ever, housing J47 169 problems at their maximum.

J47 170 Another consequence of these government housing policies is J47 171 that some African-Americans have no housing at all. It is estimated J47 172 that less than 2 percent of new housing guaranteed by Federal J47 173 Housing Authority (FHA) mortgages is available to J47 174 African-Americans. For non-government housing, the percentage open J47 175 to black people is even smaller. One-third of the 23 million J47 176 African-Americans now live below the poverty line (U.S. Bureau of J47 177 Census 1988). It is highly probable that they live in desolation J47 178 and squalor, devoting an increasing portion of their income to J47 179 rent. Waiting lists for public housing swell dramatically, forcing J47 180 many cities to close off new applicants.

J47 181 Inequality is evident in the housing shortage, in racial J47 182 exclusion, and in homelessness. The exclusion of persons from J47 183 residential areas because of their race, color, creed, or ethnic J47 184 attachment, despite their needs and ability to pay denies J47 185 African-Americans a fundamental right. While complete freedom of J47 186 selection is never achieved, compulsory or manipulated segregation J47 187 is inherently wrong, damaging both for the immediate victims and J47 188 for the general public. Housing segregation leads directly to J47 189 segregation in other areas of life: schools, churches, hospitals, J47 190 public accommodation, recreation, welfare and civic activities, and J47 191 the workplace. Although segregation of schools is a violation of J47 192 the orders of the Supreme Court, many schools of the north and west J47 193 are segregated not by law but by racial patterns of residence.

J47 194 Health J47 195 Universal health care is hotly debated in the United States. J47 196 Conservative ideology suggests that those in need must fail in J47 197 seeking help from their families and from the marketplace before J47 198 they can depend on the government for medical assistance (Enthoven J47 199 1980, Hornbrook 1983). For poor people, seeking medical care from J47 200 the marketplace drains them of hope and resources (Trevino and Moss J47 201 1983). The numbers are substantial: in 1983 the number of people J47 202 living in poverty in the United States exceeded the entire J47 203 population of Argentina, Australia, Canada, Sweden or Taiwan - in J47 204 fact of all but twenty-three nations in the world. J47 205 African-Americans represent a large number of those in poverty.

J47 206 In 1969, 19.9 percent of African-Americans sixty-five years of J47 207 age or more could not work because of ill health. (Only five J47 208 percent of whites were in the same situation.) Low incomes for J47 209 African-Americans explain many of these discrepancies, as they have J47 210 done throughout the century. At the beginning of this century, J47 211 white men outlived black men by 15.7 years; white women outlived J47 212 black women by 16.0 years. These gaps have continued to narrow J47 213 throughout the century, to become respectively 6.8 and 5.3 years. J47 214 Significant differences are evident in the proportion of each race J47 215 that lives beyond the age of 65: 74.8 percent of white and 58.1 J47 216 percent of African males; 85.7 percent of white and 74.9 percent of J47 217 African-American females (Statistics of the United States 1986).

J47 218 J47 219 J48 1 <#FROWN:J48\>Academic Achievement in Mathematics and J48 2 Science of Students Between Ages 13 and 23: Are There Differences J48 3 Among Students in the Top One Percent of Mathematical Ability?

J48 4 Camilla Persson Benbow

J48 5 Iowa State University

J48 6 The predictive validity of the Scholastic Aptitude J48 7 Test-Mathematics subtest (SAT-M) was investigated for 1,996 J48 8 mathematically gifted (top 1%) 7th and 8th graders. Various J48 9 academic achievement criteria were assessed over a 10-year span. J48 10 Individual differences in SAT-M scores obtained in junior high J48 11 school predicted accomplishments in high school and college. Among J48 12 students in the top 1% of ability, those with SAT-M scores in the J48 13 top quarter, in comparison with those in the bottom quarter, J48 14 achieved at much higher levels through high school, college, and J48 15 graduate school. Of the 37 variables studied, 34 showed significant J48 16 differences favoring the high SAT-M group, which were substantial. J48 17 Some gender differences emerged; these tended to be smaller than J48 18 the ability group differences; they were not observed in the J48 19 relationship between mathematical ability and academic achievement. J48 20 The predictive validity of the SAT-M for high-ability 7th J48 21 and 8th graders was supported.

J48 22 "Standardized testing is much in the news. New testing J48 23 programs, test results, and criticisms of standardized testing all J48 24 are regular fare in the popular media today" (Haney, 1981, J48 25 p. 1021). Moreover, "with the possible exception of J48 26 evolution, no area in the sciences has been as filled with J48 27 emotional and confusing mixtures of science, politics, and J48 28 philosophy as the field of mental testing" (Carroll & Horn, J48 29 1981, p. 1012). These remarks portray quite well the status of J48 30 mental testing at the beginning of the 1980s, yet they seem to be J48 31 equally appropriate for describing mental testing at the beginning J48 32 of the 1990s. Some might perceive this as a rather recent J48 33 development. However, concern over standardized testing has been J48 34 voiced ever since the introduction of the Stanford-Binet J48 35 Intelligence Scale and the Army Alpha test (Cronbach, 1975; Haney, J48 36 1981).

J48 37 The concerns over mental testing primarily have been threefold: J48 38 test bias against certain groups (primarily women and minorities at J48 39 present, but children from families of low socioeconomic status in J48 40 earlier decades), the role testing might play in perpetuating J48 41 social and economic injustice, and the utility of test information J48 42 (Cleary, Humphreys, Kendrick, & Wesman, 1975; Cole, 1981; J48 43 Gottfredson & Crouse, 1986; Haney, 1981; Jensen, 1980; Scarr, J48 44 1981). The questionable value of test information has been a J48 45 particularly frequent criticism levied against college admissions J48 46 tests, such as the College Board Scholastic Aptitude Test (SAT; see J48 47 Linn, 1982b, for a review). This study was conceptualized to J48 48 address the latter concern, namely, the predictive validity of the J48 49 SAT for a special population. I assess the value of the SAT, not J48 50 for high school seniors and the college admissions process, but J48 51 rather for identifying highly mathematically gifted seventh and J48 52 eighth graders and making predictions about their achievement over J48 53 a 10-year period following their SAT-Mathematics assessment. J48 54 Specifically, I asked whether the SAT-M can detect individual J48 55 differences in the top 1% of the ability continuum that bear on J48 56 subsequent academic achievement in mathematics and science.

J48 57 The use of the SAT to identify intellectually precocious J48 58 students in Grades 7 and 8 dates to 1972 when Julian Stanley J48 59 launched the first talent search (Keating & Stanley, 1972). Stanley J48 60 was interested in students who ranked in the top 1 % in J48 61 mathematical ability. Because considerable variance in academic J48 62 ability is found among students in the 99th per-centile and J48 63 because Stanley was interested in differentiating among such J48 64 students, out-of-level testing (i.e., using tests designed for J48 65 older age groups) was required. For that reason among others (see J48 66 Stanley & Benbow, 1986), Stanley chose the SAT as the instrument J48 67 with which to screen highly gifted students. Since 1972 more than J48 68 1,000,000 seventh and eighth graders have been tested with the SAT, J48 69 and more than 100,000 such students now take the SAT annually J48 70 through various talent search programs across the United States. J48 71 The distribution of scores of such students on the SAT is about the J48 72 same as found for a random sample of high school students (Benbow, J48 73 1988). The scores tend to maintain their ordinal ranking over time, J48 74 increasing 40 to 50 points per year (Benbow & Stanley, 1982; Brody J48 75 & Benbow, 1990; Olszewski-Kubilius, 1990). Thus, from a J48 76 psychometric viewpoint, the use of the SAT with seventh and eighth J48 77 graders seems justified.

J48 78 It has not been demonstrated, however, whether use of the SAT J48 79 with young but academically competent students has utility. Is the J48 80 SAT a valid tool for assessing individual differences in current J48 81 development, and can this instrument be used to refine predictions J48 82 of exceptional academic achievements? As Cronbach (1971) pointed J48 83 out, "validation is the process of examining the accuracy J48 84 of a specific prediction or inference made form a test J48 85 score" (p. 471). In assessing the validity of the SAT for J48 86 highly gifted 7th and 8th graders, I evaluated whether academic J48 87 achievement, especially in mathematics/science, during the 10-year J48 88 period after these students were identified is much higher for J48 89 those students with exceptionally high SAT Mathematics subtest J48 90 (SAT-M) scores (top quarter of the top 1%) than for those with J48 91 comparatively 'low' SAT scores who were nonetheless in the top 1% J48 92 in ability (i.e., the bottom quarter of the top 1%). I hypothesized J48 93 that meaningful differences would be detected. Several studies have J48 94 revealed that individuals with the most potential for high academic J48 95 achievement in mathematics and science are generally considered to J48 96 be those students with high ability, particularly, high J48 97 mathematical ability (Davis, 1965; Green, 1989; Walberg, J48 98 Strykowski, Rovai, & Hung, 1984; Werts, 1967). Moreover, Kuhn J48 99 (1962) noted that an overwhelming majority of "scientific J48 100 revolutions" can be ascribed to the works of mathematically J48 101 brilliant persons.

J48 102 Nevertheless, many researchers and educators, most notably J48 103 Renzulli (1986), have argued that there is a threshold effect for J48 104 ability. According to this argument, after a certain point, there J48 105 is a decline in the power of ability to influence academic J48 106 achievement and other variables, such as motivation and creativity, J48 107 become increasingly important. The precise location of this J48 108 threshold for ability has not been determined. However, it is J48 109 thought to be at some point well below the top percentile for J48 110 ability. If Renzulli and others of this viewpoint are correct, then J48 111 there should be no statistically significant differences in J48 112 mathematics/science achievement between the two high-ability J48 113 groups. All students in the top 1% should achieve highly, and J48 114 placement within the top 1% should not affect the results.

J48 115 The reasoning in the above paragraph assumes that there is only J48 116 one threshold for ability. Yet there could be a threshold effect J48 117 for ability within a certain range (e.g., between the 90th and 98th J48 118 percentiles) but not within the top 1%. That is, differences in J48 119 ability within the 90th and 98th percentiles may not relate much to J48 120 subsequent academic achievement in mathematics/science. This view J48 121 is reasonable given that the possible differences in ability within J48 122 a range, for example, within the 90th-98th or 80th-89th percentile J48 123 ranges, are small and not reliable in comparison with the ability J48 124 differences found within the top 1% when out-of-level testing is J48 125 used. I do not test this possibility in this study. If, however, J48 126 one is interested in scientific eminence or productivity, and a J48 127 threshold effect of ability for this level of achievement, it is J48 128 within the top percentile of ability that one must focus.

J48 129 Although my prediction is contrary to Renzulli's position, it J48 130 should be noted that there are data that support the validity of J48 131 Renzulli's position. For example, students who were in the top 1% J48 132 in mathematical ability in the 7th and 8th grades were studied at J48 133 23 years of age to identify those factors that affect the ways in J48 134 which childhood potential or ability is translated into adult J48 135 achievement (Benbow & Arjmand, 1990). As a group these students had J48 136 achieved academically at a very high level but not uniformly so. J48 137 When those students who were classified as high academic achievers J48 138 in mathematics/science areas (i.e., those who were attending J48 139 graduate school in mathematics/science or medical school; J48 140 n=261) were compared with those students in the sample who J48 141 were classified as low academic achievers in those areas (those who J48 142 were not attending college or had withdrawn, those who graduated J48 143 with mathematics/science major but with low grades; n=95), a J48 144 difference in previous ability between the two groups was found J48 145 (the ability difference approximated two thirds of a standard J48 146 deviation on the SAT-M). The canonical correlation (from the J48 147 discriminant analysis) between (a) 7th-grade/8th-grade J48 148 SAT-M and (b) high school SAT-M, SAT Verbal subtest (SAT-V), and J48 149 achievement group membership was .30 for male students and .29 for J48 150 female students. (Too few cases had 7th-grade/8th-grade SAT-V J48 151 scores to allow inclusion in the analysis.) Nonetheless, ability J48 152 exhibited the weakest relationship with academic achievement in J48 153 mathematics/science as compared with variables in the areas of J48 154 educational opportunity, family characteristics, and attitudes. J48 155 Similarly, Sanders, Benbow, and Albright (1991) found that among J48 156 mathematically talented female students, previous ability on SAT-M J48 157 was not a primary factor relating to choice of mathematics/science J48 158 career or to educational aspirations.

J48 159 Thus, the aforementioned studies indicate that among those J48 160 students in the top 1%, SAT-M performance was a factor but not the J48 161 major factor predicting the students' academic success. That is, a J48 162 bright mind will not make its own way. The educational J48 163 opportunities provided to gifted children make a difference in the J48 164 children's development. In the present study, I ask the central J48 165 question: Do individual differences within the top 1% in ability J48 166 make a difference in the eventual display of achievement?

J48 167 In sum, I examine whether use of the SAT in out-of-level J48 168 testing of highly gifted students yields useful information for the J48 169 prediction of academic achievement up to 10 years after assessment. J48 170 That is, is it useful to diagnose level of talent within the top J48 171 1%, as is currently being done with well over 100,000 seventh- and J48 172 eighth-grade students on an annual basis? More succinctly, is there J48 173 a benefit to knowing where in the top 1% a student's ability lies? J48 174 It has been popularly assumed that such information is not helpful. J48 175 In essence, I assess the predictive validity of the SAT for use J48 176 with gifted 7th and 8th graders.

J48 177 Method

J48 178 Subjects

J48 179 Intellectually talented students were identified by the Study J48 180 of Mathematically Precocious Youth (SMPY), in which the SAT was J48 181 administered to intellectually able 12- and 13-year-olds in the J48 182 1970s and early 1980s (Keating & Stanley, 1972). During that J48 183 12-year period, more than 10,000 preadolescents (mostly 7th J48 184 graders) participated in SMPY 'talent searches.' (Since that time J48 185 more than 1 million students have taken the SAT through other J48 186 talent search programs.) About 3,500 of the students in the talent J48 187 searches were included in the SMPY 50-year longitudinal study. As J48 188 part of this study, researchers in the SMPY are currently tracking J48 189 four cohorts of students and studying their development J48 190 longitudinally.

J48 191 Students in Cohort 1 comprised the sample in this J48 192 investigation; they were drawn from the first three talent searches J48 193 of the SMPY (i.e., those conducted in 1972, 1973, and 1974). In J48 194 those talent searches, 7th and 8th graders in Maryland were J48 195 eligible to participate if they had scored in the upper 5% (1972) J48 196 or the upper 2% (1973, 1974) nationally on any standardized J48 197 mathematics achievement test. Qualified students took the SAT-M J48 198 and, in 1973, the SAT-V also. These tests are designed to measure J48 199 developed mathematical and verbal reasoning ability, respectively, J48 200 of high school students. However, the SAT is believed to be a more J48 201 potent measure of reasoning for 7th and 8th graders than for 11th J48 202 and 12th graders (Minor & Benbow, 1986; Stanley & Benbow, 1986).

J48 203 A score of at least 390 on the SAT-M or 370 on the SAT-V in the J48 204 7th or 8th grade was required for inclusion in Cohort 1 of the J48 205 longitudinal study. These SAT criteria resulted in the selection of J48 206 2,118 of 2,582 students who, as 7th or 8th graders, scored as well J48 207 as the average high school female; the criteria also provided a J48 208 wide range of talent to study. SAT scores had been grade adjusted J48 209 (7th-grade scores had been adjusted upward to be comparable to J48 210 8th-grade scores, with the procedure outlined in Angoff, 1971). J48 211 Mean SAT scores at age 13 were as follows for male students, 556 J48 212 (SD = 73) on SAT-M and 436 (SD = 85) on SAT-V, and for J48 213 female students, 519 (SD = 59) on SAT-M and 462 (SD = 88) J48 214 for SAT-V. J48 215 J49 1 <#FROWN:J49\>CULTURE CLASH: A MODEL IN ACTION AMONG J49 2 HAWAIIAN-AMERICAN CHILDREN

J49 3 Children of color may begin the schooling process having been J49 4 socialized in a way which may be in conflict with the expectations J49 5 of the school; when this occurs, children and teachers may fail due J49 6 to the cultural incompatibility between the culture of J49 7 the school and the culture of the child. In this section, efforts J49 8 to remedy the clash between the culture of the home and the culture J49 9 of the school among one particular American minority group - J49 10 Hawaiian-Americans - will be reviewed.

J49 11 The cultural incompatibility approach has been the basis of J49 12 considerable work at the Center for Development of Early Education J49 13 (formerly the Kamehameha Elementary Education Program, KEEP), a J49 14 privately funded, multidisciplinary, educational research and J49 15 development program directed at remedying academic underachievement J49 16 of native Hawaiians. As with many other ethnic minorities in J49 17 American schools, poor school performance among Hawaiians was at J49 18 first attributed to a variety of cultural and home deficiencies. J49 19 This cultural deficit model, implying a superior-inferior J49 20 dichotomy, is unfounded, unhelpful, and often rightfully labeled J49 21 racist. All neurologically normal children have already learned a J49 22 substantial amount of relatively complex material that is specific J49 23 to their culture by the time they are of school age. Employing a J49 24 cultural incompatibility model as opposed to a cultural deficit J49 25 model implies that all children can learn prerequisite skills for J49 26 any future need, including school readiness, if given the J49 27 opportunity.

J49 28 Researchers at Kamehameha schools proposed that a school J49 29 environment that was compatible with the child's home culture could J49 30 be developed. This culturally compatible classroom might elicit J49 31 from children those skills, attitudes, and behaviors that would J49 32 contribute to the desired learning and help children achieve early J49 33 school success.

J49 34 While research findings are numerous and complex, some summary J49 35 can be attempted here. To begin with, the Hawaiian socialization J49 36 system teaches children to be contributing members of a family. For J49 37 instance, even when adolescents work outside the home, rather than J49 38 spend their hard-earned money strictly on themselves, they often J49 39 contribute to the overall family resources. the family is not seen J49 40 as a training ground for independence as is typical in many J49 41 dominant culture families. Personal independence is not a goal; J49 42 rather, interdependence is stressed. A collective orientation J49 43 develops, as opposed to the individual orientation prevalent in J49 44 middle-class caucasian society. This has implications for J49 45 motivation and instructional strategies, as well as the reward J49 46 structure in the classroom. For instance, the Hawaiian child may J49 47 not be motivated by individual rewards (gold stars, grades) as much J49 48 as a caucasian counterpart. Nor would a Hawaiian child desire to J49 49 achieve independence from the group.

J49 50 The sibling care system, whereby children from a very young age J49 51 are placed in the care of older siblings, also promotes a high J49 52 degree of interdependence by giving children early experience J49 53 caring for younger children and carrying out many meaningful family J49 54 chores. Adults tend to structure their relationships so they can J49 55 relate to the sibling group as a whole, not to individuals on a J49 56 one-to-one basis. As a result, children do not have as much J49 57 one-to-one verbal interaction with adults. In addition, because J49 58 Hawaiian children learn from peers from an early age, they are J49 59 comfortable in the role of teacher as well as in the role of J49 60 learner.

J49 61 As a result, conditions in typical classrooms may not be J49 62 sufficient to elicit and sustain appropriate learning strategies. J49 63 Sibling care and interdependence may diminish the degree of J49 64 authority alloted to any one adult. Peer orientation and J49 65 affiliation, while frowned upon in the typical classroom, has been J49 66 found to contribute to school success of Hawaiian children. J49 67 Learning stations which consider this orientation facilitate J49 68 learning. Reading instruction modeled after the culturally familiar J49 69 'talk-story' activity improves reading skill and comprehension. J49 70 Modification of instructional practice, classroom organization, and J49 71 motivation management that takes into consideration the culture of J49 72 the child has been found to make a significant difference in the J49 73 achievement of Hawaiian children in school.

J49 74 Figure 5.1 illustrates some aspects of mainstream culture which J49 75 are congruent with the culture of the school but which may be in J49 76 significant conflict with the cultural knowledge and attitudes of J49 77 Hawaiian-American students.

J49 78 Analysis of the classroom experiences of other minority J49 79 children confirms the usefulness of the cultural incompatibility J49 80 hypothesis. The KEEP model, for example, has been applied among the J49 81 Navajo at Rough Rock. While not directly transferable, there is J49 82 every indication that culture-specific modification of the program J49 83 is possible. Efforts such as KEEP should be applauded. Even if not J49 84 directly transferable to other contexts, the implications and J49 85 motivations behind such work can be applied, especially where there J49 86 is a large population of a single minority group in the schools.

J49 87 SOCIAL CLASS

J49 88 We have said that differences in school achievement may be J49 89 attributed to the cultural influences of race, ethnicity, and J49 90 gender on learning style. Other aspects of human diversity may also J49 91 be critical in determining school success. Such factors include J49 92 motivation, aptitude and achievement, self-concept, peer pressure, J49 93 family, health, teacher expectations, and socioeconomic status. It J49 94 is to issues of socioeconomic status and its influence on school J49 95 achievement that we now turn.

J49 96 Most Americans believe they live in a classless, rather J49 97 egalitarian society. At the least, American ideology promotes the J49 98 idea that, through proper attention and diligence (and some luck, J49 99 which Americans also believe in), an individual may 'rise above' J49 100 his or her social class. Part of what has been called an 'American J49 101 religion,' this faith in the reality of upward mobility may account J49 102 for the relative lack of attention to the concept of social class J49 103 in the educational and psychological literature in the United J49 104 States. Certainly it accounts for the difficulty sociology J49 105 professors encounter in helping young people understand the bases J49 106 of class differences in this society. Nevertheless, as we all know, J49 107 there are significant variations in standard of living, status of J49 108 occupation, and extent of expectations of upward mobility among J49 109 American citizens.

J49 110 Social class has been defined in a number of ways. J49 111 all of which refer to a hierarchical stratification, or 'layering,' J49 112 of people in social groups, communities, and societies. Assignment J49 113 to social class categories is one of a number of stratification J49 114 systems that can be used to distinguish one individual or group J49 115 from another in such a way as to assign 'worth.' The urge to J49 116 organize people in layers almost appears to be a culture-general J49 117 characteristic; indeed, it has been said that whenever more than J49 118 three people are in a group there will be stratification. While J49 119 many Americans would identify class membership in terms of income, J49 120 it is important to understand that money alone does not determine J49 121 one's social class. Rather, one's social class standing depends on J49 122 a combination of prestige, power, influence, and income.

J49 123 figure&caption

J49 124 Traditional class markers in the United States thus include J49 125 family income, prestige of one's father's occupation, prestige of J49 126 one's neighborhood, the power one has to achieve one's ends in J49 127 times of conflict, and the level of schooling achieved by the J49 128 family head. In other cultures, markers of social class may include J49 129 bloodline and status of the family name, the caste into which one J49 130 was born, the degree to which one engages in physical labor, and J49 131 the amount of time which one might devote to scholarly or leisurely J49 132 activities of one's choosing.

J49 133 For purposes of analysis, it is often helpful to divide J49 134 American society into five social classes. At the top there is a J49 135 very small upper class, or social elite, consisting chiefly of J49 136 those who have inherited social privilege from others. Second is a J49 137 larger upper middle class, whose members often are professionals, J49 138 corporate managers, leading scientists, and the like. This group J49 139 usually has benefited from extensive higher education, and while J49 140 family history is not so important, manners, tastes, and patterns J49 141 of behavior are.

J49 142 The third (or middle) social class has been called the lower J49 143 middle class. Members of this group are largely people employed in J49 144 white-collar occupations earning middle incomes - small business J49 145 owners, teachers, social workers, nurses, sales and clerical J49 146 workers, bank tellers, and so forth. This is the largest of the J49 147 social classes in the United States and encompasses a wide range of J49 148 occupations and income. Central to the values of the lower middle J49 149 class are a "desire to belong and be respectable .... J49 150 [f]riendliness and openness are values and attention is paid to J49 151 keeping up appearances."

J49 152 Fourth in the hierarchy of social class is the working class, J49 153 whose members are largely blue-collar workers (industrial wage J49 154 earners), or employees in low-paid service occupations. J49 155 Working-class families often have to struggle with poor job J49 156 security, limited fringe benefits, longer hours of work, and more J49 157 dangerous or 'dirtier' work than those in the classes above them. J49 158 It is not surprising, then, that members of the working class often J49 159 feel more alienated from the social mainstream.

J49 160 Finally, fifth in the hierarchy is the lower class - the J49 161 so-called working poor and those who belong to what has been termed J49 162 the underclass - a designation that refers to people who have J49 163 been in poverty for so long that they seem to be unable to take any J49 164 advantage at all of mobility options and thus lie 'below' the class J49 165 system. Clearly, poverty is both the chief characteristic and the J49 166 chief problem of this group. Webb and Sherman point out that this J49 167 simple fact needs to be underscored:

J49 168 Being poor means, above all else, lacking money. This J49 169 statement would be too obvious to mention were it not for the fact J49 170 that most Americans see poverty in other terms. Middle-class J49 171 conversations about the poor often depict them as lazy, J49 172 promiscuous, and criminal. Misconceptions about the poor are so J49 173 widespread that it is difficult to appreciate fully what life is J49 174 like at the lowest stratum of society.

J49 175 Complicating the issues of social class is the fact that in the J49 176 United States there is a large overlap between lower-middle-class, J49 177 working-class, and lower-class membership and membership in J49 178 minority groups. African-Americans, Hispanics, Native Americans J49 179 (including the Inuit and native Hawaiians) are the most J49 180 economically depressed of all groups in the United States. These J49 181 groups also have the highest school dropout rates. To the extent J49 182 that social class status depends on income and occupation (and J49 183 therefore, usually, prestige and power), women and children of all J49 184 racial, ethnic, and religious groups constitute a large proportion J49 185 of the lower classes. This is, in part, a consequence of the J49 186 descent into poverty that characterizes the lives of women who are J49 187 divorced and their children. At the present time, nearly 25 percent J49 188 of all American children under 6 years old are members of J49 189 households trying to exist below the poverty line.

J49 190 The working poor - those people who do work but in jobs that J49 191 are minimum wage or slightly above, with no benefits, and hardly J49 192 any job security - must also struggle to make it in today's J49 193 society. To reach a middle-class lifestyle, a family of four in J49 194 1987 needed an annual income of about $31,000, and inflation will J49 195 continue to raise this figure. In many cases, to reach this level J49 196 both husband and wife must work. Only 25 percent of men and women J49 197 reach this level if only one partner in the marriage earns an J49 198 income. Thurow states that

J49 199 although the dominant pattern today is a full-time male J49 200 worker and a part-time female worker, the pattern is rapidly J49 201 shifting toward a way of life in which both husband and wife work J49 202 full time .... As an increasing number of families have two J49 203 full-time workers, the households that do not will fall farther and J49 204 farther behind economically.

J49 205 Brislin comments on the effect this reality has on women and J49 206 children:

J49 207 The people left behind in the movement through social J49 208 class levels include households headed by women, and these reached J49 209 a staggering 31% of households in 1985. Dependence on one income, J49 210 combined with the well-known fact of lower salaries earned by J49 211 women, can result in poverty. Women and children constitute 77% of J49 212 people living in poverty, and 50% of these poor people live in J49 213 female-headed households with no husband present.

J49 214 Those of similar socioeconomic status, at whatever level, also J49 215 share similar cultural knowledge, attitudes, and values. J49 216 J50 1 <#FROWN:J50\>TEACHING SOMEONE TO THINK

J50 2 The most important concept in educational theory is the concept J50 3 of thinking. The concept has been clearly understood for many J50 4 years, thanks to the insights of philosopher-educators like John J50 5 Dewey and Alfred North Whitehead, but it has yet to be grasped in J50 6 the teaching/learning processes of American schools.

J50 7 Thinking, says Dewey, is what gives meaning to experience. J50 8 Through thinking we apprehend a connection in experience that J50 9 enables us to act intentionally. Thinking is the basis of J50 10 responsible action, as contrasted with capricious behavior (which J50 11 accepts things as they happen to fall out) or with routinized J50 12 behavior (which accepts things as they have always been). Education J50 13 aims at enhancing the powers of thinking in the broad sense of the J50 14 acceptance of responsibility for action.

J50 15 Dewey believes that all thinking is problem-solving. All J50 16 thinking is instrumental. The starting point of any process of J50 17 thinking is something going on, something that just as it stands is J50 18 incomplete, unsatisfactory, or unfulfilled. Its point, its meaning, J50 19 lies literally in what it is going to be, in how things are going J50 20 to turn out. To fill one's head with facts about what is going on, J50 21 says Dewey, is not to think but to function as a registering J50 22 apparatus. Thinking means considering the bearing of what is going J50 23 on upon what may be, but is not yet. It is applying what is known J50 24 about the structure of experience to the task of projecting a plan J50 25 of action for an unknown that lies beyond experience.

J50 26 The function of a good teacher, as Stanford Erickson says, is J50 27 to give voice to knowledge linking the present to the past and the J50 28 future. Dynamic teaching is not a matter of developing innovative J50 29 techniques for telling things to students. Dynamic teaching is J50 30 teaching that engages the student's thinking in the imaginative J50 31 consideration of learning. The moment of instructional truth, as J50 32 Erickson says, occurs "when the student grasps the meaning J50 33 of an important idea." Everything else in schooling is J50 34 nothing more than preparation for learning.

J50 35 The fundamental aim of all education is the enhancement of J50 36 thinking. The cultivation of skill obtained apart from thinking has J50 37 nothing to do with education; and information detached from J50 38 thoughtful action is dead (even though it may resemble knowledge) J50 39 and is a mind-crushing load. Education is the attempt to mobilize J50 40 the imagination of individuals in the activity of thinking. J50 41 "The sole direct path to enduring improvement in the J50 42 methods of instruction and learning consists," as Dewey J50 43 declares, "in centering upon the conditions which exact, J50 44 promote, and test thinking."

J50 45 For Dewey, the requirements of a dynamic educational method are J50 46 (1) that the student has a genuine situation of experience that J50 47 proposes a problem in which he is interested for its own sake; (2) J50 48 that this problem should develop within the learning situation as a J50 49 stimulus to his thought; (3) that he should have access to J50 50 information bearing on the problem, its causes and its solution; J50 51 (4) that he should be responsible for coming up with alternative J50 52 solutions to the problem and for exploring them in an orderly way; J50 53 and (5) that he should have opportunity and occasion to test his J50 54 ideas by application, to make their meaning clear, and to discern J50 55 for himself their validity.

J50 56 Human learning, as Dewey always insists, begins with J50 57 recognition of a situation that calls for some sort of doing. The J50 58 starting point of learning is a problem. Beginning with a problem J50 59 posed by experience, the student must then be helped to gain J50 60 command of data that will help him to deal with the difficulty or J50 61 the challenge that has been presented. To think effectively one J50 62 must be able to command information that bears on the problem at J50 63 hand.

J50 64 The assimilation of data in relation to an experienced problem J50 65 or concern generates suppositions, tentative explanations, and J50 66 interpretations - what Dewey calls ideas. Data are facts; the ideas J50 67 that spring from them forecast possible results. Thinking is an act J50 68 of inference that always involves an invasion of the unknown, a J50 69 leap out beyond what is known. Thinking is an incursion into the J50 70 novel and demands some measure of inventiveness.

J50 71 Ideas, as Dewey uses the term in this context, are always J50 72 generated out of the originality of the thinker. All ordinary J50 73 thinking is creative. Ideas in Dewey's sense are anticipations of J50 74 possible solutions to problems, of possible connections between J50 75 action and desired consequences. Ideas are not the final end of J50 76 learning; they are intermediate goals, significant only insofar as J50 77 they guide and organize future experience and action. The testing J50 78 of an idea, of course, involves not only assessing the adequacy of J50 79 the data that support it, but also acting upon it and seeing what J50 80 results.

J50 81 Dewey's understanding of the relationship between education and J50 82 thinking is closely akin to Whitehead's celebrated definition of J50 83 education as "the acquisition of the art of the utilization J50 84 of knowledge." Whitehead advances a passionate protest J50 85 against the notion of education as the accumulation of inert ideas, J50 86 as the transmission of scraps of information accumulated by the J50 87 learned professions. Education at every stage, he says, must permit J50 88 each individual student to experience the joy of discovery - must J50 89 permit each student, that is, to discover that ideas can provide an J50 90 understanding of the stream of events that is Life and can J50 91 establish a basis for practical decision. A merely well-informed J50 92 person is the most useless bore on God's earth. But an education J50 93 that is merely pedantry is not only useless; it is positively J50 94 harmful, for a mind loaded with the dead weight of routinized J50 95 learning is unable to think.

J50 96 Imagination is the capacity to think about the relevance of J50 97 learning for life. The "problem of problems" in J50 98 education, as Whitehead sees it, is the provision of a corps of J50 99 teachers whose learning is lighted up with imagination. A school J50 100 must be a place in which the adventure of thought meets the J50 101 adventure of action. Education is learning for life. And what is J50 102 needed for life, as Whitehead says, are ideas that provide the J50 103 basis for determining through foresight what actions are J50 104 appropriate. Education is "a preparation by which to J50 105 qualify each immediate moment with relevant ideas and appropriate J50 106 ideas."

J50 107 The basic virtue of the university, in Whitehead's view, is the J50 108 power of imagination. The primary reason for the existence of a J50 109 university is not to be found in the knowledge that it succeeds in J50 110 imparting to its students or in the scientific advances achieved J50 111 through its research. The justification for a university is that it J50 112 preserves "the connection between knowledge and the zest J50 113 for life," by uniting the young and the old in "the J50 114 imaginative consideration of learning."

J50 115 The imaginative consideration of learning, as Whitehead argues, J50 116 requires a dialectical combination of freedom and discipline. The J50 117 rhythm of education involves a series of stages in the growth of J50 118 educated imagination: (1) the awakening of interest and the general J50 119 apprehension of a subject in its vaguest details; (2) the J50 120 acquisition of specific knowledge and mastery of the relevant J50 121 details through the pursuit of an objective method; (3) the J50 122 comprehensive ordering of the subject as a whole in the light of J50 123 all relevant knowledge; (4) the understanding of the general J50 124 principle for purposes of creative application in novel situations. J50 125 The aimless acquisition through education of precise knowledge, J50 126 inert and unutilized, only paralyzes thought. The wisdom that J50 127 education seeks is the habit of the active utilization of well J50 128 understood principles.

J50 129 No area of education has more to gain from attention to the J50 130 rhythmic law of human development, as Whitehead says, than moral J50 131 and religious education. The principle of progress in moral and J50 132 religious growth is from within. The teacher of moral and religious J50 133 values has an important but limited function. He can elicit an J50 134 awakening of concern in moral issues by resonance from his own J50 135 personality and character, and he can create an environment that is J50 136 conducive to the nurture of a higher wisdom and a firmer purpose. J50 137 But the ultimate motive power for learning (whether in science, in J50 138 morality, or in religion) comes from the learner. It is the sense J50 139 of value, wonder, and reverence, the eagerness to merge one's J50 140 personality in something beyond itself. Without this sense of J50 141 value, as Whitehead says, education is a great emptiness and life J50 142 sinks back into the passivity of its lower types.

J50 143 EDUCATION AND VALUATION

J50 144 In the course of all complex human learning, questions of value J50 145 will inevitably arise. Yet the method by which intelligence J50 146 addresses problems of values is not fundamentally different from J50 147 the method by which it addresses questions of fact. Questions of J50 148 values arise in all thinking, since thinking deals with problems J50 149 arising out of human concerns and seeks a course of purposive J50 150 action that will bring about a desired circumstance. Unless it J50 151 addresses the values-related questions, education, like science J50 152 itself, is simply irrelevant.

J50 153 Valuation is a phase of the process of thinking that concerns J50 154 the formation of strategies for solving problems. Informed thinking J50 155 about alternative courses of action leads to appraisals of which J50 156 action is better or worse, more or less serviceable/feasible. Such J50 157 appraisals require an experimental justification of the same sort J50 158 as is involved in all scientific generalization. Valuations, in J50 159 other words, are concerned with what Dewey calls "rules for J50 160 the use, in and by human activity, of scientific generalizations as J50 161 means for accomplishing certain desired and intended ends." J50 162 Moral education depends on a theory of valuation that advances a J50 163 method of reaching well-founded judgments of value to guide J50 164 the intelligent conduct of human activities, both personal and J50 165 social, in the solution of human problems.

J50 166 An instrumentalist understanding of thinking opens new avenues J50 167 for the systematic application of disciplinary scholarship in the J50 168 moral domain. Of course, as Dewey himself granted, it does not J50 169 immediately resolve all the questions that philosophers and other J50 170 scholars might wish to raise. Instrumentalism remains debatable as J50 171 an epistemological or meta-ethical theory. Yet in the context of J50 172 our struggle for better answers to the problems of life, Dewey's J50 173 analysis of the way we think identifies an approach to J50 174 values-related questions that is by no means foreign to the J50 175 academy. It is, indeed, the method by which all human understanding J50 176 progresses 'from sounds to things,' from appearances to reality, J50 177 from illusion to truth.

J50 178 Thinking is an ongoing process. It is never finished, and its J50 179 conclusions are never final. Thinking is the search for greater J50 180 coherence in our expanding experience, but it is always limited by J50 181 the particular context in which it functions at any given moment. J50 182 Thinking is always contextual; so, too, is all thinking about J50 183 ethical issues. Every ethical problem arises in a particular J50 184 context of experience; and all thinking about ethical issues J50 185 proceeds by stages as the mind seeks new data and perspectives for J50 186 forming more coherent and comprehensive understandings. Moral J50 187 thinking does not deal with ultimate or intrinsic values any more J50 188 than scientific thinking deals with absolute truth. The only truth J50 189 to be found in either domain is the increasing coherence of our J50 190 expanding experience.

J50 191 A pragmatic program of moral education might well be guided by J50 192 the assumption of ethical contextualism, which views moral J50 193 reasoning as always taking place within a specific context of moral J50 194 decision. All ethical argumentation depends on a set of premises in J50 195 which at least one ethical premise is included. Contextualism is J50 196 the claim that in every context of ethical discussion we have J50 197 available some moral principles that set the problem. If these J50 198 principles are themselves challenged, we simply move to another J50 199 context of investigation, where we shall need to identify another J50 200 principle or set of principles with which to carry on the new moral J50 201 inquiry.

J50 202 Ethical contextualism is based on the understanding that there J50 203 is no working ethical premise that cannot be inferred by a process J50 204 of logical inference from other ethical premises in combination J50 205 with factual information. Such a view is analogous to the J50 206 philosophy of causality, which assumes that there is no cause of a J50 207 phenomenon that is not itself the effect of some other cause. For J50 208 ethical contextualism the logic of evaluation is coordinate with J50 209 the logic of science. J50 210 J51 1 <#FROWN:J51\>INTRODUCTION

J51 2 The Grammar of Argumentation

J51 3 It is generally regarded as mandatory for a textbook in logic J51 4 to begin with a definition of logic. With some misgivings I shall J51 5 attempt to comply with that convention. However I think that more J51 6 than a simple definition is required. The concept of logic has J51 7 undergone numerous changes since the Greeks first took up the study J51 8 in the fourth or fifth century B.C. Since then the term has taken J51 9 on new meanings which it was not originally intended to carry, J51 10 without, however, entirely loosing its original sense. The result J51 11 is that nowadays the word "logic" does not identify a single J51 12 enterprise so much as a confusing conglomeration of enterprises J51 13 related primarily by historical accident. By way of introduction, I J51 14 shall attempt to clarify the concept of logic, and introduce at J51 15 least some of the basic vocabulary of syllogistic logic:

J51 16 History of the Concept of Logic

J51 17 The people who conceived and developed syllogistic logic had J51 18 quite different views on what they were doing than the people who J51 19 call themselves logicians today. In taking up the study of J51 20 syllogistic logic - a form of logic with deep historical roots - it J51 21 may therefore be advisable to begin by acquiring some sense of how J51 22 the word 'logic' has changed its meaning over the centuries.

J51 23 The Greek word 'logos' from which the J51 24 word logic is derived, means (among other things) an 'account', in J51 25 the sense in which a naughty child might be asked to give an J51 26 account of his actions. Thus when Plato defines knowledge in the J51 27 Meno as 'true belief with a logos', he J51 28 means that a belief can properly be called knowledge only if (a) it J51 29 is true, and (b) the person holding the belief can give an adequate J51 30 reason, or an adequate account of why he holds that belief.

J51 31 Of course any account which is adequate to justify one person's J51 32 beliefs should also be adequate to convince someone else to adopt J51 33 the same beliefs. To act rationally or logically means to act in a J51 34 way that can be understood and justified. Hence, since logic was J51 35 concerned with the nature of an adequate account, the Greeks J51 36 understood logic to be concerned with the nature of persuasion, J51 37 refutation, and inquiry.

J51 38 But what does it mean to say that an account is 'adequate?' J51 39 Plato thought that one should distinguish between legitimate J51 40 persuasion and refutation, such as a genuine philosopher employs J51 41 when conducting an inquiry into the truth, and illegitimate J51 42 persuasion and refutation, such as a sophist employs in trying to J51 43 sway public opinion. Granted that people justify their beliefs and J51 44 actions by telling a story or giving an account, what sort of J51 45 account qualifies as a legitimate and proper justification and what J51 46 sort does not? It was this question which, I believe, logicians J51 47 originally set out to answer. Hence the original conception of J51 48 logic seems to have been that logic was supposed to give a J51 49 legitimate account of the way in which we legitimately account for J51 50 our beliefs and actions.

J51 51 Plato, however, offers no criteria for drawing the distinction J51 52 between legitimate and illegitimate accounts, other than to point J51 53 out that a legitimate account is easier to remember, since the same J51 54 line of thought, if it is genuinely rational, can always be J51 55 reconstructed. The only method which Plato offers for telling the J51 56 difference between legitimate and illegitimate accounts is J51 57 self-honest evaluation based upon a commitment to the truth. A J51 58 person who is being honest with himself, and is genuinely concerned J51 59 to discover the truth, will simply be able to tell the J51 60 difference.

J51 61 The problem with this account (as Plato was well aware) is that J51 62 it gives us no recourse against someone who stubbornly persists in J51 63 being irrational. In the Gorgias Socrates falls into a J51 64 discussion with such a person, and the dialogue degenerates into a J51 65 shouting match. Socrates accuses Callicles of dishonesty - of being J51 66 in disagreement even with himself - but without being able to give J51 67 a better account of logic, he has no way to make the accusation J51 68 stick. Logic may begin with self-honesty and a commitment to J51 69 the truth, but it cannot end there.

J51 70 The earliest surviving essays which attempt a systematic study J51 71 of logic are a group of lectures by Aristotle. Aristotle did not J51 72 necessarily intend these lectures to be treated as a single work, J51 73 nor did he even necessarily regard them all as essays on logic. In J51 74 fact, they cover a wide range of topics and were written at widely J51 75 different periods of his life. The Alexandrian editor, Andronicus J51 76 of Rhodes, nevertheless grouped these lectures together under the J51 77 title Organon, which means 'organ' or 'tool', and placed them J51 78 at the beginning of Aristotle's works. The implication is that J51 79 these essays were not themselves lectures on philosophy, but were J51 80 rather an explanation of the tools with which philosophers work. As J51 81 such they should be regarded as preliminary to the study of J51 82 philosophy. The subjects treated in Aristotle's lectures can be J51 83 summarized as follows:

J51 84 (1) In the Categories, a discussion of the various J51 85 types of terms, the classifications into which terms can be J51 86 divided, and the distinctive features which terms in each category J51 87 display.

J51 88 (2) In the Hermeneia, a discussion of propositions, J51 89 how they are put together and how they are related to each J51 90 other.

J51 91 (3) In the Prior Analytics, a discussion of the J51 92 formal patterns which arguments may take. It is this work in which J51 93 syllogistic logic first appears.

J51 94 (4) In the Posterior Analytics and the Topics, a J51 95 discussion of the practical application of argumentation and proof J51 96 to the acquisition of knowledge.

J51 97 (5) In the Sophistical Refutations, a discussion of J51 98 common errors (or fallacies) made in argumentation.

J51 99 Medieval logicians learned their dicipline chiefly by studying J51 100 Aristotle's Organon, so it is not surprising that they J51 101 considered logic to include the same subjects covered by J51 102 Aristotle's Organon. The medieval logicians made some J51 103 significant changes in the syllogistic logic, gave names to each of J51 104 the valid argument forms, and put the system into more or less its J51 105 modern form. They also wrote commentaries on the other books in J51 106 Aristotle's Organon, especially the Categories.

J51 107 But it was not entirely clear why logicians, concerned with the J51 108 question of legitimacy in methods of persuasion and refutation, J51 109 should take time to examine types of terms. Some conception of J51 110 logic was needed to give a sense of unity to these diverse matters. J51 111 The medieval response to this problem was to point out that all the J51 112 topics in Aristotle's Organon are concerned with 'second J51 113 intentions'. The word 'intention' may be taken as roughly J51 114 synonymous with the word 'meaning'. Let us call anything which has J51 115 an intention or meaning a 'sign'. Words, for example, are signs, J51 116 since there is some object which any word 'intends', though this J51 117 object need not actually exist. (The word 'unicorn', for example, J51 118 has a meaning or intention, since it refers to a type of being, J51 119 even though the being to which it refers does not actually exist.) J51 120 However, words are not the only signs. Thoughts are also signs, J51 121 since thoughts, like words, 'intend' some object, i.e. they have a J51 122 meaning. (To prove this to yourself, try to imagine what it would J51 123 be like to have a meaningless thought. Wouldn't that be the same as J51 124 not having a thought at all?)

J51 125 According to the medieval logicians, signs could be divided J51 126 into two groups:

J51 127 (1) those whose object is a thing or relationship that (if it J51 128 exists at all) exists in nature. Examples would include 'green', J51 129 'tree', 'unicorn', etc. These were called signs of first J51 130 intention.

J51 131 (2) those whose object is another sign. Examples would include J51 132 'noun', 'verb', 'subject', 'predicate', etc. These were called J51 133 signs of second intention.

J51 134 Logic, they thought, was the science which studied signs of the J51 135 second kind rather than signs of the first kind. Hence logic could J51 136 be defined as the 'science of second intentions'. By this J51 137 definition, of course, logic incorporates all of the linguistic J51 138 sciences, including what we would now call grammar and J51 139 semantics.

J51 140 This definition gave a sort of unity to the list of subjects J51 141 which were thought to fall within the scope of logic, but it also J51 142 caused some of the subjects to be pushed into the forefront, while J51 143 others took on a merely peripheral importance. For example, the J51 144 discussion of types of terms became particularly important, since J51 145 the purpose of the discussion was not to classify types of things, J51 146 but rather to classify what we could sensibly say about J51 147 things. It was, in other words, an attempt to give some order to J51 148 signs themselves; it was not an attempt to give order to the J51 149 objects which the signs intended. Hence the categorization of terms J51 150 fit neatly within the medieval definition of logic. The discussion J51 151 of sophistical reasoning, which would have been regarded as central J51 152 to logic according to the Greek conception, still had its place in J51 153 medieval logic but it no longer occupied a position of overweaning J51 154 importance.

J51 155 Thus while the Greeks thought that logic should be primarily J51 156 concerned with persuasion, refutation, inquiry, and the J51 157 justification of beliefs, the medieval logicians extended the J51 158 notion, so that logic was thought to be concerned with the nature J51 159 of signs in general.

J51 160 It is important to understand that the notions of 'sign' and J51 161 'intention' are very broad. They include, not only words and J51 162 language, but also thought itself. Because of this, it was easy to J51 163 make the transition from thinking of logic as primarily concerned J51 164 with language (especially language used to persuade), to thinking J51 165 of logic as primarily concerned with thought. The philosophers of J51 166 the early modern period came to regard logic as concerned with 'the J51 167 laws of thought', and this frequently meant that they regarded J51 168 logic as a sort of psychology. Logic was thought to study the J51 169 operations of the mind, specifically, of course, the operations J51 170 associated with rationality. Certain operations of the mind were J51 171 thought to be more fundamental than others; so fundamental, in J51 172 fact, that it would be contrary to human nature, or outside the J51 173 scope of human capacity, to reason in any other way. This view of J51 174 logic was held by philosophers as diverse as Leibniz, Locke, and J51 175 Kant. Even Hegel, when writing on logic, seems to have in mind a J51 176 thorough-going analysis of the structures of the conscious mind, J51 177 though he is careful to make clear that he means any J51 178 conceivable conscious mind, not simply the human mind. Hence, J51 179 Hegel does not regard logic as a type of psychology but rather as J51 180 related to (though not identical with) phenomenology.

J51 181 During the 19th century a revolution occurred in the methods J51 182 employed by logicians. George Boole was able to demonstrate that J51 183 simple mathematical operations, such as multiplication and J51 184 subtraction, could be made to parallel the familiar laws of logic. J51 185 This meant that logicians were able to use quasi-mathematical J51 186 formulas to represent or model patterns of argumentation. This new J51 187 technique was incredibly powerful, and gave logicians the tools J51 188 with which they could analyze more and more complex arguments. J51 189 However, most logicians still considered logic to be, at root, the J51 190 study of the laws of thought. They simply assumed that these laws J51 191 could be expressed mathematically. Hence mathematics changed the J51 192 methodology of logic, but not (at first) its essential subject J51 193 matter.

J51 194 However, the notion that there are such things as laws of J51 195 thought, and that it is the business of logic to study them, began J51 196 to break down with the advent of non-Euclidean geometry, and with J51 197 the many other sudden advances in mathematical theory also made J51 198 during the 19th century. It would be false to claim that the J51 199 non-Euclidean geometries destroyed the concept of 'laws of thought' J51 200 by proving that humans could conceptualize on a much grander scale J51 201 than had previously been supposed. In fact they proved no such J51 202 thing. Rather, the non-Euclidean geometries proved that even J51 203 apparently absurd presuppositions could be developed into J51 204 consistent (and perhaps even useful) systems. Hence discussion J51 205 among mathematicians concerning what was absurd or inconceivable J51 206 was supplanted by discussion concerning what could and could not be J51 207 consistently systematized. Mathematicians could no longer regard J51 208 mathematics as founded upon a priori truths, or as an J51 209 attempt to elaborate some particular system of well understood J51 210 formal relations. J51 211 J52 1 <#FROWN:J52\>Speaking of the poetic writers of this period, Harold J52 2 Bloom has argued that they characteristically exhibit a compulsion J52 3 toward priority in intellectual and artistic creation and a fear of J52 4 being regarded as derivative. Bloom terms this state of mind J52 5 "the anxiety of influence." In his study with this J52 6 same title, Bloom observes that "all quest-romances of the J52 7 post-Enlightenment, meaning all Romanticisms whatsoever, are quests J52 8 to re-beget one's own self, to become one's own Great J52 9 Original." Surprisingly, in this study Bloom repeatedly J52 10 draws on Kierkegaard's statements about artistic and intellectual J52 11 creativity to illustrate his own points about the poet's efforts at J52 12 self-creation, the poet's demand for priority and for J52 13 freedom from the influence of his precursors.

