How We Divide the World Michael Root Philosophy of Science, Vol. 67, No. 3, Supplement. Proceedings of the 1998 Biennial Meetings of the Philosophy of Science Association. Part II: Symposia Papers. (Sep., 2000), pp. S628-S639. Stable URL: http://links.jstor.org/sici?sici=0031-8248%28200009%2967%3A3%3CS628%3AHWDTW%3E2.0.CO%3B2-6 Philosophy of Science is currently published by The University of Chicago Press. Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/ucpress.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org. http://www.jstor.org Thu Mar 27 16:05:14 2008 http://links.jstor.org/sici?sici=0031-8248%28200009%2967%3A3%3CS628%3AHWDTW%3E2.0.CO%3B2-6 http://www.jstor.org/about/terms.html http://www.jstor.org/journals/ucpress.html How We Divide the World Michael Root? University of Minnesota, Minneapolis-St. Paul Real kinds or categories, according to conventional wisdom, enter into lawlike gener- alizations, while nominal kinds do not. Thus, gold but not jewelry is a real kind. How- ever, by such a criterion, few if any kinds or systems of classification employed in the social science are real, for the social sciences offer, at best, only restricted generaliza- tions. Thus, according to conventional wisdom, race and class are on a par with tele- phone area codes and postal zones; all are nominal rather than real. I propose an account of real kinds that recognizes the current reality of race but not zip codes and shows how a kind can be both constructed and real. One virtue of such an understand- ing of realism is the light shed on our current practice of racial classification. Race is not a real biological kind but neither is race a myth or illusion. However, the question of whether a social kind is real is separate from whether the category is legitimate. W. E. B. Du Bois maintained that while there are no biological races, race is real and should be conserved. My aim, in this paper, is not to argue for the legitimacy or con- servation of race but to defend Du Bois's idea that kinds of people can be both made up and real and provide an understanding of realism that does justice to the social sciences. 1. Introduction. W. E. B. D u Bois wrote that the great problem of the 20th century is the color line, the division of men into races, but he also wrote that there are no biological races and no natural differences between hu- mans marked by our system of racial classification. Racial differences, he maintained, are invented and yet no less real than any in nature (1969, 1992). How can this be? Can a category of our own making be a real kind? Most realists say "no." Were there no biological races, race, o n their view would be a myth or an illusion, and racial classification would have no place in science; for unless race is biological, 'race' is not a term of difference but only a trope (Appiah 1992). My aim, in this paper, is to explain how a social category, a kind like race, can be both invented and tSend requests for reprints to the author, Department of Philosophy, University of Minnesota, 831 Walter Heller Hall, 271 19th Avenue South, Minneapolis, MN 55455. Phrlosophy of Scrence, 67 (Proceedings) pp 9628-9639. 0031-824812000167supp-0048$0.00 Copyright 2000 by the Ph~losophy of Science Association. All rights reserved. HOW WE DIVIDE THE WORLD S629 real and how the distinction between blacks and whites can have a place in science without having one in biology. 2. Generalizations. The rates of many diseases, including major infectious diseases, many cancers, diabetes, asthma, and strokes are significantly dif- ferent between races in the U.S. today (Fein 1995). Blacks are seven times more likely to die of tuberculosis than whites, three times more likely to die of H.1.V.-A.1.D.S and twice as likely to die of diabetes (Hacker 1992, 225-236). The diseases are biological but the racial differences are not; How is this possible? How can a social fact, a person's being labeled 'black', cause a bodily ill? How can there be interaction between a person's race and her blood chemistry unless race is biological? No mystery. Race affects income, housing, and healthcare, and these, in turn, affect health. Stress suppresses the immune system and being black in the U.S. today is stressful (Cooper and Cooper 1986). As a result, race can enter into many statistically robust biomedical generalizations even though there are no biological races. * Epidemiologists routinely use racial categories in their research, but, in the social sciences, race seems to be the independent variable of choice, for school attainment, marital status, labor force participation, income, family size, arrest and conviction rates, residence, and political affiliation vary considerably with race in the U.S. By using this system of classifi- cation, by classifying their subjects by race, social scientists discover that 60% of all female-headed households are black while only 18% are white, or that the inmate population is 50% black and only 35% white. Good taxonomy supports significant generalizations, and, in this respect, racial categories are good (Whewell 1841, 495). Racial generalizations, while