The psychological construction of race

Recent racially charged events have brought renewed focus on racial tensions in the United States, effectively ending the myth of a postracial society. Moreover, the recalcitrance of race-relevant social problems suggests that current methods of addressing those problems are inadequate, or at least incomplete. Such is the case with the received view, social constructivism, according to which race results from historically and culturally specific practices and decision. New psychological research challenges the received view, suggesting that racial cognition results in part from psychological mechanisms that operate outside of the conscious awareness of human agents. This body of research has led to interactionist, complementarian constructivist models of race. Ron Mallon and Dan Kelly's position, "hybrid constructionism," advances the claim that "racial social roles are psychologically constrained." My position, the psychological construction of race, is committed to something more specific-that opaque psychological mechanisms are foundational to social construction, that is, that the individual psychology constructs the subsequent social constructions. In short, if hybrid constructionism is "constraint-ist," psychologically constructed race is purely constructivist.

THE PSYCHOLOGICAL CONSTRUCTION OF RACE by Michael B. Wilson A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Philosophy The University of Utah August 2017 Copyright © Michael B. Wilson 2017 All Rights Reserved The University of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The dissertation of Michael B. Wilson has been approved by the following supervisory committee members: James Tabery , Chair 8/5/2015 Date Approved Stephen Downes , Member 8/5/2015 Date Approved Matthew Haber , Member 8/5/2015 Date Approved Ronald Mallon , Member 8/5/2015 Date Approved Erin Beeghly , Member 8/5/2015 Date Approved and by Matthew Haber the Department/College/School of and by David B. Kieda, Dean of The Graduate School. , Chair/Dean of Philosophy ABSTRACT Recent racially charged events have brought renewed focus on racial tensions in the United States, effectively ending the myth of a postracial society. Moreover, the recalcitrance of race-relevant social problems suggests that current methods of addressing those problems are inadequate, or at least incomplete. Such is the case with the received view, social constructivism, according to which race results from historically and culturally specific practices and decision. New psychological research challenges the received view, suggesting that racial cognition results in part from psychological mechanisms that operate outside of the conscious awareness of human agents. This body of research has led to interactionist, complementarian constructivist models of race. Ron Mallon and Dan Kelly's position, "hybrid constructionism," advances the claim that "racial social roles are psychologically constrained." My position, the psychological construction of race, is committed to something more specific- that opaque psychological mechanisms are foundational to social construction, that is, that the individual psychology constructs the subsequent social constructions. In short, if hybrid constructionism is "constraint-ist," psychologically constructed race is purely constructivist. For Julie and Bill TABLE OF CONTENTS ABSTRACT .......................................................................................................... iii ACKNOWLEDGMENTS ......................................................................................vii Chapters 1. INTRODUCTION .............................................................................................. 1 2 THE DICHOTOMIZATION OF RACE: BIOLOGICAL OR SOCIAL .................. 10 2.1 Introduction .............................................................................. 10 2.2 The Natural Kinds Approach: Is Race "Really Real"? .............. 11 2.3 Three General Worries ............................................................ 16 2.3.1 Essentialist Natural Kinds .......................................... 20 2.3.2 Nonessentialist Natural Kinds .................................... 22 2.3.3 The Monism-Pluralism Problem ................................. 24 2.4 The "Dichotomous Framing of Alternatives" ............................ 27 2.4.1 Consequence of Dichotomization .............................. 38 2.5 How Gannett's Critique Motivates the Dissertation.................. 40 2.5.1 Some Ways to Approach the Dichotomy ................... 41 2.6 Conclusion ............................................................................... 43 3 A METHODOLOGICAL ALTERNATIVE TO SEMANTICISM........................... 46 3.1 Introduction .............................................................................. 46 3.2 The Semantic Strategy, Its Problems, and Its Rivals ............... 48 3.2.1 Michael Hardimon's Rationalist Semanticism ............ 58 3.2.2 Joshua Glasgow's Empirical Semanticism................. 65 3.3 The Evolutionary Cognitive Program ....................................... 69 3.4 Conclusion ............................................................................... 79 4 OPAQUE PSYCHOLOGY I: INNATE PSYCHOLOGY AND RACE ................. 81 4.1 Introduction .............................................................................. 81 4.2 Explanations of Racial Classification ....................................... 84 4.3 Race on the Received View: Social Science Explanations...... 86 4.4 Race on the ECP View ............................................................ 88 4.4.1 Races as Coalitions ................................................... 89 4.4.2 Races as Ethnies ....................................................... 91 4.4.3 Race as Interactive .................................................... 93 4.4.4 Common Themes for ECP ......................................... 96 4.4.5 The Verdict ................................................................ 99 4.4.6 Normative Consequences of ECP ........................... 107 4.4.7 A Challenge for ECP ................................................ 108 4.5 Conclusion ............................................................................. 113 5 OPAQUE PSYCHOLOGY II: IMPLICIT ATTITUDES AND RACE ................. 114 5.1 Introduction ............................................................................ 114 5.2 Explicit and Implicit Bias ........................................................ 114 5.2.1 Are Implicit and Explicit Attitudes Dissociable?........ 118 5.2.2 Can Implicit Attitudes Be Measured? ....................... 119 5.2.3 Do Implicit Attitudes Affect Judgment and Action? .. 124 5.3 Implicit Attitudes, Indirect Measures, and Race ..................... 128 5.3.1 Implicit Bias in Health Care ...................................... 130 5.3.2 Implicit Bias in Criminal Justice ................................ 132 5.3.3 Implicit Bias in Education ......................................... 135 5.4 Normative Consequences ..................................................... 137 5.5 Conclusion ............................................................................. 139 6 THE PSYCHOLOGICAL CONSTRUCTION OF RACE ................................. 143 6.1 Introduction ............................................................................ 143 6.2 What Psychological Constructivism About Race Is and What It Is Not ............................................................................... 148 6.2.1 PCR Is Not Semantic ............................................... 150 6.2.2 PCR Is Not (Merely) Social Constructivism.............. 152 6.3 What PCR Is .......................................................................... 156 6.3.1 The Makings of the Second Dichotomy ................... 156 6.3.2 Is an Interactionist Thesis Possible? ........................ 161 6.3.3 PCR as Framework and Nexus ............................... 166 6.3.4 PCR Versus Evolutionary Social Construction......... 176 6.3.5 Distinguishing PCR From Hybrid Constructionism ... 180 6.4 What Psychological Constructivism Does ............................. 189 6.4.1 PCR and Biomedical Contexts................................. 191 6.4.2 PCR and the U.S. Criminal Justice System ............. 195 6.4.3 PCR and Scientific Racism ...................................... 199 6.5 Conclusion ............................................................................ 204 7 CONCLUSION ............................................................................................... 206 REFERENCES ................................................................................................. 210 vi ACKNOWLEDGMENTS Many people contributed to the dissertation, and I wish that I could list them all, but I must limit my acknowledgments to those most directly responsible for its completion. First and foremost, I would like to thank my advisor, James Tabery, who contributed more things than I have space to name. The most important is that he never let me settle. I want express the deepest gratitude to the other members of my committee as well. I thank Ron Mallon, who was my primary subject matter expert and whose philosophical work served as the inspiration for my own. I thank Steven Downes for not only always challenging my ideas, but also shining light on the way forward. I thank Matt Haber for providing a model for delivering devastating objections in the most civil possible way. Finally, I thank Erin Beeghly, who, in spite of being a last-minute addition to the committee, was patient and conscientious enough to offer thoughtful comments that led to significant emendations. In addition, I want thank Cindy Stark and the rest of the graduate committee for their patience and for giving me the time I needed to finish, and all of my professors who, in one way or another, helped me become a better philosopher. This is especially true of Elijah Millgram, who challenged me philosophically like no one before. I owe a great deal of gratitude to my closest friends and colleagues in the University of Utah Philosophy Department─Monika Piotrowska, Matt Mosdell, and Jennifer Warriner─and to my closest lifelong friends─Avery Demby, Sean Trotter, Chris Pappas, and Major Mugrage. Each in her or his own way supported my efforts and made this accomplishment possible. In addition, Alissa Winkler's helpful suggestions, patience, and love inspired the final push toward completion. Lastly, I would be remiss if I did not thank my Master's Thesis advisor, Helen Heise, who set me on the good path. viii CHAPTER 1 INTRODUCTION To many, the election of Barack Obama, the first Black President of the United States, signaled the transition to a "postracial America." A National Public Radio broadcast considered the prospect of "A New, Postracial Political Era in America" (Siegel, 2008). The L.A. Times was more tempered, alluding only to "Obama's postracial promise" (Steele, 2008). A bit later, academics joined the fray. In "Reality or Rhetoric? Barack Obama and Postracial America," Bettina Love and Brandelyn Tosolt report that "Dominant discourse holds that Obama's election proved the end of racism," but argue that "an alternate view is possible...that Obama's election reveals less about the end of racism and more about the public's view of racism as a changing construct" (Love & Tosolt, 2010, p. 19).The strongest claim came from John McWhorter, a Columbia University linguistic professor, who wrote in Forbes that "in answer to the question, ‘Is America past racism against Black people,' I say the answer is yes" (McWhorter, 2008, n.p.). He continued by conceding that "nothing magically changed when Obama was declared president-elect," but then opined that our proper concern is not whether racism still exists, but whether it remains a serious problem. The election of Obama proved, as nothing else could have, that it no longer does...increasingly, alleged cases of racism are tough calls, reflecting the complexity of human affairs rather than the stark injustice of Jim Crow...So, if I 2 have to give a single answer, it is, yes, we can call ourselves a postracial country. (McWhorter, 2008, n.p.) Despite McWhorter's confidence, the following year, Michael C. Dawson and Lawrence Bobo published a paper under the grimmer sounding title, "One Year Later and the Myth of a Postracial Society" (Dawson & Bobo, 2009, p. 247). Finally, writing for politico.com, Roger Simon asks "What happened to postracial America?" and answers that it "Didn't last very long" (Simon, 2009, n.p.). Now, more than 6 years later and nearing the end of President Obama's second term, the claim of postracialism is hard to fathom. The last year and a half alone has seen a spate of high-profile, controversial cases of White police officers killing unarmed Black men, which led to emotionally-charged public protests. In July 2014, Eric Garner died after officers wrestled him to the ground while attempting to arrest him for selling cigarettes illegally. His final words, "I can't breathe," became rallying cry for civil rights activists and socially conscious professional athletes, among others (Newman, 2014, n.p.). In November 2014, police shot and killed 12-year-old Tamir Rice, who was playing with a fake pistol outside of a recreation center (Fitzsimmons, 2014). In perhaps the most famous case, police shot and killed unarmed Black teenager Mike Brown in Ferguson, Missouri, setting off a series of protests in August 2014 that resulted in confrontations with police, dozens of arrests, and a scathing U.S. Justice Department review of the Ferguson Police Department (Fitzsimmons, 2014). In addition to these and other high-profile police shootings of unarmed Black men, a chapter of a prominent fraternity was banished from a major university after a video showed members engaged in a racist chant (The 3 Associated Press, 2015). And, in June 2015, a 21-year-old White man entered a predominantly Black church in South Carolina, sat through a prayer meeting for nearly an hour, and then opened fire, killing nine church members (Workman & Kannapell, 2015). Not long after his arrest, The New York Times reported that the man had ties to a White supremacist organization (Schmidt 2015). In response, President Obama, who (through no fault of his own) had been characterized as the very symbol of postracialism in the United States, acknowledged that racism remains "deeply embedded in the United States as a ‘part of our DNA'" (Shear, 2015, n.p.). In sum, after President Obama's election, "the term ‘postracial' was everywhere" (Holmes, 2015, n.p.). And today? "Well, it has mostly disappeared from the conversation, except as sarcastic shorthand" (Holmes 2015, n.p.). An interesting question, then, is whether the claim of postracialism was justified even back in the bad old days of 2008? My answer is, no, and not just because of the way things turned out. To explain my answer, I want to further examine this claim that President Obama's election signaled the advent of postracial America. In fact, at the risk of appearing facile, I want to start by taking a look at the proposition itself: "Barack Obama is the first Black President of the United States." Of course, the proposition expresses a very interesting historical fact, but that's not my focus here. I think that the proposition is philosophically fascinating, a fact obscured by its socio-historical importance and a syntactic simplicity that obscures a trove of philosophical questions and commitments. So, let's ignore the historical significance for a while, and ask a maddeningly philosophical 4 question instead. What would make that proposition true? Well, Barack Obama would have to be the POTUS, of course, which would require that he be a natural born U.S. citizen, that he be at least 35 years old, and that he receive at least 270 Electoral College votes. Those bases are covered, so Mr. Obama is President. Furthermore, there were no Black U.S. Presidents prior to Mr. Obama, so if he is Black, he is the first Black POTUS. But, is he? Is Barack Obama Black? And, if so, what conditions make him Black? To be clear, with the first question, I do not aim to question Mr. Obama's "Black-ness" in terms of its authenticity or sufficiency, in the sense that one might question whether he is "Black enough" (Coats, 2007, n.p.). Rather, the second question captures what I am getting at-that Mr. Obama is Black is apparently uncontroversial, so what makes it uncontroversial? So, to my mind, the most philosophically interesting fact is not that Mr. Obama became the first Black President of the United States, but just that he is Black. What would it take for him to be White? Mr. Obama has one Black parent (his father, who was Kenyan) and one White parent (his mother, who is from Kansas). He was reared primarily by his mother's (White) family, and even many of his life experiences are more closely associated with White, rather than Black culture. He earned two Ivy League degrees, for example, and became President of the United States. So, again, by what criteria is Mr. Obama Black? One thing is certain, McWhorter can't be right if a "mixed-race" man can become the first Black president. Putting it that way, in 5 fact, renders the claim of postracialism absurd. If this question seems trivial, consider that, shortly after the 2013 Boston Marathon bombing, a controversy arose over the question whether the surviving perpetrator is White. An informal New York Times study found "substantial ambiguity about whether the Tsarnaev brothers were White" (Kteily & Cotterill, 2015, n.p.). Furthermore, the study suggests, that "Whiteness perception had the potential to play an important role in the outcome of Mr. Tsarnaev's trial. The lower that individuals rated Mr. Tsarnaev as looking White, the more willing they were to punish him severely" (Kteily & Cotterill, 2015, n.p.). When considering difficult cases such as these, one is tempted to begin with biology, and as I show later, the biological answer was the answer for people living in the European and American colonial period. Beginning with Francis Bernier's New Divisions of Earth by the different species or races which inhabit it, the 18th- and 19th-century view of race was dominated by the idea that human races constituted primordial, natural, discrete biological units tied to continent of ancestry. The certainty of those claims likely strikes the modern reader as scientifically and socially naïve, but connections between biology and race are deeply ingrained-evidence suggests that even young children tend to reason that, whatever else is true about race, any person's racial designation is the same as his or her biological parents' designation. This colonial era biological explanation of race is clearly inadequate, however. One reason among many is that scientific rigor led scholars away from the view of discreteness in biological taxa, let alone racial division, which is recognized as 6 too fluid to count even as a subspecies.1 Predictably, the response to the failure of the biological-anthropological reductive approach to race was excessively opposite. In biology's stead, social constructivist models came to dominate racial discourse. Social constructivist accounts of various stripes suggest race is best understood as the result of historical traditions and intentional individual and collective cultural choices. Today, the social project, just like its biological predecessor, faces challenges on several fronts. The most serious threat comes from Ron Mallon and like-minded philosophers who promote an "evolutionary-cognitive approach" to race that takes account of the contributions of innate psychological mechanisms to racial cognition. Mallon's position, which is detailed throughout the dissertation, is that the essentialist thinking that informs racial cognition results from an innate psychological mechanism that is characteristic of the human species and "specialized for solving" a relatively narrow range of problems. At minimum, Mallon's view suggests that to the extent that innate psychology and implicit bias contribute to racial cognition, historical and cultural factors cannot complete the story about race. Ironically, another threat to social constructivist explanations originates in contemporary biology. While geneticists, populationists, and biologists remain committed to the denial of the discrete human groupings of the 18 th- and 19th1 I will discuss and assess these various positions in great detail below, so extensive citations will be provided at that point. For introductory purposes, I'm only introducing the general ideas to be engaged later. The citations I have provided are from media articles whose authors do not reappear in the main body of the dissertation. 7 century "biologization" of race, they also recognize statistical, probabilistic, contingent connections between race and biology. Former director of the Human Genome Project Francis Collins, for example, argues that race and ethnicity have some "biological connection." Several intuitive concerns lead to doubt over the completeness of social constructivism, as well. One might wonder why different models of racial categorization exhibit stable, cross-cultural patterns. One might wonder why proposed solutions to racism have seen limited success, if race is contingent merely on historical and social convention. As I will show, some of the causes of the racially charged events described above are seated deeply in human nature. Moreover, the phenomenon of race has a broad reach, affecting a large number of important social contexts, including biomedicine, criminal justice, and education. Several years before Mr. Obama's historic achievement, for example, the Food and Drug Administration "took a controversial step...approving the first drug ever intended for one racial group" (Saul, 2005, n.p.): A study of 1,050 African-American heart failure patients showed that BiDil, a combination of two previously available generic drugs, "reduced deaths by 43%" (Saul, 2005, n.p.). Of course, one might wonder how the advents of the era of race-based medicine and the postracial era could coincide. In my view, the claim of postracialism was never viable, even given only the evidence available in 2008. More importantly, I don't think any current explanation of race and racial phenomena sufficiently accommodates these sorts of complicated cases. Even new, more sophisticated, nonessentialist attempts 8 to explain race biologically fall short, though they remain relevant. And, social constructivism, as I will show, supplies only part of the solution; taken alone, it is incomplete. Because of the incompleteness of traditional explanations, I attempt to combine insights from social constructivism and biology to form an interactionist model of race. In order to provide a medium for such interaction, I appeal to research on innate psychology and implicit bias to argue that race is psychologically constructed. I proceed as follows. In the second chapter, I motivate the dissertation with a discussion of a series of related problems revolving around what Lisa Gannett (2010) calls a "dichotomous framing of alternatives," according to which race is either biological or social with no allowance for a middle ground. Among other things, Gannett calls for an account of race appropriate to contexts in which social and biological factors interact. BiDil, of course, provides an illustrative example. In the third chapter, I use a series of papers by Mallon to argue that one traditional approach to conceptual problems, the semantic strategy, fails to meet our philosophical needs for addressing race. I then introduce Mallon's alternate strategy, which is based in the evolutionary-cognitive program. Chapter 4 serves to elaborate that program, and Chapter 5 centers on a related psychological research program, implicit bias. Collectively, I refer to these two programs of research as "opaque psychology" because the key point for the dissertation is that both innate psychological mechanisms and implicit bias work outside of the conscious awareness and control of the agent. Finally, in Chapter 6, I use those considerations of the relationship between psychology and racial cognition to 9 argue that the conception of race as socially constructed should be supplemented by the conception of race as psychologically constructed. In particular, I locate the psychological construction of race (PCR) within a family of related attempts to combine crucial insights from social constructivism and research on opaque psychology. Within that family, Mallon and Dan Kelly lay out a position called "hybrid constructionism," arguing that opaque psychology "constrains" social phenomena associated with race. PCR is committed to a more specific claim-that opaque psychology is fundamental to the social constructions, that is, that it provides a foundation on which social constructions are built. If Mallon and Kelly's hybrid constructionism is "constraintist," PCR is purely constructivist. Moreover, PCR is fashioned not as a definition of race, but as a framework on which context-dependent uses of race can be constructed to address pressing practical concerns. Before getting there, however, I'll need to introduce the sort of problem that motivates the project. CHAPTER 2 THE DICHOTOMIZATION OF RACE: BIOLOGICAL OR SOCIAL? 2.1 Introduction In a provocative 2010 paper, Gannett details a set of closely related problems for contemporary race scholarship. Each results from a traditional but, in this context, misguided metaphysical commitment that fosters a dichotomization of race concepts, according to which race is either a biological reality or a social construct (Gannett, 2010). Gannett argues that the dubious metaphysical assumptions that underwrite the race dichotomy "foreclose" the asking of certain questions, many of which are among the most socially and politically important. While her focus is race in biomedical contexts, Gannett's perspicuous analysis of the dichotomy and its consequences reveals a broader type of conceptual problem endemic throughout a host of racially relevant contexts. The problem runs deep-at stake are issues such as what, if anything, race is and what to do with ‘race'2 talk in biomedical and other crucial contexts. 2 In almost all instances I use single quotes, as I have here, to signify the concept and double quotes to signify the word that represents the concept. So, the word "race" might be said to represent the concept ‘race'. Some authors italicize words to signify the concept, so Glasgow (2009), for example, represents the concept as race. I will alert the reader when this convention comes up. 11 Given the importance of those sorts of questions, Gannett urges philosophers of science to work beyond the traditional question of whether race is "really real" and instead take a context-sensitive, pragmatic approach that focuses on its contingent, dynamic, and statistical nature (Gannett, 2010). A primary aim of this dissertation is to provide a vehicle for the strategic shift Gannett invites, so I now turn to the details of Gannett's paper, proceeding as follows. I begin with the cornerstone of Gannett's concerns, the tendency among philosophers of science to take one-or-another "natural kinds" approach to theorizing about race, the restrictive parameters of which dichotomizes race concepts into apparently mutually exclusive biological and social causal realms. I then turn to the dichotomy itself, offering several examples of scholars whose arguments foster it. Finally, I discuss some of the most pernicious consequences of race dichotomization and conclude with a summary of what philosophical projects might be motivated by Gannett's concern. 