J52 14 Although Bloom is entirely unaware of Kierkegaard's J52 15 relationship to Kant, it is not accidental that he draws on J52 16 Kierkegaard's preoccupation with originality of authorship in J52 17 describing this "anxiety of influence." As J52 18 Kierkegaard reminds us from virtually the first to the last words J52 19 of his published work, he was always a poet, and a poet writing and J52 20 thinking in an era when originality - the secure possession of J52 21 one's own daimon, one's genius - was the mark of greatness. J52 22 Even a cursory reading of Kierkegaard's Papers tell us how J52 23 much he was concerned with the singularity of his J52 24 religious-literary effort. From Either/Or and Fear and J52 25 Trembling onward the poetic Kierkegaard always remained J52 26 concerned with his place in literary and intellectual history. J52 27 Furthermore, as Christoph Schrempf reminds us in his exhaustive J52 28 biography, Kierkegaard's strong sense of sacrificial destiny - his J52 29 wish to offer up his life as the vehicle for a unique and J52 30 redemptive idea - exhibits a turn of mind characteristic of many J52 31 other 'poetic' writers of this period.

J52 32 In view of this, we might suppose that Kierkegaard also J52 33 suffered from "the anxiety of influence," that he, J52 34 too, sought to cut the ties to his predecessors and, above all, to J52 35 that predecessor on whose originality and genius his own novel J52 36 creation so depended. In the course of his discussion, Bloom J52 37 repeatedly quotes a remark by Kierkegaard that Bloom presents as a J52 38 virtual synopsis of the poet's effort to creatively appropriate the J52 39 work of his predecessor: "He who is willing to work gives J52 40 birth to his own father." If the pattern of borrowing we've J52 41 seen is indicative, Kant is, in a sense, Kierkegaard's intellectual J52 42 'father.' Understood in terms of Bloom's analysis, therefore, J52 43 Kierkegaard's authorship becomes an impassioned effort to J52 44 're-create' Kant's philosophy in a way that makes it fully the J52 45 product of Kierkegaard's own creative genius. The absence of any J52 46 tradition of scholarship relating Kierkegaard to Kant and the J52 47 difficulty many Kierkegaard biographers have had in tracing the J52 48 lines of his descent show how well Kierkegaard was able to obscure J52 49 his own intellectual paternity.

J52 50 This first explanation of Kierkegaard's deliberate effort to J52 51 erase the lines leading back to Kant assumes a measure of J52 52 concealment and intellectual ambition. A second explanation moves J52 53 in a different direction and relates Kierkegaard's handling of Kant J52 54 to the central intellectual concerns of his authorship and to his J52 55 ongoing employment of irony as a philosophical tool. Simply stated, J52 56 it sees Kierkegaard's covert use of Kant as a subtle, necessary, J52 57 and deserved trick played by Kierkegaard on his arrogant Hegelian J52 58 foes.

J52 59 To understand this second effort at explanation, it helps to J52 60 keep in mind Louis Mackey's point that Kierkegaard faced a daunting J52 61 task in taking on Hegelianism from an orthodox Christian-religious J52 62 point of view. The challenge before Kierkegaard, Mackey observes, J52 63 was to call into question a philosophy that regarded itself as the J52 64 dialectical fulfillment of human thought and as thus able to J52 65 comprehend all possible philosophical and religious positions. To J52 66 accomplish this seemingly impossible task, Kierkegaard chose to J52 67 avoid philosophical argumentation and to employ the method of J52 68 Socratic irony, arraying its "infinite negativity" J52 69 against Hegelian pretensions.

J52 70 I would deepen Mackey's observations by adding that part of J52 71 this ironic strategy for Kierkegaard may have involved a decision J52 72 to employ Kantian philosophy in his struggle against the Hegelians: J52 73 to use a thinker who had been 'transcended' (aufgehoben) J52 74 and fully assimilated into the dialectic against the discipline's J52 75 now reigning giant. Philosophy could thus be turned against its own J52 76 methodological and substantive pretensions. To prevent this from J52 77 becoming just another page in this history of philosophical J52 78 debates, however, and to avoid the Hegelians' premature dismissal J52 79 of his position, Kierkegaard would have had to conceal Kant's J52 80 presence in his own reformulated statements of Christianity. In J52 81 this way Kierkegaard could appropriate and use Kant's brilliant J52 82 destruction of the tradition of rationalism while ironically J52 83 exposing the hollowness of the Hegelians' claims to have mastered J52 84 all preceding thought.

J52 85 There are hints in Kierkegaard's writings that he was aware of J52 86 the joke he was playing on the smug but philosophically J52 87 less-than-well-trained Danish Hegelians. Kierkegaard knew that J52 88 Martensen and his acolytes, for the very reason that they did not J52 89 take the past seriously, were often ill-versed in the writings of J52 90 philosophers whose work they purported to have transcended. In a J52 91 remark to his brother in 1841 he caustically dismisses possible J52 92 criticisms of his doctoral dissertation by what he calls J52 93 "one or another half-educated Hegelian robber." J52 94 Having endured Martensen's and others' lectures on Kant, and J52 95 knowing how little the Hegelians really understood Kant's profound J52 96 ethics and philosophical theology, Kierkegaard may well have J52 97 enjoyed the one-upmanship involved in surreptitiously turning Kant J52 98 against his teachers.

J52 99 I have already mentioned one possible instance of this kind of J52 100 playfulness on Kierkegaard's part: his handling in the J52 101 Postscript of the matter of the philosophical pedigree of the J52 102 idea of the 'leap.' Although Johannes Climacus, the J52 103 Postscript's pseudonymous author, repeatedly confesses his J52 104 debt to Lessing for this idea, even giving the title 'Attributable J52 105 to Lessing' to the section where he discusses the leap, the section J52 106 itself ends with mention to the fact that Johannes de J52 107 silentio, author of Fear and Trembling, had J52 108 previously discussed a similar idea. Climacus adds that he had read J52 109 Johannes de silentio's book before encountering J52 110 Lessing's essay, and he closes his discussion with the remark J52 111 "Whether Johannes de silentio has had J52 112 his attention called to the leap by reading Lessing, I shall not J52 113 attempt to say." This comment invites the discerning reader J52 114 to ask, who, if not Lessing, might be Climacus's/Kierkegaard's J52 115 primary philosophical source for this idea? Since Fear and J52 116 Trembling centrally addresses Kant's repudiation of the leap J52 117 of faith as this was symbolized for both Kant and Kierkegaard by J52 118 the episode of Genesis 22, there can be little doubt that this J52 119 remark about influence points toward Kant. Almost as though he were J52 120 unwilling to totally obscure his debt to Kant, in other words, J52 121 Kierkegaard gives the informed reader a glimpse into the game he is J52 122 playing with the Hegelians.

J52 123 The Fragments offers another sign of this playfulness. On J52 124 at least six occasions, the author, Johannes Climacus, raises J52 125 questions of scholarly attribution and openly acknowledges the J52 126 unoriginality of his ideas and his susceptibility to the charge of J52 127 plagiarism. In almost all these instances, Climacus is able J52 128 ironically to defend himself against such accusations because the J52 129 intellectual property he is appropriating derives from the very J52 130 public domain of biblical teaching. At the same time, many other J52 131 elements in this volume are borrowed from Kant, possibly including J52 132 at least one idea attributed by Climacus to the Bible. The project J52 133 itself is stimulated, in part, by Kant's position on historical J52 134 revelation in The Conflict of the Faculties, a work we J52 135 might suppose relatively few Hegelians had read. Hence, concealed J52 136 within Climacus's openly confessed plagiarism lies a deeper level J52 137 of unacknowledged borrowing. Kierkegaard's irony and sense of humor J52 138 seem to me to be at work here. Openly signalling the derivative J52 139 quality of this work, he leaves it to the Hegelians to detect his J52 140 employment of Kant against them, while remaining confident that, J52 141 despite their vaunted philosophical erudition, the Danish Hegelians J52 142 will surely fail to see or understand the presence of the J52 143 philosopher they had 'gone beyond.'

J52 144 Was it, then, authorial ambition or a<&|>sic! only half-veiled J52 145 playfulness and irony that led Kierkegaard to obscure his debt to J52 146 Kant? In the complex world of human motives, which Kierkegaard J52 147 himself so masterfully explored, these two explanations may be less J52 148 opposed than they appear. Perhaps both motives were at work, a J52 149 youthful author's passionate quest for originality tempered by J52 150 conscientious self-disclosure and a willingness to reveal his J52 151 philosophical legerdemain to those able to appreciate it.

J52 152 Whatever Kierkegaard's motives, there can be no doubt that Kant J52 153 played a major role in Kierkegaard's intellectual formation. From J52 154 the pattern of borrowing we've seen, I am led to conclude that as J52 155 early as his student years Kierkegaard was deeply intrigued by what J52 156 he had read in Kant's mature works on religion. This interest J52 157 shaped the course of his writing and thinking in the period J52 158 immediately following. We saw that as early as 1835 Kierkegaard J52 159 expressed the wish that he might find "the idea" to J52 160 which he could give his life. Over the course of the next five J52 161 years, reading Kant in preparation for his degree examination, J52 162 Kierkegaard found this idea. Although his best teachers treated J52 163 Kantianism as merely a way station en route to Hegel, Kierkegaard J52 164 perceived that Kant's contributions to religious thought had not J52 165 even begun to be assimilated. If Kant's ethics were taken seriously J52 166 - and despite many criticisms, including Hegel's, Kant's ethical J52 167 writings everywhere commanded respect - then his contentions in the J52 168 Religion also had to be taken seriously. They had to be viewed J52 169 neither as the work of a great philosopher in his dotage (the J52 170 opinion of some philosophers) nor as a wooden and perhaps dangerous J52 171 effort to translate biblical truth into philosophical terms (the J52 172 orthodox view). Instead, they had to be seen as a rigorous J52 173 exposition of ideas that pointed inevitably to their limits and to J52 174 the requirement that they be transcended by faith. Over the next J52 175 five years, during the intensely creative period of the major J52 176 pseudonymous writings, Kierkegaard sought to develop a new form for J52 177 the communication of these ideas, and he tried to carry Kant's J52 178 beginning to the thoroughly religious or "faithful" J52 179 conclusions Kant resisted.

J52 180 A journal entry for 1836-1837 written at the time Kierkegaard J52 181 was beginning his formal study of Kant provides an early hint of J52 182 the importance Kierkegaard attached to Kant's philosophy. J52 183 Kierkegaard here presents Kant, along with Goethe and Holberg, as J52 184 belonging to the "royal procession" of thinkers J52 185 whose work represents a fundamental breakthrough in thought and J52 186 who, because of this, are resisted by their contemporaries. These J52 187 thinkers, Kierkegaard suggests, play a decisive a<&|>sic! role in J52 188 conveying the "password which God whispered in Adam's ear, J52 189 which one generation is supposed to deliver to the next and which J52 190 shall be demanded of them on judgment day." Since J52 191 Kierkegaard's own sense of personal mission was very much in J52 192 formation at this time, it is reasonable to conclude that he J52 193 regarded himself as a recipient, in his generation, of the J52 194 "password" uttered by Kant. Remarkably, in an omission J52 195 characteristic of the whole tradition of Kierkegaard scholarship on J52 196 this matter, the Hongs fail to include Kierkegaard's mention of J52 197 Kant in their otherwise meticulous translation of this entry.

J52 198 Either/Or offers a further hint that carrying on and J52 199 fulfilling Kant's initiative is an accurate depiction of J52 200 Kierkegaard's project in the pseudonymous works. In Chapter 3 we J52 201 saw that the ethical stage of life sketched there by Judge William J52 202 is broadly Kantian in nature. Kierkegaard himself shows his J52 203 sympathy for this position when he states that in every way the J52 204 ethics which Judge William champions "is quite the opposite J52 205 of the Hegelian." Yet Either/Or also seeks to point to J52 206 the possibility of a religious stage of existence where the ethical J52 207 is dialectally transcended and fulfilled. This stage is suggested J52 208 in the position presented by the unnamed country pastor whose view J52 209 is developed in the "Ultimatum" of the book, a view that J52 210 moves toward faith via a recognition of the unavoidability of sin. J52 211 The ethical must be transcended (but not eliminated), the pastor J52 212 tells us, because if ethics alone were to shape our destiny, we J52 213 would all succumb to moral despair.

J52 214 J53 1 <#FROWN:J53\>Wittgenstein on Understanding

J53 2 WARREN GOLDFARB

J53 3 Wittgenstein's treatments in the Philosophical J53 4 Investigations of the cognitive or intentional mental notions J53 5 are evidently meant to persuade us that, in some sense, J53 6 understanding, believing, remembering, thinking, and the like are J53 7 not particular or definite states or processes; or (if this is to J53 8 say anything different) that there are no particular states or J53 9 processes that constitute the understanding, remembering, etc. Such J53 10 a dark point desperately needs clarification, if it is not to deny J53 11 the undeniable. For surely we may (and Wittgenstein does) speak of J53 12 a state of understanding, or of thought-processes; surely J53 13 when one understands - understands a word, a sentence, or the J53 14 principle of a series - one is in a particular state, namely, the J53 15 state of understanding the word, sentence, or principle. Now J53 16 Wittgenstein sometimes puts the point more specifically, speaking J53 17 of "a state of a mental apparatus" (<*_>section<*/> J53 18 149), or, with respect to understanding, of "a state which J53 19 is the source of the correct use" (<*_>section<*/> J53 20 146), as that which he wishes to deny. But clearly these denials, J53 21 though they sound somewhat less paradoxical, are equally in need of J53 22 elaboration.

J53 23 However Wittgenstein's point is clarified, though, there is an J53 24 objection that will inevitably be made, an objection claimed to J53 25 arise from a scientific viewpoint. The objection asks: isn't J53 26 Wittgenstein here usurping the place of empirical inquiry? Is it J53 27 not possible that empirical science - neurophysiology, in J53 28 particular - will find specific states and processes that will fill J53 29 the bill, as far as understanding, believing, remembering, etc. are J53 30 concerned? And so, whether or not there are definite states or J53 31 processes of understanding is something that we will discover. That J53 32 there are such can certainly be entertained as a scientific, J53 33 empirical hypothesis, and its consequences discussed. To preclude J53 34 such an outcome now is just to claim that we shall never J53 35 obtain certain sorts of results in future neurophysiology. But then J53 36 Wittgenstein is simply making a bet on the future course of science J53 37 or else he is engaged in a priori anti-science, denying a priori J53 38 that certain projects could bring results, and hence they ought not J53 39 even be investigated empirically.

J53 40 For brevity I shall call this objection, that Wittgenstein is J53 41 simply ruling out something that science could discover, the J53 42 'scientific objection.' I mean it to be narrowly based on envisaged J53 43 possibilities of results in neuroscience and thus not to be J53 44 the charge that Wittgenstein fails to leave room for the J53 45 discoveries of psychology, or cognitive science, or J53 46 psycholinguistics, or some as yet unknown science of the mind. The J53 47 latter charge has been made frequently since the publication of the J53 48 Investigations, but it is not so much an objection to a J53 49 specific point of Wittgenstein's as a denial of all of J53 50 Wittgenstein's philosophy of mind. Much in the Investigations J53 51 is devoted to exposing the conceptual confusions that would be J53 52 involved in a "science of the mind"; in any case, J53 53 any such science would presuppose just the picture of mental J53 54 states or processes and the whole notion of a mental apparatus that J53 55 Wittgenstein is concerned to undercut. In contrast, the scientific J53 56 objection as I am imagining it exploits the unarguable status as J53 57 empirical science of neurophysiology and the independence from J53 58 psychology of the conceptions of state and process that it employs. J53 59 Consequently, the objection requires detailed attention to what J53 60 Wittgenstein is saying or denying about intentional notions, and so J53 61 can function to elicit clarification of Wittgenstein's dark point J53 62 about states or processes.

J53 63 The scientific objection is, it seems to me, a most natural J53 64 one, and for that reason powerful. And it puts Wittgenstein in a J53 65 poor (and stereotypical) light: as a philosopher, making ungrounded J53 66 empirical claims or, worse yet, a philosopher trying a priori to J53 67 deny the progress of science, like a late Scholastic refusing to J53 68 grant the coherence of the Copernican conception or an Idealist J53 69 'proving' the impossibility of Einsteinian physics. Indeed, the J53 70 charge that Wittgenstein is ruling out certain empirical J53 71 possibilities a priori is doubly damaging against a philosopher J53 72 whose major concern is to fight against a priorism, to demolish J53 73 pictures of how things must be, to expose "preconceived J53 74 ideas to which reality must correspond" J53 75 (<*_>section<*/> 131).

J53 76 A common view of Wittgenstein takes his response to such an J53 77 objection to proceed by noting that a physiological state would be J53 78 'hidden'; it would not reflect central conceptual features of our J53 79 notion of understanding (or believing, thinking, etc.). No such J53 80 state would have conceptual links to correct responses and the J53 81 other criteria of understanding; but the transparency of those J53 82 links is essential to our concept of understanding. In short, the J53 83 point is that no physiological state has the 'grammar' of J53 84 understanding. Hence, Wittgenstein is taken to urge, understanding J53 85 cannot be conceptually constituted by any such state.

J53 86 Although there are passages in the Investigations that J53 87 suggest this response, to my mind it fails to capture J53 88 Wittgenstein's thought. The response requires a general and sharp J53 89 distinction between conceptual and empirical, between criterion and J53 90 result of investigation. Wittgenstein does not have the resources J53 91 to make this distinction; in fact, the distinction is of a piece J53 92 with the a priorism that he wishes to attack (see <*_>section<*/> J53 93 79, last paragraph). That is, the response requires a form of J53 94 essentialism which I do not believe Wittgenstein accepts.

J53 95 Moreover, the response leaves it open that understanding, J53 96 although not 'conceptually' constituted by a neural state, might be J53 97 'empirically' constituted by one. This underreads Wittgenstein's J53 98 conclusion. Thus the suggested response does too little. Yet it is J53 99 not totally off the mark, for Wittgenstein does think that the J53 100 identification of any state of understanding would have to bear J53 101 some conceptual burden. His conclusion will be that, in some sense, J53 102 "nothing is hidden." But this is meant to result J53 103 from detailed considerations of the notion of understanding, not to J53 104 be assumed more or less as a premise, as an unargued feature of his J53 105 methodology.

J53 106 The notion that Wittgenstein dismisses or ignores questions of J53 107 'empirical' identity also reinforces the view of him as simply J53 108 anti-scientific. Now we do know that Wittgenstein was distrustful J53 109 of the claims of science and its role in contemporary culture. He J53 110 makes this quite explicit when he expresses his opposition to J53 111 "the spirit of the age" in the Foreword to the J53 112 Philosophical Remarks, and the theme is repeated J53 113 elsewhere in his work, both early and late. Some have gone on to J53 114 ascribe to him a hostility to or belittling of projects aimed at J53 115 scientific explanation, of such proportions as to depict him as J53 116 simply dogmatic and blinkered. Bernard Williams, for example, J53 117 complains that Wittgenstein's general practice and teaching serve J53 118 to "stun, rather than assist, further and more systematic J53 119 explanation," and offers as illustration a story about J53 120 Wittgenstein used by Georg Kreisel in 1960 as evidence for J53 121 Wittgenstein's hostility to legitimate scientific explanation. The J53 122 story bears repeating. Imagine a child, learning that the earth is J53 123 round, asking why then people in Australia don't fall off. I J53 124 suppose one natural response would be to start to explain about J53 125 gravity. Wittgenstein, instead, would draw a circle with a stick J53 126 figure atop it, turn it upside down, and say "Now we J53 127 fall into space."

J53 128 Now I myself do not find the story emblematic of an attitude J53 129 against science, or of a desire to 'stun' further explanation. In J53 130 fact, I think it shows Wittgenstein being highly insightful. For he J53 131 is examining the source of the child's question, in the concepts J53 132 with which the child is operating. Given those concepts, an appeal J53 133 to gravity can do nothing but mislead: the child will take it that J53 134 the antipodal people are upside down, but they have gravity shoes, J53 135 or glue, or something similar, that keeps them attached to the J53 136 surface of the earth; as for us, we are right side up, so the J53 137 problem does not arise. What Wittgenstein's trick does is precisely J53 138 to expose the conceptual confusion in the way the child is thinking J53 139 of up and down (cf. <*_>section<*/> 351). Once the child sees the J53 140 relativity in the notions of up and down, she may then go on to J53 141 ask, "Well, why don't all these objects - us, the J53 142 Australians, and the earth - go careening around J53 143 independently?" or perhaps, "Then why do objects J53 144 fall down rather than up?" At that point, explaining J53 145 gravity might well be in order; the child is now prepared to J53 146 appreciate it correctly.

J53 147 The lesson, of course, is general. It is central to J53 148 Wittgenstein's teaching that the conceptual underpinnings to a felt J53 149 need for explanation must be scrutinized, for what it is, exactly, J53 150 that wants explanation may only become clear through such an J53 151 investigation. This does not make him anti-scientific. It does make J53 152 him anti-scientistic, against the smug and unexamined assurance J53 153 that what wants explanation is obvious, and that scientific tools J53 154 are immediately applicable.

J53 155 For Wittgenstein, it is characteristic of the notions that J53 156 figure in philosophical problems - prominently, mental concepts and J53 157 linguistic concepts like meaning - that a structure is imposed on J53 158 them, without grounding in the ordinary use of these notions and J53 159 without being noticed, when they are taken to be amenable to J53 160 certain explanatory projects. (Much work in the J53 161 Investigations, on my view, is precisely devoted to getting us J53 162 to notice, to see that there is a place at the start of J53 163 philosophizing, where this imposition happens.) Hence, only through J53 164 clarification of what the legitimate questions are can proper sense J53 165 be made of the applicability of science. A scientistic viewpoint J53 166 ignores this need for clarification. As a result, for Wittgenstein J53 167 scientism is just as misguidedly metaphysical as traditional, more J53 168 transparently a prioristic, approaches.

J53 169 I am depicting Wittgenstein as thinking that conceptual work J53 170 must be done before the question of the applicability of science J53 171 should be raised. Now it is also true that whatever empirical, J53 172 scientific results may legitimately be foreseen or hypothesized J53 173 after that work is done are of little interest to Wittgenstein. He J53 174 seems to believe that what is really at issue in a philosophical J53 175 question will be answered, or dissolved, by the conceptual work, J53 176 and not touched by science. That is, science is simply not of use J53 177 in dealing with the sorts of problems with which he is concerned. J53 178 But this broad characterization can be misleading. Wittgenstein is J53 179 not being dismissive; he is not urging a distinction between J53 180 questions of the mind vs. questions of the heart. Nor is he saying J53 181 that in doing science we are talking about different things; he J53 182 would have had little patience with Eddington's 'two tables' and J53 183 with Goodman's different worlds (although the latter claim may be J53 184 thought controversial). Moreover, as I mentioned above, I do not J53 185 believe that he wishes to rely on a sharp distinction between J53 186 conceptual and empirical, that is, a Fregean divide between logic J53 187 and analysis on the one hand and 'mere' psychology and physics on J53 188 the other. Rather, Wittgenstein operates case-by-case. For each J53 189 philosophical question we treat, we are to tease out what we are J53 190 aiming for, or what we think we are aiming for, and then to come to J53 191 see how our objectives will not be served by a scientific J53 192 investigation; and we are to recognize how the inclination to look J53 193 for science for answers elides or ignores so much as to suggest J53 194 that a philosophical picture is at work.

J53 195 This point is made in <*_>section<*/> 158, the one explicit J53 196 appearance in the Investigations of something like the J53 197 scientific objection. Here he suggests that, although the objection J53 198 says that the scientific investigation may come out either way, J53 199 that it is only a scientific hypothesis or conjecture that J53 200 such-and-such a process or state will be found, at bottom the J53 201 objector is being moved by an a priori demand that things must J53 202 turn out a certain way. The claim of a modest empiricism is mere J53 203 lip service.

J53 204 With respect to understanding, the point might be phrased thus. J53 205 Wittgenstein asks us to look in detail at the range of our J53 206 practices relevant to an ascription of understanding. We find an J53 207 enormous variety of considerations that can enter, a dependence on J53 208 context that is impossible to describe accurately by any general J53 209 rules, a lack of uniformity in mental accompaniments. J53 210 J54 1 <#FROWN:J54\>While it is not likely that Greek agriculture J54 2 often produced more than limited dividends, those who were masters J54 3 of the rural landscape gained an even greater advantage - they also J54 4 controlled almost all the population of a polis either J54 5 directly or indirectly. Accordingly, aristocrats tended to hold on J54 6 to their land. It has often been argued that an ancestral plot J54 7 (kleros) could not be alienated outside the family before J54 8 the fifth century, but a variety of evidence proves otherwise. J54 9 Hesiod's father migrated from Asia Minor to Boeotia, where he J54 10 acquired a farm; the poet himself advised his auditors to honor the J54 11 gods so you may buy another's holding (kleros) J54 12 and not another yours. Aristotle's assertion that J54 13 "to part with family estate was one of the things that were J54 14 'not done,'", does not prove the inference that it could J54 15 not be done. Normally, nonetheless, formal sale or transfer of land J54 16 rights was unlikely in so rurally based a world; political J54 17 citizenship and economic security rested on an independent J54 18 connection to land.

J54 19 Various areas saw their smaller farmers of earlier times J54 20 reduced to the level of bondsmen or serfs, though true agricultural J54 21 slavery was uncommon in the Greek world. Usually rural domains were J54 22 not large in view of the fragmented geographical nature of Greek J54 23 landscapes; the largest known at Athens, as noted earlier, ran only J54 24 about 50 hectares, though "in fifth- and fourth-century J54 25 Athens there were landowners possessing from three to six estates J54 26 in different <}_><-|>part<+|>parts<}/> of Attica." Those J54 27 who could be expected to serve as hoplites probably were masters of J54 28 at least 12 hectares; free farmers after the Solonian reforms would J54 29 scarcely have been able to cope with more than 4 hectares unless J54 30 they drew in outside labor at critical points in the agricultural J54 31 cycle.

J54 32 How were the larger holdings managed? In modern times nobles J54 33 relied on stewards, factors, and the like to bear day-to-day J54 34 responsibilities; this was probably the case also in ancient J54 35 Greece, though our evidence is very limited. Cimon's liberality to J54 36 his fellow demesmen was famous, but he was absent from Athens so J54 37 often on military operations that he must have had a resident aide. J54 38 For Pericles we do have the comment in Plutarch's life that he J54 39 arranged his paternal estate so "that it might neither J54 40 through negligence be wasted or lessened, nor yet, being so full of J54 41 [public] business as he was, cost him any great trouble or time J54 42 with taking care of it." Thus he sold "all his J54 43 yearly products and profits" in a lump and bought in the J54 44 market for his household needs - obol-pinching and keeping J54 45 precise records to the discontent of his family. "His J54 46 manager in all this was a single servant, Evangelos by J54 47 name." To secure fuller information on the careful J54 48 management of agricultural resources for economic gain we must come J54 49 down to Hellenistic Egypt where the financial director for Ptolemy J54 50 II had an extremely astute manager, Zenon, many of whose detailed J54 51 records have survived on papyri.

J54 52 The picture usually drawn of upper-class concentration on J54 53 landed possessions may as a whole stand, but before we accept it as J54 54 representing the exclusive interest of the well-to-do further J54 55 reflection is necessary on the nonagricultural aspects of the J54 56 economy. For example, Demosthenes, on reaching his majority, J54 57 brought suit against his guardians for the recovery of his J54 58 inheritance. The estate he itemized to the jury under two headings J54 59 is rather surprising: "(1) the active (energa), J54 60 which included 32 or 33 slave swordmakers, bringing in 3000 J54 61 drachmas a year; another 20 slaves engaged in the manufacture of J54 62 furniture, 1200 drachmas annually; and 8000 drachmas on loan at J54 63 12%; (2) the inactive: raw materials on hand at his father's death J54 64 nine years before, worth 15,000 drachmas, the house worth 3000, the J54 65 furniture and his mother's jewelry, 8000 in cash in a strong-box at J54 66 home, a maritime loan of 7000 drachmas, and 4700 on deposit in two J54 67 banks and with a relation." Land proper does not appear at J54 68 all; Demosthenes was rather, in Bolkestein's summary, J54 69 "accustomed to make [his fortune] bear interest in many J54 70 ways" as a capital-investor, living on the interest of his J54 71 money.

J54 72 Demosthenes' family was not, indeed, of aristocratic stock, so J54 73 it may have been willing to extend the employment of its capital J54 74 more widely in the thriving Athenian industry and trade of the J54 75 fifth and fourth centuries. How far did the aristocrats do the J54 76 same, or alternatively shun this area?

J54 77 Industry may be dismissed briefly. Men of standing were not J54 78 likely to sully their fingers or break their backs in the physical J54 79 toil of stonemasons, smiths, potters, and other trades. Even so J54 80 they did have an important, twofold role: they were the consumers J54 81 who bought the wares of craftsmen, and they provided to a large J54 82 degree the capital necessary for the purchase of the slaves who J54 83 furnished a valuable share of the labor. The swordmakers and J54 84 furniture fabricators of Demosthenes have already been noted; even J54 85 more remarkable was the fact that Nicias, of aristocratic stock, J54 86 owned a thousand slaves, whom he leased out to the entrepreneurs J54 87 running the state silver mines of Laurium.

J54 88 Commerce was another matter. Retail trade, such as that in J54 89 ribbons, could be left to vendors, many of them women, but J54 90 large-scale activity especially by sea required wider attention. J54 91 One principal mark of the Aegean world in and after the eighth J54 92 century was overseas voyaging, an this was without doubt initially J54 93 in the hands of aristocrats.

J54 94 Sappho's brother Charaxus, for example, carried wine to Egypt J54 95 and there fell in love with a courtesan, to Sappho's disgust; Solo J54 96 also engaged in foreign commerce in order to recoup his father's J54 97 prodigality. Coleus of Samos, blown off course to Egypt as far as J54 98 Tartessus whence he gained so much that he had to replace his stone J54 99 anchors by silver ingots, and the later Sostratus of Aegina, who J54 100 dedicated a statue in the Greek shrine of Etruscan Pyrgi, very J54 101 probably were both men of the leading classes inasmuch as they J54 102 entered Herodotus' pages.

J54 103 It was, after all, aristocrats who had surplus resources that J54 104 could be ventured abroad and also were leaders, able to face J54 105 possible hostile resistances on foreign shores. Contrary to the J54 106 views of many modern scholars, moreover, both they and the potters J54 107 described in Hesiod's Works and Days sought earnestly J54 108 after wealth. Already in the epics Odysseus was taunted as not J54 109 looking like an athlete, that is, a man of leisure, but J54 110 "one, who faring to and fro with his benched ship, is a J54 111 captain of sailors who are merchantmen, one who is mindful of his J54 112 freight, and has charge of a home-borne cargo, and the gains of his J54 113 greed." Solon categorized the diverse ways of gaining J54 114 wealth and concluded that those who are richest "have twice J54 115 the eagerness that others have"; his contemporary Alcaeus J54 116 quoted Aristodemus - a Spartan no less - as saying that J54 117 "wealth makes the man." At first aristocratic J54 118 seafaring might not have been much distinguished from piracy and J54 119 coastal raids, but eventually it settled into more ordered J54 120 communications.

J54 121 Aristocrats were also the men most interested in the wares that J54 122 could be acquired in the advance workshops of the Near East - J54 123 ivory, glass, faience, perfumes, ointments, and spices (many of J54 124 which had names of Semitic root) - for such luxuries were the J54 125 backbone of the earliest overseas trade. Greek lands, even J54 126 including Athens down to the time of Solon, fed themselves, though J54 127 they did have need for foreign slaves, metals, wool, stone, and J54 128 other bulk items. Hesiod drank wine of Biblis while relaxing in the J54 129 heat of summer, and to a remarkable degree men and also women of J54 130 the upper classes desired wools dyed in Tyrian purple and fine J54 131 linen. Since textiles do not survive well in archeological contexts J54 132 this item is often overlooked, but even in modern times the textile J54 133 trade has been very significant; in the English colonies of North J54 134 America in the eighteenth century the main import consisted of J54 135 English, Irish, and German cloth and textiles.

J54 136 By the later sixth century aristocrats had become more J54 137 conscious of the duties and limitations of their position and J54 138 largely yielded long-distance trade to professional shippers, but J54 139 as they withdrew into the background their interest in this realm J54 140 did not disappear. The men who scurried about the Aegean and J54 141 farther afield had to have capital to outfit their ships and J54 142 finance cargoes. To an extent what we cannot measure they may have J54 143 done so out of their own resources, but at least occasionally they J54 144 had to secure a bottomry loan at rates up to 33.33% - Demosthenes' J54 145 estate, it may be remembered, included such a loan, and in one of J54 146 his orations a money-lender/banker asserted that without the J54 147 support of men of his type "no ship, shipper, or sailor can J54 148 put to sea." And who provided the money of the banker? J54 149 Undoubtedly the well-to-do of Athens; in imperial Rome as in early J54 150 modern times the rich supplied funds by the back door to J54 151 large-scale traders. Nor did aristocrats totally surrender the J54 152 field; Andocides, who traced his ancestry back to Hermes via J54 153 Odysseus and Telemachus, actually engaged in maritime commerce J54 154 throughout the winter in the late fifth century and after his J54 155 return from exile "continued to think and act like the J54 156 businessman he had turned himself into". It is unsafe to J54 157 assume that the word kerdos (profit) totally disappeared J54 158 from aristocratic lips even after the developed ethos of the class J54 159 frowned on undue interest in economic activities. Aristotle in his J54 160 Nicomachean Ethics judiciously stresses the need for a J54 161 competence but not a search for gain per se.

J54 162 Through control of the land and the revenues from investments J54 163 the well-to-do economically commanded the Greek communities, J54 164 sometimes almost completely, though at Athens only in major degree, J54 165 and used every available opportunity to enjoy an elegant, luxurious J54 166 life. Modern hostility toward elites swiftly rises into view at J54 167 this point in the common assertion that aristocrats in all ages J54 168 spend money rather than improving the economic machinery of their J54 169 world in a bourgeois fashion. Thus early modern aristocracies were J54 170 reproved as being engaged in "the unjust, unhealthy, J54 171 brilliant and anti-economic utilisation of any surplus produced in J54 172 a given society:" An interesting study of men who J54 173 participated especially in the Thirty Years' War of the seventeenth J54 174 century after Christ suggests that their wealth was committed to J54 175 building mansions and to acquiring adornments such as gold J54 176 necklaces. Only one noted scholar, to my knowledge, finds merit in J54 177 this type of expenditure. Writing about early modern English J54 178 aristocrats, G.M. Trevelyan raises the question as to how else the J54 179 English nobles could have expended their money save by building J54 180 magnificent houses in a period when stocks, bonds and general loans J54 181 were unknown and land was not easily bought - but then Trevelyan is J54 182 nowadays generally dismissed as an elitist.

J54 183 If a phenomenon recurs frequently in different historical J54 184 societies, then there must be significant reasons for its presence; J54 185 and an understanding of those reasons will be more useful than the J54 186 common expression of indignation or reproof. Braudel, just quoted, J54 187 also observes that luxury "scarcely changes at all" J54 188 as a concept accepted both by privileged and unprivileged classes J54 189 and, as he notes, both Mauss and Sombart emphasized the role of J54 190 luxury in promoting demands on artists and others in early modern J54 191 Europe. So too the aristocrats of ancient Greece stimulated the J54 192 amazing outburst of Hellenic civilization by their patronage of the J54 193 arts and crafts, as we shall see in the next chapter.

J54 194 Some modern aristocracies unfortunately have been largely J54 195 parasitical and have been overthrown in violent revolutions. In J54 196 Greece the political, social, an economic position of the upper J54 197 classes was too deeply anchored ever to be seriously threatened; J54 198 even the Athenian democracy commonly entrusted its leadership to J54 199 men sprung from aristocratic families of high standing. The upper J54 200 classes of Athens, however, paid a heavy price both personally and J54 201 economically. If they were wealthy enough to afford hoplite armor J54 202 or horses, they had to be prepared to face the dangers inherent in J54 203 the almost unceasing wars of the fifth and fourth centuries; an J54 204 inscription of 460 or 459 lists no less than 177 men of one J54 205 Athenian tribe who died in one year in Cyprus, Egypt, and J54 206 elsewhere, including two generals (an Athenian tribe might have had J54 207 in the order of 4000 adult male citizens, but the number who in J54 208 practice might be drafted is much reduced if one takes into account J54 209 the men of age and those mentally and physically incompetent). J54 210 J55 1 <#FROWN:J55\>More recent events reinforced French insistence J55 2 that 'the Boche will pay.' Rehabilitating France's ten northeastern J55 3 d<*_>e-acute<*/>partments, which had served as one of the main J55 4 theaters of the war, required an expensive infusion of capital. J55 5 France had incurred substantial war debts to her Allies, whereas J55 6 Britain's debt to the United States was offset in part by France's J55 7 debt to Britain. American refusal to provide concessional J55 8 reconstruction loans or to forgive these debts did much to harden J55 9 the French position, rendering it inevitable that German J55 10 reparations and Allied war debts would be bound up together.

J55 11 A reparations bill as large as $200 billion was contemplated at J55 12 Versailles. Ultimately, the assembled delegates were only able to J55 13 establish a deadline for the conclusion of discussions: May 1921. J55 14 Negotiations seemed to stretch on interminably. The Reparation J55 15 Commission charged with settling the matter could agree only on a J55 16 principle: that while France and her allies were authorized to J55 17 press their claims for full damages, actual transfers would be J55 18 linked to Germany's capacity to pay as gauged by the rate of growth J55 19 of her exports and her success in obtaining foreign loans.

J55 20 By linking reparations payments to the condition of the German J55 21 economy, the Allies diminished the incentive for German J55 22 policymakers to put their domestic house in order. Hyperinflation J55 23 was only the most dramatic illustration. Politicians were not J55 24 encouraged to implement painful programs designed to promote growth J55 25 by the knowledge that the fruits of their labor would be transfered J55 26 abroad. The form of the reparations bill hardened German J55 27 resistance. Including pensions, as insisted on by Britain and the J55 28 Commonwealth to inflate their share of the total, cast doubt on the J55 29 French justification for reparations based on the cost of J55 30 reconstructing devastated regions and reinforced the German belief J55 31 that the dominant Allied motives were avarice and spite.

J55 32 An unstable German economy had far-reaching economic and J55 33 political ramifications. Anything that depressed trade in Germany J55 34 depressed trade throughout Central Europe. Economic instability in J55 35 Central Europe intensified fears of a Bolshevik threat from the J55 36 east, reviving familiar Anglo-French conflicts over spheres of J55 37 influence in Eastern Europe and undermining the spirit of J55 38 cooperation developed during the war. Prospects for compromise J55 39 among the Allies grew increasingly remote.

J55 40 In the interim, Germany was instructed to begin transfers in J55 41 kind, mainly coal but also stocks of Reichsbank gold, war J55 42 mat<*_>e-acute<*/>riel, public property in ceded territories and J55 43 colonies, railway rolling stock, and ships. The coal was essential J55 44 to a French steel industry handicapped by the destruction of French J55 45 mines by retreating German armies. These 'interim payments,' J55 46 justified as a way of defraying occupation costs, were formally J55 47 distinct from other transfers, although they eventually came to be J55 48 regarded as the first installment of reparations. Transfers J55 49 completed prior to May 1921 amounted to 8 billion gold marks (marks J55 50 of prewar value). This amounted to some 20 percent of German J55 51 national income in 1921, although it represented only 40 percent of J55 52 the interim payment specified at Versailles.

J55 53 It seemed noteworthy that these sizeable interim transfers did J55 54 not destabilize the German price level or the government budget. J55 55 They were effected despite continued uncertainty about the size of J55 56 the reparations bill and despite capital flight from territories J55 57 scheduled for cession. Since a large part of the interim transfer J55 58 took the form of public property such as railway rolling stock J55 59 rather than private-sector production that the government had to J55 60 pay for by borrowing or taxing, it was relatively easy to mobilize. J55 61 But insofar as it would be necessary eventually to replace that J55 62 public property, Germany was mortgaging her future, a fact that J55 63 could not have reassured outside observers. The presence of Allied J55 64 troops along the Rhine and the Baltic and the return of domestic J55 65 political stability following the Kapp Putsch of 1920 have also J55 66 been invoked to explain the ease of transfer. But troops were no J55 67 guarantee of compliance, as the Allies would learn in 1923. Only J55 68 with benefit of hindsight could the failure of the Kapp Putsch be J55 69 seen as strengthening moderate tendencies within the military. At J55 70 the time, each of these developments, rather than reassuring J55 71 domestic and foreign observers, heightened concern over both J55 72 economic stability and Germany's fragile political equilibrium.

J55 73 More than the presence of occupation forces or the political J55 74 climate, the key factor in the interim transfer was Germany's hope J55 75 that a demonstration of good will would elicit Allied concessions J55 76 and permit the early extinction of reparations. The Allies had not J55 77 yet irrevocably committed to their excessive demands. By evincing a J55 78 willingness to pay on the scale of France's reparations after 1871, J55 79 Germany might encourage the victors to adopt a more conciliatory J55 80 stance.

J55 81 The fiscal implications of the transfer were accommodated by J55 82 tax reforms guided through the Reichstag by the finance minister, J55 83 Matthias Erzberger, over the strident opposition of a right wing J55 84 led by Helfferich. Erzberger's tax package featured an emergency J55 85 levy and transferred the income tax from the states to the Reich in J55 86 return for a commitment by the central government to redistribute J55 87 some of the revenues back to local authorities. The tax increase J55 88 was essential for maintaining fiscal balance in the face of the J55 89 interim transfer. German politicians and their constituencies J55 90 tolerated higher taxes because they anticipated that the revenue J55 91 would be transferred abroad for only a limited period of time. J55 92 Rather than provoking capital flight and other forms of evasion, J55 93 the tax increase was followed by short-term capital inflows in J55 94 anticipation of possible stabilization of the mark. Since the J55 95 interim transfer provoked neither capital flight nor currency J55 96 depreciation, the revenue base of the new income tax was not eroded J55 97 by inflation.

J55 98 Following a series of preparatory conferences, the Allies J55 99 assembled in London in 1921 to set Germany's payment schedule. The J55 100 U.S. Congress had already indicated its unwillingness to ratify the J55 101 Versailles Treaty. The American representative to the Reparation J55 102 Commission was reduced to observer status, limiting his ability to J55 103 support the British delegation in its opposition to the more J55 104 extreme demands of France and Italy. Congress's refusal to ratify J55 105 signalled the resurgence of isolationist tendencies within the J55 106 United States, which bode ill for those who hoped for war debt J55 107 cancellation. Given American inflexibility regarding war debts, the J55 108 prospects for French, Italian, and British compromise on J55 109 reparations appeared increasingly bleak.

J55 110 The negotiators at London delivered a reparations bill of 123 J55 111 billion gold marks, or 31 billion U.S. dollars. This staggering sum J55 112 was a concession relative to the Reparation Commission's initial J55 113 recommendation of 225 billion gold marks. Denominating the debt in J55 114 gold insured that inflation and exchange rate depreciation could J55 115 not be used to erode its value. Germany was to begin service J55 116 immediately on 50 billion of the 132 billion total, on which 5 J55 117 percent interest and 1 percent amortization amounted to 3 billion J55 118 gold marks (roughly 7 1/2 percent of national income). In addition, J55 119 she was charged 1 billion marks annually for occupation costs and J55 120 in settlement of prewar debts (bringing the total to perhaps 10 J55 121 percent of national income). Payment of the second tranche of 82 J55 122 billion gold marks was deferred pending an adequate increase in J55 123 Germany's capacity to pay. These contingencies heightened the J55 124 uncertainty surrounding the date at which the reparations burden J55 125 would finally be extinguished. All that was certain was that J55 126 Germany would be obligated to make substantial transfers over a J55 127 period of decades.

J55 128 No issue in twentieth-century economic and political history J55 129 has been more hotly contested than the realism of this bill. J55 130 Contemporaries gauged the burden by comparing it to the reparations J55 131 paid Germany by France following the Franco-Prussian war. France J55 132 had paid a total of 5 billion francs, roughly one-quarter of French J55 133 national income in 1872. In comparison, Germany's immediate burden J55 134 of 50 billion gold marks represented 125 percent of national income J55 135 in 1921. Including the deferred payments (known as C Bonds) raised J55 136 the ratio to the 330 percent. At 10 percent of national income, the J55 137 first year's payments under the London Schedule were very large by J55 138 prewar standards.

J55 139 Defenders of the London Schedule observed that Britain had J55 140 transferred abroad fully 8 percent of national income through J55 141 foreign lending in 1911-13. This proved, they argued, that the J55 142 balance-of-payments adjustment mechanism was capable of absorbing a J55 143 transfer on the requisite scale. But at least some British J55 144 investment abroad had returned to London as foreign deposits and J55 145 some in the form of export demands. Together these mechanisms J55 146 minimized the impact on British industry and on the balance of J55 147 payments. It was unlikely that either mechanism would operate as J55 148 powerfully to recycle German reparations.

J55 149 The politics of the two transfers were even less comparable. J55 150 Britain had not sacrificed domestic wealth in the amount of the J55 151 transfer. The British had invested abroad voluntarily with the J55 152 option of devoting those resources to future consumption. No J55 153 necessary impact on British living standards resulted. The problem J55 154 for Germany was how to mobilize for transfer 10 percent of national J55 155 income and to reduce both present and future consumption without J55 156 provoking domestic political unrest.

J55 157 Transforming 10 percent of national income into foreign J55 158 currency required an external surplus equivalent to 80 percent of J55 159 1921-22 exports. One can imagine that strict controls modelled on J55 160 wartime practice might have succeeded in reducing German imports by J55 161 80 percent. But radically curtailing imports was inconsistent with J55 162 the maintenance of exports given the economy's reliance on inputs J55 163 from abroad such as copper, cotton, and wool, a dependence that had J55 164 been heightened by war-time losses of territory and J55 165 stockpiles. Expanding exports by 80 percent required a further J55 166 increase in imported inputs, multiplying the gross increase in J55 167 exports necessary to effect the transfer. And even these J55 168 calculations left aside the implications of massive import J55 169 compression for domestic living standards.

J55 170 Even had Germany somehow been able to provide this astonishing J55 171 increase in exports, the Allies would have been unwilling to accept J55 172 it. The problem was not that the incremental exports were so large J55 173 relative to the British, French, and U.S. economies. The projected J55 174 transfer amounted, on an annual basis, to perhaps 1 J55 175 per-cent of their combined national incomes. But German J55 176 exports would be heavily concentrated in the products of industries J55 177 already characterized by intense international competition, notably J55 178 iron, steel, textiles, and coal. The same difficulties would be J55 179 posed for Allied industries if Germany instead flooded markets with J55 180 exports. Representatives of these industries were unlikely to J55 181 accede graciously to a sudden expansion of German exports. Even J55 182 while complaining that Germany's effort to meet its reparations J55 183 obligation was inadequate, the Allies raised their import barriers. J55 184 Keynes, in The Economic Consequences of the Peace, J55 185 insisted that proponents of reparations specify "in what J55 186 specific commodities they intend this payment to be made, and in J55 187 what markets the goods are to be sold." Thomas Lamont of J55 188 the U.S. delegation to Versailles brought this same point to the J55 189 attention of the negotiators. The American economist Frank Taussig J55 190 echoed the warning.

J55 191 That 1920-21 was a period of recession aggravated both J55 192 problems: those of Germany's ability to export and the Allies' J55 193 willingness to import. The Allies would have been happy to accept J55 194 additional in-kind transfers had they taken the form of raw J55 195 materials (British reservations about coal notwithstanding). But J55 196 the German economy could provide these only to a limited extent. J55 197 Transfers of raw materials disrupted Germany's capacity to export J55 198 manufactures. Proposals to import German labor for the work of J55 199 reconstruction were rejected as immoral and politically unpalatable J55 200 in light of unemployment among demobilized Frenchmen, Belgians, and J55 201 Italians.

J55 202 Hence the theoretical question of what change in prices would J55 203 be needed to clear international markets in the presence of J55 204 reparations (known as the 'transfer problem') was ultimately beside J55 205 the point. Keynes's conclusion was that to generate a trade surplus J55 206 on the order of 80 percent of initial exports, a very considerable J55 207 decline in the relative price of German goods would be needed to J55 208 switch foreign demands toward German exports and German demands J55 209 away from imports. He raised the possibility that, if demands were J55 210 sufficiently inelastic, a decline in German export prices might J55 211 reduce the value of German exports at the same time it raised their J55 212 volume, rendering the transfer impossible at any price. J55 213 J56 1 <#FROWN:J56\>Such an assessment accurately condensed the J56 2 realities of Eire's military situation - as well as its political J56 3 position at the start of the war. Others, however, noted that J56 4 buoyant morale in the Irish services outweighed mere physical J56 5 constraints, that it was guerrilla tactics that carried the day in J56 6 the War of Independence.

J56 7 After hostilities broke out, the total strength of Ireland's J56 8 army would reach only a little over half of the authorized levels. J56 9 Nineteen thousand men - or two divisions - would be in uniform J56 10 within days of the Emergency's beginning; this compared to the 136 J56 11 divisions the Germans fielded in the Battle of France. Nearly every J56 12 Irish unit was understrength. Of the eight authorized rifle J56 13 battalions, none was yet organized. The cyclist squadrons - known J56 14 as the 'Piddling Panzers' - would have posed precious little threat J56 15 to real panzers. The army had at its disposal two 'serviceable' J56 16 tanks and 21 armored vehicles, most of the latter of which were J56 17 already in 1939 antiques - 1920, and earlier, Rolls-Royces. The air J56 18 force, a branch of the army, was equally toothless, with only 24 J56 19 craft, of which 10 might be called modern. There was no navy other J56 20 than a small coast guard unit. Prior to May 1940, no strike force J56 21 capable of resisting any invader existed. Few commands had J56 22 reserves. Armaments, ammunition, and vehicles were extremely J56 23 scarce.

J56 24 But so quickly did the land war turn into Phony War that the J56 25 army's General Headquarters announced on September 21 that J56 26 "the present Emergency does not constitute a war situation J56 27 and it would not be justified in maintaining its Establishment and J56 28 that the strength should be reduced." By Christmas, most of J56 29 the army personnel who could manage the journey had gone home for J56 30 the holiday. Only an IRA raid on the central army munitions J56 31 magazine in Dublin's Phoenix Park, in which the raiders escaped J56 32 with large quantities of arms and ammunition, caused headquarters J56 33 to order troops back to their duty stations.

J56 34 During these early months of the war, Irish soldiers weren't J56 35 even sure of receiving a uniform. That was perhaps as well. With J56 36 some irony, and undisguised distaste on the other side of the Irish J56 37 Sea, the soldier of the Twenty-six Counties looked uncannily like J56 38 his counterpart in the Wehrmacht - most notably so in the same J56 39 'iron scuttle' steel helmet both forces wore. It was likely this J56 40 unfortunate symbolism rather than the uniform's scratchy J56 41 uncomfortableness that caused the Army Department to scrap it in J56 42 1940. In its place came a new uniform little distinguishable from J56 43 the British pattern, soup-plate helmet and all (the latter being J56 44 the same style helmet worn by American GIs until replaced in 1942 J56 45 with the familiar rounded, nearly brimless model).

J56 46 Not only were the soldiers dressed poorly; they were housed J56 47 poorly as well. Most of the barracks remained as leftovers from the J56 48 British regime, with little new military shelter built since. Many J56 49 derelict country houses had been fixed up to provide minimal J56 50 standards, and farmboys turned soldiers suddenly found themselves J56 51 living in once grand but now sadly dilapidated mansions.