only statistical, are broad; they cover most of American life, our health, wealth, and politics. Moreover, racial differ- ences in social or economic status or in rates of disease have a common cause; they arise from racial discrimination in employment, housing, edu- cation, health care, and the criminal justice system. That is, much of the variance between the races in socioeconomic standing, as well as health and disease, is explained by past or present acts of discrimination based on race. Social scientists routinely employ racial taxonomies in their research because their subjects routinely employ them in their lives. Had we in the U.S. not divided ourselves by race, there would be no differences in income or incarceration based on race for an economist or sociologist to discover. 1. Though there are no biological races, even genetic diseases can vary with race, for racial discrimination can distribute genetic as well as environmental risk factors (e.g., high doses of nuclear radiation as well as bacilli) unequally. S630 MICHAEL ROOT We divided ourselves where nature did not, and, though race is not na- ture's category, there are now many racial generalizations to uncover; race was unearthed rather than imposed by the social and biomedical sciences, for once we divided ourselves by race, there were races for a science to observe and investigate. Race is like crime. No one would be guilty of theft had we not invented or recognized the relevant laws or drawn any property lines. However, given that we did, given our history, social scientists divide us along dis- cernible boundaries when they employ these categories in their descrip- tions or explanations of crime and punishment here; they can predict or explain changes in the rates of theft only because we, their subjects, some time ago, made theft a crime; before we passed the laws, there were no thefts to count, but once we did, thefts were measured and recorded as routinely as the rainfall. We invented race and crime, but we did not invent them out of nothing; our categories have a history and rely on prior practices and understand- ings. Our current categories of property crimes depend on an advanced economy and system of economic regulation, and our racial categories depend on active speculation about human origins and biological differ- ences between human populations. There would be no stock fraud now had we not years ago established the Securities and Exchange Commission and regulated our financial markets, and there would be no blacks or whites had there been no theories of race in the 18th and 19th centuries or arguments concerning the native varieties of the human species. As Marx wrote in the Eighteenth Brumaire, "Men do make their own history, but do not make it just as they please; they do not make it under circum- stances chosen by themselves, but under circumstances directly encoun- tered, given and transmitted from the past" (1977, 300). Social classifica- tions are historical, for they rely on a society's having adopted a particular form of organization or body of belief in the past. As a result, invented does not mean optional or arbitrary. Our categories of race are our choice rather than nature's but are not chosen as we might choose a password or select a seat on an empty bus. We divide ourselves by race as part of our speculations about human orgins, in light of an African slave trade and as part of a long-standing interest in comparing the worth of different kinds of people. Some systems of classification used in the social sciences are simply myths or illusions. In the late 19th century, psychologists like William Krafft-Ebbing used categories of sexual deviance in their research. They divided their subjects into zoophiles and zooerasts, but these categories captured no generalizations, for they did not mark any biological differ- ences between their subjects nor any differences their subjects had them- selves invented; these categories were entirely the psychologists' invention; HOW WE DIVIDE THE WORLD S631 they were nowhere to be discovered, not in nature or in their subjects' own ways of dividing the world. As a result, these categories did not capture any generalizations about the subjects' behavior but only fed Kraftt- Ebbing's imagination. Subjects must invent their categories for researchers to discover them, but not all of their inventions are worth discovering or of any use to science. Here in the U.S., we divide ourselves by race and zip code but only race says much about us. Though we divide ourselves by both, re- search in the social sciences routinely sorts us into black or white but not postal zones, for while we differ significantly in school attainment, marital status, labor force participation, arrest and conviction rates by race, we don't by zip code.' Social scientists do not simply copy their subjects' categories; they often sharpen or redraw them. In Europe in the 16th century, people divided adults from children before the experts discovered childhood there, but having discovered the line between adult and child, the experts embellished the category and made the distinction between adults and children richer and more important (Aries 1962). Experts sorted 4-12 year olds together as a single age group in order to educate and protect them. They collected and recorded facts about children and measured differences between chil- dren and adults; public officials took their findings and devised policies for controlling or educating children, and, as a result, the lines around the children grew brighter. With the construction of children, came the dis- cipline of children and the disciplines of childhood education, juvenile justice, pediatrics, and child psychology. 