2.2 The Natural Kinds Approach: Is Race "Really Real"? For much of its history, scholars and layfolk alike commonly associated race with the biological conjectures of the European and American colonial period. After defeat of the Nazi regime near the end of World War II, however, the predominant paradigm of race shifted to social and cultural explanations. As sociologist Howard Winant notes, At the beginning of the 20th century, a nearly comprehensive view of the race concept still located it at the biological level. On this account, races were ‘natural': their characteristics were essential and given, immutable…[but] significant shifts in the early 20 th century…motivated the gradual but inexorable development of a 12 more sophisticated social scientific approach to race. (Winant, 2000, p. 172) Completion of the Human Genome Project (HGP), however, "provoked questions, both inside and outside the academy, about the status of race as a category of classification in biomedicine and as a biological phenomenon at the level of the genome" (Gannett, 2010, p. 364). Former head of the HGP and current National Institutes of Health director Francis Collins, for example, argues that "it is not strictly true that race or ethnicity has no biological connection" (Collins, 2004, p. S13). Moreover, successful incorporation of "population genetics approaches in relevant fields like DNA forensics and pharmaceuticals are taken as evidence for race's validity as a category of classification in biomedicine and reality as a biological phenomenon at the level of the genome" (Gannett, 2010, p. 364). Among the most prominent outcomes of these approaches is the controversial drug BiDil, the first medication approved by the U.S. Food and Drug Administration to target a condition in a particular race, namely, heart disease in African Americans. Evidence shows that BiDil is significantly effective in African Americans but not in other racial or ethnic groups; in fact, the "trial was terminated early because there was 43% relative mortality benefit" in African Americans (Taylor, et al., 2004, p. 2415). The drug's approval remains controversial, however, as opponents point to its potential to reify race at the biological level and to the commercial motivations for its production, among other things. Given the potential reach of a case like BiDil, Gannett considers how 13 philosophers of science can best contribute to race scholarship in light of renewed interest in the relationships between race and biology. She challenges the widespread tendency of philosophers of science to "take a metaphysical approach relying on theories of natural kinds" in debates about race's biological and biomedical significance (Gannett, 2010, p. 365). The questions about race on this line are familiar ones-Does race "cut nature at its joints"? Is race a real biological natural kind or merely a social construct? Gannett recognizes the intuitive appeal of this approach but raises several concerns about its utility, and, along the way, she undermines race's candidacy as natural kind, ultimately prodding philosophers to eschew "the metaphysical pursuit of the ‘really real'…to forego the ‘really real' for the ‘real'" (p. 365). This crucial claim demands elaboration. Gannett (2010) introduces the distinction between the real and the "really real" early in the paper, citing David Hull's view on the goal of systematics with regard to species: "systematists find species as the things that evolve to be the most ‘real' groups of organisms around, even if they are not real enough for some to count them as being really real" (Hull, 1998, as cited in Gannett, p. 364). While scientists are generally satisfied with the real, Gannett says, philosophers of science often are compelled to go farther, exploring the really real. With regard to the category "species," for example, working biologists gather data, make informed generalizations, and even adjust the parameters and definitions of "species," all without excessive worry about whether the category really cuts nature at its joints. In contrast, philosophers of science "assume [they] 14 best contribute to debates about genetics and race by providing or withholding assent to the legitimacy of biological race concepts by appeal to what is ‘really real'" (Gannett, 2010, p. 365). If Gannett's distinction remains obscure, the following two examples should help. I use the first to clarify the distinction itself, the second to spell out consequences of failing to heed it. First, Charles Sanders Peirce (1955) defends a pragmatic conception of the concept ‘force' that (implicitly) illustrates the conceptual distinction between the real and really real. In a recent admired work on Analytic Mechanics it is stated that we understand precisely the effect of force, but what force is itself we do not understand! This is simply a contradiction…if we know what the effects of force are, we are acquainted with every fact which is implied in saying that a force exists, and there is nothing more to know. (pp. 35-36) The point of supplying the example is not to defend Peirce's pragmatic method for defining scientific terms, the truth of which has little bearing on my thesis. Rather, I want to make clear that while his opponent appeals to the "really real" nature of force, Peirce argues that the proper metaphysical level of investigation of force is the "merely" real. In most contexts, so long as we can measure force for predictive purposes, its deep metaphysical nature is irrelevant. Gannett's point is made clearer by analogy. For many racialized contexts, the proper level of philosophical investigation is the "merely" real, or what is real enough to be of practical concern. Her complaint, then, is that those philosophers of science who entrench themselves in approaches that explore the deepest metaphysical nature of race miss, ignore, or even preclude questions crucial to social and political concerns. In short, whether really real or not, race is 15 at minimum an important social category with real-world effects, but Gannett charges that the natural kinds approach blinds us to many of them. As the second example will show, the risks of hypercommitment to deep metaphysics are often obvious. To take a simple (and admittedly absurd) example, when a train is hurtling toward your stalled car, you try to restart the car or you abandon it; you do not want to waste time in consideration of whether the train is mind-independent or not. Doing so is reminiscent of the mistake made by the victim in the classic Buddhist parable of the poisoned arrow. In the parable, a man is wounded with an arrow. His friends and kinsmen summon a surgeon to remove the arrow and restore his health, but the man, apparently a devout metaphysician, refuses to have the arrow removed until he knows who shot it, whether he used a longbow or crossbow, whether it was made of bamboo or some other material and so on. Gannett recapitulates the Buddha's message: Commitment to metaphysical speculation can come at the expense of more immediate practical problems. Gannett's point, then, is that when philosophers focus on race's deep metaphysical status, they make the mistake of the arrow victim-suspending important practical matters in favor of abstruse philosophical ones that may not even have principled solutions. Stated rather uncharitably, it is as though in a racialized context one were to say, "I know it appears that African Americans receive unfair treatment in the U.S. judicial system, but we can't do anything about that now. We haven't yet determined whether ‘African American' names a real category!" These examples seem fatuous, but they represent a common 16 philosophical conundrum: While metaphysical speculation is likely important for many reasons, over-commitment to deep metaphysical pursuits can lead to disastrous consequences-one might get hit by a train or die from an arrow wound or fail to take action against racism. Returning to Gannett, the specific worry is that the natural kinds approach to race precludes the asking of relevant social and political-that is, practical-questions, many of which are among the most important for our social aims. Even if race is not really real, Gannett might say, its effects are plenty real and merit the best attention and contributions philosophers of science can give to it. When debates turn on race's candidacy as a natural kind, however, the argument space is dichotomized into biological and social realms of explanation, seriously circumscribing the potential contributions of each. A bit more explanation should clarify that claim. 2.3 Three General Worries As Gannett acknowledges, the phrase "natural kinds approach" names not a single, unified strategy but a set of related (sometimes complementary, sometimes incompatible) metaphysical strategies. In the next section, I detail Gannett's analysis of several particular natural kinds approaches, but I first prime that topic with her three general concerns regarding natural kinds approaches to race. First, Gannett (2010) argues that any natural kinds approach to race "incorporates assumptions that structure the ways in which questions about genetics and race are asked, and…restrict the questions that are asked and the answers that are possible" (p. 365). Crucially, Gannett implicates not only 17 natural kinds proponents who use it to explain race, but also opponents who use it to explain race away-many biologically-minded philosophers of race defend natural kinds explanations, of course, but social constructivists3 often invoke natural kinds reasoning, as well, to argue against race's reality. A common social constructivist argumentative strategy, an example of which appears later in the chapter, goes like this: Race is either biological or social. In order for race to be biological, it would have to be a biological natural kind. If it is not a biological natural kind, then biology has little or nothing to do with race. Race is not a natural kind. Therefore, race is a social phenomenon. The worry, then, is not merely that philosophers of biology who take a natural kinds approach might be wrong about race's reality. It is rather that the question itself structures the parameters of race debates and limits the range of possible answers. As such, it fuels the next of Gannett's general concerns, the dichotomization of race. Gannett's second general worry is that, among philosophers of science and media members alike, theoretical commitment to race's candidacy as a natural kind "tends to involve a dichotomous framing of alternatives: race is either socially constructed or biological reality, a fiction like phlogiston or a genuine natural kind, a merely linguistic or a projectible predicate" (Gannett, 2010, p. 364). In short, debates involving race too often turn on whether race is a mindindependent constituent of the biological world or an illusion created through classification schemes manufactured to make sense of (and often to exploit) Authors vary in preference between the terms "constructionist" and "constructivist," so the use of both terms is unavoidable in the dissertation. I use "constructivist" in my own writing but preserve the particular author's preference in quotes and other references. 3 18 phenotypic difference among human populations. Many scholars and media members interested in race entrench themselves on one side or the other of the metaphysical debate. The focus on race's status as a natural kind leads to a rivalry over the argumentative structure of race research, resulting in two apparently mutually exclusive sides-biological realism and social constructivism. This strategic rivalry leads to Gannett's final worry. Gannett's third general concern is that the dichotomization at the conceptual level leads to undesirable consequences at the practical level. The natural kinds approach structures the conceptual space, fueling the dichotomization of race and leaving "out many other worthwhile questions." It is not simply that asking some questions inevitably leaves other questions unasked; rather, the traditional assumptions about natural kinds (‘the really real') philosophers of science are likely to call on for guidance in determining whether race is socially constructed or biological reality actually foreclose the asking of certain questions, specifically those that matter most socially and politically. (Gannett, 2010, p. 365) Among the most crucial precluded questions are those involved in "the very context-specific ways in which biological and social factors interact" (Gannett, 2010, p. 375, emphasis added). Examples of those contexts are not difficult to imagine-many biomedical contexts, such as the ones centered on the aforementioned heart medication, BiDil, demand attention to both biological and social factors. More generally, Gannett (2010) notes that "from the ‘biological' perspective socio-cultural differences…structure the distribution of genetic variants in space and time" and from "the ‘social' perspective, it should not be ignored that race is socially constructed by enlisting biological differences and 19 investing these with socio-cultural meaning" (p. 375). Although Gannett concedes that some research questions might be appropriately confined to one or the other side of the dichotomy, she argues that "there are also research questions that cannot be satisfactorily addressed unless allowance is made for the causal interactions that occur among biological and social factors" (p. 370). According to Gannett (2010), then, the problem facing philosophers of science interested in race is this: The approach that generally structures their debates about race comes with undesirable consequences. The most important philosophical consequence for this dissertation is that it generates a rivalrous dichotomy that "compromises the critical insight philosophers of science might otherwise contribute to the debate in the public sphere" (p. 365). The challenge, then, is to recommend a strategy by which philosophers of science can contribute to race matters that have practical import. So, broadly speaking, Gannett's analysis reveals a problematic conceptual dichotomy generated by dubious metaphysical framing, a situation that engenders many philosophical and practical problems involving race. As the stakes of the debates increase, disputants tend to identify with either the biological or the social side, which serves to reinforce the rigid dichotomization of conceptual alternatives and preclude exploration of contexts in which biological and social factors might interact. Though it may seem counterintuitive, hypercommitment to the metaphysical status of race is detrimental to race scholarship. In short, when philosophers of science obsess over whether race is really real, they limit their potential to contribute to more important 20 matters. Again, I offer some examples to clarify these points. 2.3.1 Essentialist Natural Kinds Each of the natural kinds approaches Gannett (2010) critiques belongs to one of two broad types: "natural kinds as essentialist" and "natural kinds as biologically real" (p. 365 ff.). According to the essentialist natural kinds approach "membership in a kind is based on properties of individuals, usually assumed to be intrinsic, which serve as necessary and/or sufficient conditions for defining natural kind terms" (p. 365). Today, few philosophers adopt this approach in support of race's biological reality. Instead, as a rule, "this approach has been used by philosophers to argue that races are not natural kinds" (p. 365). Gannett names two: Naomi Zack (2003), who argues that no necessary and/or sufficient "racial characteristics, or genes for such a characteristic" are shared by every member of a proposed race, and Michael Root (2003), who argues that contemporary biologists' general rejection of essentialist assumptions means that race in particular cannot be an essentialist natural kind (Gannett, 2010). Several other influential philosophers of race make similar claims. In fact, this strategic use of the natural kinds approach is common enough that in his Stanford Encyclopedia of Philosophy entry "Race," Michael James notes that the "ambiguities and confusion associated with determining the boundaries of racial categories have over time provoked a widespread scholarly consensus that discrete or essentialist races are…not biologically real" (James, 2012, n.p.). For example, Kwame Anthony Appiah (2006) says that "current biology, even after the genome project, is very unlikely to endorse race-like 21 categories that are essentialized" (p. 379). Ron Mallon (2007), a central figure in the chapters to follow, argues that "nonessentialism is likely true of race and gender" and points out that essentialism is not a popular view for the category species, let alone for race. "Philosophers of biology more or less universally reject the view that species are to be characterized by reference to an underlying essential property…importantly, however, the reasoning offered for being antiessentialist about race and about species is the same" (pp. 156, 158). So, as Mallon (2006) says in an earlier paper, there is an "ontological consensus" against a view called "racialism-the view that there are racial essences" (p. 528). But, even though few if any philosophers invoke essentialism in their accounts of race, as later chapters reveal, the problem of racialism is not easily dispatched. Even if biological essentialism is dead, its ghost haunts the racializing human mind. That crucial problem is a topic for later chapters; for now, I proceed to the second natural kinds approach to race. Clearly, the virtual extinction of essentialist biology does not entail the extinction of the species concept, which remains a robust and informative biological category. Analogously, the failure of racialism alone does not entail the end of speculation about ‘race' as a biological category. In fact, Gannett (2010) points out that several "philosophers have recently defended the biological reality of race in ways consistent with the modern evolutionary synthesis, by treating biological races in Homo sapiens as a kind of human population" (p. 367). These realist, but nonessentialist, biological accounts of natural kinds enjoy two primary advantages over essentialist ones: They are 22 "better informed by scientific practice" and "less metaphysically invested than essentialist and reductionistic ones" (p. 367). Despite those advantages, however, Gannett argues that nonessentialist biological accounts fare little better than their counterparts, since they discard the "really real" only to a limited extent. As a result, they dead-end in virtually the same place as their essentialist predecessors-a conceptual space from which certain important social questions cannot be addressed. To secure the argument, Gannett (2010) appeals to three realist positions on race that discard essentialism only to re-raise the once deflated metaphysical stakes in two specific ways: via "the monism-pluralism debate and dichotomization of the biological and the social" (p. 368). The latter, in particular, motivates much of the present work, so I give it considerable attention below, but first I summarize Gannett's description of the three biological realist positions on race and their relevance to the monism-pluralism debate about race. 2.3.2 Nonessentialist, Biologically Real Natural Kinds To begin, Gannett (2010) describes Robin Andreasen's cladistic account, which identifies races as clades, or "monophyletic groups; they are ancestordescendant sequences of breeding populations, or groups of such sequences, that share a common origin" (Andreasen, 1998, as cited in Gannett, p. 367). So Andreasen accounts for race by appeal to common ancestry of breeding populations. Philip Kitcher does something similar. As Gannett reports, Kitcher's (1999) notion of race is "that of an inbred lineage, where the inbreeding may initially have resulted from geographical isolation that eventually gave rise to 23 differences in phenotype and to some interference in free interbreeding, even when the geographical isolation is overcome" (Kitcher, 2007, 296). Kitcher conceives of race as a biological natural kind that results from division of a species into geographically isolated founder populations, and because of the isolation, "distinctive phenotypic traits arise and are transmitted from one generation to another" (Gannett, 2010, 367). For both Andreasen and Kitcher, then, the biological reality of race is defensible by appeal to common ancestry. In contrast, according to Massimo Pigliucci and Jonathan Kaplan's "ecotype" account, races "are local populations adapted to particular environments which differ genetically in many or only a few genes" (Gannett, 2010, 368). Gannett notes that this view does not make appeal to ancestral relations, instead explaining a shared racial trait, such as skin color, as indicative of "selective pressure, not common ancestry" (Gannett, 2010, 368). These distinctions are not crucial here, however. Instead, the examples speak to the diversity of opinions on race within biology and, more importantly, show that (at least some) philosophers of biology offer nonessentialist but nevertheless natural kinds explanations of race. Neither type of natural kinds explanation-neither the two "phylogenetic" nor the "ecological" one-pacifies Gannett's concerns, however. The point of offering these examples, then, is not to analyze their theoretical and strategic distinctions, but to show how, despite their rejection of essentialism, these biological approaches nevertheless result in common sets of undesirable consequences. Even with the advantage of their closer affiliation 24 and greater consistency with modern biological practice, nonessentialist biological realism bogs race down in metaphysical quandaries, once again circumscribing the potential contribution of philosophy of science to race issues. As Gannett says, in the ensuing debates, the "‘really real' is not discarded for long-once a biological basis for race is identified…the metaphysical stakes are raised yet again" (Gannett, 2010, 368). So, despite some initial promise, nonessentialist biological realism fares little better in opening the intellectual space for philosophers of science to contribute to issues involving race. Gannett shows this in the two aforementioned ways: via the monism-pluralism debate and via the dichotomization of race into distinct biological and social causal realms. 2.3.3 The Monism-Pluralism Problem Given three (or more) distinct biological race concepts, disagreements about which gets race right are "inevitable." Moreover, says Gannett, these debates are bound to "be shaped by metaphysical assumptions about monism vs. pluralism" (Gannett, 2010, 368). The contested question becomes whether ‘race', like ‘species', admits of "competing definitions…appropriate for use in different areas of biology…or if instead there is one basic or authoritative race concept to which others are reducible" (Gannett, 2010, 368). The source of the potential problems, then, is that one's prior metaphysical commitment to either monism or pluralism will structure the way one conceives of race: If one is monist, one will be inclined toward monist definitions, and vice versa. But, once again, circumscribing one's explanation of race in that way means leaving out important issues in particular contexts. Returning to Gannett's examples clarifies 25 the point. Although early on Kitcher (1984) defends a position on ‘species' called "pluralistic realism," by the time of the 1999 paper, he "admits of only a single race concept," the aforementioned "inbred' lineages" which are constituted by a "curiously gerrymandered" assortment of criteria, namely, "common descent, geographic isolation, and distinctive phenotype" (Gannett, 2010, p. 368). To be sure, his view is more pliant than outdated essentialist ones, but Gannett charges that it "appears to satisfy common sense intuitions rather than theoretical demands" (Gannett, 2010, p. 368). Taking that route means ignoring, among other things, "the genetic race concept which requires only genetic, not phenotypic, differences among groups" (Gannett, 2010, p. 368). Kitcher is rendered silent here, and Andreasen, who also "appears to be a monist about race," fares equally badly across many contexts. Andreasen champions genealogical definitions "because genealogy is used by systematists to define species and higher taxa" and because "when scientists debate the reality of race…they do so under the banner of systematics" (Gannett, 2010, p. 368). Gannett is dubious: She notes, for example, that a 2004 workshop resulting in the special Nature Genetics supplement, "Human Genome Variations and ‘Race,'" was bereft of systematists. The practical consequence for Kitcher and Andreasen is that they set self-imposed limits on their potential influence in weightier racial matters. Gannett says that "in order to address questions concerning race's validity as a biomedical category and its biological reality at the level of the genome, philosophers of biology need to familiarize 26 themselves with what race concepts are in actual use in various areas of the biological and biomedical sciences" (p. 369). In contrast, Pigliucci and Kaplan "attend closely to ways in which biologists use the term ‘race' in practice and therefore do not forego the ‘real' of biology for the ‘really real' of metaphysics" (Gannett, 2010, p. 369). The pluralism that informs their ecotypes approach avoids the mistakes of the phylogenetic ones, but predictably brings its own baggage. Ultimately, Pigliucci and Kaplan, too, reraise the metaphysical stakes because of their "implicit naturalistic assumption…that nonhuman biology exhausts the range of scientific race concepts that could legitimately apply to humans" (Gannett, 2010, p. 368). So, despite their differences, each of these race scholars "are careful to distinguish their projects in philosophy of biology from the projects of social scientists," and, as a result, each "contributes to the dichotomization of biological and social causation" (Gannett, 2010, p. 368). Since the dichotomization of race into biological and social causal realms and the consequences that follow from it motivate my aims, I discuss it in great detail in the next section. I begin with Gannett's description of a debate between Andreasen and Joshua Glasgow, one she takes as paradigmatic of the troublesome dichotomization. One should not, however, be given the impression that the race dichotomy is merely special problem for two rival philosophers, so, at the end of this section, I briefly discuss several other examples that reveal how commonly scholars, scientist, media representatives, and even we, the folk, reinforce the dichotomy. 27 2.4 The "Dichotomous Framing of Alternatives" The "corrosive standoff," as Gannett calls it, between Andreasen (1998, 2005) and Glasgow (2003, 2009), who promotes a social constructivist definition of race, is representative of the sterile antagonism among philosophers of race. For her part, Andreasen aims to "challenge the trend to reject the biological reality of race by arguing that cladism (a school of classification that individuates taxa by appeal to common ancestry) provides a new way to define race biologically" (Andreasen, 1998a, p. S653). She begins by recognizing the two main types of explanations of race-"biological realism and social constructivism"-which are generally considered "incompatible views about race" (Andreasen, 2000, p. S654). Reaffirming the position in a later paper, Andreasen says "‘race' is ambiguous between its scientific and [common sense4] meanings (with reasonable overlap between the two)," but "these meanings are relatively autonomous" (Andreasen, 2005, p. 104). "Today," she concedes, "most theorists favor the view that races are social constructs" (Andreasen, 1998a, p. S654). The mass conceptual migration toward social constructivism was motivated by the failure of earlier attempts to establish the biological reality of race, but Andreasen argues that is because her predecessors relied on "phenetic" classification, that is, they attempted to define taxa-including, human race-on the basis of observable similarities and differences, not on "genealogical relations among organisms" (Andreasen, 1998a, p. Although Andreasen here calls the rival to her cladistic definition a "common sense" definition of race, the latter is what is supposed to be captured by Glasgow's constructivist definition, so "common sense" and "constructivist/constructionist" are functionally synonymous in the context of this debate. 4 28 S656). Phylogeneticists, such as Andreasen, do not ignore observable characteristics, but pheneticists use "similarity to define its taxa," while "phylogenetic concept…uses similarity as evidence for group membership" (Andreasen, 1998a, p. S656). The basis of phylogenetic classification, then, is evolutionary history, not characteristic similarity, although the latter may inform the former. Phylogenetic classification, with its greater predictive power, has virtually replaced phenetic classification for systematists, but oddly not always when racial classification is concerned. Andreasen says that many biological theorists approach race as though "similarity ought to be the foundation of an objective classification scheme without considering the possibility that race can be defined historically" (Andreasen, 1998a, p. S656). So, the point is that biological approaches to race need an update. "Races can be defined in the way that cladistics determine its taxa, as sets of lineages that share a common origin" (Andreasen, 1998a, p. S655). As a result, "Contrary to popular belief, there is a biologically objective way to define race," but "races, if they exist objectively, ought to be defined historically" (Andreasen, 1998a, p. S657). In sum, Andreasen argues that if attention is paid to the correct sets of facts-ones of ancestry, not observable similarity-one can give a viable definition of human race as a biological reality. Doing so, however, generates surprising and controversial consequences. The racial categories Andreasen endorses, for example, do not match recognizable folk categories. "People standardly divide humans into three (or more) major races-Africans, Caucasians, Asians. The cladistics concept of 29 race, however, results in racial categories that cross-classify these standard groupings" (Andreasen, 1998b, p. 212). Her cladistic model, for example, delineates nine races, but the folk racial classification ‘Asian' is not among them-"‘Asian' is not a cladistic race" (Andreasen, 1998b, p. 212). So there is a mismatch between folk and cladistic races, and this, at least for Glasgow, is where the trouble starts. Glasgow attempts to undermine the cladistic and other "populationist" approaches in order to defend his "reconstructionist" definition of race (Glasgow, 2003, 2009). Glasgow (2003) "questions the viability" of populationist approaches, specifically taking on Andreasen's cladistic race concept and, to a lesser extent, Kitcher's inbred lineages. In particular, he objects to their dedication to meeting scientific demands without regard to common sense conceptions of race. Glasgow (2003) complains, for example, that "the folk notion of race does not normally contain the nine races identified" by Andreasen's account (p. 458). Mismatches between folk and biological conceptions of race over issues such as the number of races and the identification of ‘Asian' as a folk, but not cladistic, category prompt Glasgow (2003) to ask, "How revisionist can one be about the meaning of 'race' and still call it ‘race'?" (p. 462). For Glasgow, the rejection of Andreasen's cladistic definition goes hand-in-hand with prioritization of the folk race concept, which, for him, is fundamental. Glasgow's (2003) view, then, is that in order to count as theory of race at all, a biological account must "give a biological backing to our [folk] race talk" (p. 458). We start, in other words, with the races that are recognized and sustained by the folk and then ask what 30 biology can say about those races. Andreasen (2005) counters that the "cladistic concept falls outside the race constructivist's appropriate domain of inquiry" and suggests that the "disagreement between Glasgow and myself is over how meaning gets settled in the first place" (p. 102). Glasgow, she says, "defends the authority of [the common sense, or folk, concept of race] and argues that scientists are not the arbiters of the meaning of ‘race'" (Andreasen, 2005, p. 102). Andreasen(2005) responds that deviation from the common sense, or folk, concept of race provides no reason to reject the cladistic one (p. 102). After all, scientific and folk definitions are often misaligned-the folks' continuing to define whales as fish would not affect its scientific classification as a mammal. The Andreasen-Glasgow debate largely turns on what concept of race (if any) should be privileged. That type of disagreement is common to philosophy, but this case is distinguished by its strange results. Gannett (2010) reports that "Andreasen (2000) uses [Hilary] Putnam's causal theory of reference to defend the autonomy and authority of the biological race concept" (p. 372). For quick reference, philosophers Hilary Putnam and Saul Kripke famously argued that the microstructural properties of a thing are what determine its natural kind-ship- water's being H20 is often taken to be paradigmatic. In that vein, Andreasen (2000) claims the "objectivity of a kind, biological or otherwise, is not called into question by the fact that ordinary people have mistaken beliefs about the nature of that kind" (p. 662). Ironically, Glasgow also appeals to the Putnam-Kripke model, albeit to a different aspect of it. 31 Borrowing the language of Kripke, Glasgow charges that the cladistic approach "requires de-rigidifying ‘race'. On [Andreasen's] approach, ‘race' no longer picks out the same macrophysical objects (say, the three major races)…Rather, Andreasen's approach picks out different objects entirely" (Glasgow, 2003, 468, emphasis added). Let me quickly unpack that claim. Glasgow's reference is to Kripke's explanation of a "rigid designator," which is a term that picks out the same object in all possible worlds in which it exists. On this account, "water" is a rigid designator for the chemical compound H2O. This means that a differently constituted substance, even if it is otherwise identical to water, is not water-water just is H20. So, Glasgow's use of "derigidification" is not just dramatic, but literal: He charges Andreasen with changing the subject. Some quick elaboration on Glasgow's general position will make this charge clear. Glasgow (2009) calls his approach "reconstructionism," which he describes as a normative position that