J56 52 Privates earned 14 shillings a week - about one dollar in J56 53 contemporary terms, but then with strikingly greater purchasing J56 54 power, of course. The income wasn't entirely discretionary, though. J56 55 A forced haircut deduction of two pence was taken out of each pay J56 56 packet, as was six pence for laundry and another tuppence for J56 57 'social welfare.' In theory the Irish soldier was fed better than J56 58 his civilian compatriots - supposedly a daily three-quarters of a J56 59 pound of 'best home-fed beef,' a quarter pound of fresh vegetables, J56 60 a 'liberal' quantity of butter, cheese, jam, eggs, sausages, bacon, J56 61 etc. In fact, his diet consisted of the usual monotonous regime of J56 62 many armies: oatmeal, brown stew, jam rolls, bread and butter, and J56 63 tea.

J56 64 To meet the costs of their new defense requirements, the J56 65 government forced the Irish taxpayer to pay taxes higher than any J56 66 ever known. The first increase predated the outbreak of war: in the J56 67 spring of 1939, the new pounds5.5 million defense appropriation J56 68 meant jumping the income-tax rate by a half shilling to a shilling J56 69 on the pound - 5 percent. Along with this income-tax increase came J56 70 new surtaxes, as well as additional taxes on the richest J56 71 ratepayers. Two months after Germany attacked Poland, the income J56 72 tax went up another shilling on the pound, together with higher J56 73 increments in estate duties and new levies on beer and whiskey. J56 74 Through the next two years, tax increases on income could rise J56 75 until the Irish citizen was paying on average 37.5 percent of his J56 76 income to the government. Though the Twenty-six Counties remained J56 77 at peace, their government was assessing tax levies as onerous as J56 78 those of most of the countries at war.

J56 79 <*_>three-bullets<*/>

J56 80 If the island was not yet threatened by a Wehrmacht held in its J56 81 traces, it was threatened by the ruthless and bomb-prone J56 82 activities of the IRA. The Irish Republican Army's raison J56 83 d'<*_>e-circ<*/>tre was to end British control of Ulster by J56 84 whatever means necessary, however appalling or murderous. The J56 85 terrorist organization's position was, simply stated, that J56 86 "England's difficulty is Ireland's opportunity." To J56 87 achieve its goal, the organization maintained a complete, albeit J56 88 underground, government, a constitution, and some 7,500 mostly J56 89 youthful members - plus perhaps 15,000 more or less dedicated J56 90 supporters (the figures are from Time magazine).

J56 91 What popular support the IRA received in the south during J56 92 the Emergency was given almost entirely in token of the perceived J56 93 injustice of the island's partitioned status. As to the J56 94 organization's relationship with the Dublin government, its policy J56 95 provided that no terrorist activities would be carried out against J56 96 Eire, as long as the IRA was free to carry out from southern bases J56 97 operations against Ulster and that province's British targets. De J56 98 Valera refused such a concession, understanding full well the J56 99 danger of British retaliation against Eire.

J56 100 Some months before the outbreak of war, the IRA had undertaken J56 101 to traumatize the British people into demanding that their J56 102 government leave Ulster. The shock was carried to Britain itself in J56 103 the form of a series of terror bombings. In January 1939, young J56 104 Irishmen recruited in Britain set off explosions in what were, with J56 105 war approaching, the kingdom's most vulnerable sites: factories, J56 106 power stations, and telephone exchanges. At the time, even a few J56 107 well-placed blows against British defense facilities were J56 108 enormously crippling to the catch-up effort to match Germany's J56 109 industry. The campaign reached its moral nadir back home with a J56 110 bombing of the Irish country hotel where Prime Minister Neville J56 111 Chamberlain's son Francis was spending a hunting holiday, the J56 112 fortunately ineffectual assault an apparent attempt to sour the J56 113 personally cordial relationship between Chamberlain and de Valera. J56 114 Britain endured over a hundred more explosions in July alone, with J56 115 blast sites including Piccadilly Circus and Madame Tussaud's J56 116 waxworks. Harried police waded through crowds arresting anyone with J56 117 a brogue. The IRA's outrages culminated in an act that finally J56 118 crystallized public opinion and marshaled concrete action against J56 119 the outlaw organization.

J56 120 On August 25, a package-laden bicyclist made his way through J56 121 the crowded streets of Coventry, an ancient and, by 1939, heavily J56 122 industrialized Midlands city. The rider left his parcel - a J56 123 pre-fused bomb - at a caf<*_>e-acute<*/> in crowded Broadgate, in J56 124 the center of the city. To hide the device's origins, its makers J56 125 had assembled it in one place and brought it carefully to another, J56 126 where the last man in the deadly chain put it in his cycle's J56 127 carrier basket. Delayed by traffic and worried that the bomb would J56 128 blow him up along with the innocent bystanders who were its J56 129 intended victims, the anxious IRA terrorist hurriedly threw his J56 130 bicycle against the wall on the caf<*_>e-acute<*/> and left. The J56 131 explosion a few moments later blew off the front of the building, J56 132 along with the windows of the neighboring shops. Ankle-deep J56 133 debris settled over a wide area. Five people lay dead, including an J56 134 eighty-one-year-old man and a small boy. Seventy more were J56 135 injured.

J56 136 Recognizing the threat this and the earlier outrages J56 137 represented to an Anglo-Irish accord, the authorities reacted by J56 138 searching every Irish home in Coventry, jailing hundreds of J56 139 activists, and ending with the apprehension of three members of the J56 140 city's IRA unit. Two others were later arrested for complicity in J56 141 the terror attack. The dragnet resulted in an immediate and sharp J56 142 decline in IRA terror activities in Britain for the rest of the J56 143 war. But among Eire's citizenry who deplored the IRA's methods, so J56 144 deep was the vein of antipathy for Britain that when two of the J56 145 accused were hanged in Birmingham in February 1940, almost the J56 146 entire country mourned them, with flags dropping to half staff, J56 147 theaters closed, and masses offered for the repose of the executed J56 148 men's souls.

J56 149 Hitler understandably regarded people who could commit such J56 150 acts against Britain as his natural allies. In fact, Germany had J56 151 been trying to cement a relationship with the terrorist J56 152 organization since at least 1937. The military intelligence agency, J56 153 the Abwehr, directed by Admiral Wilhelm Canaris, initiated planning J56 154 in November and December of 1939 to send its agents into Ireland by J56 155 submarine to establish contacts with the IRA, with German agents J56 156 instructed to tell prospective recruits among the Irish that J56 157 Germany strongly desired a united Ireland and that the best course J56 158 for the IRA would be to join efforts with the Reich in destroying J56 159 'England,' the sooner both their goals being fulfilled.

J56 160 But Hempel warned his superiors in November 1939 that Germany J56 161 had best not rely too heavily on playing the IRA card. On the J56 162 fourteenth, he wrote to Berlin that "the I.R.A. is hardly J56 163 strong enough for action with promise of success or involving J56 164 appreciable damage to England and is also probably lacking in a J56 165 leader of any stature." He pointedly cautioned the Foreign J56 166 Ministry that open cooperation with the IRA would very likely lead J56 167 the more moderate sections of the Irish republic into blaming the J56 168 organization for making the country's national interests dependent J56 169 on Germany, which "in view of the widespread aversion to J56 170 present-day Germany, especially for religious reasons, could rob J56 171 the I.R.A. of all chances of future success." Hempel also J56 172 noted, again, that such a course would give Britain an excuse to J56 173 intervene militarily in solving its own outstanding problems with J56 174 Eire.

J56 175 Because of the IRA's potential to damage Eire's neutrality J56 176 policy more than any other group in Irish politics, the J56 177 organization's British atrocities had in June 1939 given de Valera J56 178 a good excuse to outlaw it. The move enabled him to marshal the J56 179 resources of the state in chasing down its members and generally J56 180 branding it a menace to Eire's survival in the Emergency. But by J56 181 the end of the year, the IRA openly declared its sympathies lay in J56 182 a <}_><-|>Germany<+|>German<}/> victory in the war, evidently on J56 183 the amazing deduction that such an outcome would, by Britain's J56 184 defeat, mean the end of partition. The perhaps more likely J56 185 possibility that Hitler would occupy Ireland - the whole of J56 186 the island - apparently didn't occur to the IRA leaders. Though J56 187 Nazi Germany's most scorching depravities and betrayals lay in the J56 188 future, its many double crosses up to this point should certainly J56 189 have put the IRA off any hope that it or the nation for which it J56 190 purported to fight would be treated with respect by Adolf J56 191 Hitler.

J56 192 On December 23, 1939, the organization carried off one of its J56 193 grandest coups - though one of its last - against the Dublin J56 194 government. When the IRA stole more than a million rounds of J56 195 ammunition and cases of guns from the Phoenix Park arsenal, it J56 196 looked as though the phantom army might turn itself into a real J56 197 army and attempt a coup d'<*_>e-acute<*/>tat, or even try to start J56 198 another civil war. D<*_>a-acute<*/>il member James M. Dillon warned J56 199 of the raid: "I believe the ultimate end of the activities J56 200 of these gentlemen [the IRA] must be assassination. God knows how J56 201 many of us may be victims!"

J56 202 The reaction of the government was to arrest every member of J56 203 the IRA who could be rounded up, sending 5,000 Special Police armed J56 204 with rifles to seal the frontier with Northern Ireland and hunt J56 205 down the clandestine terrorists. To further the search-and-destroy J56 206 operation's success, the government rushed through Parliament a J56 207 bill suspending the constitutional guarantee against holding J56 208 suspects for more than forty-eight hours without evidence. J56 209 J57 1 <#FROWN:J57\>Moreover, in recent years, most of those in pursuit of J57 2 a truly 'British' past have been in thrall to a series of J57 3 remarkable articles written by J.G.A. Pocock. In the Journal J57 4 of Modern History in 1975, and in the American Historical J57 5 Review for 1982, Pocock argued that British history could only J57 6 be understood as "the interaction of several peoples J57 7 and several histories." By this, it is important to note, J57 8 he meant not only the relations that existed over time among J57 9 England, Wales, Scotland, and Ireland but also the broader J57 10 connections between these four countries and North America and the J57 11 rest of Britain's 'white' empire, including Pocock's own native New J57 12 Zealand. Predictably, though, it has been his insistence on the J57 13 need for study of the four component parts of the United Kingdom - J57 14 and not his geographically more wide-ranging manifesto - that has J57 15 generated the greatest interest among British historians.

J57 16 Some of the results of this new scholarly fashion have been J57 17 entirely benevolent. Our collective consciousness has been raised, J57 18 and we are now much less likely than we were even ten years ago to J57 19 describe exclusively English events and trends as though they J57 20 were necessarily synonymous with British developments. We have J57 21 come to understand with more precision than before that Great J57 22 Britain is a composite structure forged, as France and Spain were J57 23 forged, out of different cultures and kingdoms. And by examining J57 24 how these entities effected each other in the past, we have been J57 25 able to approach familiar historical events in a different and J57 26 revealing light. Sir John Seeley remarked as long ago as 1895 that J57 27 the interaction between Scotland, Ireland, and England was so J57 28 extensive in the 1640s that he wondered whether the civil war J57 29 "had really its origins in the necessity of revising their J57 30 mutual relations." But it is only in the past few years J57 31 that this insight has been pursued to the full.

J57 32 These are substantial gains. But while acknowledging them as J57 33 such, we also need to be aware of the problems and limitations J57 34 inherent in this approach to the British past. To begin with, some J57 35 of its practitioners are undoubtedly swayed by current political J57 36 preoccupations, and this can lead to a certain amount of special J57 37 pleading. Especially since the 1960s, both the Welsh and the J57 38 Scottish nationalist movements have increased in size and J57 39 self-consciousness (as simultaneously has support for an J57 40 independent Basque country and Catalonia in Spain and for separate J57 41 Breton and Occitanian nations in France). In addition, one of the J57 42 consequences of Margaret Thatcher's long premiership, which saw a J57 43 savage reduction in Tory electoral support in Scotland, and a less J57 44 dramatic but still significant fall in Tory support in Wales, has J57 45 been the reemergence of a right-wing Little Englandism. (The Labour J57 46 party, for reasons that will become clearer later in this essay, J57 47 remains emphatically British in its electoral base and ideology.) J57 48 Put crudely, the current political situation has encouraged some J57 49 English scholars to view the Welsh and the Scots as the Other in a J57 50 more deliberate fashion than before, and vice versa. If we add to J57 51 this the fact that Protestant Britons have traditionally viewed the J57 52 predominantly Catholic Irish as the Other, and have been so viewed J57 53 in return, it is easy to see why the appeal of a Four Nations view J57 54 of the United Kingdom can seem so overwhelming quite J57 55 independent of its scholarly value. Such an approach can J57 56 reduce Britishness to the interaction of four organic and J57 57 invariably distinct nations (or three if Ireland is left out of the J57 58 story). As such, it can sit comfortably not only with Welsh, J57 59 Scottish, and Irish nationalism but also with a newly assertive J57 60 English nationalism.

J57 61 The breakup of Britain, or at the very least the emergence of a J57 62 federal Britain existing as part of a federal Europe, may well be J57 63 desirable goals for the 1990s. I am not concerned here with J57 64 vindicating unionism or to argue for its continuation in the J57 65 future. But I would argue that the Four Nations approach, if pushed J57 66 too hard or too exclusively, is an incomplete and anachronistic way J57 67 to view the British past and, also, a potentially parochial one.

J57 68 It conceals, if we are not careful, the fact that the four J57 69 parts of the United Kingdom have been connected in markedly J57 70 different ways and with sharply varying degrees of success. Most J57 71 conspicuously, Ireland, as a whole, was only part of the Union J57 72 between 1800 and 1920. It has always been divided from the British J57 73 mainland by the sea and since the sixteenth century has been J57 74 severed from it even more brutally by its strictly limited response J57 75 to the Protestant Reformation. There is considerable evidence that J57 76 at grass-roots level the Welsh, the Scottish, and the English saw J57 77 (and often still see) the Irish as alien in a way that they did not J57 78 regard each other as alien. None of this means that we should J57 79 ignore Ireland's many and important political, cultural, and J57 80 economic links with Britain. But we should recognize that, mainly J57 81 for religious reasons, the bulk of its population was never swept J57 82 into a British identity to the degree that proved possible among J57 83 the Welsh, the Scots, and the English. We also need to recognize J57 84 that, until the late nineteenth century, at least, the majority of J57 85 people in all of these countries were never simply and invariably J57 86 possessed by an overwhelming sense of their own distinctive J57 87 identity as Englishmen, as Scotsmen, as Welshmen, or even as J57 88 Irishmen. As in the rest of Europe, intense local and regional J57 89 loyalties were always there to complicate and compromise.

J57 90 Even in the early 1800s, for example, and despite the enormous J57 91 impact of Sir Walter Scott's heroic evocation of the lochs and J57 92 glens of the North, some Lowland Scots still automatically referred J57 93 to their Highland neighbors as savages or as aborigines. They J57 94 regarded them, as they had traditionally done, as impoverished and J57 95 violent, as members of a different and inferior race, rather than J57 96 as fellow Scots. Conversely, whereas the word 'sassenach' is now J57 97 one of the kinder epithets used by all Scots to refer to the J57 98 English, before 1800 the Gaelic sasunnach (meaning a J57 99 Saxon) was commonly employed by Highland Scots to refer to J57 100 Lowlanders in general as well as to the English. Quite logically in J57 101 ethnic terms, Highlanders could view both Lowland Scots and the J57 102 English as foreigners. By the same token, the inhabitants of J57 103 northern England had (and still have) far more in common with their J57 104 Lowland Scottish neighbors than with the inhabitants of southern J57 105 England. They read the same books, ate the same kind of food cooked J57 106 in similar ways, frequently intermarried, and shared similar J57 107 literacy levels. Much the same could be said of men and women J57 108 living in Herefordshire and Shropshire with regard to their Welsh J57 109 neighbors. Here, again, people living close to the border, whether J57 110 on the Welsh or on the English side, could have more in common with J57 111 each other than with the rest of their respective countrymen. As J57 112 Hugh Kearney has demonstrated, with a scrupulous honesty that J57 113 threatens at times to undermine his own arguments, imposing a J57 114 strict three- or four-nation model onto these intricate and myriad J57 115 regional alignments is difficult and distorting. In practice, men J57 116 and women often had double, triple, or even quadruple loyalties, J57 117 mentally locating themselves, according to the circumstances, in a J57 118 village, in a particular landscape, in a region, and in one or even J57 119 two countries. It was quite possible for an individual to see J57 120 himself as being, at one and the same time, a citizen of Edinburgh, J57 121 a Lowlander, a Scot, and a Briton.

J57 122 The invention of a British national identity after 1700 did not J57 123 obliterate these other, older loyalties. True, both before and J57 124 after that date, London was always ready to employ military force, J57 125 parliamentary legislation and various kinds of indoctrination to J57 126 limit the autonomy of a few, particularly dangerous regions - the J57 127 'pacification' of the Scottish Highlands after 1746 would be an J57 128 obvious example. But Britishness was never just imposed from the J57 129 center, nor can it be understood solely or even mainly as the J57 130 result of an English cultural or economic colonization of the J57 131 so-called Celtic fringe. The extent of such anglicization has, to J57 132 begin with, often been exaggerated. Scotland always preserved its J57 133 own religious, educational, and legal structures and its own J57 134 sophisticated network of printing presses and cultural centers, J57 135 while even in the 1880s, some 350 years after the Act of Union J57 136 between Wales and England, three-quarters of all Welshmen still J57 137 spoke their own language out of choice. More broadly, though, we J57 138 need to stop thinking in terms of Britishness as the result of an J57 139 integration and homogenization of disparate cultures. Of course, a J57 140 degree of integration did occur, mainly by way of the advance of J57 141 communications, the proliferation of print, the operation of free J57 142 trade throughout the island, and a high level of geographical J57 143 mobility. But what most enabled Great Britain to emerge as an J57 144 artificial nation, and to be superimposed onto older alignments and J57 145 loyalties, was a series of massive wars between 1689 and 1815 that J57 146 allowed its diverse inhabitants to focus on what they had in J57 147 common, rather than on what divided them, and that forged an J57 148 overseas empire from which all parts of Britain could secure real J57 149 as well as psychic profits.

J57 150 It is this vital and external dimension of British development J57 151 that is most likely to be obscured by too narrow a concentration on J57 152 the Four Nations model. The interaction of Wales, Scotland, J57 153 Ireland, and England is an important and fascinating theme and is a J57 154 particularly pertinent one at the end of the twentieth century. But J57 155 in the eighteenth and nineteenth centuries, Britishness was forged J57 156 in a much wider context. Britons defined themselves in terms of J57 157 their common Protestantism as contrasted with the Catholicism of J57 158 Continental Europe. They defined themselves against France J57 159 throughout a succession of major wars with that power. And they J57 160 defined themselves against the global empire won by way of these J57 161 wars. They defined themselves, in short, not just through an J57 162 internal and domestic dialogue but in conscious opposition to the J57 163 Other beyond their shores.

J57 164 II J57 165 The absolute centrality of Protestantism to the British J57 166 experience in the 1700s and long after is so obvious that it has J57 167 often been passed over. Historians, always reluctant to be seen to J57 168 be addressing the obvious, have preferred to concentrate on the J57 169 more subtle divisions that existed within the Protestant community J57 170 itself, on the tensions between Anglicans and nonconformists in J57 171 England and Wales, between Presbyterians and Episcopalians in J57 172 Scotland, and between the older forms of Dissent and newer versions J57 173 such as Methodism. These internal rivalries were abundant and J57 174 serious. But they should not obscure what remained the towering J57 175 feature in the religious landscape, the gulf between Protestant and J57 176 Catholic.

J57 177 Even after the beginning of large-scale Irish immigration, the J57 178 Catholic community on the British mainland was a small one, and its J57 179 members were usually able to socialize with their Protestant J57 180 neighbors, own land, earn a living, and even attend mass openly. J57 181 Yet in terms of prejudice, none of this mattered very much. J57 182 Irrespective of their real strength and of how they were treated as J57 183 individuals, Catholics as a category remained in popular mythology J57 184 an omnipresent menace. Every November 5 until 1859, worshipers at J57 185 virtually all Protestant places of worship in England and Wales J57 186 would be reminded that it had been a Catholic who had tried to blow J57 187 up James I and Parliament back in 1605. In England, Wales, and J57 188 Scotland, almanacs, sermons, and popular histories made the point, J57 189 year after year, that it had been a French Catholic Queen, J57 190 Henrietta Maria, together with her interfering priests, who had led J57 191 Charles I astray and the whole island into war; that the would-be J57 192 tyrant James II had been Catholic, just as those responsible for J57 193 the Saint Bartholomew's massacre in 1572, or the Irish 'massacres' J57 194 of 1641, or the Great Fire of London in 1666 had been Catholic J57 195 also. "While Britain continues to be a nation," J57 196 wrote a Scottish pamphleteer at the end of the eighteenth century, J57 197 "she ought never to forget."

J57 198 J58 1 <#FROWN:J58\>It could be argued from Hutton's own evidence that J58 2 many of the problems facing Charles II, especially in his early J58 3 years, were the result of the legacy of the civil war and of the J58 4 difficulty at the Restoration of finding a satisfactory settlement J58 5 that could accommodate the interests of former Royalists, J58 6 Parliamentarians and ex-Cromwellians, Anglicans and Dissenters J58 7 alike. Certainly the religious tensions generated by the issue of J58 8 Dissent and the unsatisfactory Church settlement perpetually J58 9 bedeviled the government in its dealings with all three realms J58 10 throughout the reign. However, Hutton sees the year 1673 as marking J58 11 a crucial turning point in Restoration politics, with the duke of J58 12 York's public profession of his Roman Catholicism after the passage J58 13 of the Test Act and his marriage to a Catholic princess ushering in J58 14 a new period of political instability, centering around the problem J58 15 of the Catholic succession. The succession issue was to lead to the J58 16 clash between Whigs and Tories after the revelations of the Popish J58 17 Plot in 1678, but whereas Hutton is prepared to accept the J58 18 existence of a "Whig party" (p. 422), he is adamant J58 19 that "there was no 'Exclusion Crisis' at all" (p. J58 20 357), because, he says, the situation never became critical for the J58 21 government. This argument is somewhat strained, since by Hutton's J58 22 own account the government faced a series of acute difficulties J58 23 during the Exclusion period that virtually crippled its ability to J58 24 govern properly: there were problems in forming and keeping J58 25 together an effective ministerial team; it was impossible to get J58 26 anything done through Parliament; Charles was unable to conduct any J58 27 meaningful foreign policy; and the government faced the opposition J58 28 of an important section of the nation.

J58 29 Although we learn a lot from this book about what Charles and J58 30 his various ministers actually did, we do not get a very clear J58 31 sense of the nature or essence of royal power at this time. The J58 32 position of the monarchy is said to have been fundamentally strong, J58 33 and even to have gained in strength in the last years of Charles's J58 34 life, but the basis of this strength is never properly analyzed. J58 35 Rather than placing sole emphasis on the inherent powers of the J58 36 Crown, I would suggest there was also a crucial ideological J58 37 dimension to the strength of the monarchy in the first half of the J58 38 1680s. After the Exclusion scare, Charles made every effort to J58 39 cultivate public opinion - through propaganda, public J58 40 pronouncements, and even royal entries and processions. He J58 41 portrayed himself as a king committed to the rule of law and the J58 42 defense of the Church of England against the subversive threat of J58 43 the Whigs and Nonconformists, thereby attempting to recapture the J58 44 soft Anglican middle ground that had become alienated from the J58 45 Crown during the 1670s. The monarchy was stronger in 1684 than in J58 46 1679 because more people supported what it was doing; that support J58 47 was achieved partly as a result of a successful public relations J58 48 exercise but also partly as a result of a shift in policy by the J58 49 king himself, with Charles at last fulfilling the role of a J58 50 Cavalier-Anglican monarch that so many of his subjects wanted him J58 51 to play. Hutton has all the material at his disposal to discuss J58 52 such questions; it is disappointing that he decided not to do so in J58 53 a direct matter. Instead he seems to have seen his task as getting J58 54 the facts straight for his readers so that they could be in a J58 55 position to draw their own conclusions from the evidence he has J58 56 chosen to present.

J58 57 Two of the other books under review look at individuals whose J58 58 lives and influence spanned the crucial period between the civil J58 59 war and the 1680s. Conal Condren focuses on George Lawson - long J58 60 thought to be significant for his criticism of Hobbes and influence J58 61 on Locke - and on his tract Politica, which was published J58 62 first in 1660 and again in 1689. Condren's book operates on a J58 63 number of levels: in addition to being a work of political J58 64 philosophy and of the history of ideas, it is also a study in J58 65 political linguistics, and it has much to say on the use of J58 66 metaphor and rhetoric and on Lawson's manipulation and subversion J58 67 of existing political vocabularies. As a result the argument is J58 68 complex and not easy to summarize. Lawson, a clergyman who worked J58 69 most of his life in Shropshire, was a supporter of Parliament who J58 70 not only found it possible to accommodate himself to the changing J58 71 political and religious regimes of the 1640s and 1650s but could J58 72 accept the Restoration of monarchy in 1660 and work within the J58 73 reestablished national Church as well. Although the Politica J58 74 was in gestation for some time before its publication, Condren J58 75 shows that its context belongs very much to early 1660; therefore J58 76 it should not (as has been suggested) be seen as a defense of the J58 77 Commonwealth or of Cromwell but, rather, as a settlement tract that J58 78 sought to diffuse the divisiveness of constitutional and religious J58 79 differences on the eve of the Restoration. Thus Lawson's ideal J58 80 governmental form allowed for any variation of rule by king, Lords, J58 81 or Commons, or even a republican elite. Likewise any church form J58 82 might do in a pinch, so long as it did not lock itself into the J58 83 rhetoric of divine origin. Condren sees Lawson's arguments as in J58 84 many ways reflecting Clubmen rhetoric, with loyalty to the J58 85 community and the country emerging as major themes in his work. As J58 86 many will know, Lawson made a distinction between real and personal J58 87 sovereignty, with the former being invested with the people as a J58 88 community, and the latter being the attribute of a particular J58 89 governmental form. His precise views on the limits of people's J58 90 subjection, however, are not easy to unravel (necessarily so, J58 91 perhaps, given that Lawson's stress was on settlement and J58 92 subjection), so Condren prefers to offer two competing readings of J58 93 Lawson's position on resistance (the 'cobweb' and the 'seesaw' J58 94 hypotheses), which he feels circumscribe the limits of J58 95 plausibility. Nevertheless, Condren maintains that Lawson J58 96 essentially had a theory of dissolution rather than resistance per J58 97 se.

J58 98 Different aspects of this book will appeal to different types J58 99 of scholars in varying ways. There is a fascinating section on J58 100 Lawson's interpretation of the civil war, which he saw as having J58 101 primarily religious causes - although Condren warns us that the J58 102 attempt to separate discrete factors (religious, constitutional, J58 103 economic) is really misguided, given the way contemporaries J58 104 conceptualized their world. He is skeptical of the impact that J58 105 Lawson is often thought to have had on Locke. Rather it is John J58 106 Humfrey, that reluctant Nonconformist, who emerges as the main J58 107 vehicle for the transmission of Lawsonian ideas between the J58 108 Restoration and the Glorious Revolution. There is also a good J58 109 discussion of the importance of the Politica in the J58 110 "Allegiance Controversy," which reminds us how J58 111 relevant the rhetoric of interregnum religious and political J58 112 discourse still was to the world of 1689.

J58 113 Toward the end of his study, Condren offers an intriguing, J58 114 though all too brief, criticism of the way other scholars have J58 115 deployed the label 'radical' and its siblings 'moderate' and J58 116 'conservative.' None of these, he argues, are particularly helpful J58 117 to our understanding of either Lawson or his text, since they are J58 118 anachronistic terms that apply to groups in nineteenth-century J58 119 politics in a way that would not be acceptable to the world of the J58 120 seventeenth. This important point has been made many times before, J58 121 though there is no harm in saying it again, so long as it does not J58 122 blind us to the fact that the late seventeenth century did have its J58 123 own concept of 'radical,' even if it was very different from that J58 124 which came into existence after the French Revolution. A theory of J58 125 government that placed ultimate sovereignty in the people was, so J58 126 far as the Tories of the Exclusion period were concerned, a J58 127 'radical' theory. As one author alleged, the Whigs did their best J58 128 to infuse into people's minds the belief "that Power is J58 129 radically and revokably in them." Thus, according to Whig J58 130 theory, "they [the People] do not absolutely part with this J58 131 their so natural Right, but commit onely<&|>sic! the Administration J58 132 of such Power as is radically in them to others. But they retain to J58 133 themselves much of this Right, as upon the Male-administration of J58 134 the Power so delegated, they may revoke the Delegation, and take J58 135 all the Power into their own hands again." Many of the more J58 136 extreme Whigs, Locke among them, would qualify as espousing a J58 137 'radical' theory in this sense of the term; whether Lawson would no J58 138 doubt partly depends on whether one has a cobweb or seesaw reading J58 139 of the Politica.

J58 140 Not all are going to accept the conclusions of this book. J58 141 Julian Franklin, a leading scholar of Lawson and Locke who comes in J58 142 for serious criticism from Condren, has already replied in print by J58 143 identifying what he sees to be a number of errors of J58 144 interpretation. More worrying, I fear, is the fact that most people J58 145 will have terrible trouble just understanding what Condren is J58 146 trying to argue. There is a certain irony in the fact that a J58 147 student of linguistics should find it so difficult to articulate J58 148 his views in a clear and comprehensible manner, and on one occasion J58 149 even be forced to give up words altogether and start writing in J58 150 mathematical equations (p. 108). This is too bad, since I fear many J58 151 will be put off by the style and therefore miss the many J58 152 stimulating and provocative ideas that Condren's book undoubtedly J58 153 contains.

J58 154 Jonathan Scott, by contrast, has produced a lively and highly J58 155 readable account of that great seventeenth-century 'radical,' J58 156 Algernon Sidney. (The term 'radical' still seems appropriate in J58 157 this case, even after reading Condren.) Although at times Scott's J58 158 prose styles is rather rough and even colloquial (contractions and J58 159 split infinitives abound), he writes with pace, verve, and a degree J58 160 of wit that ensures his reader remains perpetually enthralled with J58 161 his argument. The present book, which is the first in a two-part J58 162 study, concentrates on the years of Sidney's life up until 1677 and J58 163 contains an analysis of his other major political treatise, the J58 164 Court Maxims written in 1665. The sequel (which was not J58 165 available at the time of writing this review) is to be entitled J58 166 Algernon Sidney and the Restoration Crisis, 1677-1683, J58 167 and will contain a detailed analysis of the more famous J58 168 Discourses. Not that Scott adheres to the strict chronological J58 169 limits that the titles of his respective volumes suggest. The main J58 170 purpose of the first study is to examine what Sidney believed and J58 171 how he came to hold such beliefs, and this is achieved through a J58 172 combination of biographical narrative, analysis of his intellectual J58 173 influences, and exposition of his ideas, which takes us back and J58 174 forth across the whole of Sidney's lifetime and has much to say J58 175 about the Discourses as well. This is an admirable book: rich, J58 176 clever and provocative in its revisionism, and to judge from the J58 177 glimpses we are offered here, the subsequent volume promises to J58 178 dust off the old cobwebs and set the seesaw rocking through a J58 179 powerfully argued reconceptualization of the Exclusion period.

J58 180 The main thrust of Scott's argument, which explains why a J58 181 separate volume on the period up to 1677 is necessary, is that J58 182 Sidney was not an Exclusionist Whig whose ideas were formed in J58 183 response to the so-called Exclusion Crisis; rather, his attitudes J58 184 were critically shaped by the age of the English Republic (1649-53 J58 185 and 1659) and his struggles during the period 1635-77 more J58 186 generally. Scott shows that the traditional view of Sidney as the J58 187 great popular and patriot philosopher, who was a reformer and a J58 188 moderate rather than a radical, has got the man seriously wrong. J58 189 Not only was he "one of the most passionate and bellicose J58 190 rebels of his age" (p.3), but he was also far from being J58 191 "the perfect Englishman," since he spent half of J58 192 his life outside England, was deeply influenced by Continental J58 193 ideas and his internationalist perspective, and was engaged among J58 194 foreign princes and republics in a variety of acts of treason J58 195 against his own country. J58 196 J59 1 <#FROWN:J59\>Walter Benjamin would ultimately admonish the J59 2 modern intellectual's ambiguous politics, exemplified by this kind J59 3 of neutralization of race, gender, and class allegiances. Western J59 4 intellectuals, he observed, did not see themselves as J59 5 "members of certain professions" but as J59 6 representatives of a "certain characterological J59 7 type," a type located somewhere between the classes. J59 8 Advocating a more activist role for the intellectual, Benjamin J59 9 called "for the transformation of the forms and instruments J59 10 of production in the way desired by a progressive intelligentsia - J59 11 that is, one interested in freeing the means of production and J59 12 serving the class struggle."

J59 13 Piper's realignment of the artist/spectator relationship rests J59 14 on her desire to work beyond such a characterological suspension of J59 15 the artist's connection to the rest of humanity. This challenge to J59 16 the forms and instruments of art production resulted both in a J59 17 withdrawal from the precious art object and a travesty of modernist J59 18 conceits: the 'apolitical' intellectual, the 'classless' dandy, and J59 19 the 'objective' fl<*_>a-circ<*/>neur who wanders the streets J59 20 of Paris observing the heroism of modern life are for Piper the J59 21 subject of parody and even derision. Take, for example, her J59 22 Mythic Being. Masquerading in dark glasses, Afro, and J59 23 pencil mustache, the Mythic Being was Piper's male alter J59 24 ego. Neither dandy nor fl<*_>a-circ<*/>neur - yet strangely J59 25 reminiscent of both - the Mythic Being represented J59 26 himself as tough black street kid who engaged in charged and J59 27 sometimes hostile encounters with strangers. Piper sent him into J59 28 white middle- and upper-middle class social contexts - theaters, J59 29 gallery receptions, museum exhibitions - in order to observe racist J59 30 patterns of avoidance and aggression. After moving to Cambridge, J59 31 Massachusetts, in 1974 to begin graduate study in philosophy at J59 32 Harvard University, Piper involved her Mythic Being in J59 33 such actions as 'cruising' the streets to experience male J59 34 sexuality, wandering the Cambridge commons, and venting class J59 35 antagonisms by mugging a male accomplice in public view after a J59 36 provocative conversation.

J59 37 If Piper's 'whiteness' allowed her to escape from the cruelest J59 38 side of racism, the racial and class specificity of her male alter J59 39 ego left her particularly vulnerable. Gaining a sense of her J59 40 "own marginality as a nonwhite (but not obviously black) J59 41 member of society," Piper realized that she was now even J59 42 more threatening to white society at large. As Homi Bhabha J59 43 observes, "the black presence ruins the representative J59 44 narrative of Western personhood: its past tethered to treacherous J59 45 stereotypes of primitivism and degeneracy will not produce a J59 46 history of civil progress, a space for the Socius; its J59 47 present, dismembered and dislocated, will not contain the image of J59 48 identity that is questioned in the dialectic of mind/body and J59 49 resolved in the epistemology of 'appearance and reality.' The white J59 50 man's eyes break up the black man's body and in that act of J59 51 epistemic violence its own frame of reference is transgressed, its J59 52 field of vision disturbed."

J59 53 Ultimately, Mythic Being personifies the mythic J59 54 black man - the presence who ruins the representative J59 55 narrative of white America. He is the being for whom miscegenation J59 56 laws were invented, codes that pretended to 'protect' white women J59 57 "but left black women the victims of rape by white men and J59 58 simultaneously granted to these same men the power to terrorize J59 59 black men as a potential threat to the virtue of white J59 60 womanhood." The tragic irony of this social equation is J59 61 that black men's 'power,' though threatening to white society, is J59 62 most often fictive and allusive; since male power within patriarchy J59 63 is relative, poorer men, frequently men of color, are most often J59 64 denied the material and social rewards of their participation in J59 65 patriarchy. With her Mythic Being, Piper conflated issues J59 66 of race and gender in order to question problematic feminist J59 67 constructions that pit white women against white men in a struggle J59 68 for power. Any understanding of patriarchal relations which J59 69 examines the power of men over women outside of broader racial and J59 70 economic issues is, of course, problematic. Arguments that feature J59 71 patriarchy as the primary determinant of women's oppression fail to J59 72 see "the inapplicability of such a concept in analyzing the J59 73 complex of relations obtaining in the Black communities both J59 74 historically and at present." Such discourses of J59 75 patriarchal relations, for example, most often ignore the struggle J59 76 that men and women of color must fight together against the J59 77 white ruling class. Mythic Being - a black woman in the J59 78 guise of a black male youth - metaphorically represents this J59 79 alliance, reminding us that in Western society racism effortlessly J59 80 crosses the boundaries of gender.

J59 81 Piper's desire to speak over the oppressive master texts of J59 82 modernism results in the blasting of another aspect of the artist's J59 83 mythology - the specter of anonymity. Like Frantz Fanon's J59 84 existentialist evocation of the 'I' that restores the presence of J59 85 the marginalized (and his own presence within dominant narratives J59 86 that would silence a black man's voice), Piper at times speaks J59 87 through personal or autobiographical narratives. The Big J59 88 Four-Oh (1988), a self-portrait video installation, juxtaposes J59 89 a display of materials soaked in her bodily fluids, an open J59 90 journal, a suit of armor, forty hard balls, and a repeat videotape J59 91 of Piper with her back to the camera dancing nonstop to soul music. J59 92 The work, which represents the artist at the time of her fortieth J59 93 birthday, affirms her complexity and strength, each of its parts J59 94 serving as a metaphor for an aspect of her existence. This J59 95 installation would seem to continue the project of the earlier J59 96 Political Self Portraits (1980) in which Piper J59 97 superimposed images of her face and body over detailed J59 98 autobiographical accounts written from the three perspectives that J59 99 define her marginalization - class, race, and gender. A chronology J59 100 of the artist's life written by Piper and published in the catalog J59 101 for her recent retrospective exhibition includes the following J59 102 note: "This chronology was created solely by Adrian Piper J59 103 and is presented as part of her artistic work." Rather than J59 104 narcissistic evocations of the self or acts of romanticized J59 105 self-expression, autobiography serves a crucial J59 106 political function in Piper's <*_>oe-ligature<*/>uvre. As part J59 107 of a broader structure for dismantling oppressive systems, such J59 108 narratives acknowledge the extent to which marginalized peoples are J59 109 spoken at and for, the degree to which people of color, women, gay J59 110 men, and lesbians have been defined and judged by the narrow J59 111 standards of a dominant culture governed by white heterosexual J59 112 males. Indeed, who constructs the master narratives of culture? Who J59 113 are the patrons of academia, the publishers, the financiers of J59 114 industrial society? Who writes history?

J59 115 Marginalized peoples, of course, are generally excluded from J59 116 defining their own role in the narratives of history. Women of J59 117 color, for example, have been alternately categorized as exotic or J59 118 'pathological' or both - universalizing conditions that deny J59 119 difference as they create stereotypes of passivity or abnormality: J59 120 "But what does the sameness of the exotic women represent? J59 121 Female heroism, humor, carnivalesque gesture, triumph, movement ... J59 122 'trans-cultural, trans-historical, trans-social' - J59 123 exotic. Here exoticism marks a universality which J59 124 systematically negates the very raison d'<*_>e-circ<*/>tre of J59 125 women's different experiences, strategies and actions." J59 126 Like the mythologized dandy, these ciphers of universality, J59 127 exoticism, or sameness mask deeper political motives. As such, J59 128 autobiographical writing can serve an important therapeutic J59 129 function for marginalized peoples. The fear and uncenteredness J59 130 associated with psychic and physical oppression can often be J59 131 overcome or helped by reconnecting with the personal narratives of J59 132 the past. Remembering can be part of a cycle of reunion, observes J59 133 Bell Hooks, "a joining of fragments, 'the bits and pieces J59 134 of my heart' that the [autobiographical] narrative made whole J59 135 again." Such speech will always be difficult for the J59 136 dominant culture to accept. Representational marginalization J59 137 exempts the exotic 'others' - whether they be women, gays, blacks, J59 138 or even artists - from serious consideration by the ruling class. J59 139 To allow such peoples to speak in their own voices is to risk J59 140 hearing their oppositional speech - discourses that demand rather J59 141 than passively accept, that scream rather than whisper.

J59 142 Although Piper's project is directed at a cultural community J59 143 that is mostly privileged, she reconstructs the ideological role of J59 144 the artist in a way that directs her audience to join her in the J59 145 struggle against racism and sexism. Indeed, she is always polite in J59 146 her address, always conscious of the psychological threshold of J59 147 complacent viewers who would rather look at 'art' than confront the J59 148 painful reality of their own racism. "I can't bear the J59 149 thought of violating the norms of etiquette," Piper has J59 150 said. "Such norms help me to grease my way ... through a J59 151 hostile white world." The inherent difficulty of J59 152 disseminating upsetting information makes this politeness a matter J59 153 of packaging for Piper; like a good advertising executive, she J59 154 understands that style is often as important as content in reaching J59 155 an audience. It is through this "power of passive J59 156 provocation" that Piper hopes to transform her audience J59 157 psychologically, by presenting them with "an immediate and J59 158 unavoidable concrete reality that cuts through the defensive J59 159 rationalizations by which we insulate ourselves against the facts J59 160 of our political responsibility. I want viewers of my work to come J59 161 away from it with the understanding that their reactions to racism J59 162 are ultimately political choices over which they have control - J59 163 whether or not they like the work or credit it for this J59 164 understanding."

J59 165 To this end, Piper embarked on a series of audience J59 166 performances in which predominantly white, art world groups were J59 167 engaged in various consciousness-raising activities. The J59 168 ground-breaking Funk Lessons, for example, began as a J59 169 question in Piper's mind: Why are white people indifferent, even J59 170 hostile, to soul music? "I found that response [to funk J59 171 music] so often," Piper observes. "It seems to me J59 172 there was a gap between the purported attitude of openness and J59 173 receptivity to popular culture that is usually espoused by the art J59 174 world, according to which anything is adequate subject matter J59 175 for appropriation and reuse within the context of high culture. And J59 176 what actually seemed to be the case is that in fact only J59 177 some things can have that function, and in particular black J59 178 working-class culture cannot have that function." J59 179 Having incorporated funk music into earlier politically oriented J59 180 performances, Piper found that white audiences misunderstood her J59 181 motives or attacked her use of this music as "cheapening" J59 182 the serious political content of the performance. Such resistance, J59 183 Piper suggests, is rooted in several areas: "One problem J59 184 has to do with the overt sexuality of that music - it talks about J59 185 fucking, it talks about making love, it talks about bodies, and J59 186 it's very hard to assimilate that in a way that's not threatening J59 187 to white upper-middle-class culture. Another problem is that it J59 188 requires a very highly structured use of one's body in order to J59 189 respond to it."

J59 190 Piper decided to 'educate' white audiences about the J59 191 significant (if not always acknowledged) role played by this music J59 192 in both dominant and marginal culture. Funk Lessons was J59 193 structured in an academic format as a participatory performance J59 194 with Piper as instructor and audience members as students. Piper J59 195 distributed a bibliography and photocopied handouts that listed J59 196 some of the 'characteristics' of funk dance and music. She J59 197 proceeded to discuss certain mainstream presumptions about funk J59 198 music in an attempt to free her audience from discomforting J59 199 misconceptions. She then led the group in body isolation exercises, J59 200 discussed the structure of the music, and practiced dance movements J59 201 with musical accompaniment. (An extraordinary video version of J59 202 Funk Lessons centers on a particularly successful J59 203 performance at the University of California at Berkeley in 1983. J59 204 Piper augments film of the Berkeley evening with voice-over and J59 205 on-screen commentary and archival footage of influential soul J59 206 performers [e.g., James Brown, Little Richard, and Aretha Franklin] J59 207 and the white entertainers [Elvis Presley, Mick Jagger, The Talking J59 208 Heads] who capitalized on the black funk idiom. The video yields a J59 209 level of meaning somewhat independent of but not unrelated to the J59 210 performance - that of the mass-media suppression of working-class J59 211 black culture and its comfortable, and profitable, appropriation by J59 212 white artists.) Ultimately, Piper did not want to scare or J59 213 intimidate her audience; rather she constructed a J59 214 "comfortable and safe" format that encouraged J59 215 people to explore their apprehensions about the music and their J59 216 ability to soul dance. Individual audience reactions to Funk J59 217 Lessons varied from antagonism and resistance to jubilation. J59 218 Successful performances culminated in a celebratory dance party; J59 219 failed ones fizzled out in a morass of confusion and resentment.

J59 220 J60 1 <#FROWN:J60\>Lady Elisabeth's disappointment in him was most vocal J60 2 by the middle of that year. Talbot was undoubtedly affected, at J60 3 least to some degree, by the general state of affairs in Britain. J60 4 In 1820, the 'Swing Riots' swept through Wiltshire and other areas J60 5 of the country. They were the inevitable outcome of the intolerable J60 6 economic pressures that increasing industrialization was inflicting J60 7 on agrarian workers. As a result of his compassionate and J60 8 surprisingly expert management of Lacock, Talbot succeeded in J60 9 avoiding any local violence. He soon found that in trying to J60 10 alleviate the condition of the poor, he was out of step with the J60 11 local farmers. Reform was in the air. Henry Talbot expressed his J60 12 opinion that representation should be equitable; he stood for J60 13 Parliament in 1831 as a reformer, but was defeated on this first J60 14 try. Perhaps his entry into politics was less anomalous than it J60 15 might at first appear. Talbot's ancestors (and many of his current J60 16 relatives) had been deeply involved in politics. His cousin Kit, J60 17 Christopher Rica Mansel Talbot, had entered Parliament in 1830. J60 18 Lady Elisabeth considered her son's talents wasted in the game of J60 19 politics, but she would have encouraged him to at least do J60 20 something, anything, to break out of his ill humor. The passage of J60 21 the Reform Bill in 1832 increased his constituency, and Henry J60 22 Talbot was finally elected as a Member from Chippenham. He was to J60 23 serve only one term.

J60 24 Talbot's short political career proved unsatisfying but perhaps J60 25 also convinced him that a life in science was what he wanted. A J60 26 sure sign that Talbot was getting a grip on his doldrums came with J60 27 his March 1831 election as a Fellow of the Royal Society. Proposed J60 28 by his relatives Charles Lemon and William Fox-Strangways, Talbot's J60 29 election was supported by the testimony of Michael Faraday, Davies J60 30 Gilbert; George Peacock, Thomas Philipps, and William Whewell J60 31 (Herschel would have been happy to sign for Talbot as well save for J60 32 his continuing protest against the Society). Perhaps even this J60 33 short experience in politics was for the best; Talbot's point of J60 34 view on the relationship between science and society had been J60 35 expanded. In 1833, he wrote to the botanist William (later Sir J60 36 William) Jackson Hooker that "the difficulty which you J60 37 complain of, of getting any bookseller to publish a scientific work J60 38 on Botany is not confined to that science." Talbot J60 39 suggested that there was a societal benefit to be gleaned from J60 40 support:

J60 41 Abroad, not only is paper & printing cheaper, but J60 42 assistance is rendered by the Governments. In my opinion public J60 43 libraries ought to be established in all our principal Towns at the J60 44 national expense. A considerable sum should be voted annually for J60 45 the encouragement of science, which should be in part expended in J60 46 patronizing literary undertakings of merit. From 20 to 50 copies of J60 47 such works should be purchased by government & distributed to these J60 48 provincial libraries, which small at first, would soon become J60 49 important.

J60 50 Another sign that Henry Talbot was emerging from melancholy was J60 51 his December 1832 marriage to Constance Mundy, of Markeaton. The J60 52 youngest daughter of the MP for Derbyshire, she came from a family J60 53 possessed of more than the usual artistic interests and was a J60 54 reasonably accomplished amateur sketcher herself. Constance's J60 55 cheerful countenance was an important foil to Talbot's occasionally J60 56 brooding behavior. She was highly supportive of Henry's efforts and J60 57 proved to be a good moderator for the well-motivated ambitions of J60 58 Lady Elisabeth. However successful in this, Constance never became J60 59 the intellectual partner to Henry that Maggie was to John Herschel. J60 60 The evidence for this is distressingly clear. Since Henry spent J60 61 much time away from home, his correspondence to and from Constance J60 62 is voluminous. It is perhaps understandable that she carried on no J60 63 discourse about her husband's highly-specialized researches. But it J60 64 is extraordinary that Constance, as an amateur artist herself, J60 65 never once made any serious assessment of any of Henry's J60 66 photographs. Constance rarely commented on his subjects and J60 67 virtually never made suggestions for possible photographs. When J60 68 Constance was vacationing separately in Wales in 1835, at a time J60 69 when Henry's interest in his new discovery was at its first peak, J60 70 she wrote that "I wish you could have taken the outline of J60 71 the castle & fine elms behind just as I saw them - but I think you J60 72 told me you could not produce the desired effect by any light J60 73 except that of the sun." But even this breezy comment was J60 74 an exception. Her very occasional references to photography after J60 75 it became public in 1839 indicate that she made some attempts to J60 76 sensitize paper and make some prints; however, they betray a very J60 77 superficial understanding of how the process worked. Perhaps J60 78 Constance Talbot's artistic training exerted some influence on her J60 79 husband during his picture making, but one is disappointed in J60 80 efforts to trace hints of such an influence in their J60 81 correspondence. Constance's understanding of her husband's work, J60 82 including his photography, was conversational, and her role was to J60 83 be minimal. It would be an overstatement to call her a J60 84 photographer.

J60 85 Completing his article on colored flames in 1826, Talbot had J60 86 explained to Herschel, that "as I am not acquainted with Dr J60 87 Brewster perhaps you will have the kindness to transmit J60 88 it." The relationship between Talbot and Sir David Brewster J60 89 obviously flourished, for, judging by the surviving correspondence, J60 90 Brewster and Talbot were in regular contact at least by 1833 and J60 91 were good friends by 1836. They shared interests in light and J60 92 perhaps some personality traits as well. Brewster; as influential J60 93 as he was, struck many as an abrasive personality. He had many J60 94 admirers but few friends. Talbot was reclusive by nature but the J60 95 two men hit it off very well. A letter from Constance to Lady J60 96 Elisabeth written on the occasion of Sir David's visit to Lacock J60 97 Abbey in August 1836, is particularly revealing of Talbot' J60 98 character:

J60 99 You are perfectly right in supposing Sir D. B. to pass J60 100 his time pleasantly here. He wants nothing beyond the pleasure of J60 101 conversing with Henry discussing their respective discoveries & J60 102 various subjects connected with science. I am quite amazed to find J60 103 that scarcely a momentary pause occurs in their discourse. Henry J60 104 seems to possess new life - & I feel certain that were he to mix J60 105 more frequently with his own friends we should never see him droop J60 106 in the way which now so continually annoys us. I am inclined to J60 107 think that many of his ailments are nervous - for he certainly does J60 108 not look ill. I hear from Sir David that he distinguished J60 109 himself at the Meeting in a conversation on the Improvement of the J60 110 Telescope. ... When I see the effect produced in Henry by Sir D.B's J60 111 society I feel most acutely how dull must our ordinary way of life J60 112 be to a mind like his! - and yet he shuts himself up from J60 113 choice.