3. Realism. Were realism simply about generalizations, race would be real. However, the biomedical or social generalizations into which racial cate- gories enter are local. Blacks are seven times more likely to die of tuber- culosis in the U.S. but not in Great Britain. While in the U.S. 60% of all female-headed households are black today, many fewer here were black fifty years ago. The rates of marriage and disease vary with time and place, for, among other reasons, the categories do. An individual with sickle-cell dis- ease can be black in the U.S, but white in Brazil, for the category of black or white is defined differently here and there. As a result, rates of sickle-cell disease for blacks differ from place to place, in part because race does. Race does not travel. Some men who are black in New Orleans now would have been octoroons there some years ago or would be white in 2. Zip codes would matter only if they matched differences in race or income; in such a case, the social scientist would try to discover why a high percentage of residents in one postal zone were black and a low percentage in another. I owe this point to Naomi Scheman. S632 MICHAEL ROOT Brazil today. Socrates had no race in ancient Athens, though he would be a white man in Minnesota. Where R is a race, a person is R at a site only if R is used there to divide people. Because the ancient Greeks did not divide people by race, there were no races in Athens then and, of course, no dif- ferences between people in morbidity or mortality based on race there.3 Biological categories are different. Whether we are Homo sapiens, male, or R h positive does not depend on how we categorize ourselves or what we understand of b i ~ l o g y . ~ With sex and blood type, how people are classified, the category they belong to, does not depend on their self- conceptions or on whether they recognize the category, while with social classification self-conceptions are central. A group of people must divide themselves by race but not blood type in order to have Because social categories or human kinds are local, generalizations in the social sciences are spatially and temporally restricted. However, ac- cording to conventional wisdom, categories or kinds are real only if they enter into spatially and temporally unrestricted generalizations, i.e., laws of nature (de Sousa 1984, 575). By this measure, few categories in the social sciences are real. By this measure, race or marital status and gender are, in the U.S., no more real than postal zones or telephone area codes. But such a conception of realism is too narrow. Though well suited to many kinds in the natural sciences, e.g., the periodic table of elements, the conception prevents us from construing any of our talk about the social world realistically (Rubens 1989). Even if the categories routinely used in research in the social sciences are invented rather than natural, they are not all metaphysically the same. Some invented categories are more real and have more of a role in the social sciences than others, even if none enter into laws of nature. 3. The ancient Greeks, of course, divided people using categories and drew invidious distinctions as we do today with our racial categories, but our categories of difference, e.g., black and white, were not theirs. 4. David Hull has argued that biological species are not spatiotemporally unrestricted classes. On his view, species are historical entities defined by evolution, and as a result two genetically-identical organisms would belong to different species given different evolutionary histories. Thus, I am a human being here, but had I a genetically-identical twin on some other planet, he would not be. Given such a conception of species, species are more local than sex or blood type but not as local as race and, most importantly, membership does not depend on how the members divide themselves. Thus, even on the historical account of species, the category of species, unlike race, does not depend on the history of the category but only the history of the individuals within it. 5. I give the name 'internalism' to the doctrine that a category K divides a people only if they divide themselves by K. Internalism is a reasonable doctrine for the social but not the biological sciences, e.g., where K is a race rather than a species. According to the internalist, a person needn't believe that she is race R in order to be R at a site s, but unless some there sort by R, she is not an R at s. HOW WE DIVIDE THE WORLD S633 