advocates neither complete elimination nor conservation of status quo racial discourse. His "basic idea" is that "we should replace racial discourse with a nearby discourse" that shifts the conversation from (pseudo) biological categories to "wholly social categories" (pp. 2-3). Of course, if he wants to revise race concepts, Glasgow (2009) must allow that they could have been and likely will continue to be revised in light of shifts in values and refinements in science and philosophy, but Glasgow (2009) thinks that Andreasen and other philosophers of biology shift race so far from its original folk meaning that they effectively change the subject. Again, the worry is 32 that a switch to cladistic race precludes talking about important social issues. If ‘Asian' is not a cladistic race, for example, then cladistics is rendered silent over social issues that result in oppression of Asian Americans, for example. If that is right, one of Gannett's worries is realized: Here is a case of a natural kinds commitment compromising the potential contributions of a philosopher of science to normative race issues. Worse still, Glasgow (2003) reacts by digging in his heels. The cladistic approach, he says, not only changes the subject, but also reinforces the wrong idea about race-that it is biological, not social (p. 462). Now, Gannett's general worry is realized: At this point in the debate, the dichotomy is manifest. Still, the debate escalates. Andreasen's response to Glasgow is that that "de-rigidification" is not an issue; she "expects that ordinary usage, when mistaken, may be corrected to conform to scientific usage (like we now accept that whales are mammals, not fish)" (Gannett, 2010, p. 372). In other words, when we discover that whales are mammals and not fish, we do not change the object of discussion ("That large aquatic animal…"); we just get better at talking about it ("…is a mammal, not a fish"). Furthermore, she suggests Glasgow's claim for the primacy of the common sense concept is arbitrary. "It is likely that we do not know enough about the history of ‘race' to know what was in the minds of speakers during the baptismal procedure" (Andreasen, 2005, p. 103). For his part, Glasgow (2003) clearly wants to ensure that the "right" racial information gets tracked in order that those who are racially oppressed be correctly counted. However, from a philosophical standpoint, Andreasen's objections leave their mark. Glasgow 33 (2003) does appear presumptive in staking his claim for the priority of folk race. And even if he is right about the "baptismal moment," all parties agree that ‘race' is a dynamic, fluid concept, one that we should not expect to corral by demanding its rigid designation. Despite their best initial intentions, these scholars appear a bit derailed. Remember, Gannett is concerned not with which side is right but with the consequences of the assumptions and structure of the debate itself. She appears right about at least this-as the combatants entrench themselves more deeply into their own positions, the metaphysical stakes are heightened and the important issues obscured. Initially, Andreasen hoped only "to challenge the trend to reject the biological reality of race," but in the end, she pushes race from the social realm. Glasgow resists, aiming at a wholly social position in the face of renewed scientific interest in the possible connections between biology and race. So, as racial tension increases in the United States and around the world, and as problems multiply and intensify, Andreasen and Glasgow mire themselves in metaphysical questions about microstructural properties and derigidification! Despite their connection to the highly influential work of Putnam and Kripke, talk of baptismal procedures and even microstructural properties ought to raise suspicion in the context of race. In this case, the harm is evident- the use of these hallowed philosophical concepts fosters an "unproductive, even corrosive, standoff" between Andreasen and Glasgow that clearly reveals the race dichotomy and its consequences (p. 365). The details of the Andreasen-Glasgow debate are instructive. I have 34 tracked Gannett's appraisal of their standoff because each directly addresses the other's argument, which makes it easy to spot some of the consequences of raising the metaphysical stakes over the reality of race. Through Gannett's narrative lens, one sees the dichotomy develop organically, revealing that Gannett's poker allusion is apt-each scholar raises the other at different points in the game, and once the argument is mired in muddied disputes over baptismal procedures, it looks like both are bluffing. This is not just a local problem, however. While few of the arguments that sustain the dichotomy are as conveniently direct as this one, many scholars and media representatives promote conclusions that also (directly or indirectly) sustain the dichotomy. I briefly survey a few of them before discussing some further consequences of dichotomization in the next section. Assumptions and arguments that sustain the dichotomy are relatively common to race debates. In spite of the growing recognition of the complexity of race concepts, in one way or another, many philosophers reinforce the wall between biological and social accounts. Andreasen, for example, does not merely fuel her side of the standoff; she proffers descriptions of "relatively autonomous" biological and social realms in which distinct, perhaps mutually exclusive, race concepts operate. In other words, she advocates dichotomization, and this move is not uncommon among philosophers of race. Michael Hardimon (2013b), a key figure in future chapters, does the same, arguing that "Confusions about the place of race in medicine result in part from failure to recognize the plurality of race concepts" (p. 6). In a pair of 2012 35 papers, he focuses on two-"socialrace" (Hardimon, 2013b) and what he calls "the minimalist phenomenon of biological race," which he virtually equates with populationist accounts (Hardimon, 2013a, p. 249). As I discuss in outline below and in detail in the next chapter, Hardimon, like Andreasen, thinks these two types of race concepts are relatively autonomous and so assigns them different roles in addressing the aforementioned "confusions." Most philosophers interested in race, however, locate themselves on one side or the other. In their seminal book on racism, Racial Formation in the United States: From the 1960s to the 1980s, for example, the aforementioned Winant and Michael Omi (1986) introduce "racial formation theory," a patently social conception of race. They replace biological conceptions of race, which they consider little more than anachronistic, with the view that "the racial order is organized and enforced by the continuity and reciprocity between micro-level and macro-level of social relations" (p. 67). Race, then, is not a biological phenomenon, but an "unstable and ‘de-centered' complex of social meaning constantly being transformed by political struggle" (Omi & Winant, 1986, 68). In part owing to the influence of Omi and Winant, Paul Taylor (2000, 2004) defends a view on race he calls "radical constructionism," which forcefully rejects any biological account. In fact, when Taylor (2004) mentions biology at all, it is only to point out the "failure of racial biology" and promote the thesis that "races are social constructs. They are things that we humans create in the transactions that define social life" (p. 86). Similarly, in Philosophy of Science and Race, Naomi Zack (2002) 36 concludes that "Essences, geography, phenotypes, genotypes, and genealogy are the only known candidates for physical scientific bases of life. Each fails. Therefore, there is no physical scientific basis for the social racial taxonomy," so race "is not biologically real as most people still think" (pp. 88, 111). Instead, it "is a construction requiring constant sorting and identification"; it "is a dynamic ongoing, performative process" (p. 115). Charles Mills (1998) argues that there is no "conceptual room for the notion of race as ‘deep' and ‘metaphysical,'" concluding that "race as biology, race as destiny, has been discredited" (p. xiv). Race, he says, is "socio-political rather than biological. The cliché that has come to express this insight is that race is not natural but ‘constructed'" (Mills, 1998, p. 86). Despite the confident tone conveyed by some of these scholars, however, the conception of race as a completely social category has never been entirely secured. While instances are rarer, some scholars have used the renewed interest in biological connections of race to argue for the biological reality of race. Of course, I talked above about Gannett's description of three biological realists. Andreasen, Kitcher, and Pigliucci and Kaplan each promote theories of race as a biological reality, all the while trying to circumvent the conclusion of the scientific racism of prior generations. Other scientifically minded scholars, however, argue for both the biologically reality of race and some of its racist implications. The most famous case involves The Bell Curve, the most controversial part of which centers on authors Richard Herrnstein and Charles Murray's conclusions that Blacks are on average less intelligent than Whites and 37 that the reason for the difference is more likely genetic rather than social (Herrnstein & Murray, 1994). Along the same lines, Neven Sesardic (2010) argues that, contrary to popular expert opinion, traditional racial categories do match what is in the biological world. The "way ‘race' was defined by biologists several decades ago (by [Theodosius] Dobzhansky and others) is in no way discredited by conceptual criticisms that are now fashionable and widely regarded as cogent" (p. 143). In Making Sense of Heritability, Sesardic employs an argumentative strategy similar to Herrnstein and Jensen's to suggest that variation in average IQ among distinct racial groups is primarily attributable to biological difference. Gannett reports that Vincent Sarich and Frank Miele (2004) make similar claims. In fact, Gannett (2010) says that her description of the dichotomous framing of alternatives "mirrors Sarich and Miele's dichotomy of race as ‘a mere social construct' or ‘an underlying biological reality'" (p. 382). Their central claim is "that DNA data gleaned from the ‘latest genetic technologies' provide decisive evidence for the biological reality of race" from which they conclude that "significant cognitive differences arising during the past 10,000 years…makes it inevitable that racial groups will be disproportionately represented at the extremes of values for traits like being a criminal or having a high-paying job" (Gannett, 2010, p. 381). Gannett finds these conclusions "horrid," but invokes them in order to reveal their role in sustaining the dichotomy, the effects of which, of course, are not limited to the realm of scientists and scholars. 38 2.4.1 Consequence of Dichotomization The dichotomy is created in the academy, but the effects extend its boundaries. One duty of the news media, for example, is to report provocative results of scientific and scholarly studies for consumption by the public. Because of the technical nature of many studies, however, media reports often misrepresent scientific conclusions. Gannett (2010) cites the case of New York Times science reporter Nicholas Wade, who following the New England Journal of Medicine publication of the BiDil results asked, "Is there a biological basis for race? If there is not, as many social scientists and others argue, how can a drug like BiDil work so well in one race?" (Wade, 2004, as cited in Gannett, p. 364). Among Wade's errors, which he repeats in his 2013 book, A Troublesome Inheritance, is that his "focus on the question of the biological reality vs. social construction of race leaves many other more interesting and socially and politically important questions unasked" (Gannett, 2010, p. 364). For example, it ignores Johnathan Kaplan's charge that the approval of BiDil was commercial rather than biomedical. In general, instead of casting wide conceptual and normative nets, reporters tended to react to the BiDil patent by highlighting combustible topics. New York Times reporter Robin Henig, for example, "touted [BiDil] as the first ‘ethnic drug' in ‘the emerging field of race-based pharmacogenetics" (Henig, 2004, as cited on p. 364). While Henig does not explicitly invoke the language of the dichotomy, terms such as "ethnic drug" and "race-based pharmacogenetics" do reify race as a natural kind. One result is a second, but closely related, sort of dichotomization, this 39 time between "scientist-expert and nonscientist-commonfolk conceptual schemes" (Gannett, 2010, p. 372). The view that scientific and folk conceptions of race are wholly different gives the illusion that scientists and philosophers of science are limited in their potential contributions to social and political contexts in which folk conceptions of race operate, but "scientific and folk meanings are not wholly autonomous because science influences the folk meaning and the folk meaning often provides a starting point for scientists in their research" (Gannett, 2010, p. 375). So, however it manifests, the dichotomous framing of alternatives "compromises the critical insight philosophers of science might otherwise contribute to debate in the public sphere" (Gannett, 2010, p. 365). In particular, it blinds them to certain sorts of questions. Gannett (2010) concedes that "asking some questions inevitably leaves other questions unasked" (p. 374), but choice of research questions is not her concern. The problem "involves more-it involves privileging a certain set of questions (theoretical, metaphysical, etc.) such that others (practical, evaluative, etc.) are not merely overlooked but systematically ignored" (p. 374). In fact, "the dichotomizing of the biological and social as distinct causal realms…precludes investigating the very context-specific ways in which biological and social factors interact" (Gannett, 2010, p. 375). So, Gannett (2010) reveals a blind spot for philosophers of science, and this is unacceptable because "there are also research questions that cannot be satisfactorily addressed unless allowance is made for the causal interactions that occur among biological and social factors" (p. 370). Even if some contexts demand only one or the other of the traditional accounts of race, many crucial 40 ones do not. This is the very philosophical problem space in which I aim to work, so in the next section, I elaborate on the problem of biological-social interactive contexts and show where work is left to be done. 2.5 How Gannett's Critique Motivates the Dissertation Broad agreement exists that many important conceptions of race are complexes of apparently incongruous parts that frequently vary in proportion of influence from one context to the next. Gannett (2010) argues that our theories about race should accommodate that fact, and she takes to task those philosophers who sustain the dichotomy from the comfort of familiar theoretical bunkers when many contexts involve both sorts of causal factors. The challenge, then, is to figure out how to accommodate the strange mix of causal constituents. Even if 'race' talk is patently complex and multifactorial, explaining how and by what medium the disparate factors are put together has so far perplexed philosophers of race. Even Gannett's credulity is strained-again, Gannett charges Kitcher with conceptual gerrymandering even though his cocktail is limited to biological concepts. Biological-social interactionist conceptions of race are even more confounding. In defense of this point, Gannett cites sociologist Paul Gilroy, who conceives ‘race' as ‘an active, dynamic idea or principle that assists in the constitution of social reality': social groups are constituted by racializing some contingent combination of biological, cultural, and national differences as essential, hierarchical, primordial, authentic, historical, natural, discrete, absolute, fixed, static, immutable, and unbridgeable. (Gilroy, 2000, as cited in Gannett, p. 377) A crucial question for the dissertation, then, is how to account for that 41 dynamic interaction. I agree with Gannett that the strategies that foster the dichotomy-more precisely, the aspects of those strategies that do-should be abandoned. But Gannett is specifically interested "in how philosophers of science approach debates about genetics and race," so while she focuses primarily on two related scientific issues-biomedicine and biological reality-in my view, the general problem Gannett describes pervades virtually all racerelevant contexts. So, my interests are more general than hers. Philosophers, scientists, and layfolk alike have difficulty reasoning about race largely because of its inherent complexity. When we talk about race, we sometimes reason biologically, sometimes socially, usually both, and almost always inconsistently. The vast array of both biological and social constructivist explanations of race often perpetuate confusion. The lack of interaction between the camps blocks the path to new inquiry and, worse, retards progress toward practical solutions to contextual problems that involve race. "Indeed," says Gannett, "the statistical, contingent, accidental, localized, and interest-relative bases of such inferences serve to undercut the dichotomizing of race as either biological reality or social construct and favor the adoption of a pragmatic approach" (Gannett, 2010, p. 363, emphasis added). A survey of existing theories on race, some of which I have mentioned already, suggests that we still lack a structure for that approach. 2.5.1 Some Ways to Approach the Dichotomy Several of the positions described above hint at potential solutions to the dichotomy. In her harder stance, for example, Andreasen argues that having two 42 mutually exclusive race paradigms is fine because each is usefully fit to its own realm of discourse. At the opposite extreme, some have argued for the eventual elimination of "race" and all associated concepts. Appiah (1995), for example, calls for the relative short-term elimination of race words (such as "Black," "White," "Asian") because the "truth is that there are no races: there is nothing in the world that can do all we ask ‘race' to do for us" (p. 75). This solution is quite simple: no race, no race dichotomy. Appiah holds an important position in the history of the philosophy of race, and his arguments have wielded tremendous influence, so although they appear to fail, they merit elaboration. Appiah arrives at the conclusion that there are no races via what Ron Mallon (2006) calls a "mismatch" argument, which holds that "the true account of the extension of a term or concept x would be sharply different from what is believed about the extension of x" (p. 533). In the present case, Appiah holds that if race is real, it must be a natural (biological) kind, and since he does not recognize any tenable biological concept ‘race', Appiah reasons that race words do not name anything in the world. So, since there is a mismatch between the words we use and the things in the world, the words name nothing at all-so race is, as Gannett said, "a fiction like phlogiston." Combined with the pernicious history of race, that conclusion leads Appiah to argue that ‘race' talk should be eliminated. Again, if Appiah's view bears out, it would eliminate the dichotomy, since there would be no concept about which to contend, but Mallon argues that this "semantic approach"-in fact, any semantic approach-to race is problematic. That crucial topic is tabled until the next chapter, however, so that I 43 can discuss one other potential solution. A final, less advertised way to deal with the dichotomy is via what I will call a "matching" argument. While Appiah (1995) attempts to reveal a mismatch between our race vocabulary and what is in the world, Sesardic, as I said above, argues that our words do match what is in the world-this is what he means by saying that Dobzhansky's classification has not been discredited. What Dobzhansky said is in the world is in the world. Arguably, Sesardic's view, if proven correct, could eliminate the dichotomy as well-in this case by establishing the primacy of the biological conception. I will argue that neither these nor any existing philosophical accounts provide a suitable response to the interactionist problem, however. In fact, this is what will necessitate my thesis. 2.6 Conclusion Gannett diagnoses the dichotomy as a problem grounded in metaphysics, and her analysis paves the way for the position I defend in the dissertation. So, it bears repeating that on Gannett's analysis, the natural kinds approach to race is what "compromises the critical insight philosophers of science" and sustains the "unproductive, even corrosive standoff between biological realists and social constructionists" (Gannett, 2010, pp. 363, 365). Neither natural kinds approach-essentialist or biological-helps us here, so Gannett (2010) is "skeptical that race fulfills these basic assumptions about natural kind…this suggests that race cannot do for us what it seems many have wanted it to do. Even if one is loathe to jettison natural kinds, the category of race as a postulated natural kind must go" (p. 378). 44 She concedes that "it would be very surprising if the statistically correlated DNA markers which so impress Sarich and Miele were not to be found" but argues they will not do the work that they and like-minded scholars think. Among other things, she strongly suspects they will be "statistical not universal, interestrelative not mind independent, dynamic not static, [and] indeterminate not determinate" (Gannett, 2010, p. 383). So how do we proceed? Gannett (2010) advises worrying "less about the ‘really real' [natural kinds reasoning]" and urges that philosophers of science and of race "instead assess the appropriateness of group categories of classification relative to the purposes of specific research programs, [which] invites consideration of the social and political ramifications of drawing boundaries in one way rather than another" (p. 383). Gannett's groundwork is crucial to my project. By undermining the natural kinds approach to race by showing that it leads to a pernicious dichotomization of concepts, she has opened the space for more appropriate philosophical racial discourse. I aim to locate race somewhere in that conceptual and metaphysical space between the "really real" reality of natural kinds and the "nothingness" of social construction5. Following Gannett's recommendation, I attempt to employ a "better argumentative strategy [that] recognizes that ‘race,' as it is socially constructed, is essentialist" (Gannett, 2010, p. 371). The position I advocate is quite close to that one, but there are other roadblocks, potential dead ends, and tempting wrong turns on the way to a pragmatically useful conception of race, so As Millgram (2015) points out, "'constructivism' has in recent years come to mean almost all things to all people," so here I follow Millgram, who recommends the view on constructivism promoted by Rawls (1989). 5 45 I now I turn to a second fundamental problem for race scholarship, one that questions another revered philosophical strategy, what Mallon (2009) calls the "project of ‘semantic ascent'" (p. 1). CHAPTER 3 A METHODOLOGICAL ALTERNATIVE TO SEMANTICISM 3.1 Introduction In the preceding chapter, I detailed Gannett's description of a set of closely related problems for philosophers of race. She reveals that arguments over race's status as a natural kind lead to a conceptual dichotomy constituted by mutually exclusive biological and social constructivist explanations of race. The dichotomization of race, Gannett argues, precludes investigating important questions in contexts in which racial phenomena are constituted by the interaction of both social and biological factors. In this chapter, I focus a bit more closely on the primary source of the dichotomy problem, namely, the battle over the methods of defining the concept ‘race'. While Gannett effectively describes the pitfalls of natural kinds reasoning about race, Mallon critiques a different aspect of traditional philosophical analysis-the "project of semantic ascent," which, according to its most general description, uses "the conceptual analysis of race (and related concepts) to teach us about what race must be (if it is anything at all)" (Mallon, 2009, p. 2). Mallon charges that, despite its venerated status among philosophers, this "semantic strategy" offers little if anything to philosophy of race (Mallon, 2004, 2006, 2009).6 6 For a general treatment of semantic analysis in philosophy see Millgram (2015). 47 This might be surprising to many philosophers, since semantics has long enjoyed special status in philosophy-one of philosophy's first demands, after all, is that crucial terms be defined. In this context, then, it would seem only natural to begin with rigorous conceptual analysis to arrive at a proper definition of "race." Even Gannett appears to assume the primacy of semantics for race debates. She begins the aforementioned paper by laying out various definitions in order to reveal the limitations of each, and, by the end, she advises philosophers "to familiarize themselves with what race concepts are in actual use in various areas of the biological and biomedical science" (Gannett, 2010, p. 369). So, even though Gannett clearly questions particular semantic approaches to race, and even though she hints at a more liberal approach allowing contributions from many sources, she never explicitly calls into question the semantic strategy itself. Mallon does. In this context, Mallon is highly skeptical of any semantic approach, arguing that it is simply the wrong strategy for assessing race and issues involving race. It is not merely that he is skeptical about the possibility of arriving at the right or best definition; rather, semantic analyses of "race," even if correct, are unimportant, according to Mallon. This conclusion is even more surprising- shocking even-which is why I need to dedicate a large proportion of this chapter to describing multiple versions of the semantic strategy and the problems they face. If Mallon is right-if the semantic approach7 fails-we will be left in need of a new medium for ‘race' talk and new strategies for approaching it. In short, the 7 I use "semantic strategy" and "semantic approach" interchangeably. 48 focus of this chapter is methodology. The ultimate aim is to introduce Mallon's methodological alternative, which he dubs the "evolutionary-cognitive program." A proper introduction of Mallon's favored methodology, however, requires a detailed discussion of the entrenched methodology it is meant to replace. I begin, then, with a description of the semantic strategy and Mallon's critique of it. 3.2 The Semantic Strategy, Its Problems, and Its Rivals In general terms, the semantic strategy involves the attempt to connect an analysis of the concept ‘race' (or a definition of the word "race") and certain metaphysical commitments via one or another theory of reference. Very often, philosophers of race who employ this approach then use it to decide their normative stance on ‘race' talk. If this basic description seems a bit obscure, locating it within the broader structural context of philosophy of race helps make it clear. In recent years, several philosophers of race have explicitly described a relatively "canonical" investigative structure that revolves around four interconnected questions: the normative question, the ontological question, the conceptual question, and the methodological question (Glasgow, 2009; Mallon, 2004; Taylor, 2004). The normative question concerns whether we should "eliminate or conserve racial discourse and thought, as well as practices that rely on racial categories" (Glasgow, 2009, p. 2). The two normative positions on race, then, are eliminativism-the view that ‘race' talk should be removed from discourse- and conservationism-the view that at least some form of 'race' talk should be 49 preserved (Mallon, 2006). Many philosophers assume that the normative decision depends on the answers to the others. As Mallon (2004) says, "Normative disputes give rise to a concern with the metaphysics of race because of the role metaphysical arguments play in supporting normative conclusions" (p. 645). In other words, the answer to the ontological question is assumed to inform important aspects of the answer to the normative one. Returning to an earlier example, Appiah argues for elimination of ‘race' talk because of the combination of its pernicious past and, more to the point, the fact that it "makes reference to a set of racial properties that literally do not exist" (Mallon, 2004, p. 645). In other words, Appiah supports his (normative) eliminativism by way of his (metaphysical) skepticism. Gannett's normative concerns inspired her survey of the metaphysical landscape, but that move lead her to urge a reversal of the traditional direction of fit by subordinating the latter question to the former. Appiah and Gannett provide illustrative examples of the other's question as well. Crucial to Appiah's ontological skepticism, for example, is his answer to the conceptual question. He conceives of race in terms of "racialism," which is "the view that there are racial essences" (Mallon, 2006, p. 528). Hence, when Appiah denies the existence of race, he is denying racialist race, but I have already shown that there are other reasonable ways to conceive of race. Andreasen, for example, promotes a natural kinds view of race, but one that does not entail racialism-she, unlike Appiah, is a realist about race, but conceived cladistically, not "racialistically." Taylor (2000, 2004) is also a realist about race but in a sense different than Andreasen's. In explicit opposition to 50 Appiah's skepticism, Taylor argues that race is real as a social construct, not as a biological kind. All of this is to say, of course, that philosophers' ontological commitments are tied to their answers to the conceptual question, which simply involves attempts to determine the correct or best analysis of the concept ‘race' or definition of the word "race." To review, for Appiah, the relevant race concept is racialist, for Andreasen cladistic, for Taylor social constructivist. At first gloss, their respective methodological commitments appear as different as their conceptual conclusions. Andreasen's biological approach contrasts from Taylor's constructionism as much as it does from Glasgow's "reconstructionism." A broader look, however, reveals that each approaches the methodological question via some version of the semantic strategy. For philosophers, this likely seems natural enough: Normative concerns drive the conversation, and philosophers choose a methodology for defining contested terms so that we may determine whether those terms describe something in the world. After we arrive at answers regarding our conceptual and ontological positions, we return to our normative concerns, and using our newly formed analytic tools, determine our normative stance. It hardly seems controversial, yet Mallon shows this revered strategy fails to advance