J60 114 Henry Talbot would never achieve the same levels of fame that J60 115 Herschel (unwillingly!) reached. His scholarly contributions were J60 116 generally less influential; much of this stemmed from the fact that J60 117 he rarely took on the more general questions of the organization J60 118 and role of science that Herschel approached. Unlike Herschel's J60 119 minutely detailed and lengthy treatises, Talbot's journal J60 120 publications tended to be very short. Frequently they reflected J60 121 scientific concerns common to the day and were just as often as J60 122 suggestive as they were declarative. Even so, Talbot made two J60 123 fundamental contributions to science before photography was even J60 124 announced. These stemmed from a growing awareness in Talbot's time J60 125 of the interrelationship between forms of energy and matter. Talbot J60 126 made one very practical contribution that is still in use today. In J60 127 July 1834, he read a paper to the Royal Society summarizing his J60 128 experiments on light: "I have lately made this branch of J60 129 optics a subject of inquiry, and I have found it so rich in J60 130 beautiful results as entirely to surpass my expectations." J60 131 In this, Talbot revealed his discovery of the polarizing J60 132 microscope; his method of placing one polarizer close to the J60 133 eyepiece and another below the stage is still considered highly J60 134 effective. It provides an increase in contrast between the subject J60 135 and the background and is an important tool in the analysis of J60 136 internal structure. Talbot applied it immediately to studying the J60 137 structure of crystals.

J60 138 Whereas the polarizing microscope was a discrete invention of J60 139 Talbot's, his work in spectral analysis ( the analyzing J60 140 of the physical makeup of substances through optical means) was J60 141 more in the nature of fundamental contributions to the beginnings J60 142 of an important branch of science. As Talbot was to remember in J60 143 later years:

J60 144 About the year 1824 or 1825, Dr. Wollaston gave one of J60 145 his evening parties, to which men of science and amateurs were J60 146 invited and it was the custom to exhibit scientific novelties, and J60 147 to make them the subject of conversation.

J60 148 On the evening in question I brought as my contribution to the J60 149 meeting some very thin films of glass (such as are shown in J60 150 glass-houses to visitors by a workman, who blows a portion of J60 151 melted glass into a large balloon of extreme tenacity, and J60 152 afterwards crushes the glass to shivers.) Such a film of glass I J60 153 brought to Dr. Wollaston and his friends, and after showing that in J60 154 the well-lighted apartment it displayed a uniform appearance J60 155 without any markings, I removed it into another room, in which I J60 156 had prepared a spirit lamp, the wick of which had been impregnated J60 157 with common salt. When viewed by this light, the film of glass J60 158 appeared covered with broad nearly parallel bands, which were J60 159 almost black, and might be rudely compared to the skin of a J60 160 zebra.

J60 161 Talbot's optical 'zebra skin' in fact provided some of the J60 162 earliest evidence of the presence of the lines of sodium. These J60 163 bands had been discovered by Josef von Fraunhofer (the same man who J60 164 was the cause of Herschel's and Talbot's first meeting). Henry J60 165 Talbot's first non-mathematical journal article, his 1826 J60 166 'Experiments on Coloured Flames,' was a foundation stone for J60 167 spectral analysis. The following year, Herschel incorporated J60 168 Talbot's findings into his treatise on light. Two types of objects J60 169 can be analyzed this way. In absorbent subjects (where the J60 170 effect of incident light is analyzed), important pioneering work J60 171 was done by Sir David Brewster and others. In the treatment of J60 172 self-luminous subjects (such as the sun), Talbot and Herschel J60 173 were the pioneers. Many of their subsequent publications returned J60 174 to this subject of study.

J60 175 Henry Talbot, through his substantial family connections and J60 176 through his own efforts, now exercised a certain amount of J60 177 political influence. An excellent demonstration of this was his J60 178 effective support for the establishment of a national botanic J60 179 garden at Kew. Long known as 'the Royal Vegetable Patch', Kew had J60 180 declined by 1838 to a most precarious position and was threatened J60 181 with closure. Talbot's longtime botanical friend, Sir William J60 182 Jackson Hooker (recently knighted for his botanical contributions), J60 183 led the fight to establish a national collection. Agreeing that J60 184 "it would be a pity indeed if Kew Gardens were to be J60 185 sacrificed to a pitiful & false economy," Talbot persuaded J60 186 his influential relatives to take up the cause. He met with the J60 187 Chancellor of the Exchequer, determined that petitions to J60 188 Parliament from learned societies would have great weight, and J60 189 proposed to the Council of the Linnean Society that they present a J60 190 petition to the House of Commons. His idea was accepted J60 191 unanimously. By May, Talbot could report to Hooker that the J60 192 Chancellor of the Exchequer "spoke confidently to me last J60 193 night that something satisfactory would be done about Kew Garden. J60 194 ..." Although he was leaving town and could be of no J60 195 further assistance, "I trust that it is put into a right J60 196 train, & will issue favourably." This support was perhaps J60 197 Talbot's greatest and most effective contribution to the politics J60 198 of science. Kew was saved and revitalized that year; in 1841, J60 199 Hooker became its new director.

J60 200 When considering the dynamics of the beginnings of photography, J60 201 to paint Talbot as being junior to Herschel, as is often done, is J60 202 terribly misleading. No one was like Herschel and only a few even J60 203 approached his status. J60 204 J61 1 <#FROWN:J61\>Instead of focusing just on how her sympathy feels, J61 2 she remains interested in understanding what provokes it, what J61 3 responses it provokes in others, how it reflects childhood J61 4 experience. Somewhere along the way, however, she finds that she J61 5 takes great pleasure in contemplating (if rather coolly) the J61 6 fact of her feeling sympathy when she does; there is something J61 7 exactly right about it, she finds, that makes it worth appreciating J61 8 apart from its mere psychological significance or effect on others: J61 9 like a dancer's gesture or a poet's turn of phrase, it is, somehow, J61 10 satisfying to behold. Or perhaps the response itself is nothing J61 11 special; it becomes interesting only through beholding it as J61 12 Bullough's aesthete beholds the fog at sea: she neither anxiously J61 13 interrogates it for clues about her psychological health or her J61 14 moral worth, nor is she so detached that she is indifferent to it J61 15 altogether. Instead, she has learned how to contemplate, with J61 16 appreciation and from the 'proper psychical distance,' the mere J61 17 fact of it. This, too, counts as an aesthetic meta-response.

J61 18 No doubt, there is something vaguely troubling about routinely J61 19 aestheticizing one's feelings. But even more troubling is the habit J61 20 of aestheticizing feelings that play a central role in our moral J61 21 lives.

J61 22 IV J61 23 Feagin's failure to note the aesthetic character of some of our J61 24 meta-responses underlies a further important difference between our J61 25 views. She thinks such responses are a good thing, from a moral J61 26 point of view, while I am not convinced they are. I am skeptical in J61 27 part because the aesthetic and moral points of view often conflict; J61 28 and when they do, the aesthetic often triumphs, if in unexpected J61 29 and subtle ways.

J61 30 The extent and obviousness of the conflict vary, as do the ways J61 31 we resolve it. Formalist purists go far enough when they proclaim, J61 32 with Clive Bell, that while content may or may not be harmful to J61 33 aesthetic experience, always it is irrelevant. Formalists of this J61 34 sort pay attention only to the look or sound or structure of J61 35 things; moral concerns, like content, are beside the point. But one J61 36 can go further. Consider the Futurists, whose sensibility echoes in J61 37 confused and disturbing ways through contemporary culture. One J61 38 Futurist goal seems to be to reject morality altogether by J61 39 aestheticizing (rather than ignoring) just those aspects of a J61 40 situation which would ordinarily be most appalling:

J61 41 We will glorify war ... militarism, patriotism, the J61 42 destructive gesture of anarchist<&|>sic!, the beautiful Ideas which J61 43 kill .... Art can only be violence, cruelty, and J61 44 injustice.

J61 45 Of course, most of us are even less likely to be Futurists than J61 46 formalist purists. If we would feel uneasy ignoring moral content J61 47 for the sake of aesthetic satisfaction, we would be outraged at the J61 48 glorification of immorality, in life or art. But compromises, J61 49 especially confusing ones, can be struck.

J61 50 One sort of compromise involves a kind of response that, unlike J61 51 the formalist's, attends to morally compelling content, but not for J61 52 the Futurist purpose of rejecting morality. This seems a step in J61 53 the right direction; paying attention to a work's morally J61 54 compelling content seems to be an improvement, from a moral point J61 55 of view, over ignoring it. But, as we have seen, Danto worries that J61 56 when we confront a work of art, we respond aesthetically even J61 57 to morally compelling content. And this is wrong, for Danto, J61 58 because representations of human misery demand moral rather than J61 59 aesthetic response. Indeed, one might even argue that J61 60 aestheticizing what we should respond to morally may be worse than J61 61 ignoring it as the formalist does.

J61 62 I do not doubt that the sort of response Danto describes J61 63 sometimes occurs. But just as Bell's formalism fails as a J61 64 descriptive theory by ignoring audiences' widespread concern with J61 65 content, Danto's view fails to recognize how fictional portrayals J61 66 of human misfortune may engage most people's moral responses. Danto J61 67 seems to imply that even if one writes a play for the sake of J61 68 making a moral difference, the project is doomed from the start; J61 69 injustice, as soon as it becomes the subject of the play, is put J61 70 "at a distance which is exactly wrong from a moral J61 71 perspective." However, if my analysis is correct, we may, J61 72 after all, have our moral cake and the aesthetic pleasure of eating J61 73 it, too: We respond to the injustice morally rather than J61 74 aesthetically, but then diminish the moral significance of this J61 75 first-order moral response by aestheticizing it.

J61 76 Something like this may be going on with Alice Walker's J61 77 character. That the sight of the black woman is for her a J61 78 weepy miracle suggests not the austere ecstasy that some J61 79 formalists talk about, but a pleasure that is born, however J61 80 inappropriately, of moral sentiments, in particular those of a J61 81 white liberal woman confronting a poor black woman. Here, though, J61 82 the character knows it is wrong to treat black people as art, as a J61 83 means to those weepy but enjoyable sentiments. So the character J61 84 falls somewhat short of a full-blown aesthetic meta-response: her J61 85 guilt keeps her from fully enjoying her sensations, let alone from J61 86 contemplating how fine they are.

J61 87 V J61 88 So what exactly is wrong with aesthetic meta-responses J61 89 to tragedy? To answer this question, we must look more closely J61 90 still at what counts as such a meta-response, and we must talk J61 91 about the moral significance of sympathy itself. We must also J61 92 consider whether such responses, even if inappropriate in 'real J61 93 life', are unobjectionable in our experience of fiction. The best J61 94 way into this labyrinth is through a more thorough look at Feagin's J61 95 view.

J61 96 That Feagin doesn't consider the possibility of aesthetic J61 97 meta-responses may make it easier for her to sing their moral J61 98 praises. Given the tension between the aesthetic and the moral J61 99 points of view, it may be easier to exonerate morally, if not J61 100 celebrate, the pleasures we take in tragedy if we do not understand J61 101 these pleasures to be aesthetic ones. Indeed, Feagin avoids the J61 102 traditional notion of the aesthetic altogether. When she talks J61 103 about aesthetic response and aesthetic pleasure, she is simply J61 104 using 'aesthetic' as a synonym for 'artistic'; she is not using the J61 105 language of aesthetic-response theorists. At one point she even J61 106 implies that differentiating between direct and meta-responses to J61 107 art may allow us to avoid talk about aesthetic response altogether. J61 108 Nevertheless, some of what Feagin says about these meta-responses J61 109 seems to indicate that she has aesthetic ones in mind. This, we J61 110 shall see, makes it easier to question their moral credentials.

J61 111 Why does Feagin find meta-response to tragedy praiseworthy? J61 112 First, she claims that the sympathy we enjoy when we respond to J61 113 tragedy also underlies our capacity for moral action. But does this J61 114 establish its special moral status? Here I think we must inquire J61 115 further into the grounds of our pleasure. One might say, "I J61 116 am glad I can sympathize; sympathy helps me to be a better person, J61 117 to care about others, to do what needs to be done. Helping others J61 118 is important, and sympathy is part of this, if not a means to this J61 119 end." Insofar as we sincerely give this sort of J61 120 explanation, our meta-response is a moral rather than an aesthetic J61 121 one. Furthermore, it seems more natural here to talk about J61 122 valuing or appreciating one's capacity to be moved J61 123 by the suffering of others than about taking pleasure in J61 124 or enjoying the feeling of being moved. The latter sort J61 125 of language, which suggests that we savor a feeling in abstraction J61 126 from a larger moral context, as we would the look of a painting or J61 127 the sound of a symphony, seems to be a paradigmatic case of the J61 128 sort of quasi-perceptual aesthetic meta-response I discuss J61 129 above.

J61 130 To be fair, I should note that Feagin does once use the word J61 131 'satisfaction.' But elsewhere her emphasis is different. Consider J61 132 how she explains the moral difference between taking pleasure in J61 133 one's reaction to fictional suffering and taking pleasure in one's J61 134 reaction to the real thing. When we sympathize with merely J61 135 fictional suffering, Feagin points out, our enjoyment of our J61 136 reaction does not have the price of other's pain:

J61 137 In real life, the importance of human compassion is J61 138 easily overshadowed by the pain of human suffering. It is not J61 139 possible in real life to respond to the importance of human J61 140 sympathy as a distinct phenomenon, since that sympathy depends on, J61 141 one might even say 'feeds on,' human misery. It is not, in life, an J61 142 unequivocal good. In art, however, one experiences real sympathy J61 143 without there having been real suffering, and this is why it is J61 144 appropriate to feel pleasure at our sympathetic responses to a work J61 145 of art, whereas it is not appropriate to feel pleasure at our J61 146 sympathetic responses in reality. There the sympathy comes at too J61 147 great a cost.

J61 148 This may be true enough. But here, again, the enjoyment of our J61 149 sympathy seems to be an end in itself, far removed from its moral J61 150 context - especially any connection to the practical action J61 151 sympathy often motivates in the real world. Indeed, given that J61 152 Feagin is talking about the moral significance of our J61 153 meta-response, it is odd that she doesn't say more here about the J61 154 connection between sympathy and action, if only to strengthen her J61 155 case by pointing out that we need not worry about rescuing a J61 156 fictional victim. On the other hand, if a concern with acting, when J61 157 we can and as we should, does not somehow figure into our J61 158 meta-response, how morally significant can it be?

J61 159 Another sign that Feagin has in mind an aesthetic meta-response J61 160 is her admission that this pleasurable meta-response would be J61 161 morally inappropriate in a real situation. But unless she means an J61 162 aesthetic meta-response, there would be no moral impropriety; J61 163 after all, if our meta-response reflects only a concern with doing J61 164 the right thing, there would be nothing wrong in saying, for J61 165 example, "I'm glad I could sympathize with her; it made her J61 166 feel less alone," or even "I'm glad that after all J61 167 I've been through I can still feel for people." We need not J61 168 regard another's pain as a means to a self-congratulatory, J61 169 narcissistic end; rather, through our feelings we may come to J61 170 understand and ease another's pain and, perhaps, become better J61 171 people. But this means just that not all meta-responses to J61 172 sympathy are aesthetic ones, and that the moral (as opposed to the J61 173 aesthetic) significance of sympathy does not rest in its being J61 174 valued as an end in itself.

J61 175 Determining the moral status of sympathy is, of course, an old J61 176 and thorny problem. J.S. Mill complained that some of his J61 177 contemporaries regarded the sympathies as "necessary evils, J61 178 required for keeping men's actions benevolent and J61 179 compassionate." Such a view certainly is too harsh. But J61 180 surely the moral significance of sympathy emerges as part of a J61 181 larger moral picture, including a commitment to understanding those J61 182 with whom one sympathizes and to acting when appropriate. Without J61 183 such connections, the feeling of being at one with humankind seems, J61 184 from a moral point of view, neither here nor there. If a world of J61 185 purely rational Kantian agents, completely dutiful and completely J61 186 unfeeling, is a grim prospect, a world of sensitive souls so J61 187 involved in sympathy that they are distracted from or uninterested J61 188 in understanding and acting on the situation at hand may be even J61 189 worse.

J61 190 Indeed, in the nineteenth century, obsession with sympathy drew J61 191 criticism on just these grounds. Historian Walter Houghton puts it J61 192 this way: the cult of benevolence, typified by Dickens and Eliot, J61 193 "took a new direction in the nineteenth century when the J61 194 misery of the industrial workers became sufficiently apparent to J61 195 demand redress - and all the more so because it constituted a J61 196 threat to social order. If one solution proposed [by Carlyle and J61 197 Arnold] was a more earnest sense of social duty, another lay in J61 198 quickening the moral sensibility to an acute sympathy for suffering J61 199 humanity." Such a sensibility, however, may care more for J61 200 sentiment than for understanding and eliminating the conditions J61 201 that give rise to it. Houghton sees this tendency culminating in J61 202 the aestheticism of Walter Pater, for whom morality was J61 203 "all sympathy," contemplation the end of life, J61 204 and the job of the artist to disengage our thoughts from life and J61 205 fix them, "with appropriate emotions, 'on the great and J61 206 universal passions of men.'"

J61 207 J62 1 <#FROWN:J62\>Nevinson's Elegy: Paths of Glory

J62 2 CHARLES E. DOHERTY

J62 3 Christopher Richard Wynne Nevinson (1889-1946) was among the J62 4 first British artists to witness the horrors of the First World J62 5 War. A volunteer ambulance driver on the western front, Nevinson J62 6 observed the suffering and carnage resulting from the new trench J62 7 warfare in the winter of 1914-15. He returned to the front in 1917 J62 8 as a member of the British government's official war-artists' J62 9 program. For the program Nevinson produced a controversial J62 10 painting, Paths of Glory (fig. 1), that was banned from J62 11 his one-person show at the Leicester Galleries, London, in March J62 12 1918. The exhibition took place at a time during the war when J62 13 morale seemed at its lowest, and the government supposedly censored J62 14 the painting because it portrayed dead British soldiers. Since J62 15 other contemporaneous paintings and photographs of Allied and enemy J62 16 war dead were shown in London galleries and reproduced in the J62 17 British press, the reason for the censorship of Nevinson's J62 18 Paths of Glory demands further investigation.

J62 19 Nevinson was the son of two eminent figures in British society, J62 20 the Manchester Guardian war correspondent Henry Woodd J62 21 Nevison (1856-1914) and the suffragist writer Margaret Wynne J62 22 Nevinson (1860?-1932). The progressive Nevinson household became J62 23 involved in Futurist activities during F.T. Marinetti's prewar J62 24 visits to London. The younger Nevinson joined forces with Marinetti J62 25 to write the only English Futurist manifesto, 'Vital English Art,' J62 26 published in the Observer on June 7, 1914. Although stopping J62 27 short of the warmongering rhetoric of earlier Italian manifestos, J62 28 the English proclamation, like its Italian counterparts, urged that J62 29 society be cleansed of her ills, her tired platitudes, and J62 30 long-worshiped traditions and extolled the virtues of sport, J62 31 adventure , and the heroic instinct of discovery.

J62 32 A quest for adventure may have drawn Nevinson to the Friend's J62 33 Ambulance Service, a group of Quaker volunteers who risked their J62 34 lives tending the injured and dying in Flanders. Following his J62 35 arrival in Dunkirk on November 13, 1914, he observed death on a J62 36 daily basis; on one occasion his ambulance was blown up. His J62 37 personal experience of the grisly results of aerial bombardment and J62 38 automatic weaponry proved advantageous in producing early, J62 39 innovative, modern images of the war.

J62 40 When the fatigued Nevinson returned to London on January 30, J62 41 1915, suffering from rheumatic fever, one newspaper reporter stated J62 42 that the Futurist artist was a victim of neurasthenia, or shell J62 43 shock. Nevinson denied the accusation, calling the war "a J62 44 violent incentive to Futurism" and proclaiming, in a J62 45 quotation borrowed from Italian Futurist manifestos: "There J62 46 is no beauty except in strife, no master-piece without J62 47 aggressiveness." H. W. Nevinson's journal entry of October J62 48 25, 1914, however, indicates that his son was interested, even J62 49 prior to visiting the front, in distancing himself from the J62 50 Futurist movement.

J62 51 Following a period of convalescence, Nevinson enlisted in the J62 52 Royal Army Medical Corps and served as an orderly in a London J62 53 military hospital. He was invalided out in late 1915, due to J62 54 recurring rheumatic fever. During 1915 and 1916, he produced J62 55 Cubofuturist portrayals of fighting men, exploding shells, and J62 56 destroyed landscape. One early image, Returning to the J62 57 Trenches (fig. 2), of 1914?-15, captures the buoyant mood of J62 58 marching French soldiers, equipped with kits and bayonets. Their J62 59 feet barely touch the ground as they move in synchronized formation J62 60 toward the front.

J62 61 This oil on canvas, along with Nevinson's other representations J62 62 of marching men, is sometimes associated with Vorticism, the J62 63 short-lived British abstract-art movement. Although a woodcut J62 64 version of it, titled On the Way to the Trenches, was J62 65 reproduced in the second issue of Wyndham Lewis's Vorticist J62 66 publication Blast, the painted version, in its emphasis on J62 67 simultaneity, lines of force, and fractured planes of colour, J62 68 derives its abstract qualities more from Futurism than J62 69 Vorticism.

J62 70 Expressing none of the action and vitality of war, his La J62 71 Patrie (fig. 3) of 1916 anticipates the grim portrayal of war J62 72 dead in Paths of Glory. Both are among the most moving J62 73 British paintings of the First World War. Set in the poorly lit J62 74 wooden shed that served as a casualty clearing station in the J62 75 Dunkirk railyard, La Patrie depicts row upon row of J62 76 casualties lying on stretchers on a straw-covered floor. The J62 77 clearing station was known as 'the Shambles,' and Nevinson later J62 78 described the blood, stench, typhoid, and agony in which six J62 79 workers attempted to rescue more than three thousand maimed and J62 80 dying men.

J62 81 Despite its portrayal of grimacing faces, bloody bandages, and J62 82 squalid conditions resembling those of a temporary morgue, La J62 83 Patrie presented no problem for the British censoring J62 84 authority. Not only was its exhibition permitted twice in 1916 (in J62 85 June and September-October), it was praised by critics, reproduced J62 86 in newspapers, and included in a 1917 volume of Nevinson's war J62 87 images. One reviewer hailed the artist for his non-flag-waving, J62 88 non-drum-beating depiction of war, while the influential P. G. J62 89 Konody stated that "C. R. W. Nevinson stands alone, in J62 90 England, as the painter of modern war." Thus, an image J62 91 portraying the pain and suffering of Allied soldiers, which not J62 92 only implied the possibility of death but questioned the J62 93 invincibility of the British forces, was sanctioned by the British J62 94 government in 1917.

J62 95 Under the direction of the journalist and politician Charles F. J62 96 G. Masterman and, later, the novelist John Buchan, the government J62 97 hired artists to produce eyewitness accounts of the war after June J62 98 1916. These artists retained ownership of their work, while the J62 99 Department of Information (known as the Ministry of Information J62 100 after February 1918) had right of first refusal to purchase and use J62 101 the art to illustrate propaganda literature. Nevinson, who departed J62 102 for France and Belgium as an official war artist on July 5, 1917, J62 103 was the first of the young avant-garde artists to be hired for the J62 104 innovative program. He toured the British line until mid-August, J62 105 returned to London, and produced more than seventy-five paintings, J62 106 drawings, and prints over the following seven months.

J62 107 The controversy surrounding Paths of Glory began on J62 108 November 24, 1917, when Masterman asked Alfred Yockney, the former J62 109 editor of the London Art Journal, who served as an J62 110 advisor within the Department of Information, to show him two of J62 111 Nevinson's paintings. One of these Masterman called 'Dead Men.' J62 112 Five days later, the department's censor, Major A. N. Lee, voiced J62 113 an objection to the one portraying dead British soldiers. Although J62 114 it would not yield "military information to the J62 115 enemy," he wrote to Yockney, its "subject matter J62 116 raises a point of policy." Given a probable concern for J62 117 public morale, he deemed it necessary to consult the War Office for J62 118 their opinion, and refused to permit the work to be released.

J62 119 The elder Nevinson accompanied his son to the government J62 120 offices on December 4 for a meeting with Buchan, the director of J62 121 information. They wished to learn more about Lee's reason for J62 122 suppressing the work. They learned that two Nevinson paintings, J62 123 Paths of Glory and an oil titled A Group of Soldiers J62 124 (fig. 4), were considered hindrances to the war effort. The J62 125 latter, a painting of four Tommies standing at ease, was suppressed J62 126 because of its purportedly unflattering representation of British J62 127 soldiers, who were thought to resemble mannequins or ventiloquists' J62 128 dummies. On this painting Lee's decision was later overruled by J62 129 Masterman, who claimed that paintings should only be censored J62 130 "from a military point of view," rather than an J62 131 aesthetic one. A Group of Soldiers was approved for J62 132 exhibition and reproduction on December 13, 1917, and for inclusion J62 133 in Nevinson's one-person show the following March, but Paths J62 134 of Glory was not.

J62 135 The artist did not fare as well with Lee regarding this J62 136 painting. Although no details of the Nevinsons' meeting with Buchan J62 137 survive, most likely they were told what Yockney was reporting to J62 138 Masterman, that "representations of the dead have an J62 139 ill-effect at home" and that all such paintings were J62 140 "now rigidly suppressed." The painting was J62 141 purchased by the government in January 1918, perhaps as a means of J62 142 acquiring full control over its use and reproduction.

J62 143 The brown-and-green oil on canvas depicts two dead British J62 144 soldiers sprawled on a hillside littered with rifles, helmets, and J62 145 barbed wire. The soldiers appear, compositionally, to recede from J62 146 the lower left to the upper right of the canvas. Unlike J62 147 Returning to the Trenches, in which military might is J62 148 suggested by an infinitely long line of marching men, Paths of J62 149 Glory portrays the human cost of war, with a recessive line J62 150 implying an untold number of dead beyond the limits of the canvas. J62 151 A mood of morbidity and death is heightened through use of an J62 152 eerie, unnatural light that casts aqua-colored shadows across the J62 153 dead soldiers and the debris that surrounds them. The small zone of J62 154 background sky is transected by a skeletal web of posts and barbed J62 155 wire, and appears only at the very top of the composition, J62 156 conveying a sense of restriction, rather than liberation.

J62 157 Scenes of inglorious death in the trenches resembling that in J62 158 Nevinson's painting were commonplace along the front line that J62 159 stretched from the North Sea coast of Belgium to Switzerland. J62 160 Burial in a prepared grave was often an impossibility: bodies were J62 161 frequently left where they fell, sometimes used as shields for J62 162 trench reinforcement, as gun supports, or as guideposts. The living J62 163 became accustomed to the sight of trenches "rotthen with J62 164 dead," who looked "like ghastly dolls, grotesque J62 165 and undignified," as the poet and soldier Siegfried Sassoon J62 166 described them. Nevertheless, the British government preferred that J62 167 the public at home be shielded from knowledge of such images.

J62 168 Paths of Glory was painted in November and December J62 169 1917, during the blackest period of Nevinson's war years, when the J62 170 government was calling up discharged soldiers and previously J62 171 reflected men to replenish the drastically depleted ranks at the J62 172 front. Among those under review were men suffering from J62 173 neurasthenia and shell shock. Entries in H. W. Nevinson's journals J62 174 of the winter of 1917-18 provide proof that his son was depressed, J62 175 extremely fearful of returning to the front, and a patient of the J62 176 eminent British neurologist Sir Henry Head. The younger Nevinson's J62 177 depression during these dark days was exacerbated by his desire to J62 178 succeed in the war-artists' program, which he viewed as his only J62 179 means of avoiding conscription. He had failed in an earlier attempt J62 180 to emigrate from Great Britain after Parliament began debating the J62 181 issue of reexamining discharged soldiers. Therefore, the idea that J62 182 the Futurist sympathizer and former member of the rebel avant-garde J62 183 deliberately chose to paint an alarming or potentially J62 184 controversial work of art for political or personal reasons during J62 185 these traumatic and difficult months is not consonant with his J62 186 profound wish to remain in the good graces of government officials J62 187 at the time. His other official war art, in a more representational J62 188 and stylistically conservative mode than his 1914-16 innovations, J62 189 also bears this out.

J62 190 Nevionson's shift toward a more representational style may have J62 191 been independent of his commitment to the government-sponsored art J62 192 program. After the trauma and illness following his service at the J62 193 front, he appears to have distanced himself from the Futurist J62 194 rhetoric of 1914-15, regarded by some as having contributed to the J62 195 war's madness. Moreover, this shift is in keeping with a similar J62 196 move on the Continent toward more legible and orderly stylistic J62 197 tendencies during the mid and late teens.

J62 198 A crowd of dignitaries and luminaries from the military and art J62 199 worlds, as well as invited members of society, gathered at the J62 200 Leicester Galleries on March 1 to see Nevinson's exhibition and to J62 201 hear an opening speech by the newly appointed minister of J62 202 information, Lord Beaverbrook. Their surprise at and intrigue with J62 203 one painting in a semidisguised state was recorded in the press: J62 204 despite its proscription Nevinson had hung Paths of Glory J62 205 and placed a piece of brown paper diagonally across it, covering J62 206 the dead soldiers. Upon the paper he had written in bold letters J62 207 "CENSORED."

J62 208 The press photographed and reproduced the painting in its J62 209 altered state (fig. 5). One reviewer concluded: J62 210 "Probably no picture in the Nevinson show excites more J62 211 interest and speculation than the one which is partly obscured by J62 212 the 'censored' label." The issue was raised in the J62 213 London Mail's column, 'Things We Want to Know,' which J62 214 asked, "What is hidden by the patch?"

J62 215 J63 1 <#FROWN:J63\>Nonetheless, both ruler and ruled always benefit in J63 2 some way - intentionally, accidentally, or indirectly - because J63 3 they share some common task or purpose (Pol 1254a27-28).

J63 4 That ruling and being ruled are according to nature does not J63 5 mean that either is easy. What is according to nature appears to be J63 6 divine insofar as it appears to be in the best state possible; but J63 7 it is not "sent by the gods," or the same as J63 8 fortune, because it requires effort on our part (NE J63 9 1099b9-24). Indeed, Aristotle observes, "in general, it is J63 10 difficult to live together and be partners in any human J63 11 activity" (Pol 1263a15-16). This observation seems to J63 12 move Aristotle's notion of the household toward Arendt's J63 13 interpretation - that the household is a place of toil yielding no J63 14 real satisfaction. According to Aristotle, however, things brought J63 15 into being through effort - nature's or man's - are the greatest J63 16 and noblest of all things (NE 1099b22-24). They thus yield J63 17 much pleasure, for "actions in accordance with virtue are J63 18 by nature always pleasant" (1099b13-14). Furthermore, the J63 19 difficulty of living together decreases to the extent that the J63 20 parties recognize their common aim, a life as complete and J63 21 self-sufficient as possible (Pol 1280b33-35, 1260b13, J63 22 1254a27-28).

J63 23 THE AIM OF HOUSEHOLD RULE: VIRTUOUS INDIVIDUALS

J63 24 In that the best household's aim is to instill unqualified J63 25 moral virtue through some sort of rule, its aim appears to be J63 26 indistinguishable from that of the best regime. Moreover, the aims J63 27 of the best household and the best regime are alike in that they J63 28 both seek to acknowledge the distinctiveness of individual human J63 29 beings; according to Aristotle, diversity more than sameness gives J63 30 rise to unity (Pol 1261a29-30, 22-24). Both the household and J63 31 the city should promote similarity in the sense of virtue, but J63 32 neither should promote homogeneity (1263b31-32). "Habits" J63 33 deriving from household activities and "laws" from the J63 34 regime can together make the city "one and common through J63 35 education" (1263b36-40) without sacrificing diversity. J63 36 Nonetheless, as noted earlier, household activities are better J63 37 suited to individualized instruction and thus to acknowledgment of J63 38 individuality than is public education. Cities, then, should rely J63 39 more on households than on laws and public, institutions to J63 40 maintain diversified excellence. The question is, what should J63 41 household rule instill to achieve this diversity?

J63 42 According to Aristotle, instilling moderation and judgment J63 43 makes human beings virtuous without eradicating any distinctiveness J63 44 other than a lack of virtue. The man and the woman of the household J63 45 may exercise both moderation and judgment as well as "show J63 46 who they really and inexchangeably are" by selecting and J63 47 remaining with each other, managing the household, and caring for J63 48 their children. Likewise, children and servants may also acquire J63 49 and demonstrate moderation, judgment or understanding, and J63 50 distinctiveness by the ways they conduct themselves and respond to J63 51 the heads of the household. Indeed, the extent to which members of J63 52 the household practice moderation and judgment is itself expressive J63 53 of distinctiveness.

J63 54 TEACHING MODERATION

J63 55 All household members must learn to be moderate toward things J63 56 and each other. The various forms of household rule can teach J63 57 members moderation by revealing to them the natural ends of their J63 58 natural desires (Pol 1257b19-34). For example, household J63 59 management (rule over the material conditions of a household) J63 60 teaches that specific things must fulfill specific needs and J63 61 desires: food satiates hunger, a bed satisfies the need for sleep; J63 62 money itself cannot satisfy such needs. Thus, household management J63 63 teaches human beings to check their desire for money - itself an J63 64 unnatural, because unfulfillable, desire. The various household J63 65 relationships also teach moderation in various ways. Forming a J63 66 household entails the exercise of moderation in that it requires J63 67 limiting oneself to one out of many sexual partners and companions. J63 68 Parenthood teaches both the parents and the children moderation. J63 69 Since children's reasoning powers are not developed, parents must J63 70 find the mean between arguments and force which is effective for J63 71 teaching their children (Pol 1260a13-14, b6-7, 1332b10-11; J63 72 NE 1179b23-29). It is because children are potentially J63 73 reasoning and reasonable beings - or "free persons" J63 74 - that one ought to rule them in "kingly fashion" J63 75 (Pol 1259a39-b1, 1253b4, 1285b32). And children, who are not J63 76 inclined to be moderate, must learn to be so if they are to live J63 77 well (NE 1179b24-34). Finally, as the next chapter shows, J63 78 ruling slaves teaches both the masters and the slaves J63 79 moderation.

J63 80 Aristotle's characterization of the ideal household as J63 81 requiring the exercise of moderation contrasts with the general J63 82 contemporary liberal view according to which what goes on in the J63 83 household is entirely a matter for the (undefined) discretion of J63 84 household members. Indeed, activities are private according to J63 85 Aristotle only when the actors heed the limits established by J63 86 nature.

J63 87 The moderation learned in the household not only helps to J63 88 sustain the household but facilitates all human engagement. J63 89 Moderation is both the result of and fosters seeing what is J63 90 required for living together. It is thus neither a strictly private J63 91 nor a strictly public virtue, and so it - not courage - might be J63 92 said to be in Aristotle's eyes the political virtue par J63 93 excellence.

J63 94 TEACHING JUDGMENT

J63 95 In addition to moderation, the good household teaches judgment J63 96 (Pol 1253a15-18). Forming a household requires judgment in J63 97 that it requires choosing a good partner. Raising children involves J63 98 judgment as something to be taught. Ruling servants involves J63 99 judgment in trying to compensate for the servants' lack of it. What J63 100 is pertinent to this inquiry, however, are the ways judgment J63 101 required by the household differs from that required by the regime. J63 102 One significant difference involves natural affection; another, the J63 103 end each aims to realize.

J63 104 According to Aristotle, the end of the city is justice, which J63 105 all take to be "some sort of equality" - that is, J63 106 equal things for equal persons (Pol 1282b14-21). But this J63 107 definition encompasses both natural justice, the fundamental J63 108 principle of which is proportionality or desert, and conventional J63 109 justice, the fundamental principle of which is arithmetical J63 110 equality (NE 1134a26-28, b18-19). The regime that is J63 111 "by nature" - realized natural justice - is best J63 112 (NE 1135a5). But since realizing natural justice in a regime J63 113 presupposes many deserving human beings and the ability to detect J63 114 them - that is, requires fortune and virtue to achieve (Pol J63 115 1331b21-22, 1277a1-5) - cities should aim first to realize J63 116 conventional justice.

J63 117 Should the household also then seek conventional or ordinary J63 118 justice? In two places, Aristotle says that it should not. J63 119 "Political justice seems to consist in equality and J63 120 parity," "but there does not seem to be any justice J63 121 between a son and his father, or a servant and his master - any J63 122 more than one can speak of justice between my foot and me, or may J63 123 hand, and so on for each of my limbs. For a son is, as it were, a J63 124 part of his father" (MM 1194b23, 5-15). As he explains J63 125 in the Nicomachean Ethics, "there can be no J63 126 injustice in the unqualified sense toward what is one's own, and a J63 127 chattel or a child until it reaches a certain age ... is, as it J63 128 were, a part of oneself, and no one decides to harm himself. Hence J63 129 there can be no injustice toward them, and therefore nothing unjust J63 130 or just in the political sense. ... what is just in households ... J63 131 is different from what is politically just" J63 132 (1134b10-17).

J63 133 By proceeding immediately to discuss natural justice, Aristotle J63 134 suggests that it characterizes the household. The household appears J63 135 to be even a paragon of natural justice in that inequalities within J63 136 it are evident and determine who rules and who is ruled. And, as J63 137 Arlene W. Saxonhouse explains, "the family, because its J63 138 differences in eid<*_>unch<*/> are observable, demonstrates a J63 139 unity in diversity which perhaps becomes impossible in political J63 140 life. In the polis obvious differences in eid<*_>unch<*/> J63 141 are absent. ... The family with its definition of differences ... J63 142 attains a certainty in nature not available to the city." J63 143 Or, at least, not available to most cities. In other words, it J63 144 appears that the household, being a model of natural justice, is a J63 145 kind of model for the best regime. Aristotle would apparently like J63 146 the natural superiority holding together the (best) household to J63 147 hold together the (best) city. Indeed, he may insist on the J63 148 preservation of households (in all regimes) because they have the J63 149 potential to exemplify perfect unity or justice and by their J63 150 examples point the city toward a higher justice.

J63 151 Aiming to realize natural, not conventional, justice, the good J63 152 household ruler does not treat all members equally or give each a J63 153 turn at ruling; rather, it is incumbent on this ruler to detect the J63 154 virtues of each member and treat him or her accordingly, giving J63 155 guidance or instruction when needed and freedom to make choices J63 156 when deserved. The household is a compound of "unlike J63 157 persons" - man, woman, servants, and children - who, J63 158 moreover, have multiple functions or obligations - as husband and J63 159 father, wife and household manager, son or daughter and future J63 160 citizen (Pol 1277a5-8, 1253a4-14). There are thus not only J63 161 manly virtues, womanly virtues, servile virtues (1277b20-23), and J63 162 presumably even youthful virtues but also virtues attached to being J63 163 a husband, father, wife, and child. A household thrives when each J63 164 member performs his or her function, or upholds his or her J63 165 obligations, in accordance with the virtues proper to doing so J63 166 (NE 1098a14-15).

J63 167 The variety of virtues indicates the variety of judgment in the J63 168 household. Most notably, the judgment of those ruling differs from J63 169 that of those being ruled, as becomes clear when we take into J63 170 account the deliberative capacities of each kind of member and J63 171 Aristotle's distinctions among intellectual virtues in the J63 172 Nicomachean Ethics. One acquires prudence by repeatedly J63 173 putting into effect good judgments about at least one's own J63 174 affairs, if not the affairs of others (NE 1141b12-21, J63 175 29-1142a10). Lacking experience, the young cannot have prudence J63 176 (1142a15-16). Lacking good judgment, or the ability to detect J63 177 through deliberation what action to perform, and how and when to J63 178 perform it, the slavish, who lack the ability to deliberate, cannot J63 179 have prudence either (NE 1143a29-31, Pol 1260a12). The J63 180 nonslavish adults of the household, however, having both experience J63 181 and the ability to deliberate (Pol 1260a10-13), may have J63 182 prudence. In fact, household management requires that they do J63 183 (NE 1141b31-32). Nonetheless, the prudence of the man and the J63 184 woman apparently differ. Although it is the responsibility of both J63 185 to manage the household, the man should acquire possessions and the J63 186 woman should oversee their use and consumption (Pol J63 187 1277b24-25). It follows that the man should acquire the household J63 188 servants (Pol 1255b37-39), since they are animate possessions J63 189 (1253b32), and that the woman should command them, since their J63 190 function is to assist in the use of other possessions (1252b32-33, J63 191 1254a2). Moreover, Aristotle indicates in several ways that the J63 192 man, at least more than the woman, should guide their children; for J63 193 example, "the man rules the child" (Pol J63 194 1260a10). In addition, Aristotle assigns marital rule to both the J63 195 husband and the wife; that is, spouses rule each other (Pol J63 196 1253b9-10, 1259a39-b1, 4-10). Since the man and the woman each rule J63 197 over others, at least in part for the good of those others J63 198 (Pol 1278b32-1279a8), each has complete moral virtue, which J63 199 Aristotle calls justice and prudence (NE 1130a2-14, 1145a1-2; J63 200 Pol 1260a17-18, 1277b25-26). But because each rules over J63 201 different persons, they again exercise prudence differently J63 202 (Pol 1260a10-12, 20-24, 1277b20-23).

J63 203 In contrast to the judgment of the free adult members of the J63 204 household, the judgment of children and servants is lacking. J63 205 Children have only the potential for judgment and prudence; J63 206 servants can only follow judgment and comply with prudence J63 207 (Pol 1260a12-14, 1254b22-23).

J63 208 Variety of judgment appears naturally in the household; even J63 209 more, in the good household, those who rule acknowledge it. Good J63 210 household rulers do not command their spouse, children, and J63 211 servants in the same way (NE 1134b15-16). By way of presenting J63 212 the household, then, Aristotle suggests that private judgment J63 213 differs from the judgment required by most regimes in that it J63 214 acknowledges differences in kinds of, and aptitude for, virtue J63 215 among human beings. Moreover, in trying to promote the virtues J63 216 peculiar to each member, household rulers promote individuality.

J63 217 In addition to promoting individuality, private differs from J63 218 public judgment in not having law to aid it (Pol 1282b1-6). J63 219 Both political and household rulers must employ "knowledge J63 220 and choice" (Pol 1332a31-32) to bring about, J63 221 respectively, the city's and the household's excellence. J63 222 J64 1 <#FROWN:J64\>Complementation, re-presentation, reworking of J64 2 strategic elements - all of these are present in the finale. The J64 3 subset center C is insistently thrust into attention by the J64 4 wrong-footing opening, and at crucial articulative points the J64 5 movement also reworks the Neapolitan. In fact, the finale shows J64 6 remarkable new interpretative roles for C, which connect it closely J64 7 in pitch or function to F-natural. The second subject of the J64 8 finale, for example, has a strong Neapolitan element. In the J64 9 exposition, the second subject (mm. 70-78) is in the key of B J64 10 minor, so the Neapolitan pitch for that key is C. This means that J64 11 C, the subset pitch of the first movement and the wrong-footing J64 12 opening of the finale, now assumes a new role as the Neapolitan of J64 13 B minor (Ex. 18).

J64 14 Ingenuity is demonstrated on many fronts: the second subject in J64 15 E minor - a recapitulatory feature - comes back in the development J64 16 (m. 216), not in the recapitulation. This placement indicates not J64 17 so much an exchange of function between development and J64 18 recapitulation, but the interpolation of a recapitulation J64 19 element in the development. In a movement where surprise and J64 20 ingenuity are of the essence, this is one further instance of a J64 21 technical virtuosity, with Beethoven reordering the elements of J64 22 sonata and rondo design in an individual solution of dazzling J64 23 skill. In the E-minor appearance of the second subject, the J64 24 Neapolitan pitch is F-natural. The second subject is extended by a J64 25 strongly marked four-measure phrase, dynamically highlighting F J64 26 major and its dominant C. Since C major is also the Neapolitan of B J64 27 minor, one could say this passage encapsulates both uses of J64 28 the Neapolitan, as well as the Neapolitan and its dominant (Ex. J64 29 19).

J64 30 figures&captions

J64 31 The most stunning use of the Neapolitan, though, comes in the J64 32 coda, where it is reintegrated into E minor. Like the hammered-out J64 33 dominant minor ninth in the first movement, this Neapolitan is the J64 34 dramatic and structural hinge of the finale. It reworks precisely J64 35 elements from the scherzo - strong dynamics and sudden dynamic J64 36 change, wide registral span - but more powerfully. The extreme J64 37 range of the fortissimo Neapolitan is cut off by a shocking J64 38 measure's silence, then the V of E minor comes in softly. Silence, J64 39 as well as the pitch and harmonic elements from the first movement, J64 40 is another element in this powerful return (Ex. 20). This searing J64 41 Neapolitan figure and its resolution in E minor fulfills all the J64 42 functions of the Neapolitans from the beginning of the work - J64 43 except one. The coda has one last card to play. In the last page of J64 44 the score, one final reference to the Neapolitan appears as part of J64 45 an ascending chromatic line in the first violin from E to E'. The J64 46 movement finishes with three pairs of perfect cadences: the first J64 47 has F-sharp - G in the upper line, so raising the F-natural to its J64 48 diatonic F-sharp and resolving it to G; the second has D-sharp - E J64 49 in the upper line, reversing the E - D-sharp of the opening J64 50 i-V6 at the beginning of the work; the third pair of J64 51 cadences falls a fifth, B - E, reversing the space-opening gesture J64 52 of the first movement. The circle is now closed.

J64 53 figure&caption

J64 54 Drama, conflict, design - these features characterize the works J64 55 of Beethoven's middle period that fall within the description of J64 56 'heroic'. To say that Mozart's elegant gracefulness or Haydn's J64 57 witty equipage have vanished in the face of a more powerful musical J64 58 conception sounds like a truism, but truisms are rarely the only J64 59 truth. The description 'heroic' sits oddly on the shoulders of the J64 60 Fourth and Pastoral Symphonies, the Piano Sonata in F, op. 54, J64 61 or the C-major Razoumovsky Quartet, op. 59, no. 3, works J64 62 smaller in frame , more relaxed and graceful in style than their J64 63 bigger-boned neighbors. But drama and conflict, the characterizing J64 64 features of the heroic concept, are central to the Eroica and J64 65 Fifth Symphonies, the E minor Razoumovsky Quartet, the J64 66 Waldstein and Appassionata Piano Sonatas.

J64 67 The Appassionata Sonata

J64 68 The nature, or rather the status, of conflict is treated J64 69 differently in the Appassionata Sonata than in the other J64 70 middle-period works just cited. In the Eroica and the E-minor J64 71 Razoumovsky Quartet, first-subject elements are lightened and J64 72 partially transformed in the finale. This reworking in turn affects J64 73 the formal design of the finale where the underlying framework - J64 74 respectively variation and sonata rondo - is shaped into a highly J64 75 individual rescoring of the form. The Waldstein Sonata J64 76 releases its first-movement intensity in the finale by a J64 77 wonderfully expansive rondo. Only in the Apassionata is J64 78 conflict unresolved. Conflict in the first movement is taken up in J64 79 the finale and replayed in a correspondingly intense mode, but J64 80 while the first-movement tonal opposition of contrasted J64 81 keys is resolved, the discourse of conflict is not.

J64 82 At the beginning of the first movement, two distinct elements J64 83 are presented: First, the linearization of the tonic-triad F minor J64 84 is followed directly by the flat supertonic G-flat major, forming J64 85 the harmonic juncture i-bII. The second element is the J64 86 modern C - D-natural - C, which is a variant of the movement's J64 87 prime mordent, C - D-flat - C. Even more closely than the J64 88 Eroica or the E-minor Razoumovsky Quartet, the finale of J64 89 the Appassionata is a direct reworking of first-movement J64 90 elements in a movement of parallel intensity, rather than a J64 91 reinterpretative reworking in a lighter vein - an intentional J64 92 matching of mode and material that underscores the cyclic nature of J64 93 the work. Not only will first-movement elements, particularly the J64 94 Neapolitan and the prime mordent, return prominently in the finale, J64 95 but there will also be close parallels of formal placement and J64 96 pacing between the two movements.

J64 97 The developments of the two movements vividly illustrate these J64 98 parallels. In both instances, conflict is central to the dramatic J64 99 action of the development, with the Neapolitan its most powerful J64 100 agent for engendering tension and deflecting tonal direction. In J64 101 the first-movement development, the Neapolitan pitch, G-flat, J64 102 appears in the context of B-flat minor at m. 115. By a rising bass J64 103 motion, V-VI, the Neapolitan is then tonally reinforced in G-flat J64 104 major. Immediately, it is respelled enharmonically as F-sharp, and J64 105 for the first time in the movement, it rises to G in the key of C. J64 106 The cumulative energy of this intensification erupts into sweeping J64 107 diminished sevenths and climaxes onto hammered fortissimo D-flats - J64 108 C, the latter segment of the movement's prime mordent, now J64 109 magnified at its structural dominant. (Compare this emphatic J64 110 segment with its bleached-bones version at the end of the J64 111 recapitulation).Prolonged through the sweeping line of diminished J64 112 sevenths and the structural dominant. G resolves to F only at the J64 113 recapitulation (Ex. 21))

J64 114 The central action of the development may be seen, therefore, J64 115 as the conflict between the chromatic bII and the J64 116 diatonic supertonic. If this description is reminiscent of the J64 117 first movement of the Razoumovsky, op. 59, no. 2, what J64 118 differentiates the first movement of the Appassionata is the J64 119 way Beethoven locks the chromatic and diatonic supertonics in J64 120 conflict, rather than allowing the diatonic supertonic to form a J64 121 point of relaxation as part of the dominant harmony of F minor, so J64 122 that conflict between the two supertonics is polarized, but not J64 123 resolved.

J64 124 figure&caption

J64 125 In the finale development, the Neapolitan G-flat is again at J64 126 the center of conflict against G-natural. Just as in the J64 127 first-movement development, G-flat appears in the context of B-flat J64 128 minor, but here replays the work's other prime element, the J64 129 mordent, on the Neapolitan pitch in a sharply condensed focus of J64 130 the two main structural components of the outer movements. (See J64 131 <}_><-|>Exx.<+|>Ex.<}/> 22, 23.)

J64 132 figures&captions

J64 133 Striking parallels between the outer movements of the J64 134 Appassionata demonstrated by matching formal placement of J64 135 salient harmonic features and the contextual reworking of the J64 136 Neapolitan. Far beyond general similarities of mood and expressive J64 137 delineation, these precise parallels make viable the description of J64 138 the work as cyclic form in which the finale provides a complete J64 139 variant reworking of the first movement. In variation form, the J64 140 variation - normatively, in classical style - exhibits the same key J64 141 planning, phrase structure, harmonic progressions, and large-scale J64 142 formal sections as the theme, while elaborating texture and J64 143 changing register, articulation, and dynamics. The variant concept J64 144 also opens up a larger dimension of formal integration for the J64 145 sonata as a whole; just as prime first-movement elements are J64 146 reworked (in the manner of development and variant) in the finale, J64 147 so the principal elements of the slow movement are elaborated (but J64 148 without development) in a set of variations. While individual J64 149 variations in the slow movement are self-contained, the movement J64 150 overall is not, but the end of the slow movement leads into the J64 151 finale in a way comparable to the slow-movement link to the finale J64 152 of the F-major Razoumovsky Quartet (F-minor, op. 57, and J64 153 F-major, op. 59, no. 1; the closeness of compositional period will J64 154 make such parallels of compositional procedure understandable). Yet J64 155 the two transitions are different in the way the two movements are J64 156 connected: the slow movement of the quartet dissolves out in a J64 157 swirl of elaborate figuration that fines down onto the trill J64 158 initiating the finale. At the end of the slow movement of the J64 159 Appassionata, the opening theme returns, but its resolution is J64 160 blurred by two hazy diminished sevenths. These same chords, J64 161 hammered out thirteen times fortissimo at the beginning of the J64 162 finale, shatter the serene repose and consonance of the variation J64 163 movement. Dynamic rupture, though, is anchored by pitch continuity. J64 164 D-flat, the pitch and key center of the slow movement, is carried J64 165 through to the diminished sevenths and falls to C on the turning J64 166 sixteenth-note figure which heralds the thematic statement of the J64 167 finale. This juncture, D-flat - C, forming a hinge between the two J64 168 movements, takes up, at a larger level, the mordent intersection J64 169 D-flat - C, which is the structural hinge of the first movement.