4. Norms. How is race real in the U.S. today? Though our racial categories capture some generalizations, the generalizations do not make the cate- gories real. In the social sciences, real taxonomy is less about generaliza- tions and more about regulations; we divide ourselves not by discovering our differences but by requiring ourselves to be different. What has divided blacks and whites in the U.S. so deeply and for so long is not biology but norms of racial separation and difference. Social categories, unlike those in the natural sciences, are normative. Kinds of elements or compounds in chemistry simply are, there is no way that a drop of water or bit of hydrogen ought to be. Water and hydrogen don't enter into our laws but into nature's laws. Race is different. In the U.S. in 1970, only 1.3% of all lawyers and 2.2% of all electricians were black. This statistic describes black employment because the norms used in training or hiring directed union leaders, law school officials, or em- ployers to exclude or prefer white applicants to black. However, unlike laws of nature, norms can be observed in their breach, and so in 1970 there were some blacks in each profession and trade; to say that blacks, as a rule, were not doctors, lawyers, or university profes- sors is to express a norm rather than a generalization. Some were, but, of course, not with impunity. Blacks on the "whites only" train had to strug- gle to enter and fight to remain; the message from the white riders was clear: you don't belong here. A naturally occurring category K is real if and only if K makes extrap- olations of many discoverable traits possible across all K things. Such categories are obviously well made for scientific generalization (Wiggins 1995, 220-221). Where K is a social category, extrapolation across all instances is not possible, but normalization is, for if real, K prescribes how all K things ought to be rather than how they are. As result, these cate- gories are well made for social regulation. With classification in the natural sciences, real categories sort individuals on the basis of what they are (by nature), while in the social, they sort on the basis of how, according to the subjects, they ought to be. Society's categories are different from nature's not in being less real but in being man-made. What is unique about humans, as John Dupre ex- plains, is not that they contravene but that they create a causal order (1993, 14). They create a causal order when they invent ways of dividing them- selves and, once invented, conform to their own divisions. But the order they create is local and unstable and supports, at best, only restricted generalizations, and, as a result, the predictive power of real social cate- gories is weak. Nevertheless, some have great explanatory power, for a social scientist can explain why blacks are underrepresented in some trades or professions by citing a past or present rule or regulation which says to keep blacks out. Blacks are not underrepresented because they have the S634 MICHAEL ROOT same causal powers, e.g., an inability to learn a trade or master a profes- sion but because they are subject to the same importunities. There is order in both the physical and social world but, as Jon Elster argues, the source of the order is different (1989,287). In nature, constant conjunction ensures order, but in society order is also based on norms and regulations. Societies are held together by the statements of what ought to be and not merely by what happens to be. The norms d o not ensure the constant conjunction of race and occupational segregation or child- hood and child protection, but they require it.6 In the U.S. during the Jim Crow era, the laws of racial separation acted like cement in the southern States; they kept the blacks together and the blacks and whites apart; the races were never so real as when enforced by the law. With the elimination of these laws, the divisions between the races are less sharp and our racial norms more contested or disavowed and more often flouted. But race is still real here. Even though the conjunction of race and occupation is less constant, the reasons for hiring by race remain the same.7 5. Race. Race is not biological, but in the U.S. race is biologically rather than culturally transmitted. That is, the biological offspring of two mem- bers of race R are members of R , while the adopted children of two mem- bers of R are not unless their biological parents (or at least one of their biological parents) are. We can adopt children of other races, but, given our current system of racial classification, we d o not make them members of our race by adopting them no matter how much of our culture we impress on them or how eagerly they embrace it. Our system of racial transmission rests on the false belief that there is some genetic property, a race gene or gene cluster, that individuates the races and explains a number of alleged racial differences in behavior or morphology. There aren't such genes, but, nevertheless, the practice of classifying people by race in the U.S. proceeds as if there were. Our criteria for being black or being white includes the following two biological beliefs, one false and one true. (1) People of different races have different race genes. (2) Biological parents pass their genes to their children, while adop- tive parents d o not. The practice of basing race on one's biological rather than adoptive par- 6. Because segregation required that blacks ride in the back, many blacks were disposed to sit there and many whites to force them to, but some resisted; their resistance weak- ened the causal order but strengthened the order of regulations by making the regula- tions more visible and increasing the level of their enforcement. 