the agenda on race. The problem, according to Mallon (2006), is the semantic strategy itself, which he details in step-wise fashion. First, one "connects metaphysical claims and linguistic-conceptual practices with the assumption of a particular theory of reference for the word" (Mallon, 2006, p. 527). Then, from these assumptions, "it 51 is concluded that racial terms or concepts appropriately refer (or fail to refer) to some or other metaphysical features of the world" (Mallon, 2006, p. 527). Finally, as I indicated above, these conclusions are often used to determine one's normative stance on ‘race' talk. Crucially, then, both the arguments generated by this strategy and the normative consequences that flow from them turn on disagreements concerning either "the metaphysical features of the world…or the appropriate theory of reference for race terms/concepts" (Mallon, 2006 ,p. 527). Think again of Appiah's "mismatch argument," a paragon of the semantic strategy. Appealing to his racialist conception, Appiah argues that there are no races because nothing that exists does or even can fulfill the conditions of racialism. There is, he says, "nothing in the world that can do all we ask ‘race' to do for us" (Appiah, 1995, as cited in Mallon, 2006, p. 525). Appiah's conceptual choice is not arbitrary. For him, it is the appropriate conception because it is the one we have inherited and the one operative even today. It is W.E.B. Du Bois' conception of race, which was established at a particular time-the late 19th and early 20th centuries-in reaction to particular events-the lasting consequences of racist European and American colonialism. So, like Andreasen and Glasgow, Appiah thinks Kripke and Putnam's causal-historical theory is more appropriate than descriptivism for racial contexts, albeit for the purpose of showing that race does not exist. Since I have previously discussed some problems with this approach to race, I needn't dwell on it here. It suffices to say that Appiah offers a version of the now familiar argument that the causal-historical account best represents the ordinary or folk concept and that he then argues that the folk 52 concept attempts to pick out biological populations that do not exist. According to Appiah, however, no matter what theory of reference one chooses, racialist race fails to refer to anything real. Even if one switches to a "descriptivist" approach, according to which a "term or concept is associated with a description: a proposition or set of propositions about the properties of the referent" (Mallon, 2006, p. 530), Appiah concludes that "race" describes nothing at all. Appiah's skepticism is secured, then, because no theory of reference picks out an object to which "race" (so conceived) refers. Again, the result is a mismatch between our vocabulary and what is really in the world. Appiah is alone neither in approach nor conclusion. In Philosophy of Science and Race, for example, Naomi Zack argues that "the only known candidates for physical scientific bases of race…fails. Therefore, there is no physical scientific basis for social racial taxonomy" (Zack, 2002, 88). In short, she says, "The ordinary concept of race in the United States has no scientific basis" (Zack, 1993, 18). This line of argument has a deep intuitive pull and a rich history. Gannett's earlier allusion to phlogiston provides a useful example. "Phlogiston" is the name given in the 18th century to a theoretical element intended to explain combustion. By the turn of the 19th century, however, phlogiston theory had been abandoned and, quite naturally, "phlogiston" eliminated from discourse. So it is hard not to be surprised when Mallon argues that the semantic strategy is irrelevant for race. Nevertheless, this "venerable strategy," he says, "is problematic…race theory ought not to rely on finding the correct theory of reference to determine the appropriate use of ‘race' talk," largely 53 because it sustains unnecessary and unhelpful metaphysical disputes (Mallon, 2006, p. 528). Once again, the previous chapter provides an illustrative example-Andreasen and Glasgow began with promise for a productive debate, but it devolved into unhelpful squabbles over race's "baptismal procedures." Given missteps like this, Mallon regards the normative question, rather than the conceptual or metaphysical questions, as primary. Spelled out this way, Mallon's objection is made more intuitive, but comprehending its full effect requires elaborating on its fine points. Mallon's (2006) sustained attack on semantic approaches begins with an appraisal of the state of play in metaphysics of race. There is, he says, an "ontological consensus" that racialism is false; "there is now widespread agreement among philosophers, social theorists, anthropologists, and biologists that races do not share…biobehavioral essences" (p. 529). Even if molecular genetics provided the last "hope" for a source of racial essence, "studies of human genetic diversity suggest that genetic variation within racially identified populations is as great as or greater than diversity between populations. Thus, it is very unlikely that any interesting genetic ‘essence' will be shared by all and only members of a race" (p. 529). As a result, the contemporary debate has splintered into the three metaphysical positions on race: racial skepticism, racial constructionism, and racial population naturalism (pp. 525-526). I have already introduced each position and some of its supporters: Appiah and Zack, as I just showed, are skeptics; Glasgow and Taylor are constructionists; Andreasen, Kitcher, and Pigliucci and Kaplan are population naturalists. At first blush, the 54 "three groups…seem to disagree fundamentally on the metaphysical character of race," but Mallon argues that much "of the apparent metaphysical disagreement over race is an illusion," that the semantic strategy sustains that illusion, and that the semantic strategy is, therefore, problematic (Mallon, 2006, pp. 527-528). I elaborate on each point. Mallon regards the metaphysical disagreement as illusory because despite their differences, the competing camps actually "share a broad base of agreement regarding the metaphysical facts surrounding racial or racialized phenomena that suggests their views are complementary parts of a complex view incorporating biological, social, and psychological facts" (Mallon, 2006, 528). Mallon (2006) thereby expands the ontological consensus by divorcing metaphysical facts from "questions regarding the use of racial terms or concepts," a move which results in "an almost banal list of observations": that all parties agree that racialism is false, that there is a plurality of operative racial concepts, that a common set of criteria are used to ascribe persons to a race, and that racial classification affects persons "in both superficial and profound ways" that are sometimes "profoundly oppressive" (p. 545). Virtually everyone also agrees that past "geographic distribution of populations" likely resulted in a "significant degree of reproductive isolation," which is "partially responsible for the geographic distribution of superficial bodily features associated with race," and that racial classification affects marriage and reproduction rates (p. 546). In fact, if theorists did broadly disagree in their metaphysical commitments, one would expect arguments to revolve around those 55 disagreements, but Mallon shows that disagreements about race tend to center on the appropriateness of ‘race' talk, rather than race's metaphysical status. For example, even though "Appiah thinks races do not exist…he offers an account of racial identification to account for the constructionist intuition that racial classifications is causally important" (Mallon, 2006, p. 546). Again, racial skeptics, such as Appiah, are skeptical about racialism in particular, but so is everyone else, constructivists and populationists included. Mallon (2006) concludes that if such broad metaphysical agreement exists, "it is mistaken to view dispute among constructionists and naturalists as primarily metaphysical in character" (p. 547). As a consequence, "for a variety of important questions of public policy and applied morality, the questions may be restated without important metaphysical disagreement within different idioms of ‘race' talk" (Mallon, 2006, p. 547). For example, despite disagreement over the existence of race, skeptics and constructionists alike can justifiably "call for rectification of civil rights violations in twentieth-century America" (Mallon, 2006, p. 547). So exactly what is the fuss? Mallon argues that the problem is this very strategy of debating the semantics of race under the guise of metaphysical disagreement: "in the absence of substantial metaphysical disagreement, racial theorists have achieved alternative conclusions by making different assumptions about the correct semantics for racial terms" (Mallon, 2006, p. 547). It makes no difference whether semanticists opt to "decide which theory of reference is correct and decide what auxiliary assumptions regarding the application of such a theory are 56 needed to determine the correct referents of racial terms and concepts" or to "take Appiah's strategy and attempt to justify a conclusion in terms of every plausible theory" (Mallon, 2006, p. 548). Choosing a particular theory of reference "is obfuscating because…it makes philosophical debate over the reference of racial terms and concepts appear as a genuine metaphysical disagreement" and "ineffective because it is unlikely to be fruitful in resolving how we ought to use ‘race' talk" (Mallon, 2006, p. 548). Appiah's alternate strategy of arguing that "one's conclusions follow from all the plausible candidate theories of reference" seems to offer hope, but to no avail. To do as Appiah wishes, "we need to be able to separate the plausible from the implausible candidates" of theories of reference, but even if "we can decide on the plausible candidates, there is no reason to believe that all the plausible candidates converge on a single answer regarding whether or how race exists" (Mallon, 2006, p. 549). So, the semantic strategy's viability is dubious already, but Mallon isn't done. Suppose, he says, that we "arrived at a correct account of the reference of racial terms…yielding a definitive account of what (if anything) race is" (Mallon, 2006, p. 549). Even in that unlikely event, "it is not clear that the semantically correct account of ‘race' talk ought to dictate our use" (Mallon, 2006, p. 549). Despite the intuition that answering the normative question depends on answering the others, Mallon (2006) shows that "semantic arguments regarding the referents of ‘race' talk need not dovetail with other sorts of argument" (p. 549). Mallon's analysis makes this point surprisingly easy to swallow. If we decide that "‘race' talk is deeply oppressive, no argument to the effect that such 57 talk refers to a biological population or a social construction would be of sufficient weight to merit the continuation of this practice" (Mallon, 2006, p. 549). On the other hand, even if it does not refer to anything in the world, we still might decide that ‘race' talk is morally required for addressing existing oppression. Mallon (2006) clearly shows there is no necessary relationship between the normative and other questions, arguing that "the attempt to link the [normative and metaphysical questions] via the semantic strategy has…resulted in an illusion of metaphysical disagreement and a misplaced emphasis on metaphysical and semantic concerns" (p. 550-551). Even so, "profound disagreement" remains regarding the moral status of ‘race' talk, but in place of the semantic strategy, Mallon (2006) recommends "a complex assessment of many factors, including, the epistemic value of ‘race' talk in various domains, the benefits and costs of racial identification and social enforcement of such identification…the role of ‘race' talk in promoting or undermining racism" and so on (p. 550). It appears, then, that the semantic approach to race produces little of value, even if it hits its intended target. Still, not everyone is convinced-the semantic strategy was not abandoned after the publication of Mallon (2006). I now turn to two more recent analyses of "race," each of which represents one the two general types of semanticism about race. I first introduce rationalist semanticism via the work of Michael Hardimon, and then turn to Glasgow's empirical semanticism.8 Despite renewed vigor among these semanticists, however, none appear to overcome Mallon's objections to the semantic strategy Glasgow makes this distinction as well, but in terms of what he calls "armchair" and "empirical" analysis. I choose "rationalist" as a more neutral representation of the former. 8 58 for race. 3.2.1 Michael Hardimon's Rationalist Semanticism Appiah (1985) describes a view according to which "Understanding the idea of race involves grasping how people think about race: what they take to be central truths about races; under what sorts of circumstances they will apply the idea of race; what consequences for action will flow from that application" (Appiah, 1985, 56). Appiah calls this the "'ideational' view of meaning," and it is where I locate the rationalist semantic approach. The chief proponent of this approach with regard to race is Michael Hardimon. Hardimon's foray into philosophy of race begins with a 2003 paper in which he aims at "providing a general answer to the question: What is the concept of race?" (Hardimon, 2003, p. 437). Like most race theorists, Hardimon recognizes the existence of a multiplicity of race concepts, conceding that there "is no single concept of race that deserves the honorific ‘the'" (Hardimon, 2012, 6). Nevertheless, Hardimon (2003) initially focuses on a single one, the "important and poorly understood" ordinary or folk concept of race, which "corresponds (roughly) to the meaning of the ordinary word ‘race'…[and] bears the imprint of eighteenth- and nineteenth-century attempts to develop a scientific concept of race" (p. 437). Unlike Glasgow, however, Hardimon (2003) aims neither "to rehabilitate the word ‘race'," nor "propose to introduce a new sense of the word"; instead, his approach "takes the form of reflection on an already given concept" (p. 440). Hardimon (2003) concedes that his strategy "carries with it a certain ineliminable element of rationalization," but that means accepting only 59 "the familiar methodological assumption that appeals to intuition are a legitimate and inescapable component of the articulation of ordinary concepts" (p. 441). To even get off the ground, he says, "we must suppose that at least some of our intuitions about the concept of race are correct" (Hardimon, 2003, p. 441). His methodology, then, is to analyze ‘race' to its "logical core," that is, its "intelligible nucleus...characterized by three basic theses" that are conjunctively "necessary and sufficient for constituting the content" (Hardimon, 2003, pp. 441442). It should be obvious that Hardimon's methodology is both rationalist and semanticist-rationalist by explicit admission, semanticist by virtue of its attempt to arrive at and proceed from a definitive meaning of ‘race'. According to Hardimon (2003), the "concept of race is the concept of a group of human beings distinguished from other human beings by visible physical features of the relevant kind…whose members are linked by common ancestry" originating "from a distinctive geographic location" (pp. 442, 445, 447). In sum, Hardimon's necessary and jointly sufficient conditions for division of human populations by the logical core of ‘race' are (1) physical appearance, (2) shared ancestry, and (3) shared geographic origin. Hardimon (2003) says that the first condition "captures the basic intuition that race is essentially manifest," which means that the "ordinary concept of race requires that the distinction between racial groups be visibly marked in some way or other" (p. 442). Any explanation of race that leaves this feature out necessarily fails. "The very notion of a visually indistinguishable racial group runs counter to the idea of race" (Hardimon, 2003, p. 442). Hardimon attempts to 60 secure the point through a thought experiment intended to show that some physical features-say, skin tone-are racial, while others-such as the presence of an Adam's apple-are not. Hardimon concludes that "Racial groups are distinguished from one another by visible physical features that are racial," a claim that is supposed to be saved from the charge of tautology by the fact that "the relevant physical features of race can be picked out ‘directly' through ostension…Pointing allows us to escape the circle of words" (Hardimon, 2003, pp. 444-445). So, Hardimon (2003) does not delineate the kinds of features he takes to be racial, but he apparently knows them when he sees them. Despite its importance to the core, however, Hardimon stresses that this first thesis should not be taken to "suggest that race is a matter of physical appearance merely" (p. 445). Race "is also a matter of ancestry: who one's parents are, who their parents are, and so forth" (Hardimon, 2003, p. 445). That is, the second condition of the logical core is that to be considered a race, a human group must share a common ancestry. Although different conceptions of race might yield different "details of the nature of the ancestry that racial identity involves," Hardimon (2003) reasons from the etymological link between the words "race" and "lineage" that "it is possible to extract from the concept a number of structural features" (Hardimon, 2003, p. 445). First, the race of an individual is determined by one's "immediate ancestors: his or her parents" (p. 446). In addition, races proceed from "founders" whose line is "maintained through inbreeding. Endogamy is thus a structural characteristic of race" (Hardimon, 2003, p. 445). 61 So far, then, to qualify as a "racial lineage," a human group must have founders that "exhibit distinctive visible features of the relevant kind or possess alleles for such features"; a group does not count as a race unless its members "resemble one another in their visible features" (Hardimon, 2003, p. 447). Finally, the combination of shared visible features and lineage entail the third necessary and sufficient condition of the logical core-that the members of the group "originate form a distinctive geographic location" (Hardimon, 2003, p. 447). Hardimon (2003) again appeals to etymology and intuition to establish this point: The connection between race and geography "can be seen in the names major writers on race in the seventeenth century assigned to racial groups" (p. 447). Furthermore, this "lexical link between race and geographical area reflects the intuitive idea that racial groups have their origin in different geographic locations" (Hardimon, 2003, p. 447). Any supposed race, then, must be distinguished by "a specific geographical location…uniquely associated with it" (p. 447). In sum, the ordinary concept ‘race' corresponds to the meaning of the folk term "race," which Hardimon defines as a human group or population that is distinguished by visible physical features ("of the relevant kind") and a common ancestry uniquely associated with a geographic region. Noticeably absent is the racialist aspect of race that concerned Appiah and others. That omission is one that Hardimon vigorously defends-he intentionally deflates ‘race' in part to distance it from racialism. To this end, Hardimon (2003) devotes significant effort to distinguishing the race "concept's logical core and the racialist development of that core" (p. 442). His "minimalist" 62 account, he says, arises out of his more general philosophical commitments, the most crucial of which is a distinction between concepts and conceptions. "The ordinary concept of race and the ordinary conception of race are two different things," though they "operate in tandem. The concept of X specifies what X is. A conception of X indicates how the concept of X is to be understood" (Hardimon, 2003, p. 440). So, Hardimon allows that "thicker" explanations of race exist, but argues that they are constructed upon the foundation of the logical core-that is, as conceptions built upon the concept of race. He further claims that this distinction "makes it possible to see that much of what is commonly presented in the literature as a discussion of the ordinary concept of race is perhaps better understood as a discussion of the ordinary conception of race" (Hardimon, 2003, p. 440). The ordinary concept answers to the "logical core [and] does not require that races have essences. The ordinary concept is not essentialistic" (Hardimon, 2003, p. 449). This, argues Hardimon, is not a bug but a feature because it means the concept "is compatible with the modern view that there are no…sharp divisions between racial groups. More precisely it does not demand that each race…possesses some unique property…(visible or hidden) or set of properties " (Hardimon, 2003, p. 449). So, since racialism is a "species of essentialism," ‘race' (the concept) is not properly perceived as racialist. The association with racialism, says Hardimon, is an accident of history. "When the logical core first entered the historical scene, it was already articulated by the racialist development. The step from the logical core to racialist development…is 63 historically contingent" (Hardimon, 2003, p. 453). Since nothing about the "logical core necessitates this step, [it] could have appeared without the racialist development, if world history had gone differently" (Hardimon, 2003, p. 453). Because of the way things did go historically, however, "the logical core and racialist development appear to be a unity" (Hardimon, 2003, p. 453). The "ordinary conception of race," which appends racialism to the ordinary concept, results from that apparent unity (Hardimon, 2003, p. 451, emphasis added). In other words, Hardimon argues that while the ordinary concept ‘race' is not racialist, the ordinary conception of race is. Of course, all of this predates Mallon's 2006 attack on the semantic strategy, but in a spate of more recent papers on the same topic, Hardimon stands his ground. The thesis of his 2003 paper is foundational to a 2013 paper, for example, in which he argues that the failure to distinguish the ordinary concept from the racialist conception is what leads to "Confusions about the place of race in medicine" (Hardimon, 2013, p. 6). In fact, much of his other recent work on race centers on practical benefits of distinguishing concepts and conceptions of race. In contrast to Gannett, who argues that we must find ways to talk about racial contexts in which social and biological factors interact, Hardimon (2013) attributes the difficulty of applying racial concepts in practical contexts to the "erroneous belief that there is an amorphous thing race that is (somehow!) both social and biological" (p. 6). Race, Hardimon argues, "is not one thing. The social and biological phenomena of race are two different things" that require "two technical race 64 concepts, one registering the social phenomenon…and the other registering the possible biological phenomenon" (Hardimon, 2013, p. 7). Neither is identical with the ordinary concept. "Socialrace"-roughly, the concept of socially constructed race-is racialist and, therefore, different from the ordinary concept (Hardimon, 2014, p. 75). But, while the populationist concept is not quite equivalent to the ordinary one, it is "continuous with" it. Defined according to "the fact that human beings exhibit morphological differences…statistically associated with differences of geographical ancestry," the populationist concept, or "minimalist phenomenon of biological race," is a "scientization" of the ordinary one (Hardimon, 2013, p. 18). The point of these elaborations on Hardimon's stance is twofold. First, many of them are relevant to later parts of the chapter and dissertation, so their introduction here primes their future roles. More importantly, though, at virtually every stage, the extended version of Hardimon's story depends on the semantic strategy, as evidenced by his explicit use of "conceptual analysis of the concept race (and related concepts) to teach us about what race must be (if it is anything at all)" (Mallon, 2009, p. 2). One could hardly generate more accurate general description of Hardimon's approach. Although he recognizes a "plurality" of race concepts, Hardimon's project is founded on an attempt to cull the logical core from what is common to all ordinary conceptions of race. The relevance of Mallon's objections is obvious, and recognizing this line of criticism, Hardimon offers a curious response. He essentially bites the bullet. In the 2003 paper, he says one "of the most striking results of our account 65 of the logical core of the ordinary concept of race is that race turns out to be relatively unimportant" (Hardimon, 2003, p. 451). In 2012, he echoes this sentiment: "it is crucial to distinguish the question concerning the reality of race from the question concerning its importance" (Hardimon, 2012, p. 269). Hardimon's motivation is not hard to guess. A deflated race concept bereft of racialist baggage improves the odds of preserving ‘race' talk without racist baggage. This is something for which Hardimon clearly aims-his link between the ordinary and populationist concepts culminates with an attempt to show that latter can be of use in medicine without fostering racism (Hardimon, 2012). Even so, as Mallon's arguments reveal, one can accept all that Hardimon has claimed and still take whatever normative position on the use of ‘race' talk one wants. But Hardimon's is not the only available semantic approach to race. The history of philosophical rivalry is rife with opposing rationalist and empiricist explanations of the same phenomenon, and this context provides another example. I turn now to the empiricist semanticism of Joshua Glasgow. 3.2.2 Joshua Glasgow's Empirical Semanticism Glasgow (2009) defends a position on race according to which we would neither "out-and-out eliminate race-thinking, nor…wholeheartedly conserve it" (p. 2). Instead, we would "replace racial discourse with a nearby discourse," that is, we would "stop using terms like ‘race,' ‘Black,' and ‘White,' and so on to purport to refer to biological categories, as we currently do, and instead use them to refer to wholly social categories" (Glasgow, 2009, p. 2). The folk biologically informed concept ‘race', he argues, should be replaced by his new "de-biologized" concept 66 "race." In other words, Glasgow's goal is to redefine the folk race concept because it has been historically misconceived as biological. As a result, Glasgow (2009) considers his view neither eliminativist nor conservationist; instead he calls for "racial reconstructionism" (p. 2). Even this brief description already suggests that his methodology is semantic; the detailed description cements it. Contra Hardimon, Glasgow (2009) argues that what he calls the "armchair" approach (what I've called the "rationalist approach") is "misguided… [so] we should focus our attention squarely on how racial terms are used in contemporary mainstream discourse" (p. 8). In lieu of reflecting from his armchair, Glasgow opts for "the ‘experimental approach,' which holds not only that the meanings of racial terms are…at least partially fixed by common sense, but also that we should inform our analysis of folk racial discourse with data gathered from actual empirical research conducted in a manner consistent with the practices of the social sciences" (Glasgow, 2009, p. 8). His approach joins a "growing experimental philosophy movement in insisting that we accommodate empirical data when doing conceptual analysis" (Glasgow, 2009, p. 39). The data to which he appeals reveal interesting, if not entirely conclusive, results. Empirical research on race does converge on some relatively consistent claims. It lends evidence to what Glasgow (2009) calls the "biosocial complexity of racial discourse," which amounts to the claim that "folk racial concepts…are composed of biological elements and sometimes also social elements" (pp. 10, 78). The data say that "phenotype is in some sense central to race, without going so far as to say that one's race will always be dictated by the way one 67 looks" (Glasgow, 2009, p. 78). Despite these and several other statistical truths, Glasgow concedes that we must "settle for an incomplete analysis of" the concept ‘race' (Glasgow, 2009, p. 78). Whatever we decide the concept ‘race' includes-whatever, that is, the empirical results show-Glasgow answers the ontological question in the negative: "race is not real" (Glasgow, 2009, p. 8). For Glasgow, the "upshot" is that race is neither socially nor biologically real. Like Appiah, then, Glasgow is a skeptic; unlike Appiah, he is not an eliminativist. Despite advocating antirealism about race, Glasgow warns that "we'd be poorly advised to simply get rid of racial discourse" and suggests that we opt for his "purely social" revisionist concept (Glasgow, 2009, p. 8). Just as Hardimon does, then, Glasgow utilizes a semantic strategy. To review, Glasgow argues on the grounds of empirical research that the folk conceive of race in a certain way. He then argues that what the folk think race to be has no concomitant "object" in the world, so race is not real in the "relevant" sense. We have another mismatch argument. What distinguishes Glasgow from Appiah, then, is not the method but the conclusion. Glasgow thinks the mismatch between ‘race' and the world should motivate us to reconstruct, not eliminate, it. Mallon's general objection to the semantic strategy barely needs repeating. In fact, by agreeing with Appiah ontologically but disagreeing with him normatively, his own commitments appear to show "the semantics of ‘race' doesn't really matter to normative debates" (Mallon, 2009, p. 1). In his 2009 commentary on Glasgow's A Theory of Race, Mallon doubles down on his 68 critique of the semantic strategy. To make the case, Mallon explicitly stipulates "that Glasgow's own analysis is correct"-that is, he concedes for the sake of argument that the meaning of the folk term "race" is the correct term of analysis and that the term entails some sort of biological element (Mallon, 2009, p. 2). As a result, any "concept (e.g., Glasgow's race*) that does not entail biological reality of a certain sort is thereby a different concept" (Mallon, 2009, p. 2). From there, he argues that Glasgow's "correct" analysis nevertheless fails to lead to any interesting normative results. Consider one of Mallon's examples. Dealing first with the normative question, Mallon points out that Glasgow shares commitments with both eliminativists and conservationists. On one hand, he shares the eliminativist view that "the folk term ‘race'…entails something untrue," namely, the claim "that race has an ‘adequate biological basis'" (Mallon, 2009, p. 2). On the other, Glasgow endorses a view Mallon calls "practice conservationism": Glasgow's reconstructionism "involves combining relatively conservative reform of existing sociolinguistic practices with an anti-realist metaphysics" (Mallon, 2009, p. 3). The rub, says Mallon (2009), is that a "conservationist might combine the same reform with a realist metaphysics" (p. 3). Despite their differences, all parties could agree that "‘race' used to entail belief in a biological kind" and "‘race', henceforth means a social kind" (Mallon, 2009, p. 3). According to Mallon, Glasgow and traditional conservationists disagree only about "whether the folk term ‘race' already refers to a social kind, perhaps because they disagree about whether the common but false folk belief in racial 69 biology is constitutive of the meaning of ‘race' or not" (Mallon, 2009, p. 3). But this is just the sort of disagreement that Mallon thinks "makes no difference in practice" (Mallon, 2009, p. 3). Mallon uses a similar strategy to questions Glasgow's semantic approach to the metaphysical question, so I skip the details here; it suffices to say that Mallon argues Glasgow's answer to the metaphysical question does not lead to any substantial conclusions or distinctions. Mallon says that Glasgow's conclusion leads to nothing more than "a disagreement about the meaning of words (and not one that makes a difference to practice)" (Mallon, 2009, 4). Glasgow (2009) responds that his answer to the conceptual question "does...make a difference to practice, concerning what we should do with racial discourse" (p. 13). As the many racially charged events of 2014 and early 2015 suggest, race is not going away. So if it turns out that Mallon is right, and the semantic strategy is misguided, we will not be able to just stop talking about race. We will need a methodological alternative to advance the discussion. Mallon offers one based on what he calls the "evolutionary-cognitive program" (Mallon, 2010, 2013). 