J64 170 Just as with development sections, so the recapitulation and J64 171 finale coda both re-present and compress strategic elements from J64 172 the first movement. The coda presents the final clash of the J64 173 supertonic and wrenches it back into the constraints of F minor.

J64 174 Some of the techniques discussed here with reference to J64 175 important works in Beethoven's middle period were not in themselves J64 176 necessarily new, nor had they originated with him. For example, the J64 177 favored harmonic shift of unmediated I-ii (although not J64 178 flatII) which Beethoven used virtually as his own style J64 179 characteristic for strongly defined first-movement openings (in the J64 180 Appasssionata, and the String Quartets, op. 59, no. 2, and op. J64 181 95, as examples) had been used by Haydn at the beginning of the J64 182 first movement and also at the beginning of the second movement - J64 183 scherzando - of his C-major String Quartet, op. 33, no. 3. Expanded J64 184 dimensions and contrapuntal enrichment of development can be found J64 185 in the first movement of Mozart's C-Major String Quintet, K. 515, J64 186 in the first movements and finales of his G-Minor Symphony, K.550, J64 187 and Jupiter Symphony, K. 551. What is new in Beethoven's J64 188 middle period is the consistently expanded formal framework as J64 189 distinct from individual instances; the versatility and daring with J64 190 which he rearranges the modules of formal entities, yet retains J64 191 their underlying coherence and sense of internal logic; and the use J64 192 of strategic pitches as axial centers around which a movement is J64 193 built and from which it diverges by contextual reinterpretation J64 194 into new and surprising key areas to form, frequently, the J64 195 structural hinge of the movement.

J64 196 In the middle-period works, Beethoven differentiated the four J64 197 movements of the quartet and symphony by sharply defined J64 198 characterization, yet, conversely, drew closer relationships J64 199 between first movement and finale, binding them by similar or J64 200 related mood, and even more, by the reinterpretation of prime J64 201 first-movement elements in the finale. This compositional procedure J64 202 produces a kind of cyclic form., but one conceptually different J64 203 from Schubert's Wanderer Fantasy or from later J64 204 nineteenth-century works lie Liszt's B-minor Sonata. Beethoven's J64 205 expansion of dimensions, of dynamic and tempo ranges, has as its J64 206 context the retention of formal design and structural principles. J64 207 The strong individualization of movements is set against the J64 208 concern for the organic form of the work overall. Accordingly, J64 209 large-scale integration of the outer movements depends for its J64 210 effect on the internal formal autonomy of each movement and its J64 211 specific delineation of character and material - in the J64 212 relationship of self-standing parts to a larger whole. J64 213 J65 1 <#FROWN:J65\>5

J65 2 Conclusions

J65 3 SUE BRIDEHEAD: THE CASE FOR A FEMINIST READING

J65 4 In his depiction of Sue, Hardy shows remarkable sensitivity to J65 5 feminist issues. The novel's tragedy turns on marriage, and it is a J65 6 double tragedy. This view is augmented by looking into the J65 7 historical context of women's issues.

J65 8 In discussing Jude in the context of Hardy's fiction as a J65 9 whole, Patricia Ingham (1989) observes that over the course of his J65 10 novel writing Hardy's treatment of women increasingly diverges from J65 11 the traditional misogynist stereotype which had been J65 12 'scientifically' justified by Herbert Spencer in his popular J65 13 The Study of Sociology (1873).

J65 14 Owing to the far-reaching influence of Darwin, Spencer's J65 15 discourse was scientific rather than moral. Darwinism, with On J65 16 the Origin of Species (1859), but more particularly with J65 17 The Descent of Man (1871), gave momentum to biological J65 18 determinism as it related to the female nature and role. By J65 19 appealing to so-called objective laws, and adopting a tone of J65 20 neutral and dispassionate observation, science sought to establish J65 21 the biological link between physiology, psychology, and sociology, J65 22 and to effect the ratification of the status quo - that is, to J65 23 confirm the old stereotypes, and to reaffirm the disabilities of J65 24 being a woman. This spurious science replaced the Bible as the J65 25 underpinning of the double standard of sexual morality. Social J65 26 critics used concepts of evolution to show that sexual difference J65 27 was the result of adaptation to the conditions necessary for social J65 28 survival. Woman's position in society was seen as the natural J65 29 result of processes designed to strengthen her essential function - J65 30 maternity. Spencer claimed that women had less power than men for J65 31 abstract thinking because their vital energies went toward J65 32 nurturing offspring. As a result, women lagged behind men in the J65 33 evolutionary process, having smaller brains as well as weaker J65 34 physiques. As the weaker sex, Spencer argued, women had learned to J65 35 disguise their feelings, to please and persuade, and to delight in J65 36 submission. Because Victorian women were not only dependent but J65 37 ready to cultivate and display that dependence, a husband was their J65 38 only goal. Although in the Victorian period women were often J65 39 revered as being morally superior - more devout and devoted to J65 40 caring for others - some writers were of the opinion that woman's J65 41 reproductive capacity gave her a far more menacing nature. Havelock J65 42 Ellis, in Man and Woman (1894), stated that since J65 43 menstruation is disgusting, women are ashamed of it, and shame J65 44 makes them deceitful. This tendency toward dishonesty is, he J65 45 asserted, reinforced by the duties of maternity, and much of the J65 46 education of the young, which is entrusted to women, consists of J65 47 skillful lying.

J65 48 Childbearing, one of the few acceptable activities for women in J65 49 Victorian society, dominated women's lives. In 1900 a quarter of J65 50 all married women in England were pregnant. Most deliveries took J65 51 place at home, where the experience could be nothing but a struggle J65 52 with poverty, pain, and death. Stillbirths, miscarriages, attempts J65 53 at abortion, uncaring doctors, and incompetent midwives caused J65 54 women to fear pregnancy. The infant death rate in 1900 was 163 per J65 55 thousand, compared to 9.4 per thousand in 1985. In all, 145,000 J65 56 infants died in 1900. The medical profession did not explain about J65 57 or provide contraception or abortion. The condom and vaginal sponge J65 58 were unreliable and in any case unavailable to most people because J65 59 birth control was relatively costly. Among the working classes J65 60 there was a flourishing trade in abortion-inducing pills.

J65 61 Employment for women outside the home was effectively limited J65 62 to 'women's work' - work that required nimble fingers or no great J65 63 physical strength, for example, dressmaking, schoolteaching, bar J65 64 keeping, assisting in a shop, doing office work, or working in J65 65 domestic service. The telephone, typewriter, and bicycle widened J65 66 career possibilities for women in the 1890s, but professional J65 67 opportunities did not exist for women until after the First World J65 68 War. Therefore the ideal woman was supported by her husband and had J65 69 no independent legal existence.

J65 70 The movement to change this state of affairs came to the fore J65 71 in the last two decades of the nineteenth century, a period during J65 72 which organized labor was pressing for social and political J65 73 emancipation. Throughout Europe mass socialist and working-class J65 74 parties were organizing and demanding fundamental changes. A group J65 75 of middle-class feminists sponsored debates and campaigned for J65 76 legislation giving women access to the professions, secondary and J65 77 higher education, the right to own property, and the right to vote. J65 78 They crusaded against adult prostitution, the flourishing trade in J65 79 child prostitution, and the protection of 'innocence' that made J65 80 sexuality furtive and dismal.

J65 81 By the 1890s the woman question was being widely debated in J65 82 newspapers, journals, and novels. J.S. Mill had argued in The J65 83 Subjection of Women (1869) that the so-called disabilities of J65 84 women were maintained in order to make women servants of men who J65 85 feared the competition of women in the working place, and who could J65 86 not tolerate living with them as equals. The most popular writer on J65 87 the woman question, Grant Allen, portrayed marriage as a degrading J65 88 form of slavery. In his novel, The Woman Who Did (1895), J65 89 the heroine deliberately has a child with her lover whom she J65 90 refuses to marry, regarding herself as a moral pioneer doomed to J65 91 martyrdom.

J65 92 But the whole weight of social orthodoxy brought to bear on J65 93 maintaining the stereotype was deeply ingrained in the majority of J65 94 women as well as men, and the women's movement was not able to J65 95 reach a consensus on most of the feminist issues raised in the J65 96 1880s and 1890s. Some writers argued that promiscuity was the path J65 97 to self-fulfillment; others asserted that such freedom J65 98 could only come from celibacy. Another point of contention was J65 99 childbearing. Was it a woman's most sacred calling? Or was it J65 100 rather an aspect of her degradation? Could fulfillment come from J65 101 working in the man's world? Or was the man's world a trap for J65 102 women?

J65 103 In spite of much debate on the fundamental place of women in J65 104 society, many women maintained that there was no escape from their J65 105 established role. In June 1889 more than one hundred well-known J65 106 women signed their names to the 'Appeal against Female Suffrage,' J65 107 which was printed in a leading journal. The appeal stated that J65 108 women's direct participation in politics "is made J65 109 impossible either by the disabilities of sex, or by strong J65 110 formations of custom and habit resting ultimately upon physical J65 111 difference, against which it is useless to contend." Among J65 112 those who endorsed the appeal were Beatrice Webb, Mrs. Humphry J65 113 Ward, Eliza Lynn Linton, Mrs. Matthew Arnold, and Mrs. Leslie J65 114 Stephen; a supplementary list of two thousand names was added two J65 115 months later.

J65 116 A woman might decide to escape the life marked out for her by J65 117 "the inexorable laws of nature" - that is, the J65 118 controlling and channelling of her sexuality into marriage - by J65 119 refusing the sexual dimension of a relationship. According to Penny J65 120 Boumelha, Sue Bridehead elected this option. Boumelha contends that J65 121 Sue's situation is confused and confusing because Sue is not sure J65 122 whether she wants love without sex or sex but not marriage. Yet one J65 123 must not see her as lacking sexual feeling, Boumelha argues; J65 124 rather, Sue's actions should be seen as her response to the dilemma J65 125 of how to have love without "the penalty."

J65 126 According to Boumelha, whether Sue denies her sexuality or J65 127 risks pregnancy, she is reduced. Boumelha says that the tragedy is J65 128 not brought on by her frigidity but by motherhood: "It is J65 129 motherhood - her own humiliation by the respectable wives who hound J65 130 her and Jude from their work, Little Father Time's taunting by his J65 131 schoolmates - that convinces her that 'the world and its ways have J65 132 a certain worth,' and so begins her collapse into 'enslavement to J65 133 forms'" (Boumelha, 148). Hardy, Boumelha observes, is alone J65 134 among writers of stature in drawing attention to motherhood's role J65 135 in confining women within the nuclear family. Sue's sexuality J65 136 destroys her, whereas Arabella's, by contrast, helps her survive. J65 137 Both reject their husbands, take up with other men, sublimate their J65 138 sexuality into religiosity, and eventually return to their J65 139 husbands. Yet a crucial point that emerges from the ironic J65 140 paralleling of Arabella's life with Sue's is that Arabella never J65 141 plays a maternal role. Whereas Arabella is identified with J65 142 sexuality and fecundity - she barters her sexuality for security, J65 143 seducing Jude by flinging a pig's penis at him and pretending to J65 144 hatch an egg between her breasts - Sue assumes the role of mother - J65 145 to her own children as well as to Arabella's son. Thus, according J65 146 to Boumelha, Hardy understands that a woman's freedom depends on J65 147 remaining free of the maternal role.

J65 148 Ingham (1989) argues that Hardy gradually developed his J65 149 sensitivity to the woman question over the course of 25 years of J65 150 novel writing. She traces the emergence of metaphors in which J65 151 workingmen suffering from self-devaluation are compared to women, J65 152 demonstrating that Hardy ever more insistently subverts the social J65 153 ideal that a woman's self-fulfillment is rooted in self-denial. By J65 154 the time Hardy wrote Jude, the workingman and woman are J65 155 "two in one," twins who suffer a similar J65 156 oppression. This affinity in oppression, Ingham says, is J65 157 highlighted by the emphasis on Jude's and Sue's similarities. Their J65 158 being cousins on Jude's mother's side, children of a family doomed J65 159 by a hereditary curse, points up a sameness that is continuously J65 160 stressed: by the rhyming circumstance of each taking refuge in the J65 161 other's room, by Sue's actually appearing in Jude's clothes as a J65 162 kind of double, and most emphatically by Phillotson's view of them J65 163 as the lovers in Shelley's 'The Revolt of Islam' - transcendent J65 164 beings, martyred in the cause against tyranny.

J65 165 Jude comes to see himself and Sue as martyred pioneers. They J65 166 are yanked "back into pre-determined forms of J65 167 marriage," Boumelha notes, and this is the tragedy J65 168 (Boumelha, 150). Conventional notions of sanctity and free will are J65 169 exploded by the novel: neither the home nor the love relationship J65 170 is a protected zone, and individual acts and intentions cannot J65 171 reform society.

J65 172 Boumelha claims that Hardy understood the crucial importance to J65 173 women of socialized child care, that it was his expressed reason J65 174 for supporting female suffrage. In his 1906 letter to the Fawcett J65 175 Society he states that he hopes suffrage would tend "to J65 176 break up the present pernicious conventions in respect to manners, J65 177 customs, religion, illegitimacy, the stereotyped household (that it J65 178 must be the unit of society), the father of a woman's child (that J65 179 it is anybody's business but the woman's own), except in cases of J65 180 disease or insanity" (Letters 3:238). The view that J65 181 unless the institution of marriage is radically changed, women will J65 182 continue to be enslaved, is also expressed in the novel. Phillotson J65 183 tells Gillingham, "And yet, I don't see why the woman and J65 184 the children should not be the unit without the man" (4:4). J65 185 When faced with the possibility that Little Father Time may not be J65 186 his child, Jude makes a statement that gives Hardy's position a J65 187 sharper focus: "The beggarly question of parentage - what J65 188 is it, after all? What does it matter, when you come to think of J65 189 it, whether a child is yours by blood or not? All the little ones J65 190 of our time are collectively the children of us adults of the time, J65 191 and entitled to our general care" (5:3).

J65 192 I do not see Jude as a novel primarily about marriage, nor J65 193 do I think of it as "the Sue story" (Boumelha, J65 194 138), as Hardy called it in a letter to Florence Henniker. Yet, J65 195 Boumelha's and Ingham's arguments offer persuasive interpretations J65 196 that illuminate Sue's life.

J65 197 Sue's behavior is confused and confusing, which is to say her J65 198 behavior indicates her self-control. Marrying Jude would invite J65 199 oppression, yet loving him without marriage would invite the J65 200 penalties reserved for sinners. Ambivalent, she appears to be J65 201 coquettish, half inviting his advances, yet sidestepping them. J65 202 "[E]picene tenderness," "boyish as a J65 203 Ganymedes" (2:4) are what the love-sick man sees in her J65 204 external behavior, but he has no clue about what is at issue. Sue J65 205 wants to be loved, but she cannot bear to lose her freedom.

J65 206 SHAME J65 207 Hardy's alternation of scenes - one on a comic plane, the other J65 208 tragic; one lofty, the other low; one affirming, the other J65 209 repudiating - is the novelist's method for grasping the ambiguous J65 210 real world. J65 211 J66 1 <#FROWN:J66\>New York, however, was the magnet that drew artists of J66 2 all sorts. Space was still cheap in lower Manhattan. America's J66 3 far-flung universities had little interest in recruiting J66 4 experimentalists-in-residence, and the one serious exception to J66 5 this rule, Black Mountain College, folded in the fifties, sending J66 6 much of its teaching staff and student body to New York as well. J66 7 Bohemian Manhattan was an intimate, small-scale scene: a band of J66 8 outsiders easily recognizable by their dress and demeanor. Groups J66 9 that later would seem diametrically opposed or at least very J66 10 different - for example, the Beats and the 'New York School' of J66 11 poetry - rubbed elbows amiably and frequented the same bars and J66 12 jazz clubs. Being few in number, they were obliged to stick J66 13 together; in Eisenhower's blandly conformist America, all weirdos J66 14 were brothers until the opposite was proven. In addition, artists J66 15 shared an exhilaration born of their recent liberation from Europe. J66 16 The old American colonial complex - a sense of being on the J66 17 periphery of things, still strong among the modernists of the 1920s J66 18 - had been swept away by the triumph of abstract expressionism, by J66 19 William Carlos Williams's appropriation of American speech as a J66 20 basis for new poetry, and, of course, by jazz, the American art J66 21 form par excellence.

J66 22 Needless to say, not all artists frequented bars and jazz J66 23 joints, but a remarkable number did. The abstract expressionists - J66 24 Jackson Pollock, Willem de Kooning, Franz Kline, and others - were J66 25 hard drinkers and inveterate hangers-out. One of them, Larry J66 26 Rivers, was also an accomplished jazz saxophonist and served as a J66 27 point of intersection between the worlds of painting and jazz. The J66 28 Beats also spent a lot of time in night spots. For the New York J66 29 School (Kenneth Koch, John Ashbery, and others), a key connection J66 30 with jazz was Frank O'Hara, whose best-known poem, a poignantly J66 31 oblique homage to Billie Holiday, is entitled "The Day Lady J66 32 Died." Judith Malina's and Julian Beck's Living Theatre, J66 33 with its mixture of raw psychodrama and dreamy pacifism, sponsored J66 34 poetry readings at its headquarters on 14th Street and featured J66 35 Jackie McLean and hard-bop pianist Freddie Redd in its production J66 36 of Jack Gelber's play The Connection.

J66 37 Almost anyone's account of the era includes both this heady J66 38 mixture of scenes and the centrality of jazz as an artistic model J66 39 and jazz clubs as meeting places. Ron Sukenick's Down and In: J66 40 Life in the Underground, a combined study and memoir covering J66 41 the period from 1945 through the eighties, describes the clientele J66 42 and atmosphere at the Five Spot in the late fifties: "If J66 43 the painting seemed more consciously American after 1950, the J66 44 uniquely native American art form, jazz, became, through the J66 45 fifties, more central than ever for underground artist of all J66 46 kinds. It came together at the Five Spot, a bar on Cooper Square J66 47 where the brothers Iggie and Joe Termini hosted a basically J66 48 flophouse clientele until the artist started coming in during the J66 49 mid-fifties. Painters like Grace Hartigan, Al Leslie; David Smith, J66 50 de Kooning became habitu<*_>e-acute<*/>s. Larry Rivers, the J66 51 painter, played jazz there, poets read poetry to jazz, and J66 52 avant-garde film makers even showed their films to jazz. Writers J66 53 like Kerouac, Frank O'Hara, and Kenneth Koch moved in, and finally J66 54 the great jazzmen themselves came down to play - Charlie Mingus, J66 55 Sonny Rollins, Cecil Taylor, Thelonious Monk, Ornette J66 56 Coleman."

J66 57 In response to a question about which jazz musicians J66 58 down-town artists and intellectuals were friendliest with, J66 59 the painter Emilio Cruz (who also writes poetry and plays jazz J66 60 drums) gave this description of the scene: "In the case of J66 61 Jackie McLean, I would think that if he was around the Living J66 62 Theatre crowd, he was friendly with Judith Malina and Julian Beck J66 63 and that crowd, and a number of interesting people in that company. J66 64 Also Cubby Selby; we all called him Cubby. I never knew his real J66 65 name was Hubert Selby, Jr., until his book [Last Exit to J66 66 Brooklyn] came out. Paul Blackburn, the poet, who was a good J66 67 friend of mine, knew a lot of those people. Bob Thompson the J66 68 painter, Allen Ginsberg at times, Bob Kaufman the poet, some of the J66 69 Black Mountain poets. Rollins lived downtown and knew a number of J66 70 artists, though he was very solitary.

J66 71 "Others who were there not just because it was cool but because J66 72 they had a deep interest in the music were [Amiri] Baraka and Larry J66 73 Rivers. [Rivers] was a friend of Zoot Sims, Stan Getz I know he J66 74 considers himself as serious a saxophone player as he does a J66 75 painter. During that period I lived on Jeffferson Street, down by J66 76 the river. Pepper Adams lived in that building underneath me. In J66 77 fact, he was the only one in the building who had heat. He had a J66 78 gas heater, so in the winter when he was on the road I was a lot J66 79 colder than when he was there. Donald Byrd used to come by a lot J66 80 because they had that band together at that time. Jefferson Street J66 81 was south of East Broadway, maybe southeast, not that far from J66 82 where the old Fulton Fish Market was. That's all changed now. I J66 83 don't even know whether that street exists anymore. Ed Blackwell J66 84 used to come by that building too.

J66 85 "That was a period, in my life, when a lot of things were J66 86 integrated. Jazz was integrated within artists' lives. A lot of J66 87 people lived in lofts, and oftentimes the musicians might not have J66 88 lived in lofts themselves, but they would come over there (at least J66 89 on the Lower East Side) to play, to rehearse a band, so there were J66 90 a lot of connections because of that. That connection would then J66 91 extend what a person's capacity was, so that one person might learn J66 92 more about music, another person might learn more about painting or J66 93 poetry, and I knew a number of musicians who were interested in J66 94 learning about all of it. Like I'm friends with Grachan Moncur III. J66 95 I was his drummer last year at his workshop in New Jersey. He used J66 96 to come visit my studio all the time. There was a lot of openness J66 97 between various people. Herbie Lewis, the bass player, he lived J66 98 around the corner. He was a neighbor of mine, so we spent a lot of J66 99 time together. Miles ... I know numbers of people from the beat J66 100 generation who were friends with Miles, like Allen Ginsberg, Jack J66 101 Kerouac, Robert Frank the photographer. So there was an integration J66 102 of lots of the arts. The beat generation spent much of their lives J66 103 in clubs.

J66 104 "Now one of the things that it's very important to understand J66 105 is that there weren't thousands of people involved in the arts in J66 106 those days, so when you would walk down the street in New York J66 107 City, you would look at somebody and you would recognize them J66 108 instantly as an artist, and you would immediately find that you had J66 109 some kind of rapport, and there's another thing. I can't speak for J66 110 right now, but there was youth. Youth reaches out. There's one J66 111 thing that was consistent in all the artists in New York City at J66 112 that time. They were reaching out for new things, new ways of J66 113 expressing themselves. They were attempting to discover new values. J66 114 Those values were not necessarily ones that were supported by the J66 115 society at large, so there was this that they had in common. J66 116 Another thing was that there was an attempt to break down racism, J66 117 and politically there was a sense of hope that America could arrive J66 118 at a higher moral state. So racism was something that in that world J66 119 was frowned upon."

J66 120 What did jazzmean to those experimental artists who took J66 121 it most seriously? Part of the answer, as Sukenick indicates, has J66 122 to do with its American qualities, also underlined by Hettie Jones, J66 123 author of Big Star Fallin' Mama, a study of black female J66 124 vocalists from Bessie Smith to Aretha Franklin: "I think J66 125 jazz was the music they [downtown bohemian types] felt closest to, J66 126 the way someone feels close to music that's part of the zeitgeist. J66 127 It was American. There's that whole idea that abstract J66 128 expressionist art was the first truly American art movement; and J66 129 those people saw themselves as an avant-garde in what they J66 130 were trying to do. It was a shared feeling that they were all part J66 131 of a changing American art scene."

J66 132 Jazz was also influential in its improvisational freedom and J66 133 structural openness, as Allen Ginsberg indicates in his description J66 134 of jazz's relationship with his poem 'Howl': "In the J66 135 dedication of 'Howl' I said 'spontaneous bop prosody.' And the J66 136 ideal, for Kerouac, and for John Clellon Holmes and for me also, J66 137 was the legend of Lester Young playing through something like J66 138 sixty-nine to seventy choruses of 'Lady Be Good,' you know, J66 139 mounting and mounting and building and building more and more J66 140 intelligence into the improvisation as chorus after chorus went on J66 141 ... riding on chorus after chorus and building and building so it J66 142 was a sort of ecstatic orgasmic expostulation of music. So there J66 143 was the idea of chorus after chorus building to a climax, which was J66 144 the notion of part one of 'Howl,' with each verse being like a J66 145 little saxophone obbligato or a little saxophone chorus, as though J66 146 what I was doing was combining the long line of Christopher Smart, J66 147 the eighteenth-century poet, with notions of the repeated jazz or J66 148 blues chorus, till it comes to a climax, probably in the verse 'ah, J66 149 Carl, while you are not safe I am not safe.' And then there's a J66 150 sort of a coda from then on."

J66 151 Ginsberg also claims jazz as an important model for his work J66 152 and that of his contemporaries: "The whole point of modern J66 153 poetry, dance, improvisation, performance, prose even, music, was J66 154 the element of improvisation and spontaneity and open form, or even J66 155 a fixed form improvisation on that form, like say you have a blues J66 156 chorus and you have spontaneous improvisations, so in 'Howl' or J66 157 'Kaddish' or any of the poems that have a listeny style, 'who did J66 158 this, who did that, who did this,' you start out striking a note, J66 159 'who,' and then you improvise, and that's the basic form of the J66 160 list poem or, in anaphora, when you return to the margins in the J66 161 same phrase, 'Or ever the golden bowl be broken or the silver cord J66 162 be loosed or the pitcher be broken at the fountain,' as in the J66 163 Bible or as in some of Walt Whitman's catalogues or in Christopher J66 164 Smart's 'Rejoice in the Lamb' poem or the surrealist example of J66 165 Andr<*_>e-acute<*/> Breton's free union, 'my wife with the J66 166 platypus's egg, my wife with the eyes of this, my wife with that J66 167 and that ...'

J66 168 "It [jazz] was a model for the dadaists and it was a model for J66 169 the surrealists and it was a model for Kerouac and a model for me J66 170 and a model for almost everybody, in the sense that it was partly a J66 171 model and partly a parallel experiment in free form. The J66 172 development of poetics, as well as jazz and painting, seems to be J66 173 chronologically parallel, which is to say you have fixed form, J66 174 which then evolves toward more free form where you get let loose J66 175 from this specific repeated rhythm and improvise the rhythms even, J66 176 where you don't have a fixed rhythm, as in bebop the drum became J66 177 more of a soloist in it too. So you find that in painting, the J66 178 early de Koonings have a motif or a theme, the woman or something J66 179 like that, but it gets more and more open, less dependent on the J66 180 theme, and in poetry, where you have less and less dependence on J66 181 the original motifs and more and more John Ashberyesque J66 182 improvisational free form flowing without even a subject matter, J66 183 though I always kept a subject matter like the old funky blues J66 184 myself. It was partly a parallel development within each J66 185 discipline: painting, poetry, music. There were innovators who J66 186 opened up the thing after Einstein, so to speak - you know, J66 187 relative measure, as Williams said - which is in a sense something J66 188 that happened with bebop: not the fixed measure but a relative J66 189 measure. J66 190 J67 1 <#FROWN:J67\>One indication of poetry's place in the current J67 2 construction of African American critical theory is revealing. In J67 3 The Signifying Monkey, Henry Louis Gates, Jr., proposes a J67 4 compelling way of tracing the lines of continuity among major J67 5 African American texts. With consummate skill, he teaches us to J67 6 read this literature through the paradigms of vernacular culture as J67 7 a way of chronicling a common African American struggle for J67 8 authority and voice. Taking off from Roland Barthes, Gates defines J67 9 Hurston's Their Eyes Were Watching God as the first J67 10 instance in our written tradition of a "speakerly" text - J67 11 that is, a text "oriented toward imitating one of the J67 12 numerous forms of oral narration to be found in classical J67 13 Afro-American vernacular literature" (181). Gates's J67 14 characteristic reinvention of literary theory, grounded in the J67 15 language and rituals of African American culture, is insightful, J67 16 yet this particular moment in his revisionary strategy raises an J67 17 important concern. By slighting poetry in his newly constructed J67 18 canon, Gates neglects such "speakerly" nineteenth-century J67 19 texts as Paul Laurence Dunbar's seminal poem 'An Ante-bellum J67 20 Sermon' (1895). Elsewhere, Gates has written instructively about J67 21 Dunbar and African American poetry. His early essay 'Dis and Dat: J67 22 Dialect and the Descent' remains crucial to our understanding of J67 23 the conventions of 'dialect' and vernacular poetry. My point here J67 24 is simply that his exclusion of poetry from The Signifying J67 25 Monkey, given the ensuing authority of that work, reenacts J67 26 what has become a familiar pattern in African American literary J67 27 history. The effect of such omissions in our most influential J67 28 readings of African American literature may well serve to lessen J67 29 the perspective of poetry in current revisions of black critical J67 30 theory and the African American canon.

J67 31 In his essay 'Performing Blackness,' Kimberly Benston surmises J67 32 that it may be the "relative susceptibility" of J67 33 African American poetry "to discrete analyses [that] J67 34 resists theoretical efforts that move toward totalization, toward J67 35 recuperation and ideological closure" (184). He confronts J67 36 the risks involved in theorizing about the lines of continuity J67 37 between disparate poetic works only to underscore the necessity of J67 38 defining the shared themes, literary forms, and rhetorical J67 39 strategies that constitute a distinctive poetic tradition. In J67 40 particular, he entreats us to recognize that seemingly discrete, J67 41 fragmented expression - enacted as a mode of performance - J67 42 characterizes African American discourse. Black utterance, like J67 43 other counterdiscourses, masks its own continuity as a way of J67 44 veiling its challenges to the legitimacy of dominant political J67 45 institutions and cultural traditions. In theorizing about African J67 46 American poetry, critics might be guided by what Richard Terdiman J67 47 describes, in a different context, as the "capacity [of J67 48 counterdiscourses] to situate: to relativize the authority and J67 49 stability of a dominant system of utterances which cannot even J67 50 countenance their existence," as a way of enacting a new J67 51 realm of subjectivity (15-16). When black writers turn to African J67 52 American vernacular performance, they call into question the J67 53 authority of the literary conventions and racial ideologies of the J67 54 dominant society. Critics should neglect neither the insistence of J67 55 African American poets on challenging the authority of dominant J67 56 American institutions nor the individual black poet's struggle for J67 57 recuperation and wholeness. Their individual texts make up an J67 58 ongoing tradition of black poetic subjectivity.

J67 59 While the term performance can be applied variously to a J67 60 range of cultural and literary phenomena, I use it to designate J67 61 verbal performance viewed as a cultural event. In revising the J67 62 boundaries of what has traditionally constituted the folklore text, J67 63 recent scholars, especially Roger D. Abrahams and Richard Bauman, J67 64 have reconceptualized the distinctions between textual J67 65 representations and what Robert Georges calls "complex J67 66 communicative events" (313). Such scholars often associate J67 67 the term vernacular with the modern concept of folklore as an J67 68 intricate interaction between performer and audience that relies on J67 69 linguistic, paralinguistic, kinesic, and thoroughly contextual J67 70 codes and conventions. This implied notion of performance has J67 71 become a model for how to discuss formal literary texts. In written J67 72 texts that draw on the aesthetics of vernacular performance, the J67 73 relations of orality and literacy are continuous. The tensions J67 74 between repetition and improvisation that operate in a verbal J67 75 performance are translated into competing structures of creation J67 76 and recollection for literary artists and their audiences.

J67 77 Construing performance in this way, I argue that Paul Laurence J67 78 Dunbar's early poem 'An Ante-bellum Sermon' is an instructive J67 79 example of an African American "preacherly text." J67 80 The poetic heritage to which this poem belongs stretches back at J67 81 least to the eighteenth- and nineteenth-century sermons of African J67 82 American religious leaders, even as it looks forward to the J67 83 preacherly performances of James Weldon Johnson, Langston Hughes, J67 84 Robert Hayden, and Gwendolyn Brooks. In scaling what Hughes calls J67 85 the "racial mountain," Dunbar's vernacular J67 86 performance illustrates the solid ground of African American poetic J67 87 traditions.

J67 88 I J67 89 For black poets of the nineteenth and twentieth centuries, the J67 90 challenge has been to establish the grounds of their literate and J67 91 poetic authority without sacrificing the distinctiveness of their J67 92 experiences in the New World. These poets have often relieved the J67 93 disruptions resulting from different aspects of themselves by J67 94 locating their work in a continuum of African American expressive J67 95 culture. In their search for poetic voice, they have struggled to J67 96 define their experiences as African Americans through tropes of J67 97 indigenous and communal subjectivity. Many of them, including J67 98 Dunbar, Hughes, Hayden, and Brooks, have made vernacular sermonic J67 99 performance of their heritage a primary site of cultural authority J67 100 and artistic creativity. The narratives, rhetorical strategies, and J67 101 rituals of performance of the vernacular sermon have helped shape J67 102 the recurring aesthetic and ideological tendencies of African J67 103 American poetry. The relations of preachers and their congregations J67 104 have provided these poets with a model for what Barbara Bowen has J67 105 called "untroubled voice": the performance of J67 106 perfect continuity between artist and audience. By performing these J67 107 sermons, black poets have mended the divisions in their artistic J67 108 voices and the contradictory expectations of their various J67 109 audiences, thereby recovering for themselves and their communities J67 110 the privileges of vernacular eloquence.

J67 111 To explicate the cultural status of the vernacular sermon, I J67 112 begin with the historians of the American slave - including John J67 113 Blassingame, Eugene Genovese, Lawrence Levine, and Al Raboteau - J67 114 who have drawn on 'folklore,' interviews, autobiographies, and J67 115 various other kinds of texts to suggest that slaves were able to J67 116 survive their oppression by evolving sustaining forms of cultural J67 117 expression. The early black sermonic performance was one of the J67 118 rituals that defined for slaves and free African Americans their J67 119 participation in a unique religious fellowship. According to the J67 120 available evidence, slave preachers were esteemed as exemplary J67 121 figures in the black community, whether they preached in church, in J67 122 the quarters, or in the fields, whether they preached out of the J67 123 Bible or from the depths of their hearts. While many slaves were J67 124 required to listen to white preachers, some of whom were well J67 125 educated and well trained, they clearly preferred to hear black J67 126 preachers deliver messages from God. As Nancy Williams, and J67 127 ex-slave from Virginia, put it: "Dat ole white preachin' J67 128 wasn't nothin'. Ole white preachers used to talk wid dey tongues J67 129 widdout sayin' nothin' but Jesus told us slaves to talk wid our J67 130 hearts" (Yetman 13).

J67 131 The slave preachers did not necessarily derive their cultural J67 132 authority from their ability to read or their facility in speaking J67 133 standard English. Often the important issue was whether they could J67 134 perform sermons that moved their congregations toward freedom. One J67 135 former slave, Clara Young, told her interviewer that it did not J67 136 matter that her favorite preacher, Mathew Ewing, was illiterate; J67 137 what she cared about most in judging his sermons was his style and J67 138 his message. "He never learned no readin' and J67 139 writin'," she reminisced, "but he sure knowed his J67 140 Bible and would hold his hand out and make like he was readin' and J67 141 preach de purtiest preachin' you ever heard" (Yetman 335). J67 142 She must have been impressed with his wit in reenacting the J67 143 gestures of white preachers who read Scripture to their J67 144 congregations, but not as much as she was stirred by his eloquence. J67 145 Judging from his vast experience with slave congregations, the J67 146 white minister Charles C. Jones could say that they were natural J67 147 judges of a "good sermon," even though such a J67 148 sermon might corrupt Christian theology and the English language J67 149 (14-15). While recognizing that the slaves preferred to hear black J67 150 preachers, he was not willing to admit that these preachers were J67 151 evolving their own style of address. Jones championed the use of J67 152 religion among slaves as a means of social control without ever J67 153 realizing that slaves would use their religion as a way of fighting J67 154 oppression.

J67 155 Many black congregations expected their preachers to be J67 156 vigorous, dramatic, and instructive in language, theme, and J67 157 gesture. The textual accuracy of their preachers' readings of the J67 158 Bible may not have been a consideration, but the dynamics of J67 159 performance were essential. Even if the preachers had to create the J67 160 biblical references in their sermons, their congregations demanded J67 161 that they breathe life into those creation. In her Letters J67 162 from New York, Lydia Maria Child, the well-known abolitionist, J67 163 describes an especially dramatic sermon that Rev. Julia Pell, a J67 164 black itinerant preacher, delivered to a Methodist congregation in J67 165 1841. Although the regular minister of this church felt compelled J67 166 to apologize to the congregation for Pell's misquotations of J67 167 Scripture, her performance must have been remarkable. Even the J67 168 skeptical Lydia Child admitted that this "dusky priestess J67 169 of eloquence" made her shout and cry with religious fervor J67 170 (67). Child offers a rare early description of the rhythmic pacing J67 171 of a black sermon, noting that Pell "began with great J67 172 moderation, gradually rising in her tones, until she arrived at the J67 173 shouting pitch ... [that] she sustained for an incredible time, J67 174 without taking breath, and with a huskiness of effort." J67 175 This rhythm of performance, and the physical effort it required, J67 176 undoubtedly reinforced the effect of the sermon's thematic climax J67 177 and encouraged the enthusiastic response of the congregation. The J67 178 section of this climax that Child records can be transcribed as J67 179 lines of poetry:

J67 180 Silence in Heaven!

J67 181 The Lord said to Gabriel,

J67 182 bid all the angels keep silence.

J67 183 Go up into the third heavens,

J67 184 and tell the archangels to hush their golden harps.

J67 185 Let the sea stop its roaring,

J67 186 and the earth be still.

J67 187 What's the matter now?

J67 188 Why, man has sinned,

J67 189 and who shall save him?

J67 190 Let there be silence,

J67 191 while God makes search for a Messiah.

J67 192 Go down to the earth;

J67 193 make haste, Gabriel,

J67 194 and inquire if any there are worthy;

J67 195 make haste, Gabriel;

J67 196 and Gabriel returned and said,

J67 197 No, not one.

J67 198 But don't be discouraged.

J67 199 Don't be discouraged, fellow sinners.

J67 200 God arose in his majesty,

J67 201 and he pointed to his own right hand,

J67 202 and said to Gabriel,

J67 203 Behold the Lion of the tribe of Judah;

J67 204 he alone is worthy.

J67 205 He shall redeem my people. (65)

J67 206 While the members of the church congregation could not have J67 207 found this conversation in their Bibles, they would have understood J67 208 that the invented narrative testified to the possibility of J67 209 redemption. Although Child cannot record the sound of the sermon, J67 210 one can discern a pattern of rising and falling intonations and J67 211 perceive the shrillness of Pell's dynamic "shouting J67 212 pitch." As the black preachers tell us at the beginning of J67 213 their sermons, they preach as God's instruments. Here Pell J67 214 literally creates music with her voice. She provides the J67 215 congregation with proof that she is one of "God's J67 216 trombones."

J67 217 Pell could neither read nor write, Child reports, yet this J67 218 preacher's 'illiteracy' does not limit her talents as an artist. J67 219 Without 'book learning,' she nonetheless reads and performs the J67 220 cultural conventions of the black vernacular sermon. While J67 221 societies define and use literacy differently, J67 222 "literacy is always connected with power" (Pattison J67 223 viii). The sermon quoted above demonstrates that Pell is literate J67 224 in the rituals of African American culture, and it measures the J67 225 extent to which she exploits her power as a prophet. Unable to read J67 226 the Bible, Pell still knows what sermons her congregation needs to J67 227 hear. Such knowledge is communal. It is shared by preacher and J67 228 audience.

J67 229 Knowing the tacit rules of performance, many of the blacks in J67 230 Pell's audience would have recognized her skill in drawing on the J67 231 conventions of delivering a vernacular sermon: the conspicuous J67 232 patterns of repetition; the Old Testament imagery and diction; the J67 233 allusions to the spirituals; the preacher's freedom to assume the J67 234 identity of a biblical character; her dwelling on the inadequacies J67 235 of language in the face of divine revelation; her own version of a J67 236 biblical character's conversion, narrated through the devices of J67 237 vernacular storytelling. J67 238 J67 239 J68 1 <#FROWN:J68\>These people are observers who contribute to the J68 2 composite picture of Gary Gilmore, but they also help Mailer J68 3 achieve the broad social panorama he admires in writers as J68 4 different as Tolstoy and Dreiser. Indeed, Mailer has chided himself J68 5 for doing so little with the secondary characters in his previous J68 6 novels, a 'flaw' he hoped to correct in The Executioner's J68 7 Song. Here Mailer develops virtually every 'minor' character J68 8 and permits each to speak in something like his or her own voice, J68 9 however much the several idioms blend into the flat, colloquial J68 10 style for which the book is famous. Mailer's defense of his J68 11 unadorned prose might apply to the minor characters themselves: J68 12 "one's style is only a tool to use on a dig." Like J68 13 the style by which we know them, the secondary characters are J68 14 supposed to contribute to the book's larger formal ends.

J68 15 One such end is to 'examine' the American reality exposed by J68 16 the strange saga of Gary Gilmore. Joan Didion sees Mailer as J68 17 capturing two crucial features of western America. The first is J68 18 "that emptiness at the center of the Western experience, a J68 19 nihilism antithetical not only to literature but to most other J68 20 forms of human endeavor." The second is an inability to J68 21 direct our own lives, a failing so pervasive that all the J68 22 characters seem to share in "a fatalistic drift, a tension, J68 23 an overwhelming and passive rush toward the inevitable events that J68 24 will end in Gary Gilmore's death." I believe that Didion's J68 25 insights are exaggerated, but they do point up suggestive J68 26 connections between Gilmore and the people who surround him. Bessie J68 27 Gilmore, Brenda Nicol, Vern Damico, Kathryne Baker and her J68 28 daughters Nicole and April - all are 'trapped' in their futile J68 29 efforts to find a life worth living. Indeed, almost every woman in J68 30 the book first marries at fifteen or sixteen and eventually marries J68 31 at least three or four times, and the men seem equally caught up in J68 32 the fatalistic drift Didion notices. Didion does not do justice to J68 33 the admirable stability of people like Brenda Nicol and Vern J68 34 Damico, but the wasted lives of those around Gilmore suggest that J68 35 his own fate is only an exaggerated instance of that moral J68 36 emptiness Didion hears in the book's western voices.

J68 37 In this respect as in others, Nicole Baker is the second most J68 38 important character. Mailer has called her "a bona fide J68 39 American heroine," but most readers will think she is J68 40 rather the quintessential American victim. Promiscuous at eleven, J68 41 institutionalized at thirteen, married at fourteen and again at J68 42 fifteen, Nicole suffers three broken marriages before she is J68 43 twenty. "Sex had never been new to Nicole," we are J68 44 told (143), and it is more than plausible when she runs off with an J68 45 older man because "she didn't care where she was J68 46 going" (117-18). Yet Nicole has virtues to match her J68 47 troubling irresponsibility. As Gilmore sees, she is fearless and J68 48 fiercely loyal. These are the very qualities that Gilmore counts on J68 49 when he manipulates her toward a suicide pact. In his many letters J68 50 from jail, he pleads with Nicole not to make love with other men J68 51 (350), to give up sex altogether (403-04), and to join him on the J68 52 other side in death (472). At the end of book 1, he leads her J68 53 toward a double suicide attempt that epitomizes both his J68 54 romanticism and his selfishness, even as it climaxes Mailer's J68 55 portrait of Nicole as an endearing victim. Later Nicole will be J68 56 denied the 'clean' resolution of death, will emerge from yet J68 57 another institution to tell Larry Schiller (and Mailer) the story J68 58 of her love for Gary Gilmore, and will finally drift off to Oregon J68 59 to new lovers if not a new life. Nicole's story is a familiar one J68 60 among her family and friends: years of acute aimlessness followed J68 61 by an utterly hopeless commitment. Surely it is no accident that J68 62 Nicole comes to love Gilmore most fiercely when he is cut off from J68 63 her forever. For the Nicoles of the world (and perhaps this means J68 64 for all of us), there is no consummation except in an imagined J68 65 future.

J68 66 The stories of Nicole and the other witnesses point to one of J68 67 Mailer's most crucial decisions in structuring book 1. Rather than J68 68 trace Gilmore's grim history from reform school through his term in J68 69 Marion, Illinois, Mailer chooses to focus on Gilmore's last months J68 70 in Provo in 1976. The reasons for this no doubt include Mailer's J68 71 desire to achieve greater dramatic unity and to emphasize Gilmore's J68 72 'mystery' instead of the familiar stages of American crime and J68 73 punishment. But another important reason is to allow Mailer to J68 74 flesh out the human context in which Gilmore plays his final role J68 75 or sings his final song, as the title would have it. This context J68 76 is dominated by the same hateful 'habits' that take more J68 77 spectacular forms in Gilmore. Yet the human resources displayed in J68 78 book 1 should not be dismissed quite so easily as Didion's J68 79 formulation would suggest. Here we get example after example of J68 80 human folly, western style, but also many instances of what Mailer J68 81 calls "American virtue," the American's dogged J68 82 determination to do his or her best in the worst of circumstances. J68 83 The range of such portraits is really quite extraordinary, from J68 84 Gilmore's mother, Bessie, to Brenda Nicol, to the Damicos, to the J68 85 irrepressible Nicole. One of the earliest reviewers called The J68 86 Executioner's Song "a remarkably compassionate J68 87 work," and the truth in this judgment should remind us J68 88 that, like Mailer's portrait of Gilmore, book 1 is structured to J68 89 highlight the human frailties as well as the abominations of J68 90 American life.

J68 91 It might seem that book 2 offers a less sympathetic, more J68 92 satirical history of Gilmore's last months. The very title of part J68 93 1, "In the Reign of Good King Boaz," signals a new J68 94 kind of irony. Here lawyers and the press are omnipresent and one J68 95 eighty-two-page section, 'Exclusive Rights,' is devoted to J68 96 virtually nothing but Larry Schiller's and David Susskind's efforts J68 97 to corner the Gilmore market, so to speak, by securing exclusive J68 98 rights to his story. Packs of reporters are everywhere, confirming J68 99 Mailer's worst fears about press. The many lawyers introduced are J68 100 often distinguished by one bizarre detail or another, as when Earl J68 101 Dorius, Utah's assistant attorney general, is excited at the J68 102 prospect of an execution and proceeds to work himself into a near J68 103 breakdown to ensure that the state of Utah gets its execution on 17 J68 104 January 1977 (500), or when Dennis Boaz, Gilmore's second lawyer, J68 105 supports his client's desire to be executed until it occurs to him J68 106 that Gary would prefer to live if he could have connubial visits J68 107 from Nicole (590-91), perhaps in Mexico (611)! Gilmore's final J68 108 lawyers, Bob Moody and Ron Stanger, are a good deal less eccentric, J68 109 but they too partake in the grim legal struggle in which the state J68 110 of Utah pursues its pound of flesh, and the ACLU and other liberal J68 111 groups fight stubbornly to save a man who does not want to be J68 112 saved. The ironies here are obvious and may even seem undramatic. J68 113 In the film version of The Executioner's Song (1982), J68 114 scenarist Mailer and director Schiller chose to leave out most of J68 115 the materials of book 2, as if they were less relevant than the J68 116 more 'immediate' events of book 1.

J68 117 My own view is that book 2 is at least as interesting as book J68 118 1, a remarkable feat when one considers that the protagonist is all J68 119 but unavailable and the heroine is locked up throughout. Once again J68 120 Mailer gets great mileage from his so-called minor figures, a few J68 121 of whom (e.g., Boaz, Schiller, Barry Farrell) are among his most J68 122 memorable characters. Of real interest for their own sake, they J68 123 also provide perspective on Gilmore. For example, Gary's brother J68 124 Mikal is at first reluctant to allow his brother to die and J68 125 participates in legal actions to prevent it. When he finally talks J68 126 with Gary, however, Mikal is won over by his brother's seriousness J68 127 and depth of feeling. As they part, Gary first kisses Mikal, then J68 128 utters perhaps the most haunting words in this very long book: J68 129 "See you in the darkness" (840). A cellmate of J68 130 Gilmore's named Gibbs also effectively testifies on Gary's behalf. J68 131 A police informer, Gibbs refers to Gilmore as the most courageous J68 132 convict he has ever been (759). And Gilmore's relatives, especially J68 133 Vern Damico and Toni Gurney, find themselves moving ever closer to J68 134 Gilmore as he approaches death. Toni's relationship with Gilmore is J68 135 especially moving. She first visits him the day before he is to be J68 136 executed and is overwhelmed by his gentle affection (874-75). Later J68 137 that day, after her own birthday party, she returns to the party J68 138 Gilmore has been permitted at the prison and again experiences J68 139 Gary's new warmth (884-86). Toni is sufficiently moved to try to J68 140 attend Gary's execution (929). This sequence blends with many other J68 141 small but affecting moments to verify the change in Gilmore that is J68 142 sensed by many people during his final weeks.

J68 143 Mailer uses Barry Farrell and Larry Schiller to temper the more J68 144 sentimental implications of book 2, but ultimately these veteran J68 145 journalists also testify to Gilmore's surprising depth. The title J68 146 of book 2, "Eastern Voices," seems to refer to all J68 147 those safely established in the social system, whether in the East J68 148 or the West: lawyers, reporters, producers, assistant attorney J68 149 generals, and so on. Farrell and Schiller are such voices. Each J68 150 brings a heavy load of urban skepticism to the Gilmore assignment, J68 151 hating Salt Lake City, as Farrell does, and believing there is no J68 152 'center' to this story, nothing of real human resonance (577). When J68 153 both men come to see Gilmore in a very different light, Mailer is J68 154 able to bring his book to a genuine climax.

J68 155 Farrell is at first confident that nothing sets Gilmore apart J68 156 but his willingness to die. If Gilmore is not executed, Farrell J68 157 suggests, he will become indistinguishable from the hundreds of J68 158 others condemned to die but never executed (611). As he works with J68 159 Gilmore's responses to hundreds of questions, however, Farrell J68 160 notices that Gilmore "was now setting out to present the J68 161 particular view of himself he wanted people to keep" (711). J68 162 Later Farrell responds profoundly to Gilmore's tapes: J68 163 "Barry was crying and laughing and felt half triumphant J68 164 that the man could talk with such clarity" (804). Farrell J68 165 still believes that Gilmore "had a total contempt for J68 166 life" (805), but this makes it all the more impressive when J68 167 Gilmore responds so "humanely" to the massive attention of J68 168 his last months (805). Farrell is stunned at Gilmore's apparent J68 169 complexity. In the transcripts Farrell spots "twenty-seven J68 170 poses," twenty-seven different Gilmores ("racist J68 171 Gary and Country-and-Western Gary, artist manqu<*_>e-acute<*/> J68 172 Gary, macho Gary," 806). Farrell begins to pursue the J68 173 single Gary who presumably stands behind these multiple poses, but J68 174 he is "seized with depression at how few were the J68 175 answers" to his inquiry (811). There is an "evil J68 176 genius" in Gilmore's planning Nicole's suicide, but much J68 177 else in Gilmore's life suggests sheer ignorance (812); Gilmore's J68 178 relations with Bessie, his mother, seem a potential key, but the J68 179 answers to many related questions provide no "hope of a J68 180 breakthrough" (827; see 844). Continuing to ponder J68 181 Gilmore's transcripts just before the execution, Farrell turns to J68 182 yet another possible solution to the Gilmore mystery: Gilmore's J68 183 fascination with small children. But this 'answer' is also J68 184 unsatisfactory: "It was too insubstantial. In fact, it was J68 185 sheer speculation .... beware of understanding the man too J68 186 quickly!" (855). Beware indeed. Farrell's final comment on J68 187 Gilmore takes us back to the passage from Andr<*_>e-acute<*/> Gide J68 188 ("Please do not understand me too quickly") that J68 189 Mailer first used as his epigraph to The Deer Park J68 190 (1955). Farrell's conclusion should caution us against reductive J68 191 readings, psychological efforts to pluck out Gilmore's mystery. J68 192 Indeed, Gilmore's complexity should impress us as much as it does J68 193 Farrell, whose prolonged efforts to understand Gilmore are akin to J68 194 Mailer's.