7. Though with affirmative action the reasons change. HOW WE DIVIDE THE WORLD S635 ents rests on (2), a truth. But the reason why (2) is taken to be relevant to race is a belief in ( I ) , a falsehood. Race in the U.S. is different from citizenship, for there are both natu- ralized and natural born citizens here but only natural born blacks or whites. So while, some residents can convert from Canadian or Haitian to American citizenship, none can convert from black to white. Some can pass as black or white, but passing is not being. A naturalized American citizen does not pass as one but is one. Race is based on invented rather than natural differences, but there is a difference between race and perceived race, for a person who passes for black or white only passes. Were being and being-perceived the same, race would not be real. That is, race would not be real if to be black was simply to be believed to be, not because race is biological but because to be real implies a contrast, a difference between being and being believed to be. Difference in health or socioeconomic standing correlate with perceived race, and were racial generalizations our only interest, the difference be- tween race and perceived race, between being and passing, would not much matter, but my interest is racial regulation, and here the difference between being and being perceived is important. For a "blacks only" rule applies to blacks, not to people who are falsely thought to be, and as a result the rule is misapplied when used to exclude whites who are mistaken as blacks. Our practice of racial sorting could change. We could conserve race and let go of biological transmission. Adopted children could be classified as the same race as their adopted rather than biological parents. In ad- dition, we could develop procedures for renouncing our present race and becoming a naturalized member of another. The change in race could proceed in steps. First, you become a resident alien in your new race, and then, upon completion of a course of study or work on behalf of your new group, you become a naturalized member.8 Though, according to the present system, your racial identity is not chosen, the reality of race depends on what we (collectively) have made of race and, in particular, whether we regulate or discipline each other by race. Should we divide but not regulate by race, we would retain the races but not conserve their reality. That is, I would still be white, but being white would be no more significant than having a 612 area code. Issues of authenticity arise only where race is real. Were we to divide ourselves by race but not prescribe how the races should differ, there would be no reason to say that some blacks or whites are more authentic 8. Race has no essence, but division by race has so long been based on biological beliefs, e.g., (1) and (2) above, that naturalized race does not look like race at all, for were race culturally transmissible, race would become indistinguishable from ethnicity. S636 MICHAEL ROOT than others. T o be authentically black, we need a norm or ideal of black- ness-a proper way to look and behave. Such comments invite contro- versy. Who decides the proper way to be black or white? There is no distinctive or uncontested black or white experience or voice, but where many agree that blacks ought to favor blacks or whites favor whites, we can speak of authenticity, for, given such agreement, to mix is to be racially inauthentic and subject to disapproval. Clarence Thomas couldn't be accused of being an Oreo-a traitor to his race-if race weren't real here. Many people here identify themselves or are identified by race. Racial categories are commonly given in response to questions, reminiscent of Aristotle's queries, "Who is he?" or "What is it?". But people have no essence or single identity; they have crisscrossing identities. Each person belongs to many prominent or encompassing groups, and each category modifies or inflects the others. I am a white man, but to be so is not simply to be a man and white. My way of being white is different from my mother's or sister's. Human kinds or social categories are not simply ad- ditive. Moreover, though there are ways for whites to be white or men to be men, we often disagree over what they are or ought to be. Race is an unruly system of classification, but there are rules, even though contested, for how blacks or whites are to look or behave. Indi- viduals who flout them are black or white