3.3 The Evolutionary Cognitive Program Given that Mallon charges the semantic strategy with inefficacy, one would expect him to offer a methodological alternative that does matter to normative concerns, which is just what he does in a recent series of related papers. As a preview, the methodological shift Mallon favors is, like Glasgow's reconstructionism, informed by empirical, social scientific data, but Mallon makes quite different use of it. To this point, I have dealt only with the two most 70 commonly cited sets of race-relevant facts-the biological and the social-but Mallon complements the list with a third, the psychological. This move is the key to his methodological alternative to semanticism, so the rest of the chapter is devoted to introducing and motivating what Mallon calls "the evolutionarycognitive program" (henceforth, ECP) for philosophy of race. I proceed as follows. First, I review two of Mallon's recent attempts to explain ECP and the challenges it poses for social constructionism, which is the foundation of the received view in philosophy of race. I then conclude the section by introducing a collaborative effort between Mallon, Dan Kelly, and Edouard Machery in which they defend ECP's normative import by spelling out its implications for the ongoing debate between ‘race' talk eliminativists and conservationists. These combined efforts suffice to show ECP's greater potential to relevantly impact the normative debate than its rivals. To be clear, I use this final section to motivate inclusion of psychological research in our investigations of race; the detailed description and analysis of psychological research and its impact on philosophy of race come in the next chapter. To begin, Kelly, Machery, and Mallon note that "contemporary race theory is nearly devoid of effort to engage the burgeoning literature from social psychology and cognitive science on racial categorization and racial prejudice" (Kelly et al., 2010, p. 433). In "contrast to the attention paid to anthropological and historical factors, the philosophical literature on race fails to consider whether and how psychological factors could affect the feasibility of the various 71 normative proposals" (Kelly et al., 2010, p. 466). Taylor, in fact, not only ignores ECP but explicitly challenges its normative significance. Given the "centuries of cultural and social transformation during which we built up and forcefully promulgated comprehensive conceptions of human racial difference," Taylor wonders "why we need to appeal to some hard-wired mechanism that routinely cranks out organisms that indulge in racist exclusions" (Taylor, 2005, p. 38). Although I don't directly address Taylor's concern in what follows, showing that ECP impacts normative considerations in a way that semanticism cannot carries an implicit response-if its advocates can prove its normative weight, then Taylor's challenge is answered. Mallon, of course, takes the stance opposite Taylor's, arguing (both independently and collaboratively with Kelly and Machery) that understanding the "hardwiring" of racial cognition could be crucial to answering the normative question. Kelly, Machery, and Mallon deem the willful disregard of innate psychology in race theory "unfortunate" and "unjustified…[because] empirical research on racial cognition is directly relevant to the goals held by normative racial theorists" (Kelly et al., 2010, p. 433). They seek to rectify this omission by exploring "the intersection of…normative proposals with recent empirical work on psychology of racial cognition" and "aim…to demonstrate the need for normative racial philosophy to more closely engage contemporary psychology of racial categorization and racial prejudice" (Kelly et al., 2010, p. 433). Mallon (2010) advances the discussion in that direction, arguing that ECP disrupts the status quo. The "evolutionary-cognitive research program poses a 72 challenge to the received view about racial classification in social theory" (Mallon, 2010, p. 273). The positions that subscribe to the "received view," Mallon says, "share a commitment to ‘social constructionism" (Mallon, 2010, p. 272). This much is well-known; what is often missed is that "two quite distinct senses" of social construction operate within in race theory. The first, which Mallon calls "anti-racialism," is committed to the view "that race (the subject of racial classification) does not exist as a biological kind in the way ordinary or folk ideas of race seem to assume and is therefore ‘merely a construction'" (Mallon, 2009, p. 3). Of course, ECP theorists are a part of the ontological consensus against racialism, that is, they are as constructionist as anyone in this sense. The interesting sense of constructionism for this context, then, is the second. "Representational constructionism," Mallon says, "holds that racial classification itself is primarily the product of social and cultural practices" (Mallon, 2010, p. 272). In other words, work "in social philosophy on racial classification" is generally committed to the view that "we (as individuals and as cultural groups) have the theoretical representations we do, rather than some other theories or no theories at all because of historically and culturally specific conventions, decisions, practices, and so forth" (Mallon, 2010, p. 272). So, this type of social constructionism advances an empirical claim about how races and systems of racial classifications are formed. The evolutionary-cognitive program promotes a different explanation for racialized phenomena. So, to be clear, the issue at stake is an explanation of the root cause of human racial classification, and social theory and ECP are 73 committed to rival explanations of racialized phenomena. "Recent work by evolutionary and cognitive psychologists, anthropologists, and philosophers has posed a challenge to representational constructionism" by explaining "folk racial theories at least in part as the result of cognitive mechanisms which are culturally canalized, species-typical, and domain-specific" (Mallon, 2010, p. 272). This claim is crucial to ECP and receives thorough examination in the next chapter; here, a brief sketch of each characteristic suffices to motivate ECP's potential to contribute to race theory. First, a "culturally canalized" cognitive mechanism is one that has a property associated with (and often considered a condition for) innateness" (Mallon, 2010, p. 272). In contrast to social theory, which focuses on variation within classification schemata, ECP predicts that mechanisms associated with racial classification "develop stably across a wide range of different cultural environments" (Mallon, 2010, p. 272). Accordingly, in spite of variation in particular racial classification schemata, one should expect also to find among them many core commonalities. Second, the cognitive mechanisms implicated for race are also "species-typical" in that, "like having two arms and legs, eyes, ears, hair, and so forth, these cognitive capacities are traits that humans typically possess" (Mallon, 2010, p. 272). That is, being human, according to ECP, means coming equipped with a capacity and tendency to classify racially. Finally, "to say that they are ‘domain-specific' is to say that, unlike domain-general cognitive capacities (like memory, attention, or perception) that are employed across a wide range of problem domains, these mechanisms are 74 specialized for solving a particular sort of problem" (Mallon, 2010, p. 272). If, for example, human memory served only numerical recall, it would be domainspecific; since it instead allows recall of numbers, words, images, sounds, feelings, and so on, it is domain-general. To say, as ECP does, that the cognitive mechanisms underwriting racial classification are domain-specific is to say that they are "specialized for a particular sort of problem" (Mallon, 2010, p. 272). It is not to say, however, that they are adaptations specifically for racial cognition. As a matter of fact, the consensus within ECP is that the mechanisms are exaptations or "by-products of a mechanism that is adapted for something else" (Mallon, 2010, p. 272). Since the study of racial cognition has quickly become "an exceedingly complex affair...rife with controversy," in lieu of attempting to rule out all alternatives, Mallon (2010) offers a "series of considerations that illustrate and look to favor the idea that core aspects of racial cognition emerge stably in much of human development" (Mallon, 2010, p. 273). While Mallon says this is a "modest aim," his list raises significant obstacles for alternative approaches, which would have "to explain away" the predictive and explanatory advantages of ECP. As before, I simply introduce and briefly summarize them here, saving significant elaboration for the next chapter. First, Mallon (2010) reveals "a number of striking parallels" between folk biological and folk racial thinking. For example, research indicates that folk theories treat both biological kind membership and racial membership as "independent of superficial but prototypical properties" (Mallon, 2010, p. 75 274). Here, Mallon appeals to the work of Frank Keil (1989), who shows that preschool-aged children understand that a zebra dressed up to look like a horse remains a zebra (Mallon, 2010, p. 274). This style of essentialist biological judgment, which humans retain throughout adulthood, runs parallel to racialist judgments described earlier. Mallon (2009) offers another striking example: "seeming to have surface properties of one race but really being a member of another race is precisely what makes ‘passing' as a member of another race an apparently coherent idea" (p. 276). Racial "passing" occurs when a person of one race has features that give him or her the appearance of a member of another race, allowing that person to be accepted as a member of that other race. Following emancipation, for example, some light-skinned African Americans were able to "pass as White," conferring on them a measure of White privilege. Mallon's point, then, is that the phenomenon of passing makes sense only if judgment of racial membership runs deeper than the skin. The parallel with the zebra-horse example is patent. Folk biology says that the zebra remains a zebra, even if its appearance is that of a horse; species-typical human racial cognition says that a person of race R1 remains a member of R1, even if that person has the appearance of a member of R2. Given the vanishingly low probability of these parallels arising accidentally, Mallon (2010) suggests a more likely explanation: that they are "underwritten by a common mechanism" (p. 277). If that is the case, then long-established "evidence for the cross-cultural and early emergence of folk biology is also evidence for the canalization, domain-specificity, and species typicality of 76 mechanisms underlying aspects of folk racialism" (Mallon, 2010, p. 277). Despite its opposition to the received view, this agenda "is already being fruitfully pursued by evolutionary-cognitive theorist" in ways that motivate further study. I return to this topic at the beginning of the next chapter, so let me move on to ECP's second apparent advantage. The next consideration in favor of pursuing ECP is evidence "supporting the development of group cognition in advance of mastery of visual markers of groups" (Mallon, 2010, p. 280). Here, Mallon references two lines of work, one from Katherine Kinzler and her colleagues, whose conclusions I set aside for now. The other is from Lawrence Hirschfeld, whose work I briefly introduce here largely because of its impact on later chapters. Hirschfeld's studies result in an "extended argument that children's conceptual understanding of racial types seems to outstrip their ability to categorize people into race" (Mallon, 2010, p. 279). For example, data show that children misidentify their own race membership in perceptual tests using dolls, but not in verbal tests using labels. Furthermore, "children use race as a category in free sorting tasks, and also as a basis for racial prejudice, but they do not use racial categories in the selection of playmates" (Mallon, 2010, p. 279). Contra Hardimon, Hirschfeld believes these results suggest "children do not acquire knowledge of races by constructing categories based on perceptual difference"; instead they come equipped with a propensity that "leads them to acquire and organize beliefs regarding human groups well in advance of their ability to coordinate them with perceptual learning or with action" (Mallon, 2010, p. 280). ECP's innate cognitive 77 mechanism offers an explanation of the Hirschfeld data, and again, it is not clear how a social constructionist theory could accommodate this result. The third consideration in motivating ECP is that it provides an explanation for what Mallon calls "cross-cultural racialism" (Mallon, 2010, p. 277). Among social philosophers and social scientists, there is a "remarkable consensus" that racial classifications are temporally and geographically particular. This is a strong version of Michael Root's claim that "Race does not travel," by which he means things such as the fact that some "men who are Black in New Orleans now would have been octoroons there some years ago or would be White in Brazil today. Socrates had no race in ancient Athens" (Root, 2000, pp. S631S632). Or did he? Mallon (2010) thinks it more likely than Root because "across a broad range of (though not all) cultures, some human groups have been ‘racialized' in the sense that they have been seen as self-reproducing populations of individuals whose kind-typical differences are explained by unseen commonalities" (p. 277). To this end, Mallon (2010) makes a compelling "prima facie case that there are common, cross-cultural patterns of classification and inference" (p. 277) For example, in spite of apparent differences between ancient classification schemes and our own, ancient Greeks and Romans did label specific human groups…Crucially, even when theorists endorsed quite different accounts of the origin of human difference, they recognized that these differences were preserved in inheritance from parents to children over many generations. (Mallon, 2010, p. 277) Similarly, Medieval Europeans "recognized distinctive, reproducing human 78 groups, and they debated whether racial differences were evidence of multiple creations" (). Mallon cites similar attitudes in non-Western cultures, including evidence that historically both Chinese and Indian cultures exhibit deep similarities in racial thinking. These claims cast "some doubt on the thesis that genuinely racial thinking is a culturally local invention" (Mallon, 2012, p. 77). While social scientists tell stories about racial representations that appeal to "their cultural predecessors, historical and institutional context of their emergence, and theoretical and practical choices people have made regarding how to represent humans as members of groups," ECP "tells a different story, one on which essentialist thinking about human groups is itself, or is a product of, a psychological mechanism" with the aforementioned properties-innateness, domain-specificity, and specialization (Mallon, 2012, p. 77). So, in many cases, ECP displays more explanatory value than its social theoretical rival, which is a crucial point in its own right, but the question this chapter is intended to address remains: Mallon objected to the semantic strategy because it is normatively unimportant, even if explanatorily right. He offers an alternative approach-ECP. Let's assume it's right. Would that matter? Mallon and his collaborators say yes. "While interesting in its own right, the research on racial categorization in evolutionary psychology shows that there are some specific obstacles to the feasibility of eliminativism and conservationism that have been ignored by race theorists" (Kelly et al., 2010, 448). Kelly et al. (2010) note that arguments between the two "typically involve 79 evaluations of the costs and benefits attached to those agendas...which invariably involve background assumptions regarding the feasibility of the proposals" (pp. 433-434). While feasibility arguments are common along economic, legal, and social dimensions, ironically, "one dimension that is rarely considered in these assessments is their psychological feasibility, the ease with which eliminativists and conservationists goals can be reached given the psychological facts about human racial cognition" (Kelly et al., 2010, p. 438). If, for example, humans have evolved with species-wide innate mechanisms that are exapted for racial classification, then our ability to eliminate ‘race' talk might be compromised (Kelly et al., 2010, p. 438) So, Kelly et al. conclude that given the common goals of normative race theory, work on the psychology of racial categorization and racism is obviously relevant to assessing the ease with which (or the extent to which) such ideals can be realized. Moreover, if it turns out that certain ideals cannot be realized, that same psychological work will be useful in determining what sort of less-than-ideal goals are more attainable. (p. 433) So far, it seems ECP can do what the semantic strategy cannot-impact the normative question. 3.4 Conclusion I have covered a lot of ground in this chapter, so a quick review is in order. Mallon charges that the semantic strategy is a misguided one for race theory. In particular, it does not seem to bear on the most important questions about race, the normative ones. Left with a strategic void, Mallon (2010) recommends the evolutionary-cognitive program, which appears to this point at least to offer "an 80 explanation of important components of contemporary racial classification…that should be of interest to philosophers of race" (p. 273). So, in outline, the moves in this chapter are quite simple and straightforward, but the devil is in the details, which are taken up in the next chapter. CHAPTER 4 OPAQUE PSYCHOLOGY I: INNATE PSYCHOLOGY AND RACE 4.1 Introduction The previous two chapters were dedicated primarily to stating the negative (or critical) case that motivates my project. In the first chapter, I appealed to recent work by Gannett to show that both scholarly and folk racial discourse is "gappy"-the structure of current racial discourse is such that some important questions are precluded. In the second, I used a series of arguments raised by Mallon and other like-minded philosophers to show that one of philosophy's most venerated methodologies, the semantic strategy, cannot adequately address our normative concerns over race at the most general level-it does not, Mallon shows, significantly impact the question of whether to eliminate or conserve ‘race' talk. Of course, these gaps open space for alternate strategies that can bear the normative load. At the end of the last chapter, I introduced Kelly, Machery, and Mallon's collaborative philosophical work on the psychology of racial categorization and evaluation to transition to the positive phase of my dissertation, which aims at describing a conceptual framework for race that can do normative work. Before I can get there, though, I need to dedicate this and the next chapter to laying its foundation, which is informed by two distinct but 82 related bodies of research on the psychology of race. Specifically, the focus of Chapter 4 is the psychology of racial categorization, and of Chapter 5, the psychology of racial evaluation. To be clear, the former deals with the mechanism that lead humans to create racial classes and categorize individuals according to their relationship to those classes; the latter goes further, investigating how people judge the groups and individuals so categorized. To be clear, while the former appears to be subserved by an innate mental mechanism, the latter has no obvious connection to nativism. For convenience, however, I group them together under the heading "opaque psychology," owing to the fact that at least part of the tendencies both to classify and to evaluate by race operate outside of the conscious awareness of the agent. After laying out the evidence relevant to each topic, I conclude with a summary of consequences for the normative question. At minimum, last chapter's preview of Kelly, Machery, and Mallon's position supplied sufficient reason to further investigate possible influences of innate human psychology on racial cognition. To review, drawing on Mallon's earlier work, Kelly et al. (2010) tidily structure the normative debate in terms of two opposed camps, ‘race' talk eliminativism and ‘race' talk conservationism. Each promotes precisely what its name suggests: Eliminativists favor the relative short-term elimination of 'race' talk; conservationists counter that conserving racial categorization, albeit with modification, better serves our normative aims. In spite of their fundamental disagreement, then, both "are best thought of as revisionists: both suggest we reform our current practices of racial 83 categorization, but differ in whether it would be best to eliminate or rehabilitate them" (Kelly et al., 2010, p. 436). Casting the normative debate in this way yields dividends. Consider the following examples. First, the eliminativist-conservationist framing adds clarity to Gannett's by now familiar dichotomy concern. Since the dichotomy prevents investigation of contexts in which biological and social factors causally interact to constitute racialized phenomena, Gannett (2010) urges that we explore new, more versatile, context-sensitive ways of talking about race. So, in her treatment of the problem, Gannett promotes conservation with revision. She is, by all appearances, a pragmatic conservationist-conservationist because she clearly believes that ‘race' talk is still valuable, and pragmatic (at least in part) because she urges philosophers of science to de-emphasize metaphysics "and instead assess the appropriateness of group categories of classification relative to the purposes of specific research programs" (Gannett, 2010, p. 383). In sum, Gannett can be understood as assuming conservationism and moving on to ask "How or in what form(s) should ‘race' talk be conserved?" That is the primary question the dissertation aims to address. Second, Kelly et al. (2010) supply a normative framing that provides a proving ground on which the philosophical relevance of psychological research on race can be tested. If it could be shown that innate psychology significantly impacts racial cognition and, ipso facto, the eliminativist-conservationist debate, we would have at least one reason to think psychology of race is philosophically interesting. The choice between the two might hinge, for example, on whether 84 either normative position is even psychologically feasible. To take an extreme case, if the human propensity to classify by race runs so deep that it precludes elimination, then it is not clear that ‘race' talk eliminativism is a viable option. In other words, if racial cognition is psychologically ineliminable, then we might decide that ‘race' talk is normatively ineliminable. The central impediment to the normative aim, as I have shown, is the lack of a conceptual apparatus that can handle it. It is along these lines that I proceed in this and the next chapter. I begin this chapter with a more detailed delivery of the most important features of the evolutionary-cognitive program. First, I relate the conditions of its break from the social constructivist "received view" and then transition to some of the research from evolutionary psychology that informs ECP. I then consider both what appears to be the chief obstacle to ECP and Mallon's response to it. Finally, I review-in light of framework from Kelly et al. (2010)-the normative consequences of some of ECP's most robust conclusions about race. 4.2 Explanations of Racial Classification In the previous chapter, I introduced Mallon's claim that "the evolutionarycognitive research program poses a challenge to the received view about racial classification" (Mallon, 2010, p. 273). According to the "received view," race is a representational construct-that is, "a product of social and cultural practices" (Mallon, 2010, p. 273). More specifically, a consensus exists among social constructionists that "our racial representations are best understood by considering their cultural predecessors, the historical and institutional context of their emergence, and the theoretical and practical choices people have made 85 regarding how to present humans as members of groups" (Mallon, 2012, 77). Indeed, just as Mallon suggests, this view has long been orthodox. Consider another passage James's Stanford Encyclopedia of Philosophy entry on race The ambiguities and confusion associated with determining the boundaries of racial categories have over time provoked a widespread scholarly consensus that discrete or essentialist races are socially constructed, not biologically real…[but] scholarly debate exists concerning the formation and character of socially constructed, discrete racial categories. (James, 2012, n.p.) Yet, if Mallon, Kelly, and Machery are on the right track, the received view can't be the whole story. Kelly et al. (2010) concede that the ubiquitous tendency of humans to categorize by race "presents a puzzle for evolutionary-minded psychologists and anthropologists" (p. 439). The general modus operandus of evolutionary psychology involves describing evolved, innate, mental modules that explain some universal aspect of human behavior. The problem is that although people classify themselves and others by "putative racial properties" as a rule, evolutionary psychologists are highly dubious of the existence of a "race module"-that is, "an evolved cognitive system devoted to race and racial membership"-both because it is unlikely that morphologically different populations of humans came into frequent contact with one another and because it is "difficult to identify a selection pressure that would have driven early humans to pay attention to physical properties now associated with racial phenomena" (Kelly et al., 2010, p. 439). Given the unlikeliness of a mental module specifically devoted to race, evolutionary psychologists argue that racial cognition must be 86 "subserved by a module, but that the module in question was initially selected for some function…not related to race" (Kelly et al., 2010, p. 439). As such, disagreements among evolutionary psychologists generally range "over the nature and proper function of the cognitive system that now underlies racial thinking" (Kelly et al., 2010, p. 440). In general, then, they do not disagree about whether racial cognition is subserved by an innate psychological mechanism. So, even though details remain open for debate, ECP can and does proceed along the lines of this fundamental agreement and, as a result, "stands in contrast to previous explanations of racial categorization that have been offered in psychology and the social sciences" (Kelly et al., 2010, p. 440). The details of that contrast are informative and invite further analysis. I begin with the analysis of the received view supplied by Kelly et al. (2010). 4.3 Race on the Received View: Social Science Explanations According to Kelly, Machery, and Mallon, the social scientific paradigm of racial classification is supported by three types of explanation: socialization, perceptual salience, and group prejudice. All remain prominent in the social sciences, and, as such, provide the foil against which ECP is evaluated. The first and most recognizable social scientific explanation of racial categorization, socialization, posits that "children are either explicitly taught to draw the distinctions used in racial categorizations, or that they easily pick them up from the general social environment," even with no one "explicitly instructing them in the use of racial categories" (Kelly et al., 2010, p. 440). On this view, humans are racial blank slates at birth but learn, by both direct and indirect 87 means, to classify people by race. In essence, this view casts humans as products of the racial cultural environment in which they are born and mature. Without inculcation of culturally-specific classification schemata, folks would not classify by race simply because they would not recognize it. Second, some social constructivists claim that racial classification is a special case of a more general tendency to sort by "perceptually salient features" of the objects to be classified. A common strategy among perceptual salience proponents is to argue that "since color is a salient visual property, skin colors trigger this domain-general categorization system, and as a result, people form and rely on racial categorization" (Kelly et al., 2010, 440). Again, this theory is somewhat intuitive and popularly received. Recall from the previous chapter, for example, that the first of Hardimon's necessary conditions at race's logical core is distinction among humans by visible physical features of the relevant kind. In short, the view is that ‘color' (to name just one example) is an importantly meaningful category to humans, so since humans come in different colors, and since we place heavy importance on color, we tend to classify and judge racially. The final prominent social scientific explanation for racial classification centers on "a general tendency to form group prejudices about social groups, be they women, races, or social classes" (Kelly et al., 2010, p. 440). Importantly, on this view, race is no different from other human-kind categories. All result from a domain-general human propensity to sort ourselves and others by groups and to judge individuals based on the assumed group membership. 