J68 195 Larry Schiller's role is in part like Farrell's. Schiller also J68 196 looks for the human side to Gilmore, the "sympathetic J68 197 character" buried inside the cold-blooded killer (629), for J68 198 Schiller cannot imagine making a successful book or film unless he J68 199 first makes this discovery. J68 200 J69 1 <#FROWN:J69\>1.1 PEX and PEXlib

J69 2 PEX is the 3D extension to X. It adds over 200 X protocol J69 3 requests for defining and displaying 3D pictures. PEX provides all J69 4 the common features found in most modern 3D graphics systems, but J69 5 provides them in a way that's seamlessly integrated with X.

J69 6 The PEX graphics model is very unlike the 2D graphics model of J69 7 basic X. With PEX you can define objects in convenient coordinate J69 8 systems, rotate and move them with modeling transforms, view them J69 9 from different angles, and create lit and shaded scenes.

J69 10 1.1.1 The PEX Protocol

J69 11 As most X programmers know, the underlying mechanism of X is J69 12 the X protocol. While most of us think of X as consisting of Xlib, J69 13 which generates X protocol, and the X server, which interprets X J69 14 protocol, the essence of X is really the protocol itself. It is J69 15 what defines X and allows it to be interoperable.

J69 16 PEX is an extension to the X protocol. Like the X protocol, PEX J69 17 protocol travels between the client - the application program J69 18 - and the X server (see Figure 1-1). The server contains the J69 19 PEX server extension, which receives and interprets the J69 20 PEX protocol and executes the PEX requests. As always in X, the J69 21 client and server can be on the same machine or on different J69 22 machines.

J69 23 figure&caption

J69 24 Although PEX is defined as a protocol, it specifies a highly J69 25 capable graphics system. This system and how you use it is what J69 26 this book is all about.

J69 27 1.1.2 PEXlib

J69 28 Just as applications use Xlib to create and send X protocol, J69 29 so, too, they use PEXlib to create and send PEX protocol. Xlib and J69 30 PEXlib act as the application's agent, formatting and sending J69 31 requests to carry out the application's will. You do not need to J69 32 know PEX protocol to use PEXlib. You need only to understand the J69 33 PEX capabilities and PEXlib functions.

J69 34 Figure 1-1 shows the relationships between an application, J69 35 Xlib, PEXlib, and the X server. The application sits above Xlib and J69 36 PEXlib. It calls Xlib for the usual window management tasks, such J69 37 as making server connections and creating windows, and calls PEXlib J69 38 to draw 3D images. The X and PEX requests travel to the server J69 39 intermixed on the same communication channel. The request J69 40 dispatcher in the server routes the X requests to the server's J69 41 core, and sends the PEX requests to the PEX extension. If other J69 42 extensions are being used, then requests for them, too, are J69 43 intermixed with the X and PEX requests; the server's dispatcher J69 44 routes them to the correct portion of the server.

J69 45 In our descriptions of PEX and PEXlib, we'll refer mostly to J69 46 PEX, because that's really the system that's providing the 3D J69 47 functionality. We'll use PEXlib only when we mean specifically the J69 48 PEXlib interface to PEX.

J69 49 1.1.3 What's In the Name? Only History

J69 50 PEX is an acronym for 'PHIGS Extension to X.' This implies that J69 51 the purpose of PEX is to support PHIGS, a popular 3D graphics J69 52 standard (§1.4). But while PEX does indeed support PHIGS and has J69 53 had much of its definition taken from PHIGS, PEX goes beyond PHIGS J69 54 in its functionality. The name, at this point, is merely J69 55 historical, indicating where PEX had its origins, but not where it J69 56 is today.

J69 57 1.1.4 How Do I Know if I Have PEX?

J69 58 You can determine whether a server has a PEX extension by J69 59 invoking the xdpyinfo command and searching its output for the J69 60 string 'X3D-PEX'.

J69 61 % xdpyinfo | grep X3D-PEX

J69 62 The R5 sample sever from the M.I.T. X Consortium includes a PEX J69 63 extension (§1.5).

J69 64 1.2 PEXlib as a Graphics Library

J69 65 Putting aside PEXlib's role as an X extension, you can look at J69 66 it as simply another 3D graphics library, one that meshes smoothly J69 67 with X. Graphics libraries sit between the application and the J69 68 display device, manipulating the device in response to application J69 69 commands. Graphics libraries can be low-level or high-level (see J69 70 Figure 1-2). Low-level libraries typically provide commands for J69 71 manipulating pixels, device registers, and other features of the J69 72 display hardware. High-level libraries deal in abstractions like J69 73 geometric objects and color. They hide low-level and J69 74 hardware-dependent details.

J69 75 figure&caption

J69 76 1.2.1 PEXlib Is a High-level Library

J69 77 PEXlib is a high-level graphics library. It allows a programmer J69 78 to describe a graphic image in terms of familiar objects and J69 79 attributes, without having to deal with the details of producing J69 80 that image in a window. PEXlib lets a user say, Draw a wide J69 81 red line from this point to that point, without requiring her J69 82 to figure out which pixels to turn on and how to make them look J69 83 red. All the details of producing the picture are handled by the J69 84 PEX server. The user merely specifies the geometry, the location, J69 85 and some appearance attributes for the objects.

J69 86 This convenience extends to objects more complex than lines and J69 87 attributes more subtle than color. Say an application wants a J69 88 picture of a room full of furniture. PEX lets the user describe J69 89 where the walls of the room are, the shape, location and color of J69 90 the furniture, what types of lights are on and their locations, and J69 91 where the viewer is. It then figures out how the room looks to that J69 92 viewer - which parts are bright or dim, how the aim of the lights J69 93 varies the color across the walls and upholstery - and sets the J69 94 pixels on the screen to the values required by the picture (see J69 95 Figure 1-3). PEX frees the programmer to compose a picture and not J69 96 worry about the details of producing it.

J69 97 PEX provides a set of familiar graphics objects called J69 98 primitives, each with attributes that control its location, J69 99 orientation, color, and appearance. You can define simple models or J69 100 complex scenes with these primitives. Multiple views of the same J69 101 objects can be displayed. You can even animate a scene by setting J69 102 it in motion.

J69 103 figure&caption

J69 104 1.2.2 Displaying a Picture

J69 105 PEX gives you two ways to draw a picture: 1) Pass all the J69 106 primitives in the scene to PEXlib one by one and have PEX display J69 107 them immediately. 2) Store the primitives in a graphics J69 108 database, then tell PEX to display the database. In the first J69 109 method, commonly called immediate mode, you must send all J69 110 the primitives to PEX each time you want to change or redisplay the J69 111 picture; once PEX draws the primitives, it forgets about them. In J69 112 the second method, called structure mode, you need not J69 113 re-send the primitives for each redisplay, but merely tell J69 114 PEX to redisplay the stored data-base. You can selectively J69 115 edit the database between redisplays, changing only those parts of J69 116 the picture that should be different in each scene. The primitives J69 117 and their attributes are stored in the database in containers J69 118 called structures.

J69 119 There are advantages and disadvantages to both these methods. J69 120 Most applications are suited to one or the other. Immediate mode is J69 121 best for applications that must continuously re-specify the J69 122 complete images, such as animated simulation programs that display J69 123 different geometry with each frame. Structure mode is best for J69 124 programs such as computer-aided design programs that create a J69 125 graphics model and edit portions of it frequently or view it in J69 126 different ways.

J69 127 It's also possible to mix both immediate mode and structure J69 128 mode. You can store the static or infrequently changed parts of J69 129 your model in structures, and draw the other parts using immediate J69 130 mode.

J69 131 1.2.3 Limitations of PEX

J69 132 PEX has some limitations. It does not do ray tracing or J69 133 radiosity. It doesn't compute shadows or follow light as it bounces J69 134 from one object to another. It does not yet provide texture J69 135 mapping, so you can't tell it to accurately display your marble J69 136 table. And motion blur and realistic fog would be tough to do with J69 137 PEX.

J69 138 PEX implementations can extend PEX functionality, so some PEX J69 139 implementations may indeed calculate shadows or perform texture J69 140 mapping; but they'll do it in an implementation-dependent way that J69 141 may not be the way it's done in other implementations.

J69 142 PEX implementations may have other limits. Most restrict the J69 143 number of light sources you can use (although often to a high J69 144 number), some don't do depth cueing, and some provide only flat or J69 145 Gouraud shading. Most PEX implementations do provide the majority J69 146 of the PEX functionality, however, and offer a powerful, useful, J69 147 and portable set of graphics capabilities, even though they don't J69 148 support all PEX features. PEXlib provides functions for determining J69 149 which features a PEX server extension does support.

J69 150 1.3 Overview of PEXlib

J69 151 Table 1-1 shows you what's in PEX. It lists the major PEX J69 152 features and tells you where in this book we've put their primary J69 153 description. Examples of using the features are throughout the J69 154 book. See the Index to find the location of specific examples.

J69 155 caption&table

J69 156 1.4 The Relationship between PEX and PHIGS

J69 157 PEX owes much of its origin to PHIGS. PHIGS [14] is a 3D J69 158 graphics standard sanctioned by the International Organisation for J69 159 Standardisation (ISO). It's been available since 1988, and is used J69 160 by many graphics applications. A primary goal of PEX is to support J69 161 PHIGS in the X environment, although that is not its only goal.

J69 162 To an application, PHIGS serves much the same purpose as J69 163 PEXlib, sitting between the application and the server, as shown in J69 164 Figure 1-4. In the X environment, PHIGS sends PEX protocol to the X J69 165 server to carry out the PHIGS functions.

J69 166 With the birth of PEXlib, PHIGS implementations can themselves J69 167 use PEXlib to format and send the PEX requests.

J69 168 figure&caption

J69 169 Most of the PEX requests were designed to support the identical J69 170 functionality in PHIGS. Consequently, PEXlib and PHIGS use the same J69 171 graphics primitives and attributes, and share the same model for J69 172 defining a scene and rendering it. But PEXlib goes beyond PHIGS, J69 173 providing immediate-mode rendering and better integration with J69 174 X.

J69 175 Because the PHIGS standard was defined before X was popular, it J69 176 has no knowledge of X, specifically, the role of X in controlling J69 177 the display and input devices and the aim of X to be 'policy free.' J69 178 PHIGS defines its own ways to control the screen and the available J69 179 input devices. It defines a central graphics database rather than a J69 180 distributed one, and makes several of the decisions that a J69 181 policy-free interface would leave to the application. Therefore J69 182 it's not well suited as an extension to Xlib. PHIGS can be J69 183 implemented so that it's well-integrated within X, but parts of it J69 184 would have to be redefined to serve as an extension to Xlib - a J69 185 nearly impossible task now that PHIGS is an approved standard and J69 186 in wide use.

J69 187 PEX and PEXlib are the results of a fresh start on a 3D J69 188 interface for X. While accommodating and building on PHIGS, they J69 189 begin with the assumption that 3D graphics should be an integral J69 190 part of the window system. PEXlib uses the existing X mechanisms: J69 191 the communication channel, protocol requests and replies, the X J69 192 event queue, and error events. It avoids specifying policy, and J69 193 instead provides the functions needed by the application to execute J69 194 its own policy. By providing an immediate-mode capability, PEX J69 195 meets the needs of more graphics applications than PHIGS.

J69 196 1.5 Where to Get PEX and PEXlib

J69 197 PEX server extensions are available from most major workstation J69 198 vendors, as well as the M.I.T. X Consortium. PEXlib, too, is J69 199 available from these same sources.

J69 200 The PEX and PEXlib products provided by workstation vendors are J69 201 usually highly optimized for the vendor's computer and display J69 202 devices. These implementations typically provide the best possible J69 203 performance for those devices. Some vendors require you to buy J69 204 their graphics accelerator hardware to get reasonable or high J69 205 performance, however.

J69 206 The X Consortium's PEX implementation is known as the J69 207 sample implementation, or simply, the PEX-SI. As of J69 208 late 1992, the PEX-SI does not support some high-level PEX J69 209 features, such as shading and hidden line and hidden surface J69 210 removal. It also does not provide the highest possible performance J69 211 on many display devices. Nevertheless, it's a reliable and useful J69 212 implementation, and provides a good basis for learning PEX and a J69 213 good starting point for building more complete implementations.

J69 214 There is only a C interface for PEXlib. We have not heard of an J69 215 effort to define a FORTRAN binding or any other interface.

J69 216 J70 1 <#FROWN:J70\>Enhancing CRP values

J70 2 By Ted Hawn and Mike Getman

J70 3 Cooperative, interagency assistance to CRP J70 4 contract-holders in Montana has resulted in some model acreages J70 5 from the standpoint of soil erosion control and wildlife habitat J70 6 development

J70 7 A model Conservation Reserve Program (CRP) contract. That's how J70 8 one might describe Lee Berry's land in central Montana. Berry is a J70 9 landowner with an eye to the future. His 445 acres, enrolled in the J70 10 CRP in 1988, now provide optimal values for soil erosion control J70 11 and wildlife habitat.

J70 12 Seeding the recommended mix of three wheatgrasses, alfalfa, and J70 13 clover, Berry also planted nearly 24,000 feet of tree rows - more J70 14 than 4,400 trees. At a county tour held during the fall of 1989, J70 15 county commissioners, conservation district supervisors, and other J70 16 local agency personnel saw first-hand what it takes to J70 17 successfully establish CRP cover. At a stop on the Berry farm, tour J70 18 delegates were duly impressed when several walked into a J70 19 seven-foot-high clover and grass field resembling a willow thicket J70 20 and disappeared from view. With assistance from the U.S. Fish and J70 21 Wildlife Service (FWS), three ponds were constructed and food plots J70 22 were added. The diversity of habitat has resulted in phenomenal J70 23 changes in the wild-life species present and their J70 24 populations.

J70 25 Opportunities aplenty

J70 26 The 1985 farm bill's CRP has provided excellent opportunities J70 27 for the management of natural resources on private land. Federal J70 28 and state agencies have taken this opportunity to complement and J70 29 maximize the benefits on CRP acres. Most agencies have tailored J70 30 their programs to benefit the resource for which they are J70 31 responsible. The Private Lands Program was created by the FWS to J70 32 develop wetland habitats in conjunction with CRP with emphasis on J70 33 improving waterfowl production. Improved nesting habitat on private J70 34 land should aid continental duck populations, which are at record J70 35 lows. Grass and legume plantings on highly erodible land provide J70 36 excellent erosion control and the necessary cover for waterfowl to J70 37 nest on millions of acres throughout the region.

J70 38 The scarcity of water is a limiting factor for waterfowl J70 39 production in Montana. The Private Lands Program emphasizes water J70 40 development, and projects typically consist of restoring wetlands, J70 41 repairing dams, or constructing new impoundments. Interested J70 42 landowners also can receive funding to include certain grass and J70 43 forb species in their seeding mixtures. This assistance, in J70 44 conjunction with cost-sharing funds from the Agricultural J70 45 Stabilization and Conservation Service, generally covers most of J70 46 the cost of obtaining seed for these plantings.

J70 47 Data collected on first-year water projects in CRP fields J70 48 confirmed that ducks readily occupy newly created habitat. J70 49 Waterfowl pair counts showed an average of 2.2 breeding pairs per J70 50 wetland acre. This is similar to the pair counts during the Prairie J70 51 Pothole region's highest production levels in the 1950s and 1960s. J70 52 Production most likely will increase in the future because progeny J70 53 return to the same area to nest.

J70 54 Field observations also have shown increased benefits to J70 55 resident wildlife. An estimated 300 pheasants wintered in a 50-acre J70 56 CRP field where the diverse seed mixture provided optimal cover and J70 57 food needs. An ongoing radio telemetry study has shown that J70 58 white-tailed deer and mule deer have preference for and concentrate J70 59 in CRP fields. Elk also have been seen using these fields, J70 60 especially where sweetclover or alfalfa are planted.

J70 61 More than 150 mule deer wintered in 1988-1989 on Fred Lahr's J70 62 CRP fields just south of Denton, Montana, in an area where few deer J70 63 were seen on annual census routes. Biologists for the Montana J70 64 Department of Fish, Wildlife and Parks attributed the increased J70 65 wildlife populations to the excellent diversity of wheatgrasses and J70 66 alfalfa on Lahr's CRP acres. Soil Conservation Service J70 67 conservationists have worked closely with Lahr on his 700 CRP J70 68 acres. They assisted in designing four separate J70 69 shelter-belt and habitat plantings. In addition, FWS J70 70 constructed three dams to provide water for wildlife.

J70 71 Residual vegetation is a key factor in providing suitable J70 72 nesting cover for waterfowl, upland game birds, and other ground J70 73 nesting birds. For high quality nesting cover, seed mixtures J70 74 include rhizomatous grasses that provide more contiguous nesting J70 75 cover than does the scattered cover associated with bunchgrasses. J70 76 Grass species with fibrous stalks and leaves that are resistant to J70 77 flattening from winter winds and snow provide higher quality J70 78 nesting cover. Forbs, such as alfalfa and clover, provide diversity J70 79 for nesting cover and are preferred forage for mule deer, J70 80 white-tailed deer, and elk.

J70 81 The CRP acres on Pat Burton's Double B Farms also exemplify J70 82 what can come of extra effort to further program goals. Burton's J70 83 1,100 CRP acres are situated in the fragile semiarid area of the J70 84 rugged Missouri River Breaks and adjoining the Charles M. Russell J70 85 National Wildlife Refuge. Drought and grasshoppers plagued his J70 86 1987-1988 plantings, causing the seeding on about 620 acres to J70 87 fail. Another factor also affected seeding establishment. Elk from J70 88 the adjoining refuge sought out tender seedlings and 150 to 200 J70 89 head spent part of the winter on the fields.

J70 90 Working through the Montana Department of Fish, Wildlife and J70 91 Parks' Upland Game Enhancement Program, Burton seeded a mixture of J70 92 three wheatgrasses, clover, and alfalfa. Native upland gamebirds, J70 93 such as sharp-tailed and sage grouse, should benefit from the J70 94 improved nesting and protective cover. Also, a number of shallow J70 95 water developments will be constructed using Private Lands Program J70 96 funds.

J70 97 Willing landowners

J70 98 Landowner attitudes toward this multi-agency approach J70 99 to enhancing resource management generally have been positive. J70 100 Landowners favor programs that provide an annual income, improve J70 101 the equity in their land, and allow them a decisive role in J70 102 management. Some landowners were hesitant at the outset. But after J70 103 several projects were successfully completed and the benefits from J70 104 these programs became obvious, the interest increased to the extent J70 105 that it now exceeds the financial and manpower capabilities of the J70 106 several agencies involved in the project.

J70 107 Maximum accomplishments occur when personnel from agencies with J70 108 programs to assist landowners coordinate and cooperate with each J70 109 other. Landowners with specific objectives can be directed to the J70 110 agency that best serves his or her needs. In many situations, a J70 111 landowner's objectives can be met and funded cooperatively among J70 112 agencies in a way that multiple resource benefits can occur.

J70 113 Everyone gains through these cooperative ventures. Decisions J70 114 are made at the local level, and the land is treated according to J70 115 its suitability. Soil erosion is controlled, wildlife habitat is J70 116 developed, and a healthy working relationship is established J70 117 between landowners and the government agencies involved.

J70 118 The pressure is on all people involved in natural resource J70 119 management to be creative and flexible in managing natural J70 120 resources. Through these cooperative working relationships and J70 121 mutually beneficial programs, all of society wins.<*_>square<*/>

J70 122 Citizens in Mason County, Illinois, have found that by J70 123 working together they can deal effectively with J70 124 agricultural-related threats to local drinking water J70 125 supplies

J70 126 Local resource planning for water quality improvement

J70 127 By Dale A. Boyd

J70 128 THE scene was a local coffee shop in Mason County, Illinois. J70 129 Farmers were discussing a research study that showed high levels of J70 130 nitrates and traces of pesticides in the local aquifer that J70 131 supplies drinking water to most of the county. Nitrates exceeded J70 132 the drinking water standard in 70 percent of the samples from 10 J70 133 monitoring wells and in 39 percent of the samples from 14 generally J70 134 deeper domestic wells. Trace levels of metribuzin; atrazine; and/or J70 135 simazine, propachlor, and trifluralin were detected in groundwater J70 136 samples from 10 to 13 monitoring wells located downgradient of corn J70 137 and soybean fields (3). No trace levels were found in the J70 138 domestic wells.

J70 139 The farmers were concerned for two reasons. First, the aquifer J70 140 was their supply of drinking water. Second, they were worried that J70 141 if the problem persisted they would be forced to stop using J70 142 chemicals vital to their farming operations.

J70 143 The result of that early morning discussion was the Illinois J70 144 River Sands Water Quality Project, which has been a reality in J70 145 Mason County, Illinois, since the spring of 1990. The project came J70 146 about because of a unique grassroots effort organized by local J70 147 individuals and farm organizations to address existing and future J70 148 water quality problems. It is an excellent example of local J70 149 organizers working successfully with state and federal agencies to J70 150 solve problems within a community.

J70 151 Nationwide, 74 hydrologic-unit-area water quality projects have J70 152 received funding. The Mason County project is one of two in J70 153 Illinois. The activities of a group of concerned citizens and local J70 154 organizational representatives known as the Water Issues Resource J70 155 Planning Committee were important factors in the project's J70 156 selection. Resource planning is a process that encourages J70 157 individuals in a community to come together to identify and discuss J70 158 mutual problems and needs and, with the help of federal and state J70 159 technical advisors, develop a plan of action to address local J70 160 resource concerns

J70 161 The backdrop

J70 162 Mason County soils include large areas of sands, the remnants J70 163 of a prehistoric river bed. Water in the aquifer is held in pore J70 164 spaces in a sand layer that ranges from 60 to 200 feet thick, J70 165 depending upon the location in the county. The water table is high. J70 166 In most places it is only 3 to 12 feet below the surface. The J70 167 available, abundant supply of water has resulted in 35,000 acres of J70 168 specialty crops grown annually (1) and more than 100,000 acres J70 169 of irrigated fields. Combined with shallow domestic wells, J70 170 conditions are such that the drinking water supply is quite J70 171 susceptible to contamination by farm chemicals. According to J70 172 Hallberg (2), while nonfertilizer uses, such as septic systems J70 173 or manure disposal, may cause nitrate contamination in specific J70 174 circumstances, regional increases in nitrate levels in shallow J70 175 groundwater directly reflect increased use of nitrogen fertilizers, J70 176 especially in sensitive areas, such as western Mason County.

J70 177 Some families with small children have had to haul water J70 178 because of high nitrate concentrations. Excessive nitrates can J70 179 reduce an infant's ability to carry oxygen in its blood, causing a J70 180 condition known as methemoglobenemia or 'blue baby' syndrome.

J70 181 Concerned farmers approached the Soil Conservation Service J70 182 (SCS) district conservationist in March 1989 to ask what the agency J70 183 could do to help. SCS representatives took the case to the Mason J70 184 County Soil and Water Conservation District Board, which formed the J70 185 Water Issues Resource Planning Committee. That committee's first J70 186 meeting was held May 24, 1989. Efforts were made to have diverse J70 187 interests in the community represented. Members included J70 188 individuals from local and state governmental agencies, local farm J70 189 organizations, and agribusinesses. This initiative led to the area J70 190 qualifying for the U.S. Department of Agriculture's water quality J70 191 hydrologic-unit-area designation.

J70 192 Committee members realized that, while they understood the J70 193 nature of the problems in their community, they did not have the J70 194 technical expertise or the financial resources to arrive at a J70 195 solution by themselves. They again asked the SCS district J70 196 conservationist for help. As part of the resource planning process, J70 197 a technical advisory committee was formed and met for the first J70 198 time in June 1989. This subcommittee of the Water Issues Resource J70 199 Planning Committee included technical experts from the University J70 200 of Illinois; members of local farm organizations; and J70 201 representatives of federal, state, and county governmental J70 202 agencies. It provides technical assistance and guidance to the J70 203 local resource planning committee, which makes decisions based on J70 204 technical advisory committee findings and reports.

J70 205 The resource planning committee then met and discussed problems J70 206 and concerns about the local aquifer. Members decided that their J70 207 overall objective should be to maintain and enhance the quality of J70 208 water from the aquifer. They also decided to limit the project area J70 209 to the western two-thirds of the county, where more than 600 J70 210 irrigation systems irrigate about 100,000 acres.

J70 211 The result of this work was a resource plan. The purpose of a J70 212 resource plan is to document a resource problem in an area and J70 213 outline alternative solutions that can be used in addressing the J70 214 problem. In this case, the resource plan was used as the basis to J70 215 apply for USDA water quality hydrologic-unit-area funds.

J70 216 The technical advisory committee recommended a number of J70 217 practices, many related to agricultural production, that could J70 218 reduce the amount of chemicals reaching the aquifer. After the J70 219 planning committee chose practices from the alternatives presented, J70 220 they asked SCS, the Agricultural Stabilization and Conservation J70 221 Service (ASCS), and the Cooperative Extension Service (CES) to J70 222 assist in developing a joint plan of work to maintain and improve J70 223 the aquifer.

J70 224 J71 1 <#FROWN:J71\>Submarine Maneuver Control

J71 2 By William P. Gruner and Henry E. Payne III

J71 3 Submarine warfare is no longer limited to launching weapons J71 4 against enemy ships, submarines, and land targets and evading enemy J71 5 antisubmarine (ASW) attacks. It includes collecting intelligence J71 6 information on activities near foreign shores and at sea and J71 7 tracking, intercepting, and escaping from foreign submarines. To J71 8 meet these challenges, U.S. submarines must be able to maneuver J71 9 rapidly while engaged in their missions.

J71 10 The public generally is not aware of the close encounters our J71 11 submarines experience. Table 1 shows a few examples.

J71 12 The need for great maneuverability in these deadly games of J71 13 observe, tag, and evade is evident. To make an effective and safe J71 14 maneuver in the proximity of other submarines, ships, and mines, a J71 15 diving officer needs two things - precise control of his own J71 16 submarine and accurate knowledge of the positions and movements of J71 17 other vessels and objects. Our existing attack and ASW submarines J71 18 are hampered by three major shortcomings.

J71 19 <*_>black-triangle<*/> Although capable of speeds in excess of J71 20 30 knots, they are unable to make small-radius turns at high speeds J71 21 safely, because they become unstable.

J71 22 <*_>black-triangle<*/> The normally used four-man, manually J71 23 operated control system lacks the rapid-response capability for J71 24 directional control of an unstable submarine.

J71 25 <*_>black-triangle<*/> The captain of a submerged submarine has J71 26 little knowledge of the precise locations of other submarines and J71 27 ships in his vicinity. Consequently, he does not know how best to J71 28 maneuver.

J71 29 The Stability Problem. A submarine at rest has static J71 30 stability. When it starts to pitch or roll, moments are generated J71 31 by buoyancy and weight forces to restore it to the rest position. J71 32 When propelled, it has dynamic stability if it can be made to J71 33 follow a predictable path. However, pitch-yaw hydrodynamic coupling J71 34 (the dreaded snap roll) causes severe stability problems when J71 35 maneuvering at high speeds. Dynamic stability is a prerequisite for J71 36 directional control, and directional control is a prerequisite for J71 37 maneuverability. High maneuverability cannot now be achieved J71 38 because directional control cannot be maintained when large rudder J71 39 angles are applied at speeds in excess of 15-20 knots. Once control J71 40 is lost, the submarine may be lost if the diving officer cannot J71 41 regain control.

J71 42 Recognition of this instability problem did not come until very J71 43 powerful propulsion plants and better stream-lining made J71 44 high speeds possible. The problem was high-lighted in 1954 J71 45 when the experimental high-speed, battery-powered submarine USS J71 46 Albacore (AGSS-569), designed with a sleek body-of-revolution J71 47 hull, became operational. With a top speed greater than 30 knots, J71 48 she was equipped with specially designed control surfaces and a J71 49 fully automated control system that could be operated with from one J71 50 to four men, with or without selected automation. Instability soon J71 51 became evident. One commentator noted, "If in a melee J71 52 situation, a modern high-speed sub pilot tries to turn too sharply J71 53 at too high a speed, he might find himself in a snap roll, hanging J71 54 from his seat belt and with a loss of several hundred feet in depth J71 55 at a markedly slowed speed."

J71 56 table&caption

J71 57 An early lack of concern about the stability problem probably J71 58 was because, until the end of World War II, very few of the world's J71 59 submarines had submerged speeds in excess of ten knots. Manual J71 60 control of bow and stern planes and helm by separate operators J71 61 directed by a diving officer was considered adequate at normal J71 62 speeds of two to six knots. If 'the bubble was lost' while J71 63 increasing depth, an alert diving officer could soon regain control J71 64 by backing and blowing tanks.

J71 65 Later, when nuclear-powered submarines of the Skipjack J71 66 (SSN-585), Sturgeon (SSN-637), and Los Angeles J71 67 (SSN-688) classes - all characterized by body-of-revolution hull J71 68 designs and large sails - became operational, they, too, had J71 69 stability problems similar to the Albacore's. Their J71 70 instability is caused by external variable forces acting on their J71 71 hulls - a result of pressures exerted by the swirling masses of J71 72 water that are continuously generated by the hull as it turns and J71 73 drives its way through the ocean. These masses of water, termed J71 74 vortices, rotate inward toward the center of the hull. The higher J71 75 the submarine's speed, the more energy is imparted to them, and the J71 76 greater the pressures acting on the hull.

J71 77 Figure 1 is a computer simulation of vortices generated by a J71 78 submarine during the mid-stage of an evasive maneuver at 24 knots J71 79 with a 30° rudder angle. Note the displacements of the twin J71 80 vortices as the ship progresses through the water and the effect of J71 81 the sail on their location. Under the influence of the forces J71 82 generated, the stern will squat down and the bow pitch up. J71 83 Meanwhile, the forces will cause the submarine to sink and slow, J71 84 and it will go into a snap roll unless speed is reduced by backing J71 85 and the rudder put amidships.

J71 86 Sample patterns of wind tunnel airflow tests, using a scale J71 87 model of the Skipjack hull, are shown in Figure 2. The first J71 88 pattern shows the airflow to be quite smooth when the submarine is J71 89 on straight course. However, when the model is placed at J71 90 significant roll and yaw angles as in the second pattern, the J71 91 strong influence of the sail pulls the upper vortex out of its J71 92 place along the hull and into the wake of the sail. Comparisons of J71 93 such patterns at different roll angles, yaw angles, and speeds J71 94 reveal dramatic variation in the oscillating flow and pressure J71 95 patterns.

J71 96 Full-scale submarines experience similar flow conditions when J71 97 large rudder angles are applied at high speeds. Resulting J71 98 unbalanced pressures on the sail, for example, can produce forces J71 99 of millions of pounds, with corresponding moments great enough to J71 100 rotate a submarine about its roll axis. This results from a J71 101 submarine's relatively small righting moment (because of its small J71 102 meta-centric height) about the roll axis.

J71 103 The Directional Control Problem. When a roll begins J71 104 under these conditions, a coupling of motions about the roll, J71 105 pitch, and yaw axes occurs to cause a submarine to veer suddenly J71 106 and radically in unexpected directions and to change depth and J71 107 speed. Manual control under such conditions becomes highly J71 108 difficult, if not impossible.

J71 109 In discussing directional control of the Albacore, an J71 110 early commanding officer stated, "It was evident that the J71 111 Albacore performed significantly better without automated J71 112 controls programmed by a single operator than she did with a J71 113 standard four-man team of diving officer, helmsman, bow planesman, J71 114 and stern planesman." He likened the submarine control J71 115 problem to that of high-speed aircraft in which "designers J71 116 turn to higher performance control machines to offset human J71 117 limitations in sensing and reacting, a lack of uniformity of J71 118 performance, and limited adaptability to performing multiple J71 119 requirements simultaneously." He believed future submarines J71 120 would be "flown" with automatic controls with pilot J71 121 override.

J71 122 Limited human capabilities - the human brain and muscular J71 123 control systems operate too slowly and have other limitations - J71 124 caused airplane designers to turn to higher-performance-control J71 125 machines. Pilots are unable to keep pace with rapidly changing J71 126 situations while faced with multiple tasks during aerial combat, J71 127 low-level flying, and landing. Further, human systems, if under J71 128 stress, are likely to err, function inconsistently, and freeze. The J71 129 repeatability and reliability of modern computer-aided systems have J71 130 made them the only choice for high-speed airplane maneuvers. For J71 131 similar reasons, they are the only choice for controlling J71 132 high-speed submarine maneuvers.

J71 133 Despite the Albacore's favorable experience with the J71 134 automatic one-man control system in the mid-1950s, the system was J71 135 looked upon with suspicion. Although other work was performed to J71 136 improve the controllability of submarines at high speeds, the basic J71 137 four-man diving team used in low-speed submarines was applied to J71 138 nuclear-powered submarines, with limited provision for a one-man J71 139 control system. To minimize the risk of loss of control, J71 140 maneuverability is now limited by operational procedures that J71 141 constrain rudder angles to a very few degrees at speeds above about J71 142 15 knots. As another safety measure, the rudder control system J71 143 prevents the application of large rudder angles at high speeds.

J71 144 The Approach to Greater Submarine Control

J71 145 For our submarines to achieve greater maneuverability, they J71 146 must be modified through the application of current technical J71 147 knowledge in control systems, hydrodynamics, human engineering, and J71 148 physics. Much of this knowledge can be borrowed from aircraft and J71 149 missile engineers, who were faced with stability and control J71 150 problems as soon as Orville and Wilbur Wright demonstrated a J71 151 capability for powered flight in 1903.

J71 152 The well-known author and aeronautical engineer Nevillle Shute J71 153 described the state of airplane control during World War I, when J71 154 most airplanes flew at speeds of 60-130 knots. "We knew J71 155 that a clumsily executed turn might have the effect of putting an J71 156 aeroplane into a spinning nose-dive (a Parke's Dive, some of us J71 157 called it, because Lieutenant Parke was one of the very few people J71 158 who had come out of it alive). In general, a spin, once started, J71 159 continued to the ground, the machine hitting very violently. And J71 160 that literally, was all we knew about it."

J71 161 As airplane speeds increased, it became obvious that manual J71 162 flight control was no longer acceptable. It was recognized that, J71 163 "human pilots were incapable of adequately compensating for J71 164 the rapid changes encountered during tactical engagements, adverse J71 165 weather conditions, ground-skimming flight, and other J71 166 maneuvers. To make flight feasible, pilots were provided with J71 167 assistance in the form of automatic flight-control systems. Such J71 168 systems now are incorporated in all modern high-performance J71 169 aircraft, including C-141 and C-5 transports, B-1 and B-2 bombers, J71 170 and F-4, A-6, F-14, F-15, and F-16 fighters and fighter-bombers.

J71 171 "Individual systems vary from aircraft to aircraft, but in J71 172 general provide pitch, roll, and yaw control augmentation, J71 173 autopilot modes of operation, and altitude hold. The pilot normally J71 174 enters maneuvering orders by means of stick or control column and J71 175 rudder pedals. Major elements of the flight-control system include J71 176 control-force transducers, pressure and temperature sensors, J71 177 directional gyros, accelerometers, a central air-data computer, and J71 178 a flight-control computer. The latter interprets transducer input J71 179 from pilot actions and other data sources, processes data, and J71 180 sends commands to the various servo-actuators to assure that proper J71 181 control surface and device responses are made to pilot orders J71 182 within safe flight limits. Control system reliability is achieved J71 183 by rigid parts selection; inspection and test at component, J71 184 subassembly, and system levels; and redundancy in electronic, J71 185 hydraulic, and power supply elements."

J71 186 While scientists and engineers were developing systems to J71 187 control supersonic aircraft, others were developing control systems J71 188 for intercontinental ballistic missiles, to enable them to deliver J71 189 payloads accurately and reliably to targets thousands of miles J71 190 distant. At that time, submarine designers seemed content to modify J71 191 two-dimensional control systems for submarines operating in a J71 192 three-dimensional ocean. Since these systems do not provide the J71 193 degree of maneuverability required, a new control capability must J71 194 be developed - one similar to aircraft and missile control-system J71 195 technology.

J71 196 All bodies in motion - including submarines and J71 197 air-planes - are subject to the same inviolable laws of J71 198 physics. These laws govern relationships between mass, force, J71 199 torque, inertia, and acceleration. One important law states that J71 200 rotational acceleration about an axis is proportional to the moment J71 201 applied and inversely proportional to the inertia characteristic J71 202 (moment of inertia) of the body about that axis. The high energy J71 203 vortices noted previously and the large sail are the main causes of J71 204 the upsetting moments that cause U.S. submarines to lose control in J71 205 high-speed turns. Therefore, special attention must be paid to J71 206 reducing sail size and its resultant interaction with the hull J71 207 vortices.

J71 208 Concept for an Automatic Submarine-Control System

J71 209 Because of the complexity of submarine dynamic motions, the J71 210 rapidity with which upsetting moments are generated, the speed with J71 211 which control forces must be applied, and the inability of humans J71 212 to exercise manual control, the new control system must employ J71 213 computer technology. Figure 3 presents a concept for an automatic J71 214 submarine-control system composed of three major subsystems: an J71 215 automatic maneuver sub-system, a pressure-sensing J71 216 subsystem, and an automatic attitude-control sub-system.

J71 217 The Automatic Maneuver Subsystem. This subsystem J71 218 performs two major functions. First, it provides the man/machine J71 219 interface by which the diving officer enters maneuver instructions J71 220 and receives information. To initiate a maneuver, the diving J71 221 officer specifies the required maneuver within the framework of an J71 222 earth-oriented, north-referenced, three-dimensional orthogonal J71 223 coordinate system. Maneuver instructions may call for a simple turn J71 224 and a change in depth or for more complex maneuvers to avoid a J71 225 collision or to reach a distant weapon launch position or a J71 226 position offset from a moving-target track.

J71 227 J72 1 <#FROWN:J72\>Retrospective - 1992

J72 2 Internal operations and relations received primary J72 3 attention. The pace of labor relations activities provided a J72 4 breathing spell allowing improvements to be made to the tools for J72 5 dispatch, for allocations and for labor relations information J72 6 retrieval. Longer range assessment forecasts were introduced at the J72 7 request of the Industry, and a complete rewrite of the Tonnage J72 8 Reporting and Assessment Procedures Manual was begun to provide J72 9 clearer and more explicit reporting instructions for the J72 10 Industry.

J72 11 Labor Relations

J72 12 Labor relations in the West Coast longshore industry is J72 13 governed by the Joint Port Labor Relations Committees (JPLRC's), J72 14 the Coast Labor Relations Committee (CLRC), the Coast Steering J72 15 Committee (CSC) and the Area Sub-Steering Committees. The CSC and J72 16 Sub-Steering Committees are made up of PMA member company J72 17 executives, and the JPLRC's and CLRC are composed of Union J72 18 representatives and the PMA member company representatives.

J72 19 The meetings and activities of these Committees handle the day J72 20 to day administration of the Coast contracts, and their agenda J72 21 determine the direction and scope of much of PMA labor relations J72 22 staff activities. The bulk of labor relations activity this past J72 23 year took place at the level of the JPLRC's.

J72 24 The Joint Port Labor Relations Committees process routine J72 25 grievances and administer the agreement provisions locally within J72 26 the various ports. These committees, composed of local Union and J72 27 Employer representatives, also maintain registration lists and J72 28 rosters of casual labor pools, coordinate the travel of manpower to J72 29 ports with sufficient work opportunity, and facilitate the J72 30 participation of the work force in the General Safety Training, J72 31 Skilled Training and Drug/Alcohol Free Work Place programs. Their J72 32 efforts, with the cooperation of the Union membership, have J72 33 improved the dispatch of manpower, have resulted in fewer instance J72 34 of labor shortages on peak work days, and have produced a reduction J72 35 in the number of on-the-job injuries.

J72 36 The Coast Labor Relations Committee developed a procedure for J72 37 complying with the Americans' with Disabilities Act. Additionally, J72 38 the CLRC handled routine matters such as processing referrals from J72 39 JPLRC's seeking registration approvals and periodic updating of Low J72 40 Work Opportunity Port lists and negotiated mileage allowance J72 41 formulas. The Employers and the Union held preliminary discussions J72 42 on the topics of computerization and new technology in the J72 43 longshoring industry and the negotiation of a separate labor J72 44 agreement covering rail container transfer facilities. The J72 45 preliminary discussions did not result in follow-up meetings with J72 46 the Union, and no new understandings were reached on either J72 47 topic.

J72 48 Labor policy issues concerning several member companies J72 49 occupied the attention of the Coast Steering Committee. One issue J72 50 involved a company's negotiating a separate agreement with the J72 51 Union in violation of policy, and the agreement was eventually J72 52 nullified. Two issues concerned payroll practices that created an J72 53 underpayment of man-hour contributions. Both practices were J72 54 corrected, and the proper man-hour contributions were collected. J72 55 The Committee continued its review of new technology issues based J72 56 on periodic reports and on-site visits by its Clerks' J72 57 Sub-Committee.

J72 58 The relatively slow pace of labor relations activity provided J72 59 PMA staff with the opportunity to improve and develop new work J72 60 place tools. A five-volume index of alphabetical listings of J72 61 contract interpretations, arbitrators' decisions and historical J72 62 references covering the period from 1934 to the present was placed J72 63 into an electronic database. Labor relations personnel in each of J72 64 the Area offices can now research these data by subject, date, or J72 65 arbitrator's name quickly and easily.

J72 66 The PMA Allocator in each of the four Areas is now using J72 67 personal computer systems to perform many daily allocation J72 68 functions. (The PMA Allocator accepts manpower orders from the J72 69 Employers, prioritizes the orders and transmits the information to J72 70 the appropriate dispatch hall. This process traditionally includes J72 71 a voluminous amount of manual record keeping and data retrieval.) J72 72 These systems are in varying degrees of implementation at year-end, J72 73 and the goal is for computerized allocation to be fully operational J72 74 soon.

J72 75 Several staff members participated in labor relations and J72 76 arbitration work-shops, computer user seminars, and safety J72 77 and training conferences, and the four Area Managers visited the J72 78 ports of Rotterdam, Felixstowe, Tilbury and Bremen/Bremerhaven in J72 79 October to observe first-hand the technology and terminal J72 80 operations at these ports. Institutional and structural differences J72 81 in labor relations and operational procedures were explained by our J72 82 hosts at ECT/Sea-Land Delta Terminal, the Port of Felixstowe, the J72 83 Port of Tilbury and Bremer Lagerhaus-Gesellschaft.

J72 84 Training J72 85 A record number of employees, 7,437, completed ILWU-PMA J72 86 training programs. This represents a 40% increase over the previous J72 87 year, and 85% of the training was Safety training. A significant J72 88 number of direct employer representatives were among the trainees, J72 89 and in several programs, ILWU members and company representatives J72 90 attended classes together. A list of training programs showing the J72 91 number of participants by program is shown on page 38.

J72 92 General Safety Training

J72 93 The General Safety Training (GST), described in detail in the J72 94 1991 Annual Report, targets the entire workforce over a three year J72 95 period and addresses all legislated training requirements. More J72 96 than 5,000 individuals have already participated in the GST J72 97 program. The program underwent planned revisions designed to fine J72 98 tune the curriculum, to introduce new material and to incorporate J72 99 new regulatory requirements. Three successful custom videos were J72 100 added, 'Personal Protective Equipment,' 'Vehicle Safety' and 'Local J72 101 Area Hazards.' The popularity of this program is in large part due J72 102 to the videotaped segments depicting the longshore work force J72 103 engaged in their job functions in West Coast ports.

J72 104 Alcohol/Drug Free Workplace Training For Supervisors and J72 105 Employees

J72 106 In February, the Alcohol and Drug Free Workplace (ADFWP) J72 107 Training for Supervisors commenced, and more than one thousand J72 108 employees have completed the training. This program is a logical J72 109 extension of the nationally recognized Alcohol and Drug Recovery J72 110 Program (ADRP),and it targets both labor and management supervisory J72 111 personnel.

J72 112 Drug and Alcohol Free Work Place Training is also incorporated J72 113 into the General Safety Training.

J72 114 Winch Training

J72 115 A Winch Training Program was begun for members of Locals 13 and J72 116 46, and this marked the culmination of negotiations with the J72 117 Maritime Administration on the use of its USNS Curtiss J72 118 (T-AVB4) for ship's gear training. The Curtiss, a Ready J72 119 Reserve Fleet vessel homeported in Port Hueneme, is ideally suited J72 120 for 'yard and stay,' 'swinging boom' and 'jumbo' hands-on training. J72 121 Rooms on board provide space for classroom instruction and rigging J72 122 demonstrations. More than one hundred trainees have completed the J72 123 new program since September.

J72 124 Video Production

J72 125 Videotaped presentations of training materials are rapidly J72 126 supplanting other traditional teaching aids. Video delivers high J72 127 quality, standardized, scripted training to the work force. Several J72 128 new videos were added to our library this year. In addition to the J72 129 three previously described in General Safety Training and the one J72 130 in Alcohol/Drug Free Workplace Training, each Area contributed J72 131 ideas for videos which are in various stages of completion. The J72 132 Oregon Area completed work on 'Lumber Handling,' a ten minute J72 133 description of handling packaged lumber. 'Marine Fork lift' is in J72 134 progress in Washington, and 'Autos' and 'Lashing' are being planned J72 135 in California.

J72 136 Accident Prevention

J72 137 The revisions to the Man-hours and Injury Incidence Rate J72 138 Reports were completed this year. The new reports provide member J72 139 companies with both quarterly and year to date injury/illness J72 140 statistics so that they may more readily assess the progress of J72 141 their individual safety programs.

J72 142 The PMA Coast Accident Prevention Awards Program continued to J72 143 generate considerable interest this year. Emphasis on individual J72 144 achievement has resulted in increased participation as well as J72 145 deserved recognition. Awards information and other statistical data J72 146 for 1992 is contained on page 39.

J72 147 Several staff changes occurred during the year. Larry Hudson J72 148 was promoted to Area Supervisor, Oregon Area, replacing Dale Larson J72 149 upon his retirement. Joe Boettcher joined the Oregon staff as an J72 150 Assistant. Tony Peredo was promoted to Area Supervisor in the J72 151 Washington Area, and Fred Gordon joined the staff as his J72 152 Assistant.

J72 153 J72 154 Legal J72 155 As in the past, third-party litigation consumed the major J72 156 portion of PMA legal effort this past year. NLRB charges filed by J72 157 casuals against the 'permissive rule' are pending in the State of J72 158 Washington and in Southern California. The Golden consent J72 159 decree and its progeny required considerable attention in the J72 160 federal courts in Los Angeles in litigation involving unsuccessful J72 161 applicants for registration or transfer who attempted to use the J72 162 decree for their own purposes. Although the Golden decree has J72 163 generated considerable litigation in its ten years of life, it also J72 164 has provided a generally consistent and stabilizing forum for the J72 165 adjudication of issues concerning waterfront employment. PMA J72 166 counsel and ILWU counsel have worked together to provide an J72 167 effective defense for the parties to the collective bargaining J72 168 agreement.

J72 169 While a federal court action filed by several casuals in Oregon J72 170 claiming so-called 'contingency list' status was pending, the Coast J72 171 Arbitrator rendered a decision approving the dissolution of the J72 172 list in Portland and permitting registration based upon hours of J72 173 experience in accordance with the collective bargaining agreement. J72 174 The federal district court in Oregon dismissed the lawsuit in view J72 175 of the Coast Arbitrator's decision.

J72 176 After much administrative planning and oversight by the NLRB, J72 177 the re-registration of longshoremen at Local 18 in Sacramento J72 178 finally is underway. This should conclude the lengthy and costly J72 179 litigation and the processing of NLRB ordered deregistration and J72 180 re-registration.

J72 181 In what has become known as the 'Fish Case,' a Coast J72 182 Arbitrator's decision was received this fall involving an issue J72 183 crucial to the future membership of steamship agents in the PMA. J72 184 The Coast Arbitrator remanded the case to the Area Arbitrator J72 185 because of the absence of facts showing that the husbanding agent J72 186 had control of cargo operations for which it was held responsible. J72 187 We await anxiously the final outcome of this matter.

J72 188 For the first time in over a decade, PMA became involved in J72 189 litigation against a member company. Following a refusal to pay J72 190 interest penalties on delinquent tonnage assessments, Long Beach J72 191 Container Terminal, Inc., filed a declaratory relief action in the J72 192 California Superior Court against PMA and various agents and J72 193 foreign steamship operators. PMA and other defendants have answered J72 194 with cross-complaints to require LBCT to pay the interest J72 195 penalties. Discovery is proceeding.

J72 196 Early in the year, the NLRB sought and obtained a temporary J72 197 restraining order and injunction from the federal district court in J72 198 Los Angeles, barring the ILWU from interfering with PMA members J72 199 doing business with the Southern Pacific ICTF in the Ports of Los J72 200 Angeles and Long Beach. The PMA operations quickly returned to J72 201 normal after a one-day shutdown, with the exception of the refusal J72 202 by Local 63 Marine Clerks to perform work voluntarily during the J72 203 meal hour in a manner consistent with conduct prior to the dispute. J72 204 After a favorable area arbitrator's award finding Local 63 in J72 205 violation of the contract, the Coast Committee declared the J72 206 contract remedies exhausted. PMA, in a rarely necessary proceeding, J72 207 sought and received confirmation of the award from the federal J72 208 district court. Local 63 continued to violate the award, and PMA J72 209 moved for an order to show cause why the union should not be held J72 210 in contempt. The motion was denied. PMA will continue to monitor J72 211 compliance with the award for future enforcement. The ILWU has J72 212 appealed the district court confirmation of award to the Ninth J72 213 Circuit.

J72 214 PMA engages the services of local law firms in each of the four J72 215 port areas on the Coast. These firms continue to serve the J72 216 organization extremely well. Notwithstanding occasional exceptions, J72 217 the relationship between PMA counsel and ILWU counsel remains J72 218 professional and courteous and serves as an aid to the resolution J72 219 of legal problems in the industry.

J72 220 J72 221 Finance and Administration

J72 222 Building on the successful completion of the coastwise payroll J72 223 system in 1990, the Information Services (I.S.) Department began J72 224 the redesign and reprogramming of the Pay Guarantee Plan payroll J72 225 system to make it consistent in all four Areas. Design is J72 226 completed, and the new system should be operational in the second J72 227 quarter. Since the payroll system has been accepting automated J72 228 input, more than 90% of the payroll hours processed each week are J72 229 transmitted to PMA electronically.