inauthentically. As Anthony Appiah has written: The large collective identities that call for recognition come with no- tions of how a proper person of that kind behaves: it is not that there is one way that gays or blacks should behave, but that there are gay or black modes of behavior. These notions provide loose norms or mod- els, which play a role in shaping the life plans of those who make these collective identities central to their individual identities. (1994, 159) With systems of human classification, reality is prescriptive, and so mem- bers of a real human kind can be more or less genuine or act more or less like they should. Not every human kind enters into talk of authenticity, but the real ones do, for the forces that make them real support the dis- tinction between authentic and inauthentic members. 6. Legitimacy. In the U.S., the Bureau of the Census, Office of Manage- ment and Budget, and other federal agencies classify residents by race and collect racial statistics. The categories vary from agency to agency, but a directive of OMB, Directive 15, is meant to guide every agency, and, ac- cording to this directive, race is different from ethnicity and there are four races in the U.S. today: American Indian or Alaskan Native, Asian or HOW WE DIVIDE THE WORLD S637 Pacific Islander, Black, and White.9 The statistics gathered by these federal agencies are employed in both the biomedical and social sciences.1° Directive 15 is controversial. The four racial categories are treated as exclusive and exhaustive, and the directive spells out the categories without defining them." Moreover, the four categories d o not match the way many Americans think about race. Though there is no one way we divide our- selves by race, we often equate race and ethnicity or take Jew or Hispanic to be a racial category (Mackay and de la Puente 1996). The fact that the official categories do not match folk categories should be a special concern to social scientists, for when they classify their subjects by race, they routinely use the official categories, but the reality of those categories depends on public and not merely official recognition. If only the OMB divides Americans as black or white, then these races are illu- sions and to use them in sociology or epidemiology is not science but superstition. But the OMB's categories are not simply official inventions. Americans d o divide themselves into a black and a white race. Moreover, since the official categories enter into government policies and regulations, and many Americans conform to them, folk and official categories merge. Had the OMB's four races begun as myth, by entering into the nation's civil rights laws, system of affirmative action, housing and educational policies, and voting rights act, they would be myths no longer. By employing categories like race in their research, social scientists help to conserve them, for even if they believe that racial classification is ille- gitimate, by dividing their subjects by race, they contribute to those racial divisions. In describing criminal arrest rates by race, for example, a soci- ologist herself racializes crime, and by correlating 1Q and race, a psy- chologist deepens the line her subjects draw between blacks and whites in schooling and employment. Whether a social category is real is separate from whether it is legiti- mate. A category is legitimate when it ought to be used to sort people into 9. Hispanics, for example, are officially an ethnic group, and members can be of any race. The government's racial categories change; many of today's ethnic groups (Hindu and Mexican) were, and two of today's four races (Alaskan Native and Pacific Islander) were not, official races one hundred years ago. The OMB changed the racial categories slightly for the 2000 census. 10. The federal directives guiding birth statistics are different; they assume that race is biologically transmitted. Until 1989, only infants with two white parents were officially white, while an infant with only one white parent took the race of the parent whose race was other than white. Beginning in 1989, the infant had the race of its mother. The change increased the number of white and decreased the number of non-white infants substantially. 11. When filling out a census form, respondents are allowed to choose, but census takers sometimes decide to place the respondent into some other category. S638 MICHAEL ROOT kinds. Race is real in the U.S. today, but assimilationists argue that it ought not to be. They maintain that given the individious use of racial classification, the harms or wrongs done in the name of race, people here ought not to divide themselves by race and that a race-blind is preferable to a race-conscious society. Du Bois claimed that race was real in the U.S. in 1897, but he also argued that it ought to be conserved-that, despite America's bleak racial history, more social advancement would come with racial classification than without it. The very issue marks the difference between the categories of the social and natural sciences, for no one debates whether we should conserve the