88 Mallon (2012) argues that each of these social constructivist explanations appeals, in one way or another, to what he calls the "Conceptual Break Hypothesis" (CBH), according to which, sometime in or since the Renaissance, some fundamental change occurred in the European and American tradition of thinking about the human groups that we now call ‘races'-a change in the concept, meaning, or theory by which people represent those groups. (p. 77) The change in meaning "involves failure of conceptual identity among older and newer concepts of human groups" owing to the fact that newer ones reflect the now largely entrenched belief that "in recent centuries individuals in the European-American cultural tradition began to conceive of race in an essentialist manner" (Mallon, 2012, p. 77). ECP proponents deny-in part or in whole-each of these social constructivist claims, including the three explanations of race and CBH, offering in their stead a significantly different approach. 4.4 Race on the ECP View In spite of a significant base of shared fundamental commitments, among evolutionary psychologists, "disputes have emerged about the specific character of our capacity to make racial classification" (Kelly et al., 2010, 445). So, even though evolutionary psychologists share a commitment to an innate, specieswide mental mechanism, they promote different evolutionary stories about its origin and function. This debate is not trivial: Kelly et al. contend that "resolution of their disagreements may have an impact upon the debate between eliminativism and conservationism" (p. 445). To see how, I briefly consider 89 evolutionary psychology's three most prominent explanations of racial cognition. 4.4.1 Races as Coalitions First, Robert Kurzban, John Tooby, and Leda Cosmides (2001) "hypothesize that the (apparently) automatic and mandatory encoding of race is...a byproduct of adaptations that evolved for an alternative function," namely, "detecting coalitions and alliances" (p. 15387). According to Kurzban et al. (2001), our human ancestors "would have benefited by being equipped with neurocognitive machinery that tracks...shifting alliances" so long as it was "sensitive to two factors: (i) patterns of coordinated action, cooperation, and competition; (ii) cues that predict-either purposefully or incidentally-each individual's political agendas" (Kurzban et al., 2001, p. 15387). Because the hypothesized mental "circuitry detects correspondences between allegiances and appearance, stable dimensions of shared appearance...emerge in the cognitive system as markers of social categories" (Kurzban et al., 2001, pp. 15348715388). So, just like other "ethnographically well-known examples" of observable features (such as dress, manner, or gait), race markers (such as skin color) "acquire social significance and cognitive efficacy when it validly cues patterns of alliance" (Kurzban et al., 2001, pp. 153487-15388). On this line, then, "racial categorization results from a cognitive system whose function is to track coalitions (i.e. groups of individuals who cooperate with each other) in a given social environment," and since it tracks "coalitions in the social environment, it picks out races in those modern societies" (Kelly et al., 2010, p. 445). Moreover, according to Kelly et al. (2010), Kurzban and colleagues 90 "provide some intriguing evidence that adults' encoding of skin color and racial membership is influenced by whether racial membership is a relevant cue to coalitional membership" (Kelly et al., 2010, 446). In their experiments, Kurzban et al. (2001) showed test subjects photographs of individuals paired with a sentence uttered by that individual. They were then informed that each pictured person belongs one of two basketball teams that engaged in a fight with the other in the previous season. The subjects were then asked to "form an impression of the target individuals as they viewed the photos" by attempting to accurately match the individual to his or her uttered sentence. To test the coalitional effect, Kurzban et al. (2001) then surveyed each subject's mistakes to determine whether the subject erred in ascribing the utterance to a player of the same race or to one on the same team (Kelly et al., 2010). The results of the experiments tended to support the hypothesized conclusion. When team membership "was not emphasized, participants implicitly categorized the individuals involved in the verbal exchange according to race" (Kelly et al., 2010, p. 446). When members of each multirace team were given distinctively colored jerseys, however, "participants appeared to rely much less on race" (Kelly et al., 2010, p. 446). So, Kurzban and colleagues conclude that "in the absence of any obvious indicators of coalitional boundaries, racial membership is often taken to be a cue to coalitional membership" (Kelly et al., 2010, p. 446). This phenomenon is likely quite familiar to those who, like me, are either from or have connections to the United States' southeastern region. I can report 91 (anecdotally, of course) that many Southern folks capable of expressing the most racist attitudes seem not to mind or even notice that the player who just put their favored football team ahead is from a different race. To this point, Kelly et al. (2010) say that the coalition "hypothesis explains why, when other indications of coalitional membership are made particularly evident or social environments make coalitional boundaries more salient, people are less prone to classify into races" (p. 446). 4.4.2 Races as Ethnies A second hypothesis about innate features of racial cognition comes from Francisco Gil-White, who is motivated by a worry that "establishing the ontological fact" that races are not discrete biological natural kinds "may have clouded our understanding of local epistemologies" so that we "may be failing to take seriously that ethnic actors are themselves essentialists" (Gil-White, 2001, p. 515). In other words, Gil-White warns that the ontological consensus against racialism does not obviate the "need to investigate why ordinary people often believe" in racialism. The failure of biological essentialism about race does not render racialism moot. In fact, Gil-White's investigation leads to the hypothesis that social categories, including race, "are processed by the machinery which evolved to deal specifically with ‘natural living kinds' of the ‘folk-species' ranklevel such as BEAR or MOUSE" (Gil-White, 2001, p. 517). In other words, GilWhite concludes that "evolution has selected for an ethnic cognitive system, that is, a cognitive system whose evolved function is to identify ethnic groups" (Kelly et al., 2010, p. 446). As a result, "humans process ethnic groups...as if they were 92 ‘species'" (Gil-White, 2001, p. 515). Kelly and colleagues offer a brief but instructive synopsis of Gil-White's central argument. Gil-White's story centers on "ethnies," or ethnic groups, which "are in the first instance collections of individuals sharing a common self-ascription, but with no necessary relation to any particular content" (Gil White, 1999, p. 792). In the recent evolutionary past, human ancestors lived in ethnies "made up of (at least) several hundred or thousand culturally homogeneous members," who represented their ethnie membership "by means of specific ethnic markers" (Kelly et al., 2010, p. 456). Gil-White "maintains that it was important for our ancestors to map this dimension of the social world" (Kelly et al., 2010, p. 447). In addition, he points to evidence that suggests humans "have a privileged biological domain of cognition," more specifically, that we have a "living-kinds mental module" (GilWhite, 2001, p. 519). From there, he argues that "folk biology-the set of commonsense beliefs about animals and biological kinds together with the cognitive systems responsible for classifying and reasoning about animals and biological kinds-was recruited or ‘exapted' for the purpose" of navigating the social world by discerning ethnie-membership (Kelly et al., 2010, p. 447). In the past, when "a human perceived an ethnie, it counted it and processed it as a species. Over time...the brain evolved to...make ethnies part of the ‘proper domain' of the living-kind module, completing the exaptation" (Gil-White, 2001, p. 519). Since this living-kind module "essentializes the entities it classifies, we now tend to essentialize the groups we discern on the basis of these ethnic markers" (Kelly et al., 2010, p. 447). The human mind perceives ethnies as species and 93 races as ethnies, so since "skin color and other racial properties (such as body type) are often taken to be ethnic markers...races can be mistaken for ethnies by the ethnic cognitive system, despite the fact that they are, in general, not ethnies" (Kelly et al., 2010, p. 447). Just like the Kurzban et al. presentation, Gil-White's conclusions support familiar intuitions about racial thinking, most importantly, its close similarity and connection to biological thinking. 4.4.3 Race as Interactive Finally, Lawrence Hirschfeld (1996) offers an interactionist account of racial classification influential enough to warrant extra attention. He begins with a description of "some contradictory aspects" of race, namely, that classification often involves both "contrived taxonomies of difference directly linked to specific cultural, political, and economic traditions" and "robust, seemingly self-evident, widely rehearsed, and highly shared beliefs about the meaning and nature of human difference" (Hirschfeld, 1996, p. 2). In short, Hirschfeld (1996) reveals something akin to a paradox-racial classification schemes are clearly marked by both culturally particular and species-universal aspects. Since investigations of these apparently antinomous aspects of racial classification "remain largely independent of each other," Hirschfeld aims "at two principal goals: to reinterpret the tension between the universal and the specific in race and to provide an account of racial thinking that adequately captures both of these characteristic qualities" (Hirschfeld, 1996, p. 2). Hirschfeld's investigation is informed by his experimental research on the racial attitudes of young children, which he uses to argue that racialized 94 phenomena result from an "interaction of an innate, evolved capacity for folk sociological thinking, on the one hand, and the specific social structure in which it is operating on the other" (Kelly et al., 2010, p. 445). Noting that racial thinking stabilizes "quickly and seemingly effortlessly in the minds of children" (Hirschfeld, 1996, p. 2), Hirschfeld argues that his research indicates that "3- to 7-year-old preschoolers treat skin color differently" from other characteristics (Kelly et al., 2010, p. 445). He attributes this phenomenon to an innate folk sociological mechanism, whose function "is to identify the social groups in the social environment," an ability that was most likely selected for given the importance for humans to map their social world (Kelly et al., 2010, p. 445). One of the most important aspects of this mechanism "is that it essentializes whatever groups are salient in a given social environment" (Kelly et al., 2010, p. 445). As a result, "membership in these groups is associated with a set of immutable properties thought to be cause by some essence common to all group members" (Kelly et al., 2010, p. 445). While the first part of Hirschfeld's story of racialization centers on an innate, folk-sociological, essentializing mental mechanism, the second involves the particular environmental stimuli introduced to the innate mechanism. Putting these together, Hirschfeld argues that when "societies are divided along racial lines, the folk sociological mechanism guides us in the identification and essentialization of these groups" (Kelly et al., 2010, p. 445). Again, two causal factors-one internal, one external-interact to inform racial thinking. The social structure of a given society works in conjunction with the innate psychological 95 mechanism to determine which "different social groups will be picked out and essentialized" (Kelly et al., 2010, p. 445). While Hirschfeld (2001) acknowledges that it "is widely believed that the notion of race derives directly from the spontaneous perception of physical variation," he argues that "the idea of race involves more than an awareness of surface differences" (p. 1). The complexity of the idea of race demands "an epidemiology of racial representations," which constitutes a "first step in a jointly psychological and cultural account of why racial ideas come to be widely distributed, easily transmitted, and predictably transformed" (Hirschfeld, 2001, pp. 3, 2). Hirschfeld's epidemiology of racial representations suggests that even before humans are cognitively capable of reflecting on race, racial cognition covertly manifests as a sort of social competence. The kind of people there are is largely a function of community, and much of our social life turns on acquiring the dispositions, sentiments, and predilections (to name versions of the same thing) of the communities to which we belong. Relying on a vision too narrow or parochial-or too broad and too deracinated-one risks incorrectly inferring the community-based conventions, standards, and practices that apply. (Hirschfeld, 2012, p. 27) Again, this competence does not result, as some proponents of the received view suggest, from mere generalization over particular visual representations of race. Instead, children come to racial knowledge via both "a predilection to attend to input relevant to community standards and conventions" and "the agency of low levels of attention and engagement" (Hirschfeld, 2012, pp. 22-23). Moreover, children do not require socialization; they "seem to develop this knowledge largely on their own and do it without committing social 96 (as opposed to cognitive) resources" (Hirschfeld, 2012, p. 26). Citing a debt to Hirschfeld, Stephen M. Quintana and Clark McKown (2008) corroborate these claims, arguing that "children have an innate inclination to perceive grouping in their social environment" and that race "appears to be one of the more important groups into which children naturally sort their social world" (p. 17, emphasis added). 4.4.4 Common Themes for ECP Although significant disagreement remains regarding how the "cognitive system believed to now underlie racial categorization is structured, and what it initially evolved to do," the three positions limned above share a robust core of common commitments (Kelly et al., 2010, p. 447). At the most general level, for example "each of the evolutionarily informed positions...sees racial categorization as a by-product of a fairly specialized cognitive system that evolved to deal with some specific aspect of the social environment" (Kelly et al., 2010, p. 441). But there are other, more specific points of comparison. The first is one that, perhaps surprisingly, ECP shares with the received view: the recognition of the tendency of race to be essentialized. The question of the origin or cause of the essentialist thought, however, generates considerable debate. In spite of broad scientific rejection of biological essentialism, "the folk belief that racial groups…have essences is widespread," so any viable account of race must be able to account for that belief (Hirschfeld, 2008, p. 41). Again, according to the conceptual break hypothesis favored by social constructionists, "essentialist thinking about race is a relatively recent phenomenon" resulting from 97 "a culturally local product of the modern era" (Mallon, 2012, p. 78). This constructivist consensus "suggests that our racial representations are best understood by considering their cultural predecessors, the historical and institutional context of their emergence, and the theoretical and practical choices people have made regarding how to represent humans as members of groups" (Mallon, 2012, p. 78). In contrast, ECP predicts that racialism "should emerge relatively robustly across human cultures and history" instead of being "a culturally and historically local product of modern European and American thought" (Mallon, 2012, p. 78). Clearly, then, the ECP account of racial essentialism "is at odds with understanding the content of a Conceptual Break as essentialist," a fact that fosters "reason to doubt the truth of the Conceptual Break Hypothesis" (Mallon, 2012, p. 79). Essentialism is a foundational component of racial categorization for each account, but their differential causal stories predict different outcomes. A consequence of this is the creation of a standard by which we may adjudicate between the commitments of the competing research programs: if the empirical evidence better supports one or the other, we will have a basis upon which to choose (at least preliminarily). Tabling that point for the moment, I can move to the more detailed account of the core commitments of ECP, namely, the three properties of the posited innate mechanism that I previewed in the last chapter. To review, the evolutionary-cognitive program characterizes the mental module in question "at least in part as the result of cognitive mechanisms which are culturally canalized, species-typical, and domain-specific" (Mallon, 2010, p. 98 272). First, a mechanism that is culturally or environmentally canalized is one that is "roughly speaking...the same across different environments and environmental variables" (Kelly et al., 2010, p. 444). Or, as Mallon (2010) says, calling "them culturally canalized here is to say they have a property associated with (and often considered a condition for) innateness: they develop stably across a wide range of different cultural environments" (p. 272). To be clear, however, innateness does not entail immutability. To say that the mechanism is "innate is not to say that it is unchangeable. ‘Innateness' concerns only the process by which a trait develops or is acquired" (Mallon, 2012, p. 78). Although it is true that assertions of innateness are meant to communicate "a strong degree of developmental invariance across a range of cultural environments," Mallon says that "such invariance says nothing about whether the process or processes that produce it can be interrupted or their outcome altered" (Mallon, 2012, p. 78). This, of course, could turn out to be important for normative concerns. Second, ECP holds that the mental mechanism implicated in racialization is domain-specific. As I indicated in the previous chapter, racialization is in this respect unlike domain-general cognitive capacities, such as memory, attention, or perception, in that it is "specialized for solving a narrower problem or problems" (Mallon, 2010, p. 272). The faculty of attention, for example, is designed to make us aware not only of potential threats to our lives (from predators, say), but also to keep us alert to potential mates or beautiful vistas or, later on, conceptual problems. The mental mechanism that gives rise to racial 99 cognition, in contrast, appears to be designed for a small, closely related set of problems (such as categorizing organisms in terms of usefulness), even though racial categorization was not among the original set of problems. Lastly, being equipped for racial cognition is, according to ECP, little different from having "two arms and legs, or eyes, or ears, or hair"-they are, in other words, speciestypical (Mallon, 2010, p. 272). Put simply, the cognitive capacities in questions are ones "that humans usually possess" (Mallon, 2010, p. 272). With these properties described and with the controversy over the sources of racialism established, I can now turn to the advantages that ECP appears to hold over the received view. 4.4.5 The Verdict In promoting ECP, Kelly et al. (2010) offer "five lines of argument that undermine the socialization, perceptual salience, and group prejudice explanations" (p. 441). This section is devoted to a review each. I begin with the argument from Kelly et al. (2010) that "group prejudice is unable to account for the differences between different types of social classification and the different types of prejudices associated with each" (p. 442). Recall that group prejudice explanations count racial cognition as a subspecies of more a general classification strategy that picks out social groups. Kelly et al. argue that if such a domain-general capacity accounted for a broad range of social groupings, one would expect uniform stereotypes for each group, but this is not what one finds. Instead, "stereotypes about social groups vary substantially from one type of group to the next" (Kelly et al., 2010, p. 442). 100 For example, the stereotypes associated with racialized groups and political groups bear little in common-races are "thought to be biological kinds," while Democrats and Republicans are not. In a paper linking implicit attitudes, epidemiological studies, and physician recommendations, Kevin Schulman and colleagues provide additional prima facie evidence for this distinction, arguing that "the race and sex of a patient independently influence how physicians manage chest pain" (Schulman, 1999, p. 618, emphasis added). So, even the two most prominent social categories, race and sex, appear to generate different stereotypes. This important point has lurked behind the scenes since the more general discussions of social construction in the first two chapters. Borrowing from that language, the very structure of many things social theorists treat as representationally constructed appears different, but the group prejudice explanation is "unable to account for the different types of social classification and the different types of prejudices associated with each" (Kelly et al., 2010, p. 442). As a result, saying that race is socially constructed is explanatorily insufficient because races are imbued with far different properties than (say) bankers, an uncontroversially constructed kind. In sum, "if all prejudicial stereotypes were produced by a unique cognitive system, or were driven by a single, general tendency to form stereotypes about social groups, we should not expect to find such differences" as ones between race and political parties, bankers, and so on (Kelly et al., 2010, p. 442).9 We tend to use the term "stereotype" as though it were only one sort of thing and stereotyping one sort of process, but as these distinctions show, "stereotype" is not univocal. In a 9 101 Second, and surprisingly, in spite of the universal association between race and skin color, perceptual salience explanations fare little better than group prejudice explanations, primarily because "they take for granted one of the very things they are supposed to be explaining, namely why people classify each other on the basis of phenotypic properties like skin color" (Kelly et al., 2010, p. 441). Recall that, like group prejudice, perceptual salience theories invoke domain-general human tendencies to explain the propensity to classify racially. Again, color is an important general category to humans-I have favorite colors, I favor certain colors of cars or houses or shirts, and at the market, I select produce (in part) by its color. Perceptual salience theories explain racial categorization in like manner: since color is an important general category, the facts that humans sort by color and come in different colors make racial categorization virtually inevitable: "since color is a salient visual property, skin colors trigger this domain-general categorization system, and as a result, people form and rely on racial categorization" (Kelly et al., 2010, 440). But, Kelly et al. counter that color "is not always intrinsically salient, or an important feature for categorization purposes" (p. 443). They point out that even when we "pay attention to the color of artifacts, we rarely treat it as a property that is important for classificatory purposes" and that children trained to use a tool of a particular color generally show a preference for tools of similar shape, rather similar color forthcoming paper, Erin Beeghly offers both a perspicuous account of four (relatively) distinct scholarly explanations of stereotypes and a compelling argument, based on her "descriptive account," that stereotyping is not intrinsically morally objectionable. I say more about this point in the next chapter, which centers on implicit bias and so offers a better context in which to discuss some finer points of about stereotypes. 102 (Kelly et al., 2010, p. 443). Given the inconsistency of color's status as a categorization stimulus, "the salience and importance of skin color needs to be explained, not assumed" (Kelly et al., 2010, p. 443). Hirschfeld (2012) makes a similar claim, calling the reduction of race to skin color and other salient physical properties the "perceptual myth" and arguing that children do not "discover race by opening their eyes and looking" (p. 22). Naturally, Hirschfeld (2012) does not deny that visual input is relevant. "It is not that perceptual input is completely irrelevant to emerging racial categories. Rather...in building racial categories, obvious surface cues like skin color or the shape of the facial features or qualities of hair are not defining for young children" (p. 32). So, perceptual cues matter to racial cognition, but "they simply underdetermine them" (Hirschfeld, 2012, p.32). Although anecdotal evidence about children's reactions support this "myth" and despite the fact that "the idea that race is a visual phenomenon is almost universally held," Hirschfeld says, "there are few studies actually assessing whether visual information is initially relevant to learning about race" (Hirschfeld, 2012, p. 23). In fact, his own 1993 study suggests another environmental cue to racialization: children appear to attend more closely to linguistic than to visual information regarding race (Hirschfeld, 2012, p. 23). While interpretations of that study are many and controversial, Hirschfeld says, "early representations of race seem to be based on information children acquire by listening to those around them talk about social differences rather than by attending to physical differences that ‘cry out to be named'" (Hirschfeld, 2012.). 103 Third, Kelly et al. (2010) use ECP to challenge the view that children classify racially because they are socialized to do so. A common intuition along these lines is that children simply adopt the racial classification schemes of their parents, but the evidence suggests otherwise. If socialization were the correct explanation, "one would expect children's beliefs about races to be similar to their parents' beliefs. However, this is not true" (Kelly et al., 2010, p. 441). Again, Hirschfeld corroborates the claim. "Several research studies have demonstrated that children's racial and ethnic biases are not reliably associated with the beliefs and attitudes of parents or peers. Parents and teachers, in this case as in others, wildly overestimate their influence in shaping children's beliefs" (Hirschfeld, 2008, p. 42). Of course, the "dissociation between parents and their children constitutes…evidence against socialization explanations of the dispositions to categorize racially" (Kelly et al., 2010, p. 441). Furthermore, in "Seven Myths of Race and the Young Child," Hirschfeld reports that we tend to overestimate children's racial naiveté, thinking that children are "by nature innocent of race," that "even if they are aware of race, they are without prejudice," that they "have to be taught prejudice," and that they "believe race is a superficial quality, literally just skin deep" (Hirschfeld, 2012, pp. 22-28). But research in cognitive psychology suggests instead "that this portrait of young children's beliefs about race grossly underestimates what young children understand about race" (Hirschfeld, 2012, p. 28). Fourth, Hirschfeld's research on children's understanding of race provides compelling and provocative evidence that skin color is a special property for 3-to 104 7-year-old preschoolers, who "treat skin color differently from other properties-in particular, they expect skin color to be constant over lifetime" (Kelly et al., 2010, p. 443). Echoing a claim in the previous section, Hirschfeld argues that the special status children confer to race reflects "a kind of intuitive essentialism" (Kelly et al., 2010, p. 443), even though, as Hirschfeld (1996) says, "we now know that one that cannot explain a racialist mode of thought is the phenomenon of race itself. Humans are not discriminable into discrete, self-evident biological kinds" (p. 3). Finally, ECP's most controversial commitment is to innate psychology, which is supported by evidence for culturally-canalized racialism, which in turn suggests that "some aspects of contemporary racial representations are developmentally invariant across cultures" both geographically and temporally (Mallon, 2012, p. 78). Proponents of the received view struggle to account for this aspect of racial cognition for the very reasons named above. Given their commitment to CBH, social constructionists explain racialism by appeal to "cultural, historical, and institutional contexts, and the theoretical and practical choices people intentionally make in representing others and themselves as members of groups" (Mallon, 2012, pp. 77-78). Recall that this claim has been cited twice already in different forms: Gannett highlights the argumentative advantage of the claim that race as it is socially constructed (rather than biologically constituted) is essentialist, and James references the scholarly consensus that "essentialist races are socially constructed, not biologically real" and another, closely related, consensus about 105 the modern invention of race, given the presumed fact that references to race concepts are absent from ancient cultural and religious traditions. None of these groups, says James, "sought to classify humans into discrete racial categories...differences such as skin color and hair texture were noticed but did not ground discrete categories of biological difference" (James, 2012, n.p.). ECP is committed to the opposite claim, however. "Evolutionary psychologists hold that people in many cultural and historical epochs...have relied on classifications that are similar to modern racial categories" (Kelly et al., 2010, p. 441). Those classifications are similar in that they are "supposed to be based on phenotypic properties" and are "assumed to map onto biological categories" (Kelly et al. 2010, p. 441). In short, a large number of current and historical cultures "have relied on skin color and other bodily features to classify their fellows, and have further believed that such classifications also group together people who share underlying biological commonalities" (Kelly et al., 2010, p. 441). To be clear, Kelly, Machery, and Mallon are not blind to variations in racial categorization; instead, they want only to stress that "core elements of racial categorization are not a merely parochial cultural phenomenon" (Kelly et al., 2010, p. 442). Given its commitment to innate, species-wide tendencies to racialize, the universal aspects of racial cognition are just what ECP would predict, while "socialization accounts cannot explain why these core elements should recur across times and cultures," and, as a result, the received view is "at best incomplete" (Kelly et al., 2010, p. 442). To summarize, evidence generated by the evolutionary-cognitive program 106 suggests, in direct opposition to the received view, that racial categorization "develops early and reliably across cultures…does not depend entirely on social learning…[and] is, in some respects, similar to biological classification" (Kelly et al., 2010, p. 443). In other words, the evidence "is best explained by the hypothesis that racial categorization results from a specialized, species-typical system underlying racial thought" (Kelly et al., 2010, p. 444). Kelly et al. do not deny "that socialization plays some role"; they only "insist it is not the whole story" (p 440). The other part of the story, of course, is that racial classification "is underwritten by an evolved cognitive system, whose development in children is to a large extent independent of teaching and socialization" (Kelly et al., 2010, p. 440). In addition, they recognize the importance of perceptual salience in race thinking but object to "the idea that racial categorization can be explained merely by the perceptual salience of skin color" (Kelly et al., 2010, p. 441). Finally, even allowing that nonracial social classifications share many properties with racial ones, they question the general group prejudice explanation, arguing "that not all social classifications and prejudices behave the same" and "that racial cognition should be distinguished from other forms of group-related cognition" (Kelly et al., 2010, p. 441). In place of these explanations, ECP favors ones that describe the human propensity to categorize racially as resulting from "a cognitive system that has evolved to deal with a specific domain in the social world, rather than with categories or perceptual salience in general" (Kelly et al., 2010, p. 440). 