J72 230 I.S. staff continued their involvement with several projects J72 231 related to stream-lining the dispatch process. Significant J72 232 progress has been made in Los Angeles/Long Beach where the UTR J72 233 board in the Wilmington Longshore Dispatch Hall was the first to be J72 234 modernized. In addition, telephone check-in for registered J72 235 longshoremen should be operational in early 1993.

J72 236 J72 237 J73 1 <#FROWN:J73\>SUBSONIC AIRCRAFT PROPULSION/AIRFRAME J73 2 INTEGRATION

J73 3 The technology of integrating propulsion systems and airframes J73 4 involves the ability to assess and control the development of wave J73 5 drag, induced drag, and profile drag. Advances in CFD over the past J73 6 decade have contributed greatly to this technology. It is J73 7 anticipated that ongoing CFD developments will lead to even further J73 8 refinements.

J73 9 Two areas remain in which technology improvements are needed. J73 10 One is the development of wind tunnel test techniques and powered J73 11 propulsion simulators to better represent installed power effects J73 12 of the forthcoming generations of very high bypass ratio engines in J73 13 wind tunnel testing. The other is the need to predict the installed J73 14 characteristics of thrust reversers, both computationally and with J73 15 wind tunnel testing techniques. These are areas in which NASA can J73 16 make important contributions.

J73 17 AERODYNAMIC CRUISE PERFORMANCE

J73 18 Although the fundamental physical principles of subsonic and J73 19 supersonic airflow around aircraft are the same, design approaches J73 20 to minimizing drag are greatly affected by the cruise speed. This J73 21 section of the report discusses cruise performance in the two speed J73 22 ranges separately.

J73 23 Subsonic Aircraft Cruise Performance

J73 24 Long-haul subsonic transports are now, and will be for the J73 25 foreseeable future, the major product of the civilian aviation J73 26 industry and infrastructure. As noted in Chapter 2, from 1975 to J73 27 1995, aerodynamic efficiency will have increased by approximately J73 28 10 percent, and if the current rate of improvement is maintained, J73 29 another 5-10 percent is projected by the year 2020. However, J73 30 ordinary development or evolution alone will not keep the United J73 31 States at the forefront in the world market. Although continued J73 32 evolutionary advances in methods and processes (experimental, J73 33 theoretical, and computational) are needed to provide continued J73 34 improvement of aerodynamic design technologies, demonstrated J73 35 innovative technologies are necessary in the longer term to provide J73 36 opportunities for significant improvements in performance.

J73 37 Laminar Flow Control

J73 38 The flow on most of the surfaces of an aircraft is turbulent. J73 39 Laminar flow control (LFC), hybrid laminar flow control, and J73 40 natural laminar flow are promising sources of skin-friction drag J73 41 reduction on aerodynamic surfaces. Laminar flow nacelles are also J73 42 being studied by NASA. Laminar/turbulent transition of the airflow J73 43 next to the aircraft surface is delayed through a combination of J73 44 pressure gradient tailoring of the wing and control such as suction J73 45 through the skin. If full-chord laminar flow can be maintained in J73 46 this fashion, fuel savings of up to 25 percent could be J73 47 realized.

J73 48 Transition is extremely sensitive to freestream conditions J73 49 (e.g., freestream turbulence and acoustics) and surface roughness J73 50 (e.g., rain and ice crystals, insect debris, surface finish, and J73 51 fasteners); lack of confidence in these issues has hindered the use J73 52 of this concept on vehicles. Also, of perhaps greater significance J73 53 have been the questions of fabrication cost and operational cost J73 54 and maintainability.

J73 55 Engineering and optimization tools have outpaced the state of J73 56 the art in transition prediction theory. Thus, the design of LFC, J73 57 hybrid laminar flow control, and natural laminar flow systems J73 58 depends on empirical bases to determine transition. This method is J73 59 also limited because it cannot account for the effects of surface J73 60 roughness and freestream disturbances.

J73 61 Knowledge of transition - so very important to the success of J73 62 LFC techniques - is, in general, limited to the simplest of J73 63 geometries. Efforts to better understand the transition flow J73 64 physics are under way to provide valuable guidance for the surface J73 65 roughness and freestream disturbance problems.

J73 66 Only a limited number of flight tests have been flown since the J73 67 original and successful X-21 program of the 1960s; these are the J73 68 JetStar (NASA/Langley) and Boeing 757 (NASA/Boeing). In both cases, J73 69 extensive laminar flow was successfully achieved on the upper J73 70 surface of the swept wing through the use of suction. Very low J73 71 suction levels were required, with power penalties of the order of J73 72 1 percent. Studies with engine noise indicated no effect. The use J73 73 of a Krueger nose flap eliminated a potential buildup of insect J73 74 debris on the leading edge.

J73 75 The remaining challenges to the implementation of laminar flow J73 76 technology in large subsonic transport designs include validation J73 77 of the technology in actual airline service operating environments J73 78 and exploration of the technical issues associated with making J73 79 laminar flow operate effectively on the inboard portion of the J73 80 wings of very large aircraft. Recognizing the challenges, during J73 81 1990 NASA and the industry developed a cooperative research plan; J73 82 however, these efforts have been delayed by overall program J73 83 constraints. Meanwhile, the Europeans have rapidly advanced their J73 84 laminar flow efforts. Airbus plans for laminar flow technology J73 85 validation include extensive large-scale testing, targeting J73 86 technology validation as early as 1993.

J73 87 Turbulent Drag Reduction

J73 88 The most promising technique demonstrated thus far has been J73 89 passive control by riblets, tiny streamwise grooves on the aircraft J73 90 surface. This device is useful for surfaces on which laminar flow J73 91 is very difficult to achieve (e.g., the fuselage). The approach was J73 92 used successfully on the U.S. entry in an America's Cup Race and J73 93 then flight-tested on a portion of a business jet, achieving a J73 94 reduction in local skin-friction drag of 8 percent.

J73 95 The state of the art in turbulence predictions depends on J73 96 empirical correlations and models, usually developed for one set of J73 97 flow conditions or a very simplified model. Here also, efforts at J73 98 understanding the basic physics of turbulent flow are under way. J73 99 Prediction and control have been hindered by the lack of reliable, J73 100 efficient models of turbulence for complex geometries.

J73 101 Advanced Supercritical Airfoils

J73 102 Advanced supercritical airfoils, which reduce the shock J73 103 strength on transonic airfoils, have contributed to drag reduction J73 104 and have been used on all commercial transport aircraft developed J73 105 since 1975. Further modifications with reduced moments and weaker J73 106 shock waves are under study by NASA for use with LFC systems.

J73 107 Improved understanding of shock/boundary-layer interactions has J73 108 led to new opportunities to greatly improve airfoil design concepts J73 109 and procedures.

J73 110 Wing Design

J73 111 The improvement of theoretical analysis tools and CFD, coupled J73 112 with a better understanding of flow physics, has enabled the design J73 113 of more aerodynamically efficient wings with greater thickness and J73 114 reduced sweep. This allows a wing weight reduction or higher aspect J73 115 ratio. Substantial improvements in cruise Mach number and critical J73 116 Mach number have also been realized. New opportunities exist to J73 117 significantly improve the design optimization procedures for wings J73 118 that incorporate laminar flow systems along with advanced high-lift J73 119 systems.

J73 120 Winglets J73 121 Winglets, or wingtip extensions, which first appeared on J73 122 business jets, are now used on various versions of commercial J73 123 transports (e.g., the Boeing 747-400 and MD-11). These effectively J73 124 increase the aspect ratio of the wing. Advancements in J73 125 understanding of the 'nonlinear' effects of wing-wake-deflection J73 126 and roll-up have created opportunities to improve the design J73 127 optimization procedures for winglets and other wingtip devices for J73 128 drag reduction. In each of these technology topics, significant J73 129 opportunities have developed to advance the state of the art; J73 130 however, constraints in the aeronautics program have limited NASA's J73 131 ability to support the needed advancements in experimental or J73 132 computational capabilities and in ground and flight validation of J73 133 these technologies.

J73 134 figure&caption

J73 135 Status of Subsonic Technology

J73 136 As noted earlier, the past decade has seen large improvements J73 137 in wing design technology, the successful demonstration of laminar J73 138 flow control on a Boeing 757, the successful demonstration of J73 139 riblets on a business jet, the successful demonstration of natural J73 140 laminar flow on a business jet, and the use of supercritical J73 141 airfoils and winglets on transports in everyday commercial J73 142 service.

J73 143 Advances in computer and quiet-tunnel/instrumentation J73 144 capabilities have allowed details of the fundamental flow physics J73 145 of transition and turbulence to be studied. Computational time for J73 146 these efforts, however, is too long for extensive use in the design J73 147 process at present.

J73 148 NASA should play a leading, but not exclusive, role in the J73 149 development of enabling technologies. There should be a cooperative J73 150 effort among NASA, industry, and academia to research complicated J73 151 flow physics with the goal of predicting, modeling, and controlling J73 152 such flows. Research should combine theory, careful experiments and J73 153 CFD; duplication of the results by another technique or in another J73 154 facility, as recommended by the U.S. Transition Study Group, is J73 155 also desirable. NASA is an appropriate organization to encourage J73 156 flight testing of enabling technologies and the development of J73 157 advanced diagnostic instrumentation for nonintrusive testing. NASA J73 158 is also an appropriate organization to provide high Reynolds J73 159 number, quiet testing opportunities, as well as the most current J73 160 computational facilities and techniques.

J73 161 To implement the above concepts, a better understanding of flow J73 162 physics is required. Intensified efforts to develop useful, J73 163 accurate engineering models that can be used for design are J73 164 particularly necessary. This work, although seemingly basic in J73 165 nature, must be actively pursued and must include companion J73 166 theoretical, computational, and experimental efforts conducted J73 167 under careful, well-documented conditions. A better understanding J73 168 of flow physics will also afford the opportunity for effective use J73 169 of flow control. Issues to be addressed should include J73 170 implementation, reliability, and maintenance of the mechanical J73 171 systems, as well as implications of the loss of the system in J73 172 flight. Continuing advances in CFD, as well as high Reynolds J73 173 number, low-disturbance experimental capabilities, are also needed J73 174 in support of the evaluation and design of these concepts.

J73 175 NASA needs the following resources to accomplish the J73 176 foregoing:

J73 177 <*_>bullet<*/> high Reynolds number facilities (simulate J73 178 flight),

J73 179 <*_>bullet<*/> low-disturbance freestream facilities (simulate J73 180 flight),

J73 181 <*_>bullet<*/> full-scale Reynolds number flight research J73 182 capability,

J73 183 <*_>bullet<*/> nonintrusive instrumentation,

J73 184 <*_>bullet<*/> faster and bigger computers for flow physics J73 185 (model development),

J73 186 <*_>bullet<*/> more efficient CFD, faster algorithms and grid J73 187 setups,

J73 188 <*_>bullet<*/> companion theory/computation/experiment efforts J73 189 (validation and guidance), and

J73 190 <*_>bullet<*/> NASA/industry/university cooperation on J73 191 appproximately equal levels. NASA's role in the university training J73 192 of future engineers through research funding must never be J73 193 overlooked.

J73 194 The Europeans are using much of the technology described above J73 195 - particularly LFC - even though by U.S. standards the technology J73 196 is often untested or unproven. On an overall technical basis, J73 197 current European offerings are equal to those of the United States. J73 198 However, because Europeans are incorporating the technology faster J73 199 than the United States is, their rate of improvement is J73 200 significantly greater. In particular, within the BRITE EURAM J73 201 consortium involving government, industry, and academia in Europe, J73 202 a highly organized effort has been developed to advance the state J73 203 of the art for laminar flow engineering design capabilities. Their J73 204 efforts combine the best talent and facilities in all of Europe to J73 205 develop and validate transition prediction tools for integration J73 206 with industry engineering design methods. (BRITE EURAM efforts J73 207 address all other aeronautical disciplines as well.) Although J73 208 progress is being made in the United States on understanding J73 209 boundary-layer transition physics, no similar design-tool-focused J73 210 effort is being funded here. The problem in this country is not J73 211 lack of opportunities, but rather the lack of priorities on J73 212 resources.

J73 213 Supersonic Aircraft Cruise Performance

J73 214 U.S. expertise in supersonic aircraft performance stems from J73 215 the earlier Supersonic Transport (SST) program, the ongoing Phase I J73 216 High Speed Research Program, and various high-speed fighters. In J73 217 cruise, the recognized promising technologies for the HSCT are the J73 218 same as for the subsonic case: hybrid laminar flow control by J73 219 suction and pressure gradient. Unlike the subsonic speed regime, J73 220 this technology has not been demonstrated in flight. The benefits J73 221 of lower drag (as described in subsonics) and thermal requirements J73 222 associated with laminar flow can be realized in this flight range J73 223 as well. In addition, technology advances in the fundamental J73 224 understanding of high Reynolds number effects on leading-edge J73 225 vortex formation and the ability to eliminate unacceptable J73 226 characteristics, such as low speed pitch-up, could allow the J73 227 utilization of highly swept, high-performance wing planforms. This J73 228 would provide substantial increases in cruise lift-to-drag ratios J73 229 over the currently favored planform concepts.

J73 230 Laminar Flow Control

J73 231 At present, our knowledge of high-speed transition is even less J73 232 developed than our knowledge of subsonic flows; designs depend on J73 233 existing theories that are not compatible with design and J73 234 optimization tools (as described in subsonics). In the supersonic J73 235 range, however, the effects of freestream conditions (e.g., J73 236 freestream turbulence and noise) and surface roughness may be more J73 237 severe than in subsonic flows. Efforts are under way to understand J73 238 and predict the transition behavior in supersonic flight. The J73 239 exciting progress that has been made in subsonic laminar flow J73 240 technology required nearly four decades of concentrated, if not J73 241 continuous, effort. Although some of the subsonic lessons learned J73 242 may apply to supersonic laminar flow challenges, it would be J73 243 overly optimistic to expect supersonic laminar flow technology to J73 244 mature after a few years of effort. J73 245 J74 1 <#FROWN:J74\>Laser/Vision - 100% Inspection Systems

J74 2 The growth of automation requires that most fastener J74 3 manufacturers implement methods for providing defect-free material J74 4 to their customers

J74 5 The demands for stricter quality control have shifted from a J74 6 high level AQL (Acceptable Quality Level) to three parts per J74 7 million or zero defects product levels.

J74 8 The implementation of sophisticated statistical process control J74 9 has helped to insure quality of manufacture, but does not address J74 10 product mix, and foreign material contamination as a result of J74 11 plating, heat treating, packaging, and other bulk processes.

J74 12 A growing trend for high-speed sorting and 100% gaging is J74 13 emerging within industries because of the high levels of automation J74 14 employed in manufacturing today. The human factor will always be J74 15 present regardless of the SPC or quality control levels adopted for J74 16 a particular manufacturer. The same customer that was talking AQL J74 17 yesterday is demanding zero defects today at a reduced price.

J74 18 The non-contact laser gaging technology available today has J74 19 been in place for 20 yrs. Only the appliance, automotive, and J74 20 aerospace industries have sought to minimize downtime by inspecting J74 21 key attributes of material used in automatic assembly processes. A J74 22 task originally performed as incoming inspection by the end user is J74 23 now demanded of the supplier.

J74 24 Solving the quality control problems for today's manufacturer J74 25 requires systems designed with the flexibility for large product J74 26 ranges, simplicity of setup, and the ability to work in harsh J74 27 production environments.

J74 28 Perhaps the most over-looked area regardless of the system J74 29 chosen is the maintenance and level of personnel to perform that J74 30 maintenance. The laser sorters utilize a single card to perform all J74 31 the data processing relating to the inspection tasks. An operator J74 32 is capable of correcting most all problems with this type of J74 33 inspection equipment.

J74 34 Once the decision to purchase a system for 100% inspection is J74 35 made, a part packaging system integrated with the inspection J74 36 element will eliminate any potential mixing due to the human factor J74 37 or bulk packaging. Several technologies are offered to assist in J74 38 solving specific problems of the various industries.

J74 39 Our Vision Inspection System is an automatic, non-contact, J74 40 stand-alone inspection and measuring system. The standard J74 41 version with the appropriate options is capable of simultaneously J74 42 making both two-dimensional gaging measurements and surface flaw J74 43 analysis of each field of view (FOV) as well as real-time image J74 44 enhancements. This system is an easy-to-use, menu-driven system. J74 45 The menu system is a flexible job setup tool with screen prompts to J74 46 assure correct entering of parameters. No programming experience is J74 47 required to learn the systems menu. The menu allows the operator to J74 48 run the system on a turnkey level. Also, if new tolerance or J74 49 inspection data are required, they can be inserted through the J74 50 menu.

J74 51 The typical system configuration includes an inspection module J74 52 with the specific application programmed, optics, and camera(s), J74 53 light source, part detector and associated cabling, mounts, J74 54 brackets, etc. The camera is aligned to view one specific J74 55 inspection point in the material handling system. During the actual J74 56 inspection process, the camera captures an image of each part as it J74 57 passes this inspection point.

J74 58 This system utilizes an IBM PC compatible 486 processing J74 59 platform for maximum throughput. A keypad is used by the operator J74 60 to select and change inspection parameters as required. The J74 61 inspection tools used to interpret the image and provide J74 62 orientation information are adjustable in size and position to J74 63 provide the flexibility to detect many configurations without J74 64 changing software.

J74 65 During the setup procedure, the part detector is aligned to J74 66 monitor the areas adjacent to the inspection point. During the J74 67 inspection process, the detector continuously monitors the material J74 68 handling system. As each part reaches the inspection point the J74 69 detector sends the 'Part Detect' signal to the vision inspection J74 70 module. Upon receipt of this signal, the vision inspection module J74 71 actuates the image capture/analysis process. A strobe light source J74 72 is used, as is necessary with a moving part, to illuminate the J74 73 part. A visual image of the part is then captured by the camera.

J74 74 figure&caption

J74 75 A small rectangular element in the camera houses an array of J74 76 tiny electronic devices called photodiodes or pixels. Their J74 77 function is to detect the amount of light reflected from the part J74 78 to the camera. The size of the photodiode array can range from a J74 79 matrix of 256 pixels to 512 pixels (vertical columns and horizontal J74 80 rows). Each of the pixels continuously transmits an analog signal J74 81 to the vision inspection module which assigns a value of either 0 J74 82 or 1 to the binary image acquired.

J74 83 The vision inspection module then transfers the binary values J74 84 to the digital memory. The module can now inject the binary data J74 85 into a variety of complex mathematical routines. These can perform J74 86 the following analytical operations on the captured image: linear J74 87 measurement, flaw analysis, and real-time image enhancement. The J74 88 information obtained by these operations is then used for J74 89 accept/reject decisions. These decisions can be sent out via J74 90 RS-232, through discrete I/O.

J74 91 The 'Magic Black Box' that inspects every dimension on every J74 92 part does not exist, quality control personnel must identify and J74 93 target specific areas for inspection and sorting equipment to be J74 94 implemented in the most cost effective manner. The bulk handling J74 95 for plating, heat treat, and packaging will remain the major areas J74 96 of part contamination due to high volumes and materials handling J74 97 methods.

J74 98 Safe and Fast

J74 99 In the 60's scientists used a laser and a passive corner J74 100 reflector to measure the distance from the earth to the moon within J74 101 six inches. Since then the technology transfer program helped put J74 102 the laser to work for non-aerospace uses. In the past five J74 103 years the development of low-power, personnel-safe gas J74 104 lasers contributed substantially to their applications on J74 105 production and assembly lines. More specifically, gas lasers have J74 106 been adapted successfully to other tasks including automatic J74 107 dimensional and surface inspection in quality control.

J74 108 Lasers are fast, reliable, and non-contact. Properly applied, J74 109 the automatic laser system can completely eliminate manual J74 110 inspection and reduce test times by as much as 90%. Inspection J74 111 speeds can exceed 600 ppm. Studies supporting automation show that J74 112 human inspectors can miss as many as 20% of defective parts when J74 113 operator fatigue sets in. In contrast, the use of automatic laser J74 114 inspection eliminates most of these shortcomings.

J74 115 Operating Principles

J74 116 The beam configurations generated by various existing J74 117 laser-based systems include static spot, circular spot, linear J74 118 scan, and multiple spot. Because the geometry of the laser beam J74 119 determines the inspection system capability and its ultimate use, J74 120 laser systems are classified according to the type of their J74 121 beam.

J74 122 In operation, the transmitted laser beam interacts in various J74 123 ways with the target being inspected. Through this interaction the J74 124 beam at the receiver input will be encoded by various target J74 125 parameters such as its dimensions, surface quality, etc. The J74 126 receiver extracts or decodes this information and uses the J74 127 resulting data for making decisions in inspection. In particular, J74 128 consider the several ways that an incident laser beam can be J74 129 encoded by its target and produce different detected signals at the J74 130 receiver. It can be any of the following:

J74 131 Sequentially interrupted - the detected signal is J74 132 digital. It is either on/off or a go/no-go type. Application is in J74 133 simple length sorting.

J74 134 Reflected - this results in an analog detected signal J74 135 whose amplitude and duration are used in the analysis of the J74 136 inspected part. This approach is used basically in areas of surface J74 137 flaw detection.

J74 138 Shadowed - this produces a digital signal which is used J74 139 for more accurate dimensional gaging, for example, where the J74 140 sequentially interrupted method becomes inappropriate.

J74 141 Transmitted - when detected, this is either a digital or J74 142 an analog signal and is basically used in applications where the J74 143 material to be inspected is clear or transparent. Using transmitted J74 144 signals, parts can be inspected for surface defects, missing J74 145 operations as well as dimensional accuracy. For example, threaded J74 146 fasteners can be searched for surface problems, missing internal or J74 147 external threads, damaged threads, or even mixed sizes of threads. J74 148 The laser beam is scanned in a vertical plane so that it 'walks' J74 149 down the threads. At the same time, the horizontal beam determines J74 150 the location of the part under inspection. The inspection of the J74 151 peaks and valleys in the thread creates a pulsing return signal J74 152 that is picked up by the receiver/detector. The signal contains J74 153 information on the quality and size of the thread.

J74 154 Manual inspection methods using mechanical gaging still hold J74 155 their own in accuracy and resolution capabilities. Thus, even J74 156 though a laser-based inspection system can gage dimensions with an J74 157 0.0004" accuracy, this figure can be exceeded by traditional J74 158 methods. But for industrial applications where this lower gaging J74 159 tolerance is adequate, laser systems win hands down by virtue of J74 160 their highest speed. Today, laser inspection systems are used in J74 161 industries making fasteners, bearings, automotive and aircraft J74 162 components, glass and pharmaceuticals. Laser inspection functions J74 163 range from simple dimensional gaging to providing feedback signals J74 164 for automatic process control.

J74 165 System Components

J74 166 The basic non-contact laser inspection system consists of J74 167 several basic components: a laser used as a target illuminating J74 168 source (the word laser is an acronym of Light Amplification by J74 169 Simulated Emission of Radiation); a receiver to sense the optical J74 170 radiation that carries the required dimensional information; and J74 171 processing circuitry to extract this information in preparation for J74 172 further use. This radiation can be reflected, scattered, or J74 173 shadowed by, or transmitted through the target. Other parts include J74 174 lenses, mirror, prisms, polarizers, choppers, and filters.

J74 175 In practice, the gas laser provides high beam brightness, good J74 176 beam collimation, small beam diameter, and long beam life. These J74 177 laser qualities are used for very high speed, tight tolerance, and J74 178 non-contact inspection on a wide range of parts and materials.

J74 179 Parts made of metals, plastics, glass, rubber, or wood can be J74 180 inspected by a laser without tool changes usually required with J74 181 mechanical methods.

J74 182 Functions Performed

J74 183 Almost all laser system suppliers provide equipment to perform J74 184 some combination or all of the following functions:

J74 185 Gaging - most laser systems are used for gaging length, J74 186 width, height, diameter of parts, and combinations of these at high J74 187 speeds. A benefit of 100% part size qualification is uninterrupted J74 188 production.

J74 189 Operation Voids - the laser system will inspect parts J74 190 for missing operations such as threads, slots, and holes. Or it J74 191 will locate specific indicators to help in aligning parts for J74 192 assembly operations. A laser system is used in the assembly of an J74 193 automotive thermostat to first locate a scribe mark on one of the J74 194 parts. Then using the mark as a reference, to position the part for J74 195 a subsequent staking operation. The laser is used to define the J74 196 exact position. This operation runs about 120 parts/hr. The laser 2 J74 197 ft away positions the scribe mark within 0.001".

J74 198 Surface Flaw Detection - Laser scanning capability is J74 199 used to identify surface trend indicators. While tighter tolerances J74 200 are possible with mechanical inspections, the laser method is much J74 201 faster and more reliable and it works equally well for metallic or J74 202 non-metallic surfaces.

J74 203 Process Control - This function combines gaging, J74 204 surface detection, and identification to provide an on-line output J74 205 that can be used as a feedback signal to an automatic control J74 206 system to provide corrective action. A typical application would be J74 207 in weld seam inspection to generate axial information to guide the J74 208 weld track and flux flow.

J74 209 What of the Future?

J74 210 The future of laser-based inspection seems bright. In addition J74 211 to the already mentioned advantages of speed and reliability, these J74 212 <}_><-|>system<+|>systems<}/> are also predicable in terms of J74 213 pricing. For example, the cost of a laser system can be determined J74 214 accurately for planning purpose. Thus, hard figures are available J74 215 to management for comparison to alternate-method costs.

J74 216 For more information contact the author or Circle J74 217 259.

J74 218 J74 219 Developments in Self-Locking Fasteners

J74 220 Some Background

J74 221 Developed almost 50 yrs ago, the idea of engineering a nylon J74 222 locking element to solve specific fastening problems, Figure J74 223 1, is still going strong today. This is reflected not only in J74 224 growing domestic use, but in a wide range of applications J74 225 throughout the world. Global acceptance of the patented process J74 226 that provides a non-metallic prevailing torque element in both male J74 227 and female threads has been further strengthened by the J74 228 introduction and use of new technology that upgrades the J74 229 performance of the self-locking process.

J74 230 figure&caption

J74 231 For example, a new patented nylon powder dispensing system that J74 232 produces a patcxh-type locking element with more consistent J74 233 performance and better reusability features. J74 234 J75 1 <#FROWN:J75\>The active double-star architecture has J75 2 received considerable research attention. As shown in Figure 2.1, J75 3 the architecture calls for feeder fiber to connect the central J75 4 office to the RDU, which might serve around 1000 subscribers. By J75 5 equipping the RDU with switching capability, the RDU electronics J75 6 direct individually selected channels to each home from the many J75 7 signals multiplexed on the feeder cable. Multiplexing the signals J75 8 between the central office and RDU reduces costs relative to the J75 9 switched-star alternative in which individual channels are J75 10 switched only at the central office for transmission over separate J75 11 fiber cables all the way to the home.

J75 12 The first proposals for an IBN assumed an active double-star J75 13 architecture due in part to its resemblance to the current DLC J75 14 system. The telephone industry continues to cite this architecture J75 15 as a likely approach [Shumate, 1989], and much of the ongoing J75 16 effort to develop prototype IBN components assume this J75 17 architecture, although other architectural alternatives are now J75 18 receiving more serious consideration.

J75 19 Passive optical networks (PONs) are another class of J75 20 networks now receiving a large amount of research attention. PON J75 21 architectures have single fibers emanating from the central office, J75 22 and these fibers fan out via passive optical splitters similar to a J75 23 tree-and-branch topology. In this way, PON networks achieve a high J75 24 degree of shared plant throughout the network. The lowest branch in J75 25 the distribution tree connects to the individual homes. Figure 2.1 J75 26 shows an example of a passive double-star architecture in J75 27 which the RDU now houses an optical power splitter.

J75 28 The optical power budget determines the number of successive J75 29 splitting nodes in a PON network. Larger bandwidth signals have J75 30 smaller power budgets, so broadband services cannot be split as J75 31 often as narrowband services. This leads to the same problem faced J75 32 by hybrid networks to provide a mix of narrowband and broadband J75 33 services. Again, WDM or coherent transmission techniques would have J75 34 to be employed in the future to increase the bandwidth of the J75 35 system.

J75 36 An advantage of PON networks is that they provide a transparent J75 37 path between the central office and the subscriber. This provides a J75 38 degree of flexibility because the architecture is dependent upon J75 39 the power budget and not the transmission format. In addition, the J75 40 use of passive devices reduces the number of active components J75 41 throughout the network, thereby decreasing the number of potential J75 42 faults.

J75 43 There have been few proposals calling for a bus topology in the J75 44 subscriber loop. One problem is accessing the information on the J75 45 bus. If the signal is tapped by bending the fiber and detecting the J75 46 light that escapes - called a nonintrusive tap - then the J75 47 system provides very little flexibility for evolution to new J75 48 services because the taps are optimized to one service bandwidth. J75 49 Another method is to receive and retransmit the information bus at J75 50 every node. This doubles the number of optical transmitters and J75 51 receivers in the field, raising more concerns about the J75 52 environmental protection, reliability, and maintenance cost of the J75 53 remote nodes. However, one could consider the passive double-star J75 54 as a bus topology with a passive splitter as the only node. J75 55 Configuring the network as a double-star rather than a bus J75 56 simplifies maintenance by reducing the number of nodes. Also, there J75 57 is less excess signal loss from a single 1:n splitter than from a J75 58 succession of taps on a bus, thus allowing more subscribers to be J75 59 served within the available power budget.

J75 60 In summary, an all-fiber, active double-star network is still J75 61 under consideration as a viable option for the subscriber loop. J75 62 However, the number of competing alternatives is growing. The J75 63 development of PON and hybrid architectures have served as J75 64 milestones to shift attention away from the active double-star to a J75 65 richer variety of options. PON networks take more advantage of the J75 66 properties of optical transmission, whereas network hybrids offer J75 67 an interim solution before more fiber can be placed in the J75 68 network.

J75 69 Today's Cable Television Networks

J75 70 Cable operators install a tree-and-branch architecture J75 71 using coaxial cable to provide distributed video services, as shown J75 72 in Figure 2.4. The headend receives video signals from local J75 73 studios, over-the-air broadcasts, or microwave and satellite J75 74 sources and then combines and retransmits these signals over the J75 75 trunk cable of the network.

J75 76 figure&caption

J75 77 Portions of the signal on the trunk are split off every few J75 78 hundred meters to feeder cable, where in turn the signal is split J75 79 to drop cable to directly serve the household. Because these J75 80 systems use the standard NTSC AM-VSB transmission format, a J75 81 television set used for over-the-air reception can be connected J75 82 simply to the cable system through a simple frequency converter J75 83 box. This makes for low entry and exit costs to cable services.

J75 84 The high transmission and splitting losses that arise from a J75 85 coaxial cable tree-and-branch network require the placement of J75 86 amplifiers throughout the network to boost the signal. Coaxial J75 87 cable introduces transmission losses of about 1 dB every 30 meters J75 88 (optical fiber introduces about 0.1 dB every kilometer). This loss J75 89 and the amplifier gain determine the amplifier spacing. Given a J75 90 gain of about 20 dB, trunk amplifiers are typically situated about J75 91 every 600 meters. Signal quality also declines as the number of J75 92 amplifiers in cascade increases, because noise and distortions J75 93 accumulate with each amplifier. To maintain signal quality, no more J75 94 than 30-40 amplifiers are generally placed in cascade. In addition, J75 95 the series connection of amplifiers limits the reliability of cable J75 96 service. If one amplifier in the cascade fails, then all subsequent J75 97 network branches downstream of the failure lose their signal, J75 98 resulting in a loss of signal over a potentially large area.

J75 99 The tree-and-branch approach is well suited to the J75 100 characteristics of distributed video services. Each subscriber J75 101 receives the same group of video channels, which favors an J75 102 architecture with a high degree of shared network resources, such J75 103 as the tree-and-branch topology. The bandwidth of the amplifiers J75 104 limits the number of channels carried by the system. Improvements J75 105 in technology have increased amplifier bandwidth from around 200 J75 106 MHz (30 NTSC channels) to 550 MHz (80 NTSC channels) today. The J75 107 bandwidth of the coaxial cable is about 1 GHz (150 NTSC channels) J75 108 and represents the upper bound of capacity on typical cable J75 109 systems.

J75 110 Fiber Backbones

J75 111 The cable industry is also exploring architectures that would J75 112 incorporate fiber into their networks [Chiddix and Pangrac, 1988]. J75 113 In assessing fiber alternatives, cable operators enjoy the J75 114 advantage of already having a broadband connection to the J75 115 subscriber. Fiber installed in any portion of the cable network can J75 116 immediately improve broadband services. Consequently, cable J75 117 industry proposals are exclusively hybrid networks and do not J75 118 include plans to build something like an all-fiber IBN in the near J75 119 future.

J75 120 A common approach for adding fiber into cable television J75 121 networks is the fiber backbone strategy. The idea is to J75 122 replace the primary trunk lines of a current coaxial cable system J75 123 with fiber backbones that extend to within a few hundred feet of J75 124 the subscriber - the remaining distance to be connected using the J75 125 existing coaxial cable. The fiber backbone could entirely replace J75 126 the trunk line, thereby eliminating all the trunk amplifiers, or J75 127 replace a significant portion of the trunk line, thereby shortening J75 128 the number of trunk amplifiers in cascade to a small number.

J75 129 Deployment of fiber backbones reduces the number of trunk J75 130 amplifiers in cascade between the headend and subscriber. Using J75 131 fewer amplifiers in cascade improves network reliability and J75 132 picture quality, increases system capacity, and J75 133 <}_><-|>provide<+|>provides<}/> greater network flexibility for J75 134 offering new broadband services. Network reliability is an J75 135 important attribute, as frequent loss of television signal J75 136 resulting from the long amplifier cascades continues to be a common J75 137 consumer complaint about current cable television services. The J75 138 increase in system capacity made possible by fiber backbones could J75 139 enable cable systems to carry HDTV signals over the cable network J75 140 without having to reduce the number of offered channels.

J75 141 2.4 MARKET STRUCTURE FOR THE SUBSCRIBER LOOP

J75 142 The previous sections described a framework characterizing the J75 143 mechanisms of network evolution, starting with the demand for J75 144 network-based services. Network planners assess the technology J75 145 alternatives and select the approach best suited to meet current J75 146 and future demands. The cost characteristics of the chosen J75 147 technology indicate the appropriate market structure - the most J75 148 efficient form of industrial organization of firms for production - J75 149 for the network-based services. Typically, network-based services J75 150 exhibit economies of scale in production. However, in a J75 151 multiproduct environment, natural monopoly requires the dual J75 152 presence of economies of scale and scope in the production of J75 153 services.

J75 154 Public telecommunications networks provide the communications J75 155 infrastructure vital to the activities of everyday life, and the J75 156 need to establish policies favoring the development of this J75 157 infrastructure is therefore clearly linked to the notions of J75 158 economies of scale and scope. That is, presumably the justification J75 159 for a public communications network, and regulation of the J75 160 network in general, stems from strong economies of scale and scope J75 161 arising in the use of such a resource. Telephone companies propose J75 162 that IBNs will serve as the future communications infrastructure J75 163 for society by providing a diverse set of services based upon the J75 164 strong economies of scope inherent in the services provided by J75 165 fiber-based networks.

J75 166 2.4.1 The Importance of Economies of Scope

J75 167 Without any regulatory barriers, a network operator will offer J75 168 a new service if that service can be offered at a profit for an J75 169 incremental cost lower than that of competitors. Economies of scope J75 170 between new and existing services would lower the incremental cost J75 171 of the new service. As pointed out earlier, the strength of J75 172 economies of scope between narrowband and broadband subscriber loop J75 173 services should ultimately determine the success of IBNs. Whether J75 174 an IBN will achieve economies of scope depends upon the J75 175 transmission characteristics of the network services and the J75 176 technological alternatives to transport these services J75 177 separately.

J75 178 Figure 2.5 qualitatively illustrates this point by plotting the J75 179 hypothetical cost functions for three transmission technologies J75 180 versus three levels of network-based services. Assume that the J75 181 hypothetical cost functions represent the costs of building a J75 182 network using that transmission technology. The cumulative levels J75 183 of service are narrowband, distributed video, and switched video. J75 184 The shaded portion of the graph shows the area in which cumulative J75 185 economies of scope are present; its boundary represents the J75 186 combined cost of separate networks to provide each service J75 187 category. Any cost curve passing through the shaded region exhibits J75 188 economies of scope between the service categories, because the cost J75 189 of an integrated network is less than the cost of separate J75 190 networks.

J75 191 figure&caption

J75 192 The three technology curves depicted in Figure 2.5 are J75 193 presented in part to motivate discussion, but also because they are J75 194 indicative of the actual cost characteristics of the representative J75 195 technologies [Johnson and Reek, 1990]. The curve for copper J75 196 technology illustrates that this technology provides the lowest J75 197 cost alternative for narrowband services, but as the capacity J75 198 requirements of the services increase to broadband levels, copper J75 199 wire pairs quickly becomes uneconomic versus alternative J75 200 technologies and provides no economies of scope for these services. J75 201 The curve for hybrid networks shows that these systems provide J75 202 economies of scope between narrowband and distributive video, but J75 203 economies are lost in upgrades to switched video status. The curve J75 204 for an all-fiber network suggests that these systems offer no J75 205 economies of scope between narrowband services and distributed J75 206 video services (that is, a fiber network carrying these services is J75 207 more expensive than separate copper and coaxial cable networks), J75 208 but if switched services are incorporated into the network, then J75 209 the fiber alternative captures more economies of scope than the J75 210 other alternatives.

J75 211 Although somewhat contrived, these curves do succeed in J75 212 conveying the complexity of the network evolution problem. J75 213 Different technologies are most efficient for transporting J75 214 different combinations of services. No simple solution exists that J75 215 will dominate over the entire range of services. Network planners J75 216 must plan a network that can deliver an efficient mix of services J75 217 given a highly uncertain environment regarding future technology J75 218 developments, market opportunities, and government regulations.

J75 219 One aspect of the deployment of IBNs that is not considered in J75 220 the analysis is any potential economies of scope between businesses J75 221 and residential applications. Businesses may have a much higher J75 222 demand for broadband applications such as high speed date, J75 223 electronic publishing, or video imaging services. Strong demand for J75 224 broadband services by business users located in residential areas J75 225 could provide sufficient economies of scope to facilitate the J75 226 introduction of fiber into these areas. J75 227 J76 1 <#FROWN:J76\>Chapter IX

J76 2 The Insulating Refractory Product Line

J76 3 APPLICATIONS AND APPLICATION CRITERIA

J76 4 Thermally insulating refractories function by providing J76 5 stagnant or 'dead' gas space, which is to say they contain large J76 6 volume fractions of voids. Since it is impossible to build J76 7 closed-cell structures into high-void-volume ceramics, these J76 8 materials are all 'open': susceptible to permeation and saturation J76 9 by hot process liquids and to chemical attack by aggressive gases. J76 10 It follows that they are not willingly exposed directly to liquids J76 11 of any kind, nor to condensible vapors, nor to gases of more than J76 12 minor chemical reactivity.

J76 13 Ergo, corrosion resistance is downgraded as an index of merit. J76 14 The prime criterion for material selection is refractoriness and J76 15 dimensional stability sufficient for the assignment. The first J76 16 property needed for insulating refractory qualification is J76 17 accordingly the service temperature limit, which is related to J76 18 composition, sintering temperature, and void volume.

J76 19 The composition classes of Table VIII.1 turn out to be more J76 20 than ample. Virtually all refractories in the insulating product J76 21 line are oxidic, for the obvious reason that nonoxides are on the J76 22 whole innately efficient conductors of heat. Relief from the J76 23 necessity of resisting corrosion by liquids eliminates some of the J76 24 more expensive oxidic types and their composites.

J76 25 Insulating refractories are grouped at the bottom of Table J76 26 VIII.2, under Porosity -now expanded to mean void volume fraction. J76 27 But Table VIII.2 does not address the microstructures of insulating J76 28 materials nor the manufacturing methods employed specifically to J76 29 create them. Some further subclassifications are needed. To put J76 30 those in perspective, let us first examine where and how insulating J76 31 refractories are used.

J76 32 Duplex Linings; Steady-State Usage

J76 33 There are two reasons for interpolating an insulating layer J76 34 between a hot working lining and the 'outside.' These are: (a) to J76 35 cool the back face, e.g., to preserve the mechanical integrity of J76 36 an enclosing metal shell or for reasons of safety outside a wall or J76 37 roof; and (b) to reduce the heat flux J through the lining and J76 38 hence improve process fuel economy. Both motives may apply at the J76 39 same time, though the second usually predominates.

J76 40 In the simple case of a plane wall at steady state, where a J76 41 hot-face temperature Th is fixed by a given J76 42 operation, then:

J76 43 formula.

J76 44 This equation is set up for a working lining 'w' of mean J76 45 thermal conductivity <*_>unch<*/>w and thickness J76 46 Zw, an insulating lining 'i' of (low) mean thermal J76 47 conductivity <*_>unch<*/>i and thickness Zi, J76 48 and metal shell 's' of (high) mean thermal conductivity J76 49 <*_>unch<*/>s and thickness Zs. Th is J76 50 the (fixed) temperature of the working hot face; Ti is J76 51 that of the interface between linings 'w' and 'i'; Tb is J76 52 the refractory back face temperature or that of the interface J76 53 between lining 'i' and shell 's'; and To is that of the J76 54 outside of the shell. Jo is the heat flux to the J76 55 'outside,' existing by virtue of water-cooling or forced or J76 56 convective air-cooling of the shell.

J76 57 The equation is solvable, given all k's, once To J76 58 or Jo is fixed. This may be done by applying J76 59 practical criteria to Jo with a known relation J76 60 between this To, or by arbitrarily limiting To J76 61 itself (e.g., to <*_>approximate-sign<*/>100<*_>degree<*/>C if J76 62 water cooled, to 40<*_>degree<*/><*_>unch<*/> if exposed to human J76 63 contact, or to some safe limit for mechanical integrity of the J76 64 metal). Ordinarily the shell thickness Zs will be fixed J76 65 by structural design considerations; and Zw may be J76 66 limited by considerations of working lining corrosion or its J76 67 end-of-life minimum allowable thickness, for example. If a metal J76 68 shell is absent, delete the third equality and replace Tb J76 69 by To. If an insulating backup refractory is absent, J76 70 delete the second equality and replace Ti by J76 71 Tb. If both are absent, delete both of those equalities J76 72 and replace Tiby To. The equation is thus J76 73 versatile, and comparisons may be made with it among various lining J76 74 options.

J76 75 Computer programs for solving this steady-state type of J76 76 equation are in common use. Depending on their sophistication, J76 77 provisions may be made for: (a) inserting each k as f(T) J76 78 instead of estimating a mean; (b) inserting additional functions J76 79 for interface conductivity or temperature drop, where applicable; J76 80 and (c) inserting appropriate relations between Jo J76 81 and To for various modes of outside cooling. Comparable J76 82 equations are suited to cylindrical or spherical geometry.

J76 83 For hand calculations, reasonable estimates can be made for the J76 84 independent parameters. Perhaps the only one not self-evident is J76 85 the connection between Jo and To for air J76 86 cooling. A rough empirical curve for unimpeded convective cooling J76 87 of vertical exterior surfaces by ambient air at about J76 88 77<*_>degree<*/>F or 25<*_>degree<*/>C, used for estimating J76 89 purposes, is approximated by:

J76 90 formula (Jo in Btu/hr ft2 and J76 91 To in <*_>degree<*/>F);

J76 92 or, in convenient metric units:

J76 93 formula (Jo in J76 94 kJ/h<*_>dot<*/>m2 and To in J76 95 <*_>degree<*/>C).

J76 96 This rough guide applies to a refractory cold face up to some J76 97 600<*_>degree<*/>F or 300<*_>degree<*/>C. It should do about as J76 98 well for the outside of a steel shell.

J76 99 For a conservative example of what thermal insulation can do, J76 100 suppose the hot zone of a tunnel kiln, firing pottery, averages J76 101 1000<*_>degree<*/>C at the hot face. Assume its working refractory J76 102 sidewalls and roof are 9" or 22.86 cm thick, exposed to the air J76 103 outside, constructed of super-duty firebrick whose mean J76 104 thermal conductivity is 9.5 Btu <*_>dot<*/> J76 105 in./ft2hr<*_>degree<*/>F or 490. kJ <*_>dot<*/> J76 106 cm/m2hr <*_>degree<*/>C. By entering these quantities J76 107 into the above equations and solving simultaneously:

J76 108 J =490 (1000-To)/22.86 and formula,

J76 109 one obtains To =236<*_>degree<*/>C and the heat loss J76 110 J =16,380 kJ/m2hr. Now add only about 2" or 5 cm of J76 111 lightweight insulation to the outside, using a mean thermal J76 112 conductivity of about 0.6 Btu <*_>dot<*/> in./ft2 J76 113 hr<*_degree<*/>F or 30 J76 114 kJ<*_>dot<*/>cm/m2hr<*_>degree<*/>C. What will be the new J76 115 To and heat loss? One solves:

J76 116 formula

J76 117 simultaneous with the above air-cooling equation for J76 118 Jo, obtaining Ti =804<*_>degree<*/>C, J76 119 To =105<*_>degree<*/>C and the heat loss J =4,190 J76 120 kJ/m2hr. The saving in lost heat at steady state is J76 121 (16,380-4,190)/16,380 or very close to 75%. If the kiln hot zone J76 122 dimensions are 80 ft. by 10 ft. wide by 12 ft. high (24.4 x 3. x J76 123 3.7 meters), the total heat-loss area is about 250 m2 and J76 124 the saving in lost heat is about 3 million kJ/hour or 73 million kJ J76 125 per day, or 69 million Btu per day. That is worth about $120,000 a J76 126 year in 1990 U.S. dollars.

J76 127 In general, interpolating an insulating refractory layer or J76 128 increasing its effectiveness: (a) increases Ti and J76 129 decreases J at a fixed value of To; or (b) Increases J76 130 Ti and decreases Zw and To at a fixed J76 131 value of J. These effects on Ti, which is the J76 132 cold-face temperature of the working lining, make that lining J76 133 increasingly vulnerable to corrosion. Of the two effects on J76 134 Ti recited here, the first is much the more pronounced. J76 135 Where corrosion is already economically limiting, in fact, the use J76 136 of an insulating backup lining may be contra-indicated. Examples J76 137 are in the O2-blown steelmaking furnaces and the lower J76 138 parts of the ironmaking blast furnace.

J76 139 For the same basic reason, insulation is never placed on the J76 140 outside of a metal shell to decrease Jo unless J76 141 the shell temperature is already quite low, and never without J76 142 proving analytically that the metal will not be heated above its J76 143 safe limit. These same equations serve also in that proof.

J76 144 Thermal Conductivity, Void Volume, and Bulk Density

J76 145 Armed with the above means of computing and a list of qualified J76 146 insulating refractory materials and their properties, the system J76 147 designer engages iteratively in an approach to both material J76 148 selection and the determination of Zw and Zi. J76 149 The thermal conductivity of the insulating layer becomes J76 150 both a qualifying and a design property; but it is not much used J76 151 for classification. Suffice it for now that values of J76 152 ki can be had from close to those of working J76 153 refractories to nearly two orders of magnitude lower. This k J76 154 depends on the void volume fraction, fv.

J76 155 The void volume fraction is related to measurable quantities J76 156 from which two corresponding densities are obtained at room J76 157 temperature. Every defined solid has a theroretical J76 158 density, <*_>rho<*/>th, obtainable by x-ray J76 159 diffraction for crystalline species or by weight and volume J76 160 measurements for bulk glasses. The latter method is also used for J76 161 multi-phase materials after their maximum consolidation by, e.g., J76 162 fusion casting or hot pressing, etc. Sectioning and counting up of J76 163 pore areas under the microscope can be used to improve this measure J76 164 theoretical density.

J76 165 Likewise, each insulating material comprised in part of such a J76 166 defined solid and in part of void space has a bulk J76 167 density, <*_>rho<*/>b, obtained by measuring its J76 168 weight and its 'bulk volume.' Measurement of the bulk volume is not J76 169 always easy. In concept if not in actual practice, this is J76 170 obtainable by gently shrink-wrapping or enveloping a weighed J76 171 quantity of material in an infinitely thin plastic film and then J76 172 measuring the volume by water displacement in a graduated cylinder J76 173 or pycnometer.

J76 174 Though there are standard ASTM procedures for obtaining both of J76 175 these densities, the above conceptual definition avoids confusion. J76 176 It is important for present purposes that the 'bulk volume' shall J76 177 contain all of the void volume within a specimen, including J76 178 the interstitial packing space in the case of unconsolidated J76 179 granular materials.

J76 180 With these two named densities in hand and denoting the void J76 181 volume fraction by fv, this quantity is:

J76 182 formula.

J76 183 But in the classification of insulating refractories, this J76 184 parameter fv is rarely stated. Instead, the bulk J76 185 density is used for classification. Although industry J76 186 practices vary, it is uncommon in the U.S. for the theoretical J76 187 density to be reported, or even for the phase composition of the J76 188 material to be disclosed. On the other hand, the precise J76 189 computation of k from fv is all but futile anyway; J76 190 k is measured empirically vs T and is reported for each J76 191 commercial and research product. We shall deal with the measurement J76 192 of k along with its numerical cataloguing in Chapter XI.

J76 193 Working Configurations; Cyclic Usage

J76 194 Dramatic improvements in processing economy have also been made J76 195 by the use of insulating refractories in the working J76 196 configuration -that is, directly exposed to the process J76 197 environment. A first requisite is that this environment must J76 198 consist only of 'clean' (i.e., relatively dust-free) gases. A J76 199 second is that, since high-void-volume refractories are to a J76 200 greater or lesser degree friable, the usage must entail J76 201 correspondingly low mechanical wear. Numerous applications meet J76 202 these two requisites, however. Some that exemplify cyclic J76 203 process operation include the walls and roofs of batch driers, J76 204 ovens, and heat-treatment furnaces, batch or periodic kilns, J76 205 insulating lids or 'hot tops' for metal casting, the upper portions J76 206 of a limited number of metallurgical furnaces, some kiln furniture, J76 207 and much hot-gas ducting. Air, moist air, and clean oxidizing or J76 208 neutral combustion products are generally acceptable. Redox cycling J76 209 of the atmosphere disqualifies some insulating refractories, and J76 210 others may not tolerate the production of soot. Chemicals vaporized J76 211 from the charge are usually somewhat harmful. These have to be J76 212 evaluated individually depending on chemical identity, partial J76 213 pressure or transfer rate, and specific rate of attack on the J76 214 selected refractory. Dust, likewise, need not always be nil but J76 215 calls for careful evaluation.

J76 216 What is the peculiar virtue of low-density refractories in J76 217 these cyclic situations?

J76 218 It is intuitive after Chapter IV that the analysis of thermal J76 219 transients entails use of the thermal diffusivity, formula, J76 220 where c = specific heat in kJ/kg<*_>degree<*/>C or J76 221 J/g<*_>degree<*/>C and <*_>rho<*/>b = bulk density as J76 222 defined in the preceding section. Since <*_>rho<*/>b in J76 223 high-void-volume refractories range from low to very low, the J76 224 product c<*_>rho<*/>b runs correspondingly low while J76 225 k does likewise, relative to these properties of dense J76 226 refractories. Computerized mathematical methods of performance J76 227 analysis using <*_>delta<*/> are widely practiced, but can J76 228 not<&|>sic! be demonstrated here. Instead, some approximations and J76 229 manual calculations will be used to illustrate the economic J76 230 benefits of insulating working linings when the temperature is J76 231 cycled.