periodic table or our four blood types. Since to persist, social divisions must be continuously recognized, whether to continue to rec- ognize the races is a reasonable question. Social scientists can employ racial classification in their research and remain silent on the questions of conservation and legitimacy, but their use of race does contribute to the racial consciousness of their subjects. Were we to conclude that race should not be conserved in the United States, we would want the social sciences to stop their practice of classi- fying us by race. Informing us that 60% of all female-headed households are black, while only 18% are white, or that the inmate population is 50% black and only 35% white, increases the likelihood that the racial future will be like the past even if that is no part of a social scientist's intention. That is, the social sciences are not value-neutral even when they only say what is and not what ought to be, for where K is a real human kind, to say that most K's are F, where F is a socially or biomedically significant trait, gives a person a reason to believe that K is legitimate. 7. Conclusion. There are no biological races, but that does not answer the question whether race is real or simply a myth. Race is real in the U.S. as long as we are race-conscious here. I have offered an account of realism that is friendly to the social sciences and explains how, with social or human kinds, with kinds like race, what is real is dependent upon rather than independent of how we think and talk about ourselves. While na- ture's kinds are real when members share causal properties, social kinds are real when they are subject to the same importunities. Laws of nature d o not make race real, we do. While the social sciences can explain how or why we so divide ourselves, we need to decide whether we ought to. REFERENCES Appiah, Kwame Anthony (1994), "Identity, Authenticity, Survival: Multicultural Societies and Social Reproduction", in Amy Gutman (ed.), Multiculturalism. Princeton: Prince- ton University Press, 149-163. HOW WE DIVIDE THE WORLD S639 , (1992), "Illusions of Race", in In M y Father's House. New York: Oxford University Press, 2 8 4 6 . Aries, Philippe (1962 [1960]), Centuries of Childhood: A Social History of Family Life. New York: Alfred Knopf. Translated by Robert Bladick. Originally published as L ' E n f ~ n t et h vie familiale sous l'ancien regime. Paris: Plon. Cooper, R . S. and David R . Cooper (1986), "The Biological Concept of Race and its Ap- plication to Public Health and Epidemoloay", Journal o f Health Politics, Policy, and -. LOW 11: 97-1 16. De Sousa, Ronald 11984). "The Natural Shiftiness of Natural Kinds". Canadian Journal o f Philosophy 14: 5611581. DuprC, John (1993), The Disorder of Things. Cambridge: Harvard University Press. Du Bois, W. E. B. (1969), The Souls of Bbclc Folk. New York: Signet. . (1992), "The Conservation of Races", in Howard Brotz (ed.), African-American Social and Political Thought 1850-1920. New York: Transaction Publishers, 483492. Elster, Jon (1989), The Cement of Society: A Study ofSocial Order. New York: Cambridge University Press. Fein, Oliver (1995), "The Influence of Social Class on Health Status: American and British Research on Health Inequalities", Journal of General Internal Medicine 10: 577-586. Hacker, Andrew (1992), Tbvo Nations. New York: Ballantine Books. Hull, David (1978), "A Matter of Individuality", Philosophy of Science 45: 355-360. Mackay, Ruth B. and Manuel de la Puente (1996), "Cognitive Research on Designing the CPS Supplement on Race and Ethnicity", in Proceedings of the Annual Research Con- ference (1995), U.S. Bureau of the Census. Rosslyn VA, 326-337. Marx, Karl (1977), "The Eighteenth Brumaire of Louis Bonaparte", in David McLellan (ed.), Karl Murx: Selected Writings. New York: Oxford University Press, 300-325. Ruben, David-Hillel (1989), "Realism in the Social Sciences", in Hilary Lawson and Lisa Appignanesi (eds.), Dismantling Truth. London: Wedenfeld and Nicholson, 58-75. Whewell, William (1847), The Philosophy of the Inductive Sciences, I. London: John Parker. Wiggins, David (1995), "Substance", in A. C. Grayling (ed.), Philosophy. New York: Oxford University Press. You have printed the following article: How We Divide the World Michael Root Philosophy of Science, Vol. 67, No. 3, Supplement. Proceedings of the 1998 Biennial Meetings of the Philosophy of Science Association. Part II: Symposia Papers. (Sep., 2000), pp. S628-S639. Stable URL: http://links.jstor.org/sici?sici=0031-8248%28200009%2967%3A3%3CS628%3AHWDTW%3E2.0.CO%3B2-6 This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR. References A Matter of Individuality David L. Hull Philosophy of Science, Vol. 45, No. 3. (Sep., 1978), pp. 335-360. Stable URL: http://links.jstor.org/sici?sici=0031-8248%28197809%2945%3A3%3C335%3AAMOI%3E2.0.CO%3B2-X http://www.jstor.org LINKED CITATIONS - Page 1 of 1 - http://links.jstor.org/sici?sici=0031-8248%28200009%2967%3A3%3CS628%3AHWDTW%3E2.0.CO%3B2-6&origin=JSTOR-pdf http://links.jstor.org/sici?sici=0031-8248%28197809%2945%3A3%3C335%3AAMOI%3E2.0.CO%3B2-X&origin=JSTOR-pdf