107 4.4.6 Normative Consequences of ECP My primary focus in this dissertation is, of course, the conceptual question, but, unlike Hardimon, I aim to develop a conceptual framework in such a way that it can be used to address the normative question. So, even though I only touch on some specific normative issues and only at the very end, it is worth reiterating that Kelly, Machery, and Mallon are offering not merely an account of racial classification, but also an investigation of whether and how ECP impacts the normative debate between eliminativists and conservationists. I covered the general outline of this argument in the previous chapter, so I will not rehearse it in its entirety here. In general, it should suffice to say that our ability to eliminate, conserve, or revise ‘race' talk depends on the psychological feasibility to do so (Kelly et al., 2010, p. 448). If the innate structures that inform racialization are ineliminable or unrevisable, then eliminativist aims will be moot, but Kelly and colleagues hasten to add that "without further argument, such an evolutionary account of racial categorization in no way implies that racial categorization cannot be eliminated or modified" (Kelly et al., 2010, p. 444). By way of comparison, they point out that the human taste for sweetness, although innate, species-wide, and domain-specific, does not entail that a "person's taste for sweetness is...inevitable or completely impervious to modification" (Kelly et al., 2010, p. 444). Lots of things, many of which are under my control, will determine how much I can control my sweet tooth. Remember also, however, that there is not just one evolutionary story; there are (at least) three, and, although the "dust has not settled yet...the 108 resolution of the [ECP theorists'] disagreements may have an impact on" the resolution of the normative question. Since Hirschfeld's and Gil-White's respective views "suggest that racial categorization and essentialism...are the product of the same cognitive system...conserving racial categorization while reforming its normative connotations may be hindered by the nature of the evolved cognition system" (Kelly et al., 2010, pp. 450-451). In contrast, since Kurzban et al. (2001) argue that essentialism and the salience of racial properties result from distinct cognitive mechanisms, there is nothing in "the nature of human psychology to prevent the dissociation between racial categorization and its essentialist implications" (Kelly et al., 2010, p. 451). Since each view implicates innate structures in racialization, however, each has similar implications for eliminativism. 4.4.7 A Challenge for ECP As I suggested in the last chapter, the evolutionary-cognitive program faces significant resistance-recall that Taylor (2005) depicts it as a red herring of sorts, something that distracts us from real-world individual and institutional racial oppression. There is admittedly something counterintuitive and even offputting about the notion that my behavior is influenced by something that originates in me but resides outside of my consciously aware "self." Answering Taylor's criticism is an aim of the dissertation itself, so I forestall its direct address for later. Counterintuition and queasiness are not genuine objections, of course, but they can serve as useful alarms, so before shifting the subject from racial classification to racial evaluation, I want to consider a couple of challenges to the 109 story so far presented. The most obvious concern is that evolutionary psychology is itself very controversial, so positions that depend on its results inherit that controversy.10 The rebuttal is simply that mine is not one of those positions. Although I do draw on evolutionary psychology, the position that I promote is dependent neither on its general program of research nor on any of its particular theses. For my purposes, any theoretical successes of evolutionary psychology program would bolster my position, but it depends only on convincing evidence of universal aspects of human schemes of racial categorization, and that much, I believe, has been secured. For reason made clearer in later chapters, even the unlikely event of the absolute failure of evolutionary psychology would not doom my position. The more interesting and formidable challenge results from cultural variation in racial classification. That such variation exists is indubitable. Misalignment between U.S. and Brazilian racial classification, for example, has been well-rehearsed at least since the Carl Degler's 1971 Pulitzer-Prize-winning study, Neither Black nor White. The problem this poses for evolution-based solutions is patent: ECP posits a species-wide racializing mechanism despite the existence of multifarious racial schemata. The uncontroversial fact of cultural variability poses the biggest challenge to ECP. For details of those controversies, see Stephen Downes' lucid treatment of evolutionary psychology in The Stanford Encyclopedia of Philosophy. Especially informative is his fourth section, "Philosophy of Biology vs. Evolutionary Psychology," in which he notes that most critics of the latter are "philosophers of biology who argue that the research tradition suffers from an overly zealous form of adaptationism" (Downes, 2014, n.p.). Among other things, Downes notes that "Many philosophers object to evolutionary psychologists' over attribution of adaptations on the basis of apparent design" (Downes, 2014, n.p.). 10 110 At first blush, Mallon's response is shocking. He does not, per impossible, try to talk away the variation. He instead, embraces it, arguing that "cultural variation is to be expected on (some versions of) the evolutionary-cognitive account" (Mallon, 2012, p. 85). ECP, in other words, accommodates cultural variability. Mallon's defense of this claim is formidable. He begins by appealing to considerable evidence of parallels between biological and racial cognition-most importantly, each involves "assumptions about underlying essences…[and] inheritance thinking" (Mallon, 2012, p. 81). That is, both prescientific biological cognition and racial cognition assume "a wide range of different kinds have ‘underlying' natural properties" that are passed from generation to generation (Mallon, 2012, p. 85). While it is possible that these shared assumptions developed independently, a more plausible explanation is that a single psychological mechanism generates both. Mallon (2012) suggests, for example, that "whatever mechanism underlies racial cognition was initially adapted to thinking about (certain kinds of) biological kinds, and then perhaps later exapted for thinking about social kinds" (p. 85). Mallon (2012) then asks us to suppose that the proposed mechanism "was originally adapted to thinking about members of fully or largely reproductively isolated biological populations" so that its "proper domain" is "classic species" (p. 85). Now, in contrast to the proper domain of the mechanism (again, species), "humans of different races are not members of different species" because "although human populations sometimes exhibit partial reproductive isolation...human populations frequently and readily interbreed" (Mallon, 2012, p. 111 85). So, on Mallon's interpretation, "the mechanism is originally adapted to populations that rarely interbreed [species], but it is being applied to populations that frequently interbreed [human races]" (Mallon, 2012, p. 85). As a result, folk racial classifications mirror folk biological classifications, even though human races do not mirror species, and this mismatch helps to explain universal aspects of human racialization. "Evolutionary cognitive accounts suggest that we ought to expect strong overlap in classification judgments of nonmixed race persons between the United States and Brazil, and this is just what we find" (Mallon, 2012, p. 85). For example, in each country "a child of two parents of the same race is typically classified as a member of that race" (Mallon, 2012, p. 85). Culturally variant racial classification is real; nevertheless, a "number of experiments suggest that lineage essentialism exists in a range of cultures" (Mallon, 2012, p. 85). One could take the acknowledgement of this variation as vindicating "social constructionist explanations at the expense of evolutionary psychological ones," but although evolutionary psychologists insist "on a kind of developmental regularity...they need not insist that this regularity cannot be, and is not sometimes altered" (Mallon, 2012, p. 86). In short, ECP predicts broad regularities in cross-cultural racial classification without denying that social practices, historical accidents, and political dynamics (to name a few factors) could forge particular differences among the schemata. This alone should secure the point that ECP (and, hence, psychological constructivism) and social constructivism are not mutually exclusive research programs. I pick this crucial point back up in Chapter 6. 112 Here (as well as in number of other cases), it is useful to compare race to language. Recognizing universal regularities among human languages and even positing, as Noam Chomsky does, a species-typical language-generating mechanism in no way entails a denial of the differences among particular languages. In fact, given our understanding of human evolution and the partial isolation of populations throughout human development, an evolutionaryuniversalist linguist is bound to predict differences in languages despite their shared evolutionary origin. Likewise, an ECP theorist should expect racial classifications to reflect differences in detail to be constructed upon broadlyshared foundations. As Mallon (20012) says, "evidence shows that the fact of cultural variation by itself does not undermine the evolutionary-cognitive thesis that the mind is evolved with a predisposition to essentialize such identities" (p. 85). ECP theorists recognize that "a wide range of factors determines what individuals and groups believe and do, but influence by the sort of mechanisms posited by evolutionary-cognitive theorists is surely one of them" (Mallon, 2012, p. 86). Mallon (2012) neatly summarizes the situation: "In effect, the content of our racial representations is determined by a range of forces that include both domain-specific predispositions to believe (of the sort evolutionary-cognitive theorists emphasize), but also cultural and social forces (of the sort that social constructionist theorists emphasize)" (p. 86). 113 4.5 Conclusion Many of the general conclusions that I want to draw on for the position I defend are also informed by the content of the next chapter, which switches from racial classification to racial evaluation. Because of that, I delay many of the important conclusions until then. For now, a summary of the commitments commonly shared amongst ECP theorists will suffice. First, humans appear to be biological essentialists about race despite the fact that biology does not support essentialism. Moreover, evidence suggests the tendency to racialize is subserved by a mental mechanism that is culturally-canalized (or innate), domain-specific, and species typical. Although each claim informs my position, it will turn out that even they are more specific than my thesis necessitates. As I will show at the end of the next chapter, I need to establish only that a significant part of racial cognition, representation, and classification is caused by mechanisms and processes that operate outside of human consciousness so that social construction, while remaining relevant, cannot be the whole story. CHAPTER 5 OPAQUE PSYCHOLOGY II: IMPLICIT ATTITUDES AND RACE 5.1 Introduction In the previous chapter, I focused on the relationship between innate psychology and racial categorization. Another interesting set of questions at the nexus of psychology and race concerns racial evaluation, that is, judgment, bias, and stereotype with regard to race. Especially interesting are the cases in which those attitudes are implicit rather than explicit. 5.2 Explicit and Implicit Bias Most of us are familiar with explicit racial bias, which is marked by a range of intentional, conscious attitudes-explicit biases are just those that "can be consciously detected and reported" (Amodio & Mendoza, 2010, p. 355). Explicit bias is "controllable, intended, made with awareness, and requires cognitive resources" (Nosek, 2007, p. 65). To take an obvious example, Klu Klux Klan members exhibit explicit racial bias in favor of Whites and against non-Whites. Explicit bias is generally measured directly through self-report questionnaires, such as the one that informs the "Modern Racism Scale," which "poses statements explicitly about racial issues...and allows participants to react to each 115 statement by selecting, at their leisure, one of the responses, which range from Strongly Disagree to Strongly Agree" (Kelly et al., 2010, 455). Research on explicit bias is important, of course, but direct measures have significant shortcomings. According to Ohio State University's Kirwan Institute for the Study of Race and Ethnicity, the "downfalls of self-reports have been well-documented" (Staats, 2013, p. 15). In general, the aims of self-reports are frustrated by "impression management," which arises from "the desire to be perceived positively" and "can influence people to distort their self-reported beliefs and attitudes" (Staats, 2013, p. 15). Impression management is "particularly likely when individuals are questioned about politically or socially sensitive topics such as interracial or intergroup behaviors" (Staats, 2013, p. 16). So, although early research on race and ethnicity relied on self-report, "as norms discouraging prejudice gained society traction, straightforward approaches to measuring bias became less useful and increasingly suspect" (Staats, 2013, p. 16). In spite of those norms against prejudice, racism and racial tension persists, of course, so the unreliability of explicit measures necessitates complementary measures-indirect ones that aim at uncovering implicit bias. Implicit attitudes are those that are not explicit, that is, ones that are not intentional or consciously made.11 Harvard University's Project Implicit treats 11 In a 2015 paper, Christopher Jenson reports that psychological studies informed by self-report and nonverbal behavior indicate that "‘belief is fragile," that is, "multiple independent means of detecting them produce highly variant results," which makes ‘belief' "a strong candidate for elimination...from cognitive science" (Jenson, 2015). Of course, Jenson's thesis, if vindicated, could bear on what is said here about the connections between race and implicit attitudes. For now, however, I proceed with the standard conceptual vocabulary, including "belief." 116 implicit attitudes as "positive and negative evaluations that occur outside of our conscious awareness and control" (https://implicit.harvard.edu). The Kirwan Institute adds that the "main distinction between implicit and other types of bias centers on level of awareness" (Staats, 2013, p. 12). So, while subjects can consciously detect and report on explicit biases, implicit biases "occur without introspective awareness"-they are, for the person experiencing them, undetectable and, therefore, unreportable (Staats, 2013, p. 12). Since the experiencer cannot detect them, researchers attempt to reveal implicit bias through indirect measures, that is, "techniques...that do not rely on introspection or self-report" (Kelly et al., 2010, p. 452). Indirect measures are intended to "bypass" the obstacles to conscious and authentic self-report, such as the aforementioned impression management phenomenon (Kelly et al., 2010, p. 452). The following two examples-one informal and intuitive, the other formal and analytic-illustrate the difference and bring clarity to the phenomenon of implicit attitudes. For the first example, imagine that I'm hosting a company picnic in San Luis Obispo, a town on California's Central Coast, roughly equidistant from San Francisco (to the north) and Los Angeles (to the south). For the example, it is important to know that baseball fandom in SLO is roughly evenly divided between the noble and mighty San Francisco Giants and the lowly and iniquitous Los Angeles Dodgers. As a loyal Giants fan, I am, of course, honor-bound to "hate" the Dodgers, and so I occasionally and in jest exhibit explicit bias in favor of the Giants and against the Dodgers, just as I did when I described them above 117 as "noble" and "iniquitous," respectively. Suppose the guests at my picnic, like the town itself, are split evenly with regard to their loyalty to one or the other of these bitter baseball rivals. Now, I love baseball and especially the Giants, but I maintain a reasonable attitude toward those affections-baseball is entertainment; it is not to be taken too seriously. As a result, I don't make conscious decisions to favor one person over another based on the team they choose to support (even if it is the Dodgers). In fact, I'm so off put by aggressive sports fan behavior that I amend the company's bylaws to formally eliminate obnoxious fan-based favoritism. I tease my Dodger-fan co-workers and friends, but I do not take their fandom into account when I hire, fire, promote, praise, or criticize, nor do I choose friendship with Mr. Matthewson over Mr. Robinson simply because the former loves the Giants and the latter the Dodgers. To my shock, however, Dodgers fans at my picnic notice a troublesome trend: Many little things seem to at least slightly favor the Giants fans in attendance. It's a hot day, for example, and the Giants are seated in the shade. Though I did not consciously intend it, for some reason or another, the Giants fans are generally first in line and get the better food choice. Worse, all disputes in the sack race, three-legged race, and balloon toss are adjudicated (by me) in favor of teams composed of Giants fans. Even more disturbingly, in the days after the picnic, some folks begin to notice subtle forms favoritism at work-Giants fans who need office equipment, for example, receive their wares on average 24 hours before Dodgers fans do. Even if I can honestly say that-to the best of my knowledge-I do not make 118 professional decisions based on team preference, the more I look, the more I discover gaps in treatment. So, what gives? How can I explain this pattern of outcome, if it is granted that I'm being honest about my explicit attitudes, and if I make explicit effort to promote justice at work, in life, and when officiating competitive picnic games? One possible and increasingly momentous explanation is that while I do consciously avoid Giants-fan favoritism, I might hold and act on unconscious or implicit attitudes about Giants and Dodgers fans. Project Implicit offers a similar, simpler example: "Even if you say that you like math (your explicit attitude), it is possible that you associate math with negativity without knowing it. In this case, we would say that your implicit attitude toward math is negative" (https://implicit.harvard.edu). I might even be good at math, be on the math team, and sport T-shirts emblazoned with the explicit identifier "MATHLETE." And, yet, I might also unthinkingly shudder when Lumosity.com presents me with math games and routinely opt for its linguistic and spatial alternatives. Despite these intuitive examples, the very notion of attitudes invisible to the bearer may initially appear too ethereal to matter, but a mounting body of evidence suggests that implicit attitudes can be dissociated from explicit attitudes, can be measured, and are implicated in important ways in human behavior. These three points are crucial to my purposes, so I elaborate on each. 5.2.1 Are Implicit and Explicit Attitudes Dissociable? First, in order to talk substantively about implicit attitudes, they must be dissociable from explicit attitudes. Evidence suggests that they are: The claim 119 that "implicit and explicit attitudes do not align" is supported by a "vast body of empirical literature" (Staats, 2013, p. 13). Patricia Devine (1989), for example, offers a dissociation model that distinguishes between automatic processes that are "unintentional or spontaneous" and controlled processes that are "intentional and require...active attention" (pp. 5, 6). Devine's research shows that "automatic and controlled processes can be dissociated" (Staats, 2013, p. 13). In fact, at this point, the dissociability thesis has been so thoroughly corroborated that it is taken for granted. As Kelly and colleagues say, "the fact that implicit and explicit...biases can be dissociated is no longer the subject of much controversy" (Kelly et al., 2010, footnote 23). But dissociability entails neither measurability nor consequence. These questions about implicit bias remain open and require separate justification. 5.2.2 Can Implicit Attitudes Be Measured? The primary difficulty in measuring implicit biases is patent: If they are invisible even to their bearer, then self-report will be inefficacious. The hidden attitudes must be teased out. If this challenge could be met, however, indirect measure would result in several significant advantages over direct ones. Again, direct measures of explicit attitudes can be corrupted by insincerity of self-report, self-deception, and incapacity for clear introspection. Since indirect measures, by contrast, are "measurement methods that avoid requiring introspective access, decrease the mental control available to produce the response, reduce the role of conscious intention, and reduce the role of self-reflective, deliberate processes" (Nosek et al., 2007, p. 267), they have "the advantage of bypassing 120 all three of these obstacles" (Kelly et al., 2010, p. 452). Even so, measuring implicit bias remains quite challenging. As a result, researchers have invented sophisticated measurement methods, three of which-physiological approaches, priming methods, and Implicit Association Tests (IATs)-have contributed significantly to race scholarship. Because of its influence and popularity both within the academy and in the media, I begin with the last, focusing on it before briefly returning to the others for purposes of comparison. Kelly, Machery, and Mallon describe the IAT as essentially a "sorting task" that involves making associations between two pairs of dichotomous categories, the "target concepts" and the "attribute dimensions" (Kelly et al., 2010, p. 453). For example, race IATs might pair the target concepts ‘Black' and ‘White', represented by pictures of Black and White faces, with attribute dimensions ‘good' and ‘bad', represented by individual words such as "wonderful" or "terrible," respectively. Subjects are given multistage tests in which they "sort the exemplars from the four categories using only two responses" (Kelly et al., 2010, p.453). In the first stage, subjects might be prompted to press keyboard letter "e" when shown either a Black face or a good attribute or "i" when shown either a White face or bad attribute. "Crucial to the logic of the test" is that in subsequent stages, the response options (the "e" and "i" keys) are assigned to different categories" (Kelly et al., 2010, p. 453). Subjects are directed to answer and correct sorting errors as quickly as possible before moving on to the next prompt. The test computer measures and records reaction times. Nosek and colleagues explain that "the logic of the IAT is that this sorting 121 task should be easier when two concepts that share a response are strongly associated" and that "ease of sorting can be indexed both by speed of responding (faster indicating stronger associations) and the frequency of errors" (Nosek et al., 2007, p. 270). In short, differences in response times and accuracy are correlated with magnitude of bias, and "This is called the IAT effect" (Kelly et al., 2010, p. 453). Since IATs bypass the subject's conscious awareness-and so, for example, remain "resistant to social desirability concerns" (Staats, 2013, p. 25)-"the associations thus revealed are taken to be indicative of processes that function implicitly and automatically" (Kelly et al., 2010, p. 453). For example, if a subject consistently sorts more quickly and accurately when White faces, rather than Black faces, are paired with good attributes, the result "is interpreted as an indirect measure of a stronger association between the two categories good and White, and hence an implicit preference for White, or, conversely, an implicit bias against Black" (Kelly et al., 2010, p. 453). Other indirect measurement techniques corroborate IAT results. As Mahzarin Banaji reports, "the family of implicit attitude measures is becoming increasingly diverse" (Banaji, 2001, p. 122). In addition to IATs, experimenters use the startle eyeblink test, which measures affective responses by determining the magnitude of eyeblink response to a given stimulus, and priming tests. In addition to the IAT, Banaji focuses on evaluative priming. In evaluative priming, a subject is shown an object (a "prime") about which s/he has pre-established attitudes (Banaji, 2001, p. 122). The prime is presented to the subject for a brief 122 time and is "followed by an evaluatively congruent or incongruent but semantically unassociated a target," that is, an evaluative word such as "good," "enemy," "anguish" (Banaji, 2001, p. 123). The point is to reveal the extent to which the prime "influences how an individual later responds to a different stimulus" (Staats, 2013, p. 24), and, just as with the IAT, the "speed to judge the target is taken as an indicator of the strength of evaluative association between attitude object and evaluation" (Staats, 2013, p. 24). The foundational assumption of evaluative priming "is that if an attitude object (e.g., flowers, ice cream) evokes a positive evaluation, it should facilitate response to other evaluatively congruent (i.e., positive) co-occurring information" and vice versa (Banaji, 2001, p. 122). The interpretation of priming techniques as a measure of "automatic attitude," says Banaji, appears "to be well accepted" (Banaji, 2001, p.122), and Kelly et al. (2010) report that "a wide variety of other indirect measures, including evaluative priming…the startle eyeblink test...and EMG measures" support IAT data (p. 456). Mention of electromyography (EMG), which tracks facial movement in response to a given stimulus, supplies a segue to physiological approaches, the last of the aforementioned indirect measures. EMG and other physiological measures "assess bodily and neurological reactions to stimuli" (Staats, 2013, p. 22). Eric Vanman and colleagues, for example, "investigate the relationship of implicit racial prejudice to discriminatory behavior" by showing subjects "photos of Blacks and Whites" and using EMG to measure movements in muscles used in smiling and frowning (Vanman et al., 2004, p. 711). Another common 123 physiological technique uses functional Magnetic Resonance Imaging (fMRI) to record correlations between strategically selected stimuli and activation of the amygdala, a part of the brain implicated in decision making, memory, and emotional learning and evaluation (Staats, 2013, p. 22). Elizabeth Phelps and colleagues, for example, show "that representations of social groups that differ in race evoke differential amygdala activity and that such activation is related to unconscious social evaluation" (Phelps et al., 2000, p. 733). Phelps' study reveals that "members of Black and White social groups can evoke differential amygdala activity" (Phelps et al., 2000, p. 734), and, more specifically, that "White subjects generally showed greater amygdala activation when exposed to unfamiliar Black faces compared to unfamiliar White faces" (Staats, 2013, p. 22). Despite differences of opinion over cause and consequence, additional studies once again corroborate this result-distinct racial stimuli yield different strengths of amygdala activation, although the subject remains unaware of it. Moreover, researchers have revealed correlations between fMRI data on amygdala activation and other measures of race evaluations. Phelps et al., for example, were the first to relate "indirect behavioral measures of social evaluation to neuronal activity," showing that "the strength of amygdala activation to Black-versus-White faces was correlated with...indirect (unconscious) measures of race evaluation," including IATs, "but not with the direct (conscious) expression of race attitudes" (Phelps et al., 2000, p. 734). The point of all this detail is, of course, to show that implicit attitudes can be measured in a variety of ways which yield relatively consistent results. 