J76 232 Consider a periodic shuttle kiln, again firing pottery at J76 233 1000<*_>degree<*/>C. Each charge of ware plus kiln furniture J76 234 consumes 20 million kJ in firing, and an additional 20 million kJ J76 235 goes up the stack if it is not recovered. The entire cycle occupies J76 236 22 hours, leaving two hours per day for charging and discharging. J76 237 The cycle consists of 12 hours heat-up plus 4 hours steady-state at J76 238 1000<*_>degree<*/>C, plus 6 hours of slow cooling (burners off).

J76 239 J76 240 J77 1 <#FROWN:J77\>Colorfastness to Fulling

J77 2 Developed in 1954 by AATCC Committee RR22; revised 1957, J77 3 1972; editorially revised 1981, 1984, 1986; reaffirmed 1958, 1975, J77 4 1978, 1983, 1989; editorially revised and reaffirmed 1988.

J77 5 1. Purpose and Scope

J77 6 1.1 This test method is intended for evaluating the J77 7 colorfastness of dyed wool fabric and yarn to mill fulling.

J77 8 2. Principle

J77 9 2.1 Dyed wool test specimens, in contact with the desired J77 10 undyed textiles of choice and steel balls, are enclosed in an J77 11 undyed test cloth bag and fulled in a soap solution in a metal J77 12 container in a Launder-Ometer by procedures varying in temperature J77 13 of the bath and time.

J77 14 3. Terminology

J77 15 3.1colorfastness, n. - the resistance of a material to J77 16 change in any of its color characteristics, to transfer of its J77 17 colorant(s) to the adjacent materials, or both, as a result of the J77 18 exposure of the material to any environment that might be J77 19 encountered during the processing, testing, storage or use of the J77 20 material.

J77 21 3.2 fulling, n. - a textile finishing process in which J77 22 cloth is subjected to moisture, heat, friction and pressure.

J77 23 4. Safety Precautions

J77 24 NOTE: These safety precautions are for information purposes J77 25 only. The precautions are ancillary to the testing procedures and J77 26 are not intended to be all inclusive. It is the user's J77 27 responsibility to use safe and proper techniques in handling J77 28 materials in this test method. Manufacturers MUST be consulted for J77 29 specific details such as material safety data sheets and other J77 30 manufacturer's recommendations. All OSHA standards and rules must J77 31 also be consulted and followed.

J77 32 4.1. Good laboratory practices should be followed. Wear safety J77 33 glasses in all laboratory areas and a single use dust respirator J77 34 while handling powder dyes.

J77 35 4.2 All chemicals should be handled with care.

J77 36 4.3 The dyes listed in this method belong to the following J77 37 chemical classes:

J77 38 C.I. 62106: (C.I. Acid Blue 78), anthraquinone

J77 39 C.I. 42645: (C.I. Acid Blue 15), tri-phenylmethane

J77 40 C.I. 43830: (C.I. Mordant Blue 1), triphenylmethane

J77 41 5. Apparatus and Materials

J77 42 5.1 Launder-Ometer (see 14.1)

J77 43 5.2 Stainless steel specimen containers 8.9x20.3 cm (3.5x8.0 J77 44 in.) (see 14.1)

J77 45 5.3 Stainless stell balls, 6.3 mm (1.25 in.) (see 14.1)

J77 46 5.4 Soap, neutral granular (Fed. Spec. P-S-566 or ASTM D 496) J77 47 (see 14.2)

J77 48 5.5 Sodium carbonate, anhydrous, technical

J77 49 5.6 Phenolphthalein indicator solution

J77 50 5.7 Standard dyeings (see 14.3)

J77 51 5.8 Worsted test cloth with 12 effect floats (see 14.4)

J77 52 5.9 AATCC Chromatic Transference Scale (see 14.1)

J77 53 5.10 Gray Scale for Color Change (see 14.2)

J77 54 6. Test Specimens

J77 55 6.1 Form a bag by folding together a 2-gram dyed wool test J77 56 specimen and a 2-gram cutting of the test cloth [approx. 7.6x10.2 J77 57 cm (3.0x4.0 in.)] with the dyed piece innermost, and stitching two J77 58 of the open edges. Insert ten 6.3 mm (0.25 in.) stainless steel J77 59 balls and stitch the remaining open edges. To test dyed yarn, braid J77 60 the yarn with undyed wool, cotton or other yarn of interest [5.1 cm J77 61 (2 in.)] and tie the braid at the ends. Enclose the braid and ten J77 62 6.3 mm (0.25 in.) stainless steel balls in a test cloth bag formed J77 63 as above.

J77 64 7. Test Conditions

J77 65 table

J77 66 8. Procedure

J77 67 8.1 Prepare a test solution containing 37.5 grams soap and 15 J77 68 grams sodium carbonate per liter.

J77 69 8.2 Place each test bag in a 20.3 cm (8 in.) stainless steel J77 70 container. Add 100 stainless steel balls and 8 mL of the test J77 71 solution, and secure the covers. Place the tubes in the J77 72 Launder-Ometer which has been heated to the desired temperature, J77 73 and run for the required time (see table above). Remove the tubes, J77 74 release the covers, empty the contents into a colander, and rinse J77 75 the test bags in several changes of water at 43C (110F) (until the J77 76 rinse water is neutral to phenolphthalein). Open the bags, remove J77 77 the balls and the specimens, and dry the specimens and the test J77 78 cloth.

J77 79 9. Evaluation

J77 80 9.1 Evaluate the specimens by comparing them with the J77 81 appropriate Standard dyeing which has been subjected to the same J77 82 fulling procedure. These Standard dyeings represent the minimum J77 83 fastness requirements for their respective classes for alteration J77 84 in color and staining of accompanying undyed textiles.

J77 85 10. Alternate Evaluation Method for Color Change

J77 86 10.1 Define the effect on the color of the test specimens by J77 87 reference to the Gray Scale for Color Change.

J77 88 Class 5 - negligible or no change as shown in Gray Scale Step J77 89 5.

J77 90 Class 4 - a change in color equivalent to Gray Scale Step 4.

J77 91 Class 3 - a change in color equivalent to Gray Scale Step 3.

J77 92 Class 2 - a change in color equivalent to Gray Scale Step 2.

J77 93 Class 1 - a change in color equivalent to Gray Scale Step 1.

J77 94 11. Alternate Evaluation Method for Staining

J77 95 11.1 Staining can be evaluated by means of the AATCC Chromatic J77 96 Transference Scale or the Gray Scale for Staining (see 14.5).

J77 97 Class 5 - negligible or no staining.

J77 98 Class 4 - staining equivalent to Row 4 on the AATCC Scale or J77 99 Step 4 on the Gray Scale for Staining.

J77 100 Class 3 - staining equivalent to Row 3 on the AATCC Scale or J77 101 Step 3 on the Gray Scale for Staining.

J77 102 Class 2 - staining equivalent to Row 2 on the AATCC Scale or J77 103 Step 2 on the Gray Scale for Staining.

J77 104 Class 1 - staining equivalent to Row 1 on the AATCC Scale or J77 105 Step 1 on the Gray Scale for Staining.

J77 106 12. Report

J77 107 12.1 Report the class determined for color change or J77 108 staining.

J77 109 12.2 Report evaluation method used (Sections 9, 10 or 11).

J77 110 12.3 For staining report whether Gray Scale for Staining or J77 111 AATCC Chromatic Transference Scale was used.

J77 112 13. Precision and Bias

J77 113 13.1 Precision and bias have not been established for this test J77 114 method.

J77 115 J77 116 Colorfastness to Bleaching with Chlorine

J77 117 Developed in 1927 by AATCC Committee RA34; revised 1942, J77 118 1947, 1955, 1956, 1957, 1962, 1972; editorially revised 1974, 1983, J77 119 1984, 1986, 1991; reaffirmed 1975, 1979; editorially revised and J77 120 reaffirmed 1985, 1989.

J77 121 1. Purpose and Scope

J77 122 1.1 This test method is applicable to cotton and linen textiles J77 123 and mixtures thereof whether dyed, printed or otherwise colored, J77 124 which may be subjected to solutions containing up to 0.3 per cent J77 125 available chlorine.

J77 126 2. Principle

J77 127 2.1 A test specimen and cuttings of the appropriate control J77 128 dyeings are washed in hypochlorite solution under controlled J77 129 conditions.

J77 130 3. Terminology

J77 131 3.1 colorfastness, n. - the resistance of a material to J77 132 change in any of its color characteristics, to transfer of its J77 133 colorant(s) to adjacent materials, or both, as a result of the J77 134 exposure of the material to any environment that might be J77 135 encountered during the processing, testing, storage or use of the J77 136 material.

J77 137 4. Safety Precautions

J77 138 NOTE: These safety precautions are for information purposes J77 139 only. The precautions are ancillary to the testing procedures and J77 140 are not intended to be all inclusive. It is the user's J77 141 responsibility to use safe and proper techniques in handling J77 142 materials in this test method. Manufacturers MUST be consulted for J77 143 specific details such as material safety data sheets and other J77 144 manufacturer's recommendations. All OSHA standards and rules must J77 145 also be consulted and followed.

J77 146 4.1 Good laboratory practices should be followed. Wear safety J77 147 glasses in all laboratory areas.

J77 148 4.2 All chemicals should be handled with care.

J77 149 4.3 Use chemical goggles or face shield, impervious gloves and J77 150 an impervious apron during preparation of sodium hydroxide J77 151 solutions. Use in an adequately ventilated laboratory hood.

J77 152 4.4 An eyewash/safety shower should be located nearby and a J77 153 self-contained breathing apparatus should be readily available for J77 154 emergency use.

J77 155 5. Apparatus and Materials

J77 156 5.1 Launder-Omeer (see 13.7).

J77 157 5.2 Flat iron

J77 158 5.3 Stainless Steel Cylinder, 500 mL 7.5x12.5 cm (3.0x5.0 in.) J77 159 (see 13.7).

J77 160 5.4 AATCC Chromatic Transference Scale (see 13.7).

J77 161 5.5 Gray Scale for Color Change and Gray Scale for Staining J77 162 (see 13.7).

J77 163 5.6 Distilled water, pH 6.8-7.2.

J77 164 5.7 Sodium hypochlorite (4-6 per cent available chlorine, pH J77 165 9.8-12.8) (see 13.2).

J77 166 5.8 Sodium hydroxide

J77 167 5.9 Sodium carbonate

J77 168 5.10 Sodium bicarbonate

J77 169 5.11 Sodium bisulfite

J77 170 5.12 Sodium arsenite

J77 171 6. Test Specimens

J77 172 6.1 One specimen should be used for each of the types of test J77 173 to be applied (four types are described below) and one specimen J77 174 should be reserved for comparison with the tested specimens.

J77 175 6.2 Specimens of not less than 2 or more than 6 grams should be J77 176 taken. If less than 2 grams are available, make up with dyed J77 177 controls or boiled-out unbleached cotton.

J77 178 7. Tests

J77 179 table

J77 180 8. Procedure

J77 181 8.1 Solutions of sodium hypochlorite (NaOCl) containing 0.01, J77 182 0.1, 0.2, 0.3 per cent available chlorine are made up and adjusted J77 183 to a pH of 11.0<*_>unch<*/> 0.2 by means of the proper buffers (see J77 184 13.3).

J77 185 8.2 The test specimen is thoroughly wet out at 25-30C (78-85F) J77 186 in distilled water, or in event that it has been given a water J77 187 repellent finish (see 13.4), it is wet out in a 0.5 per cent J77 188 neutral soap solution at 25-30C (75-85F). The surplus liquor is J77 189 removed permitting the specimen to retain approximately its dry J77 190 weight of liquor and in this condition it is then placed in a J77 191 one-pint glass jar containing 50 times its dry weight of the sodium J77 192 hypochlorite solution adjusted to the required temperature.

J77 193 8.3 The tightly capped jar is immediately placed in the J77 194 Launder-Ometer and the machine run until one full hour has elapsed J77 195 from the time that the specimen was placed in the jar. The J77 196 Launder-Ometer is maintained at a temperature of 27 <*_>unch<*/> 3C J77 197 (80 <*_>unch<*/> 5F) during the test.

J77 198 8.4 The specimen is removed from the jar and rinsed thoroughly J77 199 in cold running tap water for 5 minutes, squeezing or agitating at J77 200 intervals. The surplus water is removed by any convenient means J77 201 after which the specimen is placed in 50 times its dry weight of a J77 202 1.0 per cent sodium bisulfite solution of 27 <*_>unch<*/> 3C (80 J77 203 <*_>unch<*/> 5F) for 10 minutes with occasional agitation. It is J77 204 then removed and again rinsed in cold, running tap water for 5 J77 205 minutes, with occasional squeezing or agitation. After removal of J77 206 the surplus water by any convenient means, it is pressed dry J77 207 between white cotton with an iron having a temperature not above J77 208 152C (305F).

J77 209 8.5 If the test is used for arbitration of a dispute, then the J77 210 test specimen should weigh 4 grams and no tolerances in temperature J77 211 or pH are permissible.

J77 212 9. Controls (see 13.5)

J77 213 9.1 Bleached muslin or light weight cotton cloth covered with J77 214 -

J77 215 No.1 4 per cent Vat Violet BN 10 per cent Paste (CI 68700)

J77 216 No.2 4 per cent Vat Brilliant Violet RK 10 per cent Paste (CI J77 217 63365) or their equivalents in any strengths of these dyes.

J77 218 10. Evaluation Method for Color Change

J77 219 10.1 The effect on the color of test specimens, by each of the J77 220 four tests with different concentrations of available chlorine, is J77 221 evaluated and the colorfastness classified by comparison with the J77 222 Gray Scale for Color Change (see 13.7)

J77 223 Class 5 - negligible or no change as shown in Gray Scale Step J77 224 5.

J77 225 Class 4 - a change in color equivalent to Gray Scale Step 4.

J77 226 Class 3 - a change in color equivalent to Gray Scale Step 3.

J77 227 Class 2 - a change in color equivalent to Gray Scale Step 2.

J77 228 Class 1 - a change in color equivalent to Gray Scale Step 1.

J77 229 10.2 As stated in Section 13.6, it is prudent to include J77 230 Control No. 2 (see Section 9.1) in the tests being made as a means J77 231 to establish that the tests have been made satisfactorily.

J77 232 10.3 When subjected to the four tests and the effect on the J77 233 color evaluated by comparison with the Gray Scale; Control No.2 J77 234 should merit the following classification:

J77 235 table

J77 236 11. Evaluation Method for Staining

J77 237 11.1 In practical process bleaching with chlorine, no visible J77 238 staining is permitted; however, if desired in testing, staining may J77 239 be evaluated and classified by using multifiber test cloth (see J77 240 13.7) attached to the best specimens and classifying the staining J77 241 by comparison with the AATCC Chromatic Transference Scale or the J77 242 Gray Scale for Staining. The means used should be indicated when J77 243 reporting the test results (see 13.8).

J77 244 Class 5 - negligible or no staining.

J77 245 Class 4 - staining equivalent to Row 4 on the AATCC Chart or J77 246 Step 4 on the Gray Scale for Staining.

J77 247 Class 3 - staining equivalent to Row 3 on the AATCC Chart or J77 248 Step 3 on the Gray Scale for Staining.

J77 249 J78 1 <#FROWN:J78\>The new detector was a Hammamatsu 928R 'extended J78 2 red sensitivity' photomultiplier (PMT), pin-for-pin compatible with J78 3 the previous detector, except for a change in bias voltage. Thus, J78 4 the new detector was able to make use of the transimpedance J78 5 amplifier built into the base of the pre-existing detector mount. J78 6 To accommodate the single fibers used in the temperature monitor J78 7 system, two ST bulkhead connectors were mounted near the input end J78 8 of the PMT, angled so that the cones of light emitted by the fibers J78 9 fell on the first dynode of the PMT. Between the PMT and the ST J78 10 mounts was an optical path selector consisting of a stepper-motor J78 11 driven shutter that alternately blocks the path of light coming J78 12 from the sensor fiber or the 'reference fiber.' In the UV monitor, J78 13 the position of this selector was verified by two photoelectric J78 14 sensors; these had to be replaced with magnetic reed switches, J78 15 since the new PMT proved quite sensitive to the wavelength of J78 16 operation of the photoelectric sensors.

J78 17 The grating in the monochromator was replaced with a 930 J78 18 grooves/mm holographic focusing grating blazed for operation at 750 J78 19 nm. This grating has an effective dispersion of 8 nm/mm at the J78 20 output focus of the monochromator. The monochromator itself was J78 21 actually operated (without an output slit) as a spectrograph, with J78 22 two single optical fibers held at the output focus by a special J78 23 mounting system (Figure 2). The fibers were potted with TraCon J78 24 F-230 epoxy in 27 gauge hypodermic tubing inside 22 gauge tubing J78 25 and polished, then inserted into a 0.5" inch section of 0.5" J78 26 diameter aluminum rod having two holes parallel to the rod axis. J78 27 The rod was held in a 1.375" long section of .625" O.D. tubing that J78 28 slid into another tube directly attached to the monochromator body. J78 29 Because of this arrangement, light of two different wavelengths is J78 30 coupled into the two fibers; the spacing between the holes J78 31 determined the wavelength separation between the light coupled into J78 32 the two fibers. A slight adjustment in wavelength could also be J78 33 accomplished by rotating the cylindrical fiber mount. No difficulty J78 34 was encountered in precisely tuning the output to the two J78 35 wavelengths of importance to the temperature monitoring measurement J78 36 (828 and 856 nm – see Figure 3). Between the monochromator output J78 37 hole and the fibers was another shutter assembly acting as a J78 38 wavelength selector that could block either one, or both (to J78 39 provide dark current readings) of the fibers; this output shutter J78 40 assembly differs from the one used in the UV monitor.

J78 41 As a final improvement, a general reduction in dark currents J78 42 was achieved by making extensive use of baffling material in the J78 43 interior of the optics module. The custom analog circuit board J78 44 (incorporating the stepper motor driver, PMT power supply, and J78 45 programmable preamplifier) developed for the UV absorbance monitor J78 46 was used without modification.

J78 47 A commercial 2x2 fused fiber coupler was used to couple light J78 48 at both wavelengths to the temperature sensor fiber, and to the J78 49 reference fiber (see Figure 1). The coupler was chosen because of J78 50 its low temperature sensitivity. To further isolate the coupler J78 51 from warm-up effects, etc., it was mounted in a separate enclosure J78 52 attached to the outside of the main optical module housing. During J78 53 early experiments, considerable 'drift' and sensitivity to fiber J78 54 bending were observed. These effects appear to have been due to the J78 55 presence in the sensor fiber and coupler of light carried by 'leaky J78 56 modes.' In addition, the fiber lengths used in the apparatus are so J78 57 short that the equilibrium distribution of energy between true J78 58 guided modes, each of which has different bending loss sensitivity, J78 59 is never reached under normal transmission conditions. This problem J78 60 was essentially eliminated by the use of mode stripper/mode mixers J78 61 spliced to the coupler. The stripper/mixers were composed of three J78 62 sections: 1 m of 50 <*_>mu<*/>m (core/125<*_>mu<*/>m (cladding) J78 63 step-index fiber, followed by 1 m of 50/120 graded index J78 64 fiber, followed by 1 m of 50/125 step index fiber, all wound inside J78 65 a 3" diameter holder.

J78 66 figures&captions

J78 67 3.2 Fiber

J78 68 Since the temperature measurement algorithm employed by the J78 69 system is based on changes in the intensity of optical absorbance J78 70 peaks (i.e., since the system must work in a high-loss region of J78 71 the fiber transmission curve), it is important to maximize the J78 72 amount of light transmitted by the fiber. To this end, a large-core J78 73 multimode fiber is preferred over the more common single mode J78 74 rare-earth doped fibers. The fiber used in this system was J78 75 a 50<*_>mu<*/>m core/125 <*_>mu<*/>m cladding graded index silica J78 76 structure with a numerical aperture of 0.24, and an effective index J78 77 difference <*_>DELTA<*/>n=0.02. The fiber was designed to give J78 78 optimum performance for spatially averaged temperature measurements J78 79 over a 12 meter path. Based on a conservative system operating J78 80 margin of 20 dB total loss, and on the known loss behavior of J78 81 Nd3+ doped fibers, this means that a Nd3+ J78 82 concentration of approximately 50 ppm was necessary to achieve the J78 83 desired performance. This concentration was obtained in a specially J78 84 prepared MCVD preform by using a carefully controlled J78 85 heated-ampoule system to deliver Nd3+ during deposition J78 86 of the core region of the preform, and resulted in losses of 1.19 J78 87 dB/meter at 828 nm and 1.55 dB/meter at 856 nm. The host glass in J78 88 the core was silica, containing 17% (wt) Al2O3 J78 89 to raise the index of refraction. After collapse, the preform was J78 90 etched in hot hydrofluoric acid until the desired aspect ratio was J78 91 achieved. For use in the temperature monitoring system, the fiber J78 92 was drawn with a standard coating, and a 12 meter length was J78 93 terminated with ST style connectors.

J78 94 3.3 Electronic module

J78 95 The only modification required in the pre-existing (UV monitor) J78 96 electronics package was the replacement of the high-voltage lamp J78 97 power supply with a low-voltage, high current supply. Power J78 98 supplies to drive the stepper motor controller and DC-to-DC J78 99 converter (PMT power supply) in the optics module, together with J78 100 hardwired safety interlocks to prevent operation of the system of J78 101 the failure of cooling gas or opening of the lid of the optics J78 102 module or of the electronics module, were all left as originally J78 103 designed. The remote display unit of the UV monitor system was used J78 104 in the temperature monitoring system to display the calculated J78 105 average temperature. The main component of the electronics module J78 106 is a Pro-Log System 2 Model 10 single-board CPU, complete with J78 107 MS-DOS operating system that was re-programmed for the temperature J78 108 monitoring application.

J78 109 3.4 Software

J78 110 The software used in the distributed average temperature J78 111 monitoring system is directly descended from software developed for J78 112 the UV absorbance monitor. At any given time the program can be in J78 113 any of the fiber different modes: RUN, STANDBY, IDLE, AUTOZERO, or J78 114 TEST. Switching between modes is controlled by operator interaction J78 115 or by software detection of certain error conditions. The software J78 116 for each mode is responsible for determining any condition which J78 117 should cause that mode to exist. Upon exit from any mode, control J78 118 is returned to a dispatch routine, which determines what mode to J78 119 activate next. Figure 4 shows a diagram of the overall software J78 120 architecture used in the distributed temperature monitor.

J78 121 figure&caption

J78 122 On startup the main program enters an initialization routine J78 123 which initializes the display, the analog board, and the shutters, J78 124 turns on the lamp and high voltage to the photomultiplier tube, and J78 125 waits a specified time for the system to warm up. It also reads J78 126 some values from a parameter file that can be adjusted under J78 127 program control.

J78 128 When the initialization is done, the dispatch routine is J78 129 entered. This routine reads the digital status lines and the J78 130 operator switches and decides which mode to activate. When the J78 131 selected mode is terminated the dispatch routine starts again and J78 132 selects a new mode based on the same input or on newly generator J78 133 error codes.

J78 134 During normal operation the system uses RUN more to J78 135 continuously make measurements, calculate the temperature, and J78 136 output the results. STANDBY mode leaves the monitor in an inactive, J78 137 but ready state. IDLE mode is similar to STANDBY mode, except the J78 138 high voltage and the lamp are turned off. AUTOZERO mode is used to J78 139 automatically compute the calibration constant C for the J78 140 temperature calculation. This is done each time the system turned J78 141 on, after the normal warm-up period. Recalculation of C will also J78 142 be necessary when switching the monitor to a new fiber since the J78 143 coupling efficiencies for the fibers to the optics will be changed. J78 144 The 'AUTOZERO' calculation is done while holding the fiber at a J78 145 known temperature. TEST mode is an interactive diagnostic routine J78 146 which allows the operator to exercise all subsystems of the J78 147 monitor. The user communicates with TEST mode via a portable J78 148 terminal to move the shutters, read the photomultiplier, do an J78 149 A/D-D/A check, calculate the temperature, do an auto zero J78 150 calculation, change the calibration temperature used in the J78 151 AUTOZERO routine, or perform other operations.

J78 152 Embedded in the code for all five modes are routines which J78 153 continuously check the monitor for various error conditions. Each J78 154 error is assigned a unique error code which is displayed as a hex J78 155 digit on the lower display. If unexpectedly high or low PMT J78 156 readings are encountered, if the detector shutter is out of J78 157 position, or if the calculated temperature is out of the expected J78 158 range, only the error code is displayed. If the air flow to the J78 159 optics module is interrupted, or if the lamp and PMT supply J78 160 voltages are turned off, the system displays error codes and enters J78 161 IDLE mode. If the cover of the module is opened, the system goes J78 162 into STANDBY mode.

J78 163 3.4.1 Data acquisition

J78 164 In order to do a temperature calculation, six photocurrents J78 165 must be measured: the fiber current for both wavelengths, the J78 166 bypass current for both wavelengths, and the dark current for both J78 167 the fiber and the bypass. The data acquisition routines must J78 168 position the shutters to select the appropriate light path and read J78 169 the output of the photomultiplier tube. A layered software J78 170 structure has been designed to accomplish these tasks. The lowest J78 171 layer is the hardware specific routine which reads the analog to J78 172 digital converter. The highest level routine is the routine which J78 173 coordinates the measurement of all six different photocurrents.

J78 174 This highest level routine oversees the collection of all the J78 175 data for both the temperature determination and the autozero J78 176 function. Each call to this routine results in all six J78 177 photocurrents being measured once and the proper value for each J78 178 being assigned to its associated global variable, where it can be J78 179 accessed by the calculational routines. To do this, the routine J78 180 first initializes the integrating amplifier timer, then calls the J78 181 shutter positioning routine in the proper sequence for each of the J78 182 six photocurrents, waits for the shutters to come to rest, checks J78 183 for shutter positioning errors, and finally calls the next lower J78 184 acquisition routine, to get a value for the photocurrent.

J78 185 The second highest level data acquisition routine handles the J78 186 processing functions that are necessary to improve signal quality. J78 187 Five measurements are made for each photocurrent, sorted by the J78 188 routine, and the median three are summed for the final result, J78 189 which is then returned to the highest level routine.

J78 190 Each of these five measurements are made by a call to the next J78 191 lower level routine, which uses a TTL I/O board to put an J78 192 integrating amplifier through one complete charging cycle. The J78 193 integrating amplifier works by using the output voltage of the J78 194 photomultiplier base to charge a capacitor for an accurately known J78 195 time interval, through a fixed resistor and an operational J78 196 amplifier. The voltage developed across the capacitor is further J78 197 amplified by a second amplifier stage and presented to the analog J78 198 to digital converter, where it is read by the lowest level data J78 199 acquisition routine. To cycle the integrating amplifier and make a J78 200 measurement, the computer first issues a TTL signal to the JFET J78 201 switch on the optics module circuit board which, in turn, connects J78 202 a small resistor across the charging capacitor, causing it to J78 203 discharge. This switch is then opened and an accurately timed J78 204 signal is applied to another JFET switch which allows the capacitor J78 205 to begin charging. At the end of the time interval, this second J78 206 JFET switch is opened and the voltage is read by the lowest-level J78 207 routine.

J78 208 J79 1 <#FROWN:J79\>Seven

J79 2 Nuclear Safety

J79 3 Scientific, industrial, and political leaders involved with the J79 4 peaceful development of nuclear energy understood from the very J79 5 beginning that safety was essential to the success of the new J79 6 technology. While sound radiation protection standards were not J79 7 fully developed at the outset, the importance of isolating workers, J79 8 the public, and the environment from radioactive materials was J79 9 recognized. Further, it was clear that the peaceful benefits of J79 10 nuclear power should be shared among nations, and that technical J79 11 assistance with the development of civilian nuclear power programs J79 12 might be used as an incentive to accept international safeguards, J79 13 thus discouraging the spread of nuclear weapons. It was also known J79 14 that nuclear energy would have to be economically competitive for J79 15 any large-scale deployment to occur. However, the early developers J79 16 of nuclear energy were aware that they could not predict all the J79 17 problems the budding technology eventually would face.

J79 18 One of those problems has turned out to be public concern over J79 19 nuclear energy - a technology that scientists see as safe, J79 20 nonpolluting, and capable of enriching the lives of people in many J79 21 areas of the world.

J79 22 President Eisenhower's 'Atoms for Peace' speech on December 8, J79 23 1953, opened the door to domestic and international development of J79 24 nuclear energy. By that time, a large amount of data was already J79 25 available from programs on nuclear safety sponsored by the handful J79 26 of nations with active development programs. The United Nations J79 27 sponsored its first conference on the "Peaceful Uses of J79 28 Atomic Energy" in Geneva in 1955. No less than 200 (out of J79 29 a total of 1132) papers on nuclear safety and protection of the J79 30 environment, and a comparable number of papers on medical uses of J79 31 radiation, were presented at the conference.

J79 32 graph&caption

J79 33 Radiation J79 34 Any concern over nuclear safety is based on the potential for J79 35 release of radiation or radioactive materials. The design and J79 36 operation of nuclear power plants is intended to prevent harmful J79 37 releases under all circumstances, from normal operation to highly J79 38 improbable accidents. Nevertheless, radiation is an inevitable J79 39 by-product of nuclear power as well as a natural component of our J79 40 environment. The fraction of our annual radiation exposure that J79 41 comes from the generation of electricity by nuclear power is J79 42 approximately 0.1%, which is insignificant relative to even the J79 43 variations in the background of natural radiation.

J79 44 Radiation and its effects on man continue to be the subjects of J79 45 study by researchers and review by prestigious scientific panels. J79 46 The annual average exposure from natural radiation for a typical J79 47 American was estimated to be about 100 millirem per year until J79 48 1987, when it was revised upward to 360 millirem because of a new J79 49 appreciation for the contribution from naturally occurring radon J79 50 gas. Of the 360 millirem, 18% comes from man-made radiation, J79 51 primarily from voluntary exposure associated with medical J79 52 procedures. Consumer products such as smoke detectors and luminous J79 53 watch dials contribute about 3% of the exposure. Fallout from J79 54 atmospheric testing of nuclear weapons accounts for less than 0.3%. J79 55 The natural radioactivity in our own bodies contributes J79 56 approximately the same amount that we receive from medical x rays. J79 57 Cosmic radiation, streaming in from outer space, averages 8% of the J79 58 total.

J79 59 Individual radiation doses are unevenly distributed, depending J79 60 on where we live, types of houses and work buildings, types of J79 61 medical treatment, frequency of high-altitude flights, and other J79 62 factors. The 0.1% of the average exposure due to the nuclear fuel J79 63 cycle is received almost entirely by workers in the industry; the J79 64 rest of the population gets effectively nothing from the generation J79 65 of nuclear power.

J79 66 Most data on the adverse health effects of radiation have come J79 67 from studies of the survivors of the atomic bombs dropped on J79 68 Hiroshima and Nagasaki. Individuals receiving short-term exposure J79 69 to radiation on the order of 100 rem (100,000 millirem) are subject J79 70 to acute illness. Acute exposures of 500 rem are sufficient to kill J79 71 a substantial fraction of individuals within weeks. But there is J79 72 controversy in how to extrapolate back by a factor of 1000 or more J79 73 to the very low dose rates associated with normal living, medical J79 74 treatment, and occupational exposure. One theory is that the effect J79 75 is linear, allowing a simple extrapolation to be done. A second J79 76 theory is that there is a threshold below which there is J79 77 effectively no damage, or the damage is such that it can be J79 78 repaired by the body. A small minority holds the view that low J79 79 radiation doses actually cause proportionately more damage. J79 80 Regardless of which theory is correct, the radiation emitted from J79 81 operation of a nuclear plant is far too insignificant to have any J79 82 impact on public safety.

J79 83 It might be expected that because there is such a large J79 84 variation in state-to-state natural radiation exposure, the J79 85 importance of low radiation doses could be inferred from J79 86 state-to-state variations in the cancer rate. But the states with J79 87 the highest radiation exposures happen to have the lowest cancer J79 88 rates. The risk of cancer from normal radiation exposure is simply J79 89 insignificant compared to such other causes as smoking, industrial J79 90 pollution, and life-style.

J79 91 Reactor Safety

J79 92 In the United States, the government sponsored aggressive J79 93 reactor safety experiments prior to the start of commercialization J79 94 activities. Approximately 900 square miles of isolated desert land J79 95 in Idaho were selected to be the site for the National Reactor J79 96 Testing Station. One reason for selection of this remote location J79 97 was so that any accidental release of radioactive materials from J79 98 tests to study reactor accidents would not affect the public. J79 99 Safety testing continued well after the commercial nuclear power J79 100 era began, often with international partners. Other Western nations J79 101 recognized the imperative of safe operation of nuclear power plants J79 102 and conducted independent reactor safety experiments. The safety J79 103 philosophy that evolved in the former Soviet Union and Eastern J79 104 European nations was apparently less conservative, but there is a J79 105 current initiative to 'upgrade' some Soviet-built reactors to J79 106 Western operating and design standards. Nevertheless, some J79 107 Soviet-built PWRs such as those in Finland and Hungary have J79 108 excellent records for operating efficiency.

J79 109 The early development period in the United States was J79 110 characterized by a broad-based AEC program that included uranium J79 111 exploration, enrichment, power reactors, space propulsion, waste J79 112 management, biology, medicine, and peaceful applications of nuclear J79 113 explosions. The parallel naval reactors program focused on J79 114 production of nuclear propulsion systems. Development of safety J79 115 technology was a key focus of the U.S. Atomic Energy Commission's J79 116 (AEC's) civilian and naval reactor programs. Radiation-effects J79 117 biology programs were carried out; fundamental and applied nuclear J79 118 data were generated; a number of special safety-experiment reactors J79 119 were built and operated; and several series of major reactor safety J79 120 experiments were carried out. Fundamental radiation standards for J79 121 all radiation application were developed.

J79 122 From the very beginning, reactor designers understood that a J79 123 nuclear power plant could not explode like a nuclear bomb - the J79 124 hazard was from the 'ashes' of the chain reaction, i.e., the J79 125 fission products. Reactor safety was based on containment of any J79 126 credible release of radioactive material. The concept of 'defense J79 127 in depth' became standard practice to meet the general nuclear J79 128 safety objective, which is to protect individuals, society, and the J79 129 environment by establishing and maintaining in nuclear power plants J79 130 an effective defense against radiological hazard. First and J79 131 foremost, defense in depth emphasizes the importance of preventing J79 132 accidents. However, in the event of an accident, defense in depth J79 133 emphasizes mitigation of damage and doing everything possible to J79 134 minimize injuries to workers and the public. This philosophy has J79 135 continued to evolve over the years into internationally accepted J79 136 basic safety principles.

J79 137 figure&caption

J79 138 In 1957 the first bounding estimates of the consequences of a J79 139 large nuclear accident, based on an arbitrarily large release of J79 140 fission products, were published by the AEC in a landmark report J79 141 called WASH-740. It was not until 1975, when WASH-1400 (better J79 142 known as the Reactor Safety Study, or Rasmussen report, J79 143 after MIT Professor Norman Rasmussen who headed the study) was J79 144 issued, that a truly quantitative estimate was available for the J79 145 probability of a major reactor accident in a U.S. nuclear plant. J79 146 Over the years, the Reactor Saftey Study has been challenged by a J79 147 number of critics, and new safety issues requiring investigation J79 148 have arisen, but nothing has happened to change the basic J79 149 conclusions of the report.

J79 150 In any type of risk analysis, consequences become more severe J79 151 as more and more improbable circumstances and events are J79 152 postulated. For most reactor accidents, WASH-1400 showed that the J79 153 most probable result would be property damage to the plant, but no J79 154 deaths or acute illnesses due to radiation exposure of workers or J79 155 the public.

J79 156 The worst type of reactor accident is the so-called reactor J79 157 'meltdown,' which might be expected to happen once in 20,000 years J79 158 of reactor operation. The study showed that there would be no J79 159 detectable deaths in 98 out of 100 reactor meltdowns, over 100 J79 160 detectable deaths in only 1 out of 500 meltdowns, and 3500 J79 161 detectable fatalities in only 1 out of 100,000 meltdowns. In J79 162 addition to deaths immediately attributable to the accident, the J79 163 cloud of radioactive material released in some cases would expose a J79 164 large population to small doses of radiation. Using the J79 165 conservative linear extrapolation model for radiological effects, J79 166 some small fraction of the population would be expected to develop J79 167 cancer. The average impact would be 400 fatalities over several J79 168 decades. For the worst meltdown accident considered, there would be J79 169 an estimate 45,000 additional cancer deaths in an affected J79 170 population of ten million people. This corresponds to an increase J79 171 in the probability of an individual's dying from cancer by about J79 172 0.5%, which is significantly less than the state-to-state variation J79 173 in the normal cancer mortality rate. If there were a reactor J79 174 meltdown every five days, the effect would be comparable to the J79 175 estimated 30,000 deaths in the U.S. each year as a result of J79 176 pollution from burning coal

J79 177 Antinuclear activists often talk only of the most severe J79 178 postulated accident, the 1 case in 100,000 meltdown accidents, J79 179 leaving the impression that all major reactor accidents would J79 180 result in disaster. Clearly, with fewer than 500 nuclear power J79 181 plants operating worldwide, there are not going to be 100,000 J79 182 reactor meltdown accidents, or even 100. The equivalent cannot be J79 183 said of other energy technologies. There have been incidents such J79 184 as dam failures and excessive pollution that have caused a large J79 185 number of fatalities. Perhaps the best known incident happened in J79 186 1952, when pollution from coal burning under unusual atmospheric J79 187 conditions in London caused 3500 fatalities within a few days.

J79 188 People are subject to all types of risks, which can best be J79 189 understood in relative terms. For example, a person would have a J79 190 20,000 times greater chance of being killed by lightning than by J79 191 the largest reactor accident described above. If all electricity in J79 192 the U.S. were generated by nuclear power, it would represent the J79 193 same risk as a regular smoker indulging in an extra cigarette once J79 194 every 15 years, as an overweight person putting on an extra 0.012 J79 195 ounce, or increasing the highway speed limit from 55 to 55.006 J79 196 miles per hour. For any reactor accident, the probability of J79 197 occurence is much less than other man-made or natural J79 198 accidents with similar consequences. The relative risk of nuclear J79 199 power, to either workers or the public, is simply quite small.

J79 200 The type of analysis that went into WASH-1400 has continued to J79 201 be refined. The U.S, Nuclear Regulatory Commission (NRC) now J79 202 encourages all nuclear utilities to conduct probabilistic risk J79 203 assessments (PRAs) that are plant-specific. The largest J79 204 benefit from these expensive and lengthy studies has been the J79 205 ability to identify the systems or operations that pose the most J79 206 risk for bringing about an accident. Money for modifications and J79 207 maintenance can then be effectively directed, resulting in real J79 208 safety improvements. As a result, today's plants, even older ones, J79 209 pose less risk than the plant used in the Reactor Safety Study.

J79 210 Reactor Accidents

J79 211 There have been two major accidents in nuclear electrical J79 212 generating stations plus a major accident at a large J79 213 plutonium-production reactor. There have also been a number of J79 214 lesser accidents with no off-site consequences, a substantial J79 215 number of accident 'precursors' - events that could lead to an J79 216 accident if corrective action were not taken, and a few serious J79 217 accidents at government-owned nuclear facilities. J79 218 J80 1 <#FROWN:J80\>Floodlighting had its merits for early warning, but J80 2 the rotating beam had two enormous advantages: there is a greater J80 3 concentration of energy, and plan position indicators (PPIs) could J80 4 be employed. The advantage of PPI display had been appreciated in J80 5 both Germany and Britain since 1935; but it was only practical at J80 6 much higher frequencies with a rotating beam system. The time-base J80 7 of the PPI radiates from the centre of the screen in J80 8 synchronization with the beam. The target appears as a spot of J80 9 light in the direction of the echo at a distance from the centre J80 10 that represents the range of the echo, as in Fig. 11.5. When the J80 11 immediate priority of an early-warning system was satisfied, J80 12 efforts in ground radar turned to CHL (chain home low), GCI, and J80 13 gunlaying systems.

J80 14 Narrow beams required antenna arrays large in relation to J80 15 wavelength. At the CH frequencies (and also those used by the J80 16 German system Freya), vertical antenna arrays, 70 m and more J80 17 in height, were required to obtain the necessary discrimination - J80 18 in particular to avoid the swamping of the picture by ground J80 19 returns. Something quite different was required for ship borne J80 20 radar, to say nothing of aircraft equipment. An October 1935 J80 21 statement of the operational requirement for naval radar specified J80 22 microwave equipment, not only to reduce the size of the ship's J80 23 antenna, but to avoid the ship's radar becoming a homing beacon for J80 24 enemy bombers. The technology to home on to microwave transmissions J80 25 came only a few years later.

J80 26 figure&caption

J80 27 There was nothing new about VHF: the original Hertz oscillator J80 28 probably transmitted on a wavelength of about 4 metres, had he the J80 29 means to measure it. In the 1920s it had been thought that ordinary J80 30 thermionic valves could not operate on centimetric wavelengths, J80 31 because the transit time of the electrons from cathode to anode was J80 32 too long. The real, unrecognized, problem was the stability of the J80 33 transit time. Several exotic microwave transmitting devices J80 34 appeared, some of them producing a fraction of a watt. In 1936 the J80 35 16 cm obstacle detector on the Normandie produced 10 J80 36 watts. As many kilowatts were sought. One important J80 37 development was the magnetron, a term used since 1921 to J80 38 describe a vacuum electron tube (usually a cylindrical diode) in J80 39 which plate current is controlled by magnetic fields. In the early J80 40 1930s Philips of Holland produced a magnetron which gave 80 watts J80 41 on 13 cm continuous wave. A Philips magnetron was delivered to the J80 42 German navy in 1933.

J80 43 The difficulties due to the inability to produce power on J80 44 centimetric wavelengths, and therefore to discriminate with small J80 45 antenna arrays, are clearly seen in early British aircraft J80 46 interception (AI) radar operating on 1.5 metres. The transmitting J80 47 array was in the nose, two quarter-wavelength antennas on a wing J80 48 acted as elevation antennas, and a half-wavelength dipole with J80 49 director on each wing provided the azimuth array. The polar diagram J80 50 is illustrated in Fig. 11.6, the display in Fig. 11.7. The signal J80 51 received from the elevation antennas (after amplification) is fed J80 52 to either side of the vertical display; and, similarly, the signals J80 53 from the azimuth arrays to either side of the horizontal display. J80 54 Thus the orientation of the target is indicated by the imbalance of J80 55 the blip, about 20<*_>degree<*/> below and 30-40<*_>degree<*/> to J80 56 the left in the illustration. There was no range scale, nor any J80 57 need for one, because the ground return swamped the screen at all J80 58 ranges greater than the aircraft height, which it sharply J80 59 defined.

J80 60 AI MK IV described above (and installed in RAF Beaufighters in J80 61 1940) is interesting also because it created a human requirement in J80 62 terms of spatial visualization and quick thinking J80 63 <}_><-|>unparalled<+|>unparalleled<}/> in the history of J80 64 navigation. As the armament was fixed forward, the object of the J80 65 navigation was to get on the target's tail. If the target's range J80 66 is allowed to exceed aircraft height, it is lost. The target's J80 67 course can be inferred indirectly from the motion of the 'blip' J80 68 down the time-base and the change in its shape. Assume that the J80 69 Beaufighter is heading south. If the target is heading east, a J80 70 sharp turn to the left is required at mximum power before the J80 71 target gets lost in the ground return. If the target is heading J80 72 west, the only hope is to throttle right back, drop the J80 73 undercarriage to increase drag, and turn right, losing the blip J80 74 temporarily, and hoping to come out of the turn behind the target J80 75 and not in front of it. There are only a few seconds to decide, J80 76 and, when the correct decision is taken, the enemy unkindly changes J80 77 course! Only a few had the knack of it.

J80 78 figure&caption

J80 79 The desperate need for powerful microwave transmitters was met J80 80 by the invention of the multi-cavity magnetron at Birmingham J80 81 University by Henry A. Boot and John T. Randall at the beginning of J80 82 1940. The university had been given a contract by the Admiralty to J80 83 research microwave transmission, and work had concentrated on J80 84 developing a high-power version of the klystron invented at J80 85 Stanford University, California. Randall and Boot were assigned the J80 86 less promising task of trying to do something with the magnetron. J80 87 Apparently they shared the research philosophy of Watson Watt and J80 88 Wilkins, for in 1976 they were to write: "Fortunately we J80 89 did not have the time to survey all the published papers on J80 90 magnetrons or we would have been completely confused by the J80 91 multiplicity of theories of operation."

J80 92 The story is related that Randall went back to the original J80 93 experiments of Hertz and, in his mind, extended the Hertz resonant J80 94 ring into a cylinder with a slit in it, as in Fig. 11.8. He then J80 95 saw how this could be developed into the six-cavity figure J80 96 illustrated. Early in 1940 the laboratory model produced over 1 kw J80 97 pulse power at a wavelength of about 10 cm. The first production J80 98 model generated 10 kw, which was soon increased to 100 kw. At the J80 99 end of the war there was megawatt power at that wavelength, and J80 100 wavelengths down to 3 cm could be transmitted at less power.

J80 101 figure&caption

J80 102 That was the beginning of microwave radar as we now know it. J80 103 The Royal Navy seized on the magnetron (and on related work on the J80 104 klystron to provide a local oscillator for centimetric receivers) J80 105 to produce the first fully operational centimetric radar in the J80 106 world. From its formation early in 1941, the Admiralty Signals J80 107 Establishment made major contributions to coastal and naval radar. J80 108 Perhaps the improvement in AI is as dramatic an illustration as any J80 109 of the potency of the multi-cavity magnetron. Ground returns could J80 110 be put where they belonged, and any of the three displays shown in J80 111 Fig. 11.9 were at the AI designer's disposal. In Fig. 11.9(a) the J80 112 display shows the target as the eye sees it, in correct azimuth and J80 113 elevation, but with no indication of range, since a 2D display J80 114 cannot show three independent variables. This can be supported by a J80 115 second display in which range is shown against either azimuth or J80 116 elevation, as illustrated in Figs. 11.9(b) and 11.9(c).

J80 117 figure&caption

J80 118 A scarcely less dramatic outcome of the magnetron was J80 119 terrain-mapping radar, in the form of H2S and later J80 120 derivatives. In the early years of the war metric naval radar and J80 121 ASV (aircraft to surface vessel radar) displayed coastlines, but no J80 122 metric radar could effectively distinguish the features of terrain. J80 123 Ten-centimetre H2S enabled the bomber to identify cities J80 124 and some topographical features of enemy territory by night or J80 125 through cloud. Later 3 cm equipment gave even greater detail. J80 126 Accurate altitude above terrain for precision bombing was also J80 127 provided.

J80 128 Shortly after the achievement of Boot and Randall all J80 129 resistance to Hitler on the continent of Europe ceased. In the J80 130 opinion of many, not least the US ambassador to Britain, the defeat J80 131 of Britain by Germany was imminent. Certainly Britain lacked the J80 132 production resources its situation needed. The USA was not to J80 133 become (by courtesy of Japan) an ally for another year and a half. J80 134 Churchill chose this moment to do a curious thing. On 8 July 1940 J80 135 President Roosevelt received a letter from the British ambassador J80 136 to Washington offering to disclose all Britain's technical secrets J80 137 to the American government without reciprocal undertakings from the J80 138 US; but there was an implied quid pro quo. The letter concluded: J80 139 "We for our part are probably more anxious to be permitted J80 140 to employ the full resources of the radio industry in your country J80 141 with a view to obtaining the greater power possible for the J80 142 emission of ultra short waves than anything else."

J80 143 The timing was perfect. Eleven days earlier Roosevelt had J80 144 created the National Defence Research Council (NDRC), and one of J80 145 its first acts was to establish a microwave committee. A British J80 146 mission led by Tizard (Sir Henry at that date) departed on 30 J80 147 August for Washington bearing, inter alia, the J80 148 multi-cavity magnetron, described in 1946 by one enthusiastic J80 149 American scientist as "the most valuable cargo ever brought J80 150 to our shores".

J80 151 The word 'magnificent' appears infrequently in this book; but J80 152 the American response was magnificent. In October the J80 153 Radiation Laboratory was founded at the Massachusetts Institute of J80 154 Technology (MIT), and eminent American scientists flocked to it. J80 155 The Laboratory's given priorities, agreed between the Tizard J80 156 mission and the NDRC, were in order:

J80 157 1. microwave AI;

J80 158 2. precision gunlaying radar; and

J80 159 3. a fixing system requiring no response from the ship or J80 160 aircraft (see Chapter 12).

J80 161 In the same month the first contracts were being settled with J80 162 Bell Laboratories, General Electric, Sperry, Bendix, RCA, and J80 163 Westinghouse.

J80 164 In December 1940, for the first time ever, radar was fitted to J80 165 an American aircraft. It was a 1.5 m pre-magnetron British ASV Mk J80 166 II (see below), fitted to a US Navy PBY. Only three months later an J80 167 American-designed and built 10 cm radar employing magnetron J80 168 technology was flying in a US B18. At that precise date (10 March J80 169 1941) the staff of the Radiation Laboratory had already reached J80 170 140. In effect, American aircraft overflew pre-magnetron radar. J80 171 When the US finally became a belligerent in December 1941, its J80 172 leadership in the field was assured.

J80 173 No small group of people ever win a major war; but sometimes J80 174 quite small groups can prevent it being lost. It might be said that J80 175 the group associated with ASV (aircraft to surface vessel) were in J80 176 that category. ASV worked on similar principles to AI Mk IV except J80 177 that only a horizontal display was required and the sea return, J80 178 unlike the ground return, did not swamp the picture. The equipment J80 179 did not require the special aptitude and skill required by Mk IV J80 180 AI, but did require intense concentration. The early J80 181 unsuccessfulness of the Mk I ASV in anti-U-boat operations has been J80 182 variously attributed to the equipment itself, poor training or J80 183 unsuitability of aircrew, and defects of aircraft and weaponry. In J80 184 the quarter ending with February 1941 96 Allied ships were sunk by J80 185 U-boats without loss. If that rate had continued, Britain would J80 186 have quickly lost the war by attrition.

J80 187 figure&caption

J80 188 Figure 11.10 (based on Figure 6.1 of Bowen 1987) shows the J80 189 effect of the introduction of ASV Mk II, which had the same 7 kW J80 190 pulse output as ASV Mk I, but was better engineered and had a J80 191 longer pulse and a lower PRF. Presumably improvements in aircrew, J80 192 aircraft, and weaponry also played their part. The rise in Allied J80 193 shipping losses in early 1942 seen in the figure has been J80 194 attributed in part to the rich pickings for U-boats in American J80 195 waters immediately after the USA entered the war, but was also J80 196 partly due to the German development of ASV detectors and other J80 197 U-boat counter-measures. The signal received at the target is of J80 198 course several orders stronger than the echo signal received back J80 199 at the radar transmitter. The range of receivers designed to give J80 200 warning of radar surveillance may therefore exceed the range of the J80 201 radar itself. German equipment to warn submarines of ASV and night J80 202 bombers of AI became a technology in its own right.

J80 203 After the introduction in March 1943 of microwave ASV Mk III, J80 204 an H2S derivative, the U-boat became an ineffective J80 205 weapon and the coffin of its crew. J80 206 J80 207