124 5.2.3 Do Implicit Attitudes Affect Judgment and Action? As Kelly et al. (2010) say, "a natural question to ask...is whether or not the biases revealed by indirect measurement techniques have any influence on judgments or ever lead to any actual prejudicial behavior, especially in real world situations" (p. 456). Skepticism over this issue manifests in two ways: dismissal as "mere linguistic associations or inert mental representations" and, even to those congenial to indirect measures, doubt that implicit biases "are powerful enough to make any practical difference in day-to-day human affairs" (Kelly et al., 2010, 456). Kelly et al. argue "that such skepticism is unjustified" (p. 456). Indeed, as I hinted above, indirect measures are in many cases complementary and mutually corroborative. Moreover, indirect measures of racial bias, in particular, offer explanations of continued racial problems, despite apparent decreases in explicit racial bias. Likely more important, however, is "evidence of race and racial bias influencing real-world situations" that measures of explicit bias are at a loss to explain. One of the most popular and accessible examples of this effect is Joseph Price and Justin Wolfers' 2007 statistical analysis of National Basketball Association (NBA) officials. Price and Wolfers discovered that in the NBA, the world's top professional basketball league, more personal fouls are called against players when they are officiated by an opposite-race refereeing crew than when officiated by an own-race crew. These biases are sufficiently large that we find appreciable differences in whether predominantly Black teams are more likely to win or lose, based on the racial composition of the refereeing crew. (Price & Wolfers, 2007, p. 1859) Specifically, the study, which traced foul calls over 12 seasons, "found 125 evidence that White referees called slightly but significantly more fouls on Black players than White players, as well as evidence of the converse" (Kelly et al., 2010, p. 457). Price and Wolfers argue that this effect is explained by implicit bias. Indeed, a good case can be made for ruling out widespread explicit bias. First, the NBA is a highly multinational, multiracial, multiethnic professional entertainment organization that far outpaces its industry rivals in socially relevant ways. For example, the University of Central Florida's Institute for Diversity and Ethics in Sports' "2013 Racial and Gender Report Card" awarded the NBA an overall grade of 90.7 (out of 100), including an A+ for racial hiring practices and B+ for gender hiring practices (Lapchick, 2013, p. 1). Lapchick et al. say that "the NBA remains the industry leader among the men's sports for racial and gender hiring practices. No one else reaches the same points for race, gender or the combined score" (Lapchick, 2013, p. 1). Moreover, although more than 70% of NBA players are African American, the NBA boasted a record 101 international players representing 37 countries and territories on the opening day of the 2015 season (nba.com). So, few explicit racists would likely pursue a career in such an organization; ones who did would likely be quickly discovered and ousted. In addition, NBA "referees are subject to constant and intense scrutiny by the NBA itself, so much so that they have repeatedly been called ‘the most ranked, rated, reviewed, statistically analyzed and mentored group of employees of any company in any place in the world' by [former] commissioner David Stern" (Kelly et al., 2010, p. 457). 126 Finally, even if that thinking appears naïve, explicit attitudes likely do not impact the foul rates simply because of the speed at which the game is played and officiating decisions must be made. Implicit bias is a more likely cause of the racial disparities because of the "types of split-second, high-pressure evaluations required of NBA referees" (Price & Wolfers, 2010, p. 1861). Initially, Price and Wolfers defend the somewhat modest claim that "these results...are at least suggestive that implicit bias may play an important role" (Price & Wolfers, 2010, p. 1885). By the next year, however, Price and Wolfers are ready to put their money where their thesis is, joining co-author Tim Larsen in arguing that "in games where the majority of the officials are White, betting on the team expected to have more minutes played by White players always leads to more than a 50% chance of beating the spread" (Larsen et al,. 2008, p. 1). They add that the "probability of beating the spread increases as the racial gap between the two teams widens" (Larsen et al., 2008, p. 1). Later studies promote similar conclusions. For example, Christopher Parsons and colleagues found that Major League Baseball (MLB) "umpires express their racial/ethnic preference when they evaluate pitches. Strikes are called less often if the umpire and pitcher do not match race/ethnicity" (Parsons, Sulaeman, Yates, & Hamermesh, 2011, p. 1410). Just as in the NBA case, explicit bias is an unlikely culprit here. Both by rule and in practice, the boundary between a ball and strike is vague, so in many cases, MLB umpires must make instantaneous judgments of borderline pitches, all under close official scrutiny. Studies like these both reaffirm the (now largely uncontroversial) view that 127 implicit and explicit attitudes are dissociable and support the view that the former as well as the latter impact behavior. So, they are important. They also come with limits. Both leave open several questions important to the dissertation topic. Even after establishing the potency of implicit bias in real-world contexts, both the NBA and MLB studies are silent with regard to the controlled, clinical settings of indirect measures. In short, these studies suggest that implicit judgments affect behavior but leave open the question whether the attitudes they reveal are related to those uncovered by IATs. Perhaps the two data sets are unrelated. Investigation into a possible connection between real-world settings and IAT performance suggests the opposite, however. In fact, researchers have been successful in forging connections between investigations like the NBA study and data from indirect measurements. "Indeed, the sorts of real-world findings coming from these sorts of statistical analyses and field studies, on the one hand, and the types of automatic and implicit mental processes revealed by the likes of the IAT, on the other, appear to complement each other quite nicely" (Kelly et al., 2010, p. 457). Furthermore, "researchers have begun to...explicitly link indirect measures with behavior in controlled settings" (Kelly et al., 2010, p. 457). A striking example involves evidence that research subjects are "more apt to misidentify a harmless object as a gun if they are first shown a picture of a Black, rather than a picture of a White" (Kelly et al., 2010, p. 457). B. Keith Payne argues that this phenomenon, now known as the "weapons bias effect…correlates with performance on the racial IAT" (Kelly et al., 2010, p. 458). Echoing earlier claims, Payne (2005) reports that "several lines of research 128 have shown that group stereotypes may be activated outside the awareness and may influence behavior without the knowledge or intent of the perceiver" (p. 181). Payne supplements that research with his own by integrating "insights and techniques from social and cognitive psychology to help understand" important real-world issues, including weapons bias. This particular real-world effect allows me to steer back my primary target-the relevance of implicit bias to race matters. 5.3 Implicit Attitudes, Indirect Measures, and Race Today there is broad scholarly agreement that IATs reveal attitudes that affect real-world outcomes. In fact, legal scholar Jerry Kang and psychologist Kristin Lane say that "After a decade of research, we believe that the IAT has demonstrated enough reliability and validity that total denial is implausible" (Kang & Lane, 2010, p. 477). Moreover, Greenwald, Banaji, and Nosek's metaanalyses of earlier and apparently contradictory studies reveal that the IAT effect is "large enough to justify concluding that IAT measures predict societally important discrimination" (Greenwald et al., 2015, 559). IAT analysis implicates implicit attitudes as factor in several important areas, including formation of political attitudes (Hawkins & Nosek, 2012) and voting decisions (Friese et al., 2012), in choice of college major (Greenwald & Nosek, 2009), and probability of receiving research grants from the U.S. National Institute of Health (Ginther et al., 2011). This delivers us to the heart of the present matter-the connection between implicit attitudes, IATs, and race. The connection is robust. "Some of the first and most consistently 129 confirmed findings yielded by these tests...center on racial bias" (Kelly et al., 2010, p. 455). In fact, some of those findings are now decades old. Spurred by the fact that racial stereotypes persist even as they "become more subtle" (Gaertner & McLaughlin, 1983, pp. 29-30), Samuel Gaertner and John McLaughlin, for example, show that "research subjects, regardless of their personal prejudices, were reliably faster at pairing positive attitudes with Whites than Blacks" (Staats, 2013,p. 13) and that "negative traits are more highly associated with BLACKS than positive traits are, while the reverse is true for WHITES" (Gaertner & McLaughlin, 1983, pp. 29-30). Later studies reveal similar effects in a broad range of racialized contexts, some of the most important of which have already come up-weapons bias and research grant approval, to name two examples. So, when it comes to the connection between implicit bias and race, the "real-world effect…is increasingly difficult to deny" (Kelly et al., 2010, p. 459). Furthermore, since "it is psychologically possible to be, and many Americans actually are, explicitly racially unbiased while being implicitly racially biased" (Kelly et al., 2010, p. 450), exploring the connection could add clarity to remaining mysteries surrounding racial inequality and disparity in treatment between Whites and racial minorities. This sort of investigation could, for example, "help explain familiar anecdotes of sincerely egalitarian people who are surprised when they are called out for racist behavior or biased decision-making, especially when such accusations turn out to be legitimate" (Kelly et al., 2010, p. 459). They have the potential to explain even more serious matters. I elaborate on three: implicit bias in criminal justice, in health and health care, and in 130 education. These are among the most salient contexts and are crucial to the development of the position I develop in final chapter. 5.3.1 Implicit Bias in Health Care The first context of concern, health care, should come as no surprise. The opening chapter dealt with Gannett's concerns over race in biomedical contexts. Health care is an increasingly salient issue both in the U.S. and internationally, and, at least in contemporary America, the effect of race on health and health care is a growing concern. As Staats (2013) reports, "The presence of prevalence of racial disparities in health and health care across a wide array of ailments have been documented extensively" (p. 47). The effect likely results from a combination of several factors-lifestyle choice, biomedical factors, and social/environmental factors (Staats, 2013, p. 47). The Kirwan Institute report that Staats authors focuses on the last of those factors because they provide the initial medium through which implicit bias is realized in real-world medical situations. Substantial research backs correlations between implicit bias and racial disparities in health and health care. Michelle van Ryn and colleagues, for example, report that "thousands of studies have demonstrated that Blacks receive lower quality medical care than Whites" and that "Despite this, there has been little progress toward eradicating these inequalities" (van Ryn et al., 2011, pp. 199-200). Crucially, according to van Ryn et al., these disparities "might be influenced by implicit racial biases," as empirical evidence shows that White medical clinicians "hold negative implicit racial biases...that persist independently 131 of and in contrast to their explicit (conscious) racial attitudes" and that White clinicians "can be influenced by racial bias in their clinical decision making" (van Ryn et al., 2011, p. 200). Moreover, just as in the criminal justice context, "studies have documented the presence of implicit bias in a variety of facets of the health/health care industry" (Staats, 2013, p. 47). In a 2009 study, for example, Janice Sabin and colleagues argue that health care providers' "implicit attitudes about race contribute to racial and ethnic health care disparities" (Sabin et al., 2009, p. 897). According to Sabin et al. (2009), studies show that "internal medicine and emergency medicine physicians hold strong implicit attitudes favoring White Americans over Black Americans" (p. 898). In particular, multiple studies have shown that physicians implicitly associate Black patients with being "less cooperative" and/or White patients with being "compliant" (Sabin et al., 2009, p. 898). Gordon Moskowitz and Jeff Stone, along with colleague Amanda Childs, found that doctors show "an implicit association of certain diseases with African Americans. These comprised not only diseases African Americans are genetically predisposed to, but also conditions and social behaviors with no biological association" (Moskowitz et al., 2012, p. 996). They conclude that "diagnoses and treatment of African American patients may be biased, even in the absence of the practitioner's intent or awareness" (Moskowitz et al., 2012, p. 996). Staats (2013) reports that studies also support the conclusion that "implicit biases have been shown to affect the type(s) and quality of care that patients of various races receive" (p. 48). Among the most interesting is Alexander Green 132 and colleagues' use of the IAT to supply "compelling evidence of implicit bias among physicians" (Staats, 2013, p. 49). Green et al. argue that IAT scores used to measure physicians' implicit race bias predict differential racial treatment. "As physicians' proWhite implicit bias increased, so did their likelihood of treating White patients and not treating Black patients with thrombolysis" (Green et al., 2007, p. 1231). Similar results hold in studies of pediatricians' treatment recommendations for pain, urinary tract infection, attention deficit hyperactivity disorder, and asthma (Sabin & Greenwald, p. 2012). Finally, implicit attitudes appear to factor in "doctor-patient communication" (Staats 2013 p. 50) with studies linking unconscious bias to "White physicians'...less positive interactions with Black patients" (Staats, 2013, p. 50). Again, in at least some cases, IAT results are consistent with real-world outcomes. Lisa Cooper and colleagues use two IATs to show that "Among Black patients, general race bias was associated with more clinician verbal dominance, lower patient positive affect, and poorer ratings of interpersonal care" among other things (Cooper et al., 2007, p. 979). Adding all this together, Staats says the "impact of implicit biases in healthcare should not be understated" (Staats, 2013, p. 51). 5.3.2 Implicit Bias in Criminal Justice The second contextual issue, implicit bias in criminal justice, is the most topical and inflammatory. From 2014 to 2015, there were several highly controversial cases of unarmed Black men being killed by White police officers in various U.S. cities. Although the police harassment and brutality of Blacks and other racial minorities is nothing new in the United States, the heightened 133 publicity fueled by bystander smartphone videos of these most recent cases has finally brought the issue to the forefront, increasing both racial awareness and racial tension at the national level. In this section, I focus on the role implicit racial bias plays in the differential treatment of racial minorities, especially Black males, in the U.S. criminal justice system. Before getting to the role of implicit bias, however, I am compelled to say a word about recalcitrant explicit bias. By focusing on implicit bias, I do not wish to reinforce the misguided view that explicit racism is virtually extinct. Though it is arguably to some extent muted in contemporary American culture, explicit racism remains rampant. A striking and representative example is supplied in the Department of Justice's report on the Ferguson, Mo. Police Department after the murder of Michael Brown, an unarmed Black man, by a White police officer: Ferguson's law enforcement practices are shaped by the City's focus on revenue rather than by public safety needs. This emphasis on revenue has compromised the institutional character of Ferguson's police department, contributing to a pattern of unconstitutional policing, and has also shaped its municipal court, leading to procedures that raise due process concerns and inflict unnecessary harm on members of the Ferguson community. Further, Ferguson's police and municipal practices both reflect and exacerbate existing racial bias, including racial stereotypes. Ferguson's own data establish clear racial disparities that adversely impact African Americans. The evidence shows that discriminatory intent is part of the reason for these disparities. (United States Department of Justice, 2015, emphasis added) Clearly, explicit bias remains operative, and each source of racial bias merits attention and analysis. But, while vigilance against explicit racism is now the norm-the fact that the U.S. Justice Department publicly addressed the Michael Brown incident is testament to that-the role of implicit attitudes, by 134 contrast, has been almost entirely neglected, particularly outside of the academy. Evidence suggests that we neglect that effect to our detriment. The case for implicit bias as a causal factor in police action and criminal justice is robust and broad. I have already touched on the most famous of these cases-the weapons bias effect, which, again, is the tendency of both Whites and Blacks to misidentify harmless objects as weapons if held by a Black rather than White subject. But the concern runs much deeper. Staats reports that implicit bias "can surface in the criminal justice system in a variety fashions, all of which may potentially taint the prospect of fair outcomes" (Staats, 2013, p. 36). What is worse, implicit bias can have a cumulative effect in the criminal justice system so that "even small implicit biases can accumulate over the course of legal proceedings, thereby amplifying the effect" (Staats, 2013, p. 36). Kang et al. (2012) report that that within the criminal justice system "implicit bias is pervasive...and predicts certain kinds of real-world behavior" (pp. 11301131). Even when these "effects are deemed ‘small' by social scientists," they "may nonetheless have huge consequences for the individual, the social category he belongs to, and the entire society" (Kang, et al., pp. 2012 1143). Implicit bias can intervene during the initial police encounter and accumulate through the charge and plea bargain phase, and the trial phase-via both jury and judge-all the way through sentencing. "In each of the stages of the criminal trial process...the empirical research gives us reason to think that implicit biases...could influence how defendants are treated and judged" (Kang et al., 2012, p. 1151). In sum, 135 For a single defendant, these biases may surface for various decision makers repeatedly in policy, charging, bail, plea bargaining, pretrial motions, evidentiary motions, witness credibility, lawyer persuasiveness, guilt determination, sentencing recommendations, sentencing itself, appeal, and so on. Even small biases at each stage may aggregate into a substantial effect. (Kang et al., 2012, p. 1151) Moreover, Kang et al. (2012) indicate through "meta-analysis of IAT studies...that implicit attitudes as measured by the IAT predicted certain types of behavior, such as anti-Black discrimination or intergroup discrimination, substantially better than explicit bias measures" (p. 1131). Although work remains, it appears safe to conclude both that implicit bias is relevant to racial dynamics in the criminal justice system and that IATs at least partially track its effects. This relatively moderate conclusion will turn out to be sufficient for my purposes, so I turn now to the last of the contexts. 5.3.3 Implicit Bias in Education Finally, the effects of implicit attitudes are relevant to educational contexts, and just as before, implicit bias appears to affect educational settings in several ways, "all of which can yield disadvantageous consequences for students of color" (Staats, 2013, p. 30). This topic centers on three key themes: "teacher expectations of student achievement, teacher perceptions of student behavior, and students' self-perceptions" (Staats, 2013, p. 30). Since this topic is organized as those above, I limit my explication of each to a few brief conclusions. Staats (2013) reports that a Dutch study12 combining teacher self-reports 12 See van den Bergh et al. (2010). 136 and IATs indicates both that teachers tend to have "differential expectations of students from different ethnic groups" and that "implicit attitudes were responsible for these differential expectations as well as the ethnic achievement gaps" (p. 31). Another13 reports that teachers tend to hold lower expectations" for non-White children, a phenomenon that has been tied to academic performance (Staats, 2013, p. 31).14 With regard to teacher perception of student behavior, Staats reports studies have associated implicit bias with negative evaluations of "Black walking style" and with the perception that Black students are more threatening than White students (Staats, 2013, p. 32).15 Moreover, these attitudes "may be further amplified by a cultural mismatch that exists between Whites and their students of color, and this mismatch can lead to teachers misinterpreting student behavior (Staats, 2013, p. 32). Finally, says Staats, studies suggest that fears "of being viewed through the lens of negative stereotype," a phenomenon known as "stereotype threat," manifest in "lower performance by the stereotyped group, even when the stereotyped group and non-stereotyped group...have been statistically matched in ability level" (Staats, 2013, p. 33-34).16 In sum, students, like medical patients and persons subject to the criminal justice system, are susceptible to not only individual, but also cumulative effects of each vector of implicit bias. Clearly, the list of studies backing the real-world effects of implicit bias is 13 See Tenenbaum & Ruck (2007). See Rosenthal & Jacobson (1968). 15 See Neal, McCray, Webb-Johnson, & Bridgest (2003) and Hugenberg & Bodenhausen 14 (2003). 16 See Steele & Aronson (1995) and R.F. Ferguson (2003). 137 impressive, but all parties concede that implicit attitude research is still new and its conclusions relatively tentative. Much work remains. "To be forthright, the psychological story is still far from complete, and in a number of ways" (Kelly et al., 2010, p. 460). Open questions include the generalizability of IAT results from culture to culture and beyond Black and White, ontogenesis of implicit bias, and the details of the cognitive architecture, including whether implicit biases of different types (e.g., race, gender, and age biases) are generated by the set of mechanisms (Kelly et al., 2010, p. 461). Even so, the empirical research establishes a number of claims: A large body of evidence indicates that implicit biases exist, and are fairly prevalent…[that] They are different from, and can coexist with, their explicit counterparts…[and that] field studies...complement work done in controlled experimental settings, strongly suggesting that implicit biases indeed affect judgment and behavior, even in real-world situations. (Kelly et al., 2010, p. 462) Implicit bias almost certainly affects racial judgments that causally contribute to race-relevant actions, which suggests that they are normatively relevant. 5.4 Normative Consequences In some ways and to some extent, then, the normative case makes itself. Discrepancies in treatment and care that lead to oppressive and unequal outcomes based on morally arbitrary features such as race are simply intolerable to a society that aims at any significant level of egalitarianism. But, while specific study results, like those addressed above, importantly spark and fuel normative concerns, I want to redirect the course of the dissertation back to more general 138 normative concerns. That aim can be facilitated, of course, by returning to Kelly et al. (2010) who just as before describe the normative consequences of the connection between implicit attitudes and race in terms of the consequences for eliminativists and conservationists. Recall that the conclusions of ECP appeared to raise greater problems for eliminativism than conservationism-if humans are equipped to categorize by race, then it's not clear that eliminativism is even possible. Moreover, eliminating or even sublimating 'race' talk if racial categorization is in some measure inevitable appears to carry substantial risk-we might worry, for example, that it would be easier to implement racist policy if we don't track racial demographic information. Although Kelly et al. (2010) strike a balanced chord, mostly resisting advocacy of either normative positions, eliminativism appears less tenable in light of similar evidence regarding racial attitudes. Unless the case can be made that eliminating ‘race' talk substantially decreases troublesome implicit racial bias, eliminativism is difficult to defend. Even though these results "tilt the balance of consideration toward conservationism," Kelly et al. (2010) caution that "the conservationist goal of reducing negative racial evaluations has problems of its own-problems that the disregard of psychology has kept from being addressed" (p. 452). Broadly speaking, "to the extent that implicit biases have not been systematically taken into account, the feasibility of achieving [conservationists'] professed ideals remains largely unknown" (Kelly et al., 2010, p. 462). There is potential payoff, however, for those conservationist who do take implicit bias into their accounts because implicit biases appear relatively 139 manipulable. While "implicit biases operate beyond the direct conscious control of the participants themselves, they can be rather dramatically influenced by manipulating aspects of a person's immediate environment" (Kelly et al., 2010, p. 459). Priming persons with faces of admired "other-raced" individuals- celebrities such as Denzel Washington, for example-appears to decrease implicit bias, and persons appear to be able to exert a measure of self-control over implicit bias once it is revealed (Kelly et al., 2010, p. 49). That conclusion is supported by Stone and Moskowitz's aforementioned study of measures to reduce implicit bias in medicine, which concludes that "Contemporary training in cultural competence is insufficient to reduce these problems...However, these problems can be reduced by workshops and learning modules that focus on the psychology of non-conscious bias" (Stone & Moskowitz, 2011, p. 768). In short, implicit bias affects judgment and action and is, therefore, "directly relevant to the normative debate over race, and is especially important for conservationists" (Kelly et al., 2010, p. 465). Since it is, furthermore, measurable and manipulable, the (first) trick is simply to pay attention to it. 5.5 Conclusion It might appear, at this point, that I have arrived at yet another pernicious dichotomy-this time between social constructionism and evolutionary-cognitive explanations of race-but that conclusion is too quick. It is crucial to my thesis that innate psychological structures are the foundation of the human tendencies to racialize and racially evaluate. But in spite of that commitment and the challenges ECP poses for the received view, I resist the claim that innate 140 psychology completely accounts for racial cognition and evaluation and that social constructionism simply gets it wrong. Fortunately, that claim is rather easier to resist than one might think, primarily because the theorists who defend ECP present a more inclusive view than their constructivist rivals. As many of the protagonists of this and earlier chapters say, at least some aspects of the evolutionary-cognitive and social constructionist models of race appear to be resolvable into complementary pieces of a more complete and versatile picture of race. Mallon (2012), for example, might be the received view's most critical antagonist, but his view is that the "content of our racial representations is determined by a range of forces that include both domainspecific predispositions to believe (of the sort evolutionary-cognitive theorists emphasize), but also cultural and social forces (of the sort that social constructionist theorists emphasize)" (p. 86). As one would expect, Kelly et al. (2010) draw a similar conclusion, saying that "racialized phenomena result from an interaction of an innate, evolved capacity for folk sociological thinking, on the one hand, and the specific social structure in which it is operating on the other" (Kelly et al., 2010, p. 445). In a slightly different tone, Kang (2012) suggests that "even if nature provides the broad canvas, nurture paints the detailed pictures-regarding who is inside and outside, what attributes they have, and who counts as friend or foe" (p. 134). Now, one could take Kang, as Staats does, as taking "sides with nurture" over nature-or in this context, constructivism over nativism-in explaining racialized phenomenon. Given the evidence from ECP and implicit 141 attitude research on race and Kang's reliance on it, however, I think he is likely doing something more similar to what Mallon is doing, namely, pointing out that neither innate psychological structures nor historical practices tells the entire story. But it is Hirschfeld who advances the most forceful presentation of the interactionist thesis. Hirschfeld (2008) argues that racial cognition, just like all other all other "special purpose competencies," depends on the presence of "relevant input" (p. 38). So, even if we are naturally equipped to racialize, actual schemata of racial classification and the evaluations they invite will depend in detail on certain types of environmental input. Drawing on Hirschfeld's conclusions, Quintana and McKown (2008) conclude that "Children's developmental understanding of race appears to be a dynamic interaction of their natural curiosity about their social world and the complex ways in which they are exposed to race" (p. 17, emphasis added). In sum, while humans have an "innate inclination" to sort themselves into groups, only through interaction with certain localized input do robust conceptions of race form. So, in fact, ECP reflects the need for the second part of the equation. In the next chapter, I operate in philosophical space created by the combination of overwhelming evidence for the role of opaque psychological structures in racialization in conjunction with the "concession" that innate psychology can't explain everything. Mallon says that ECP scholarship should be of interest to philosophers, and along with Kelly and Machery, looks to "help to shape novel suggestions proposed in the conservationist spirit" (Kelly et al., 142 2010, p. 435). I provide my novel suggestion-a shift from social to psychological constructivism-in the next chapter. There, I argue that to the extent that races are constructed, they are constructed on a foundation formed outside of human consciousness. CHAPTER 6 THE PSYCHOLOGICAL CONSTRUCTION OF RACE 6.1 Introduction With the four foundational chapters in pocket, the time has come to describe and defend my thesis. Having covered so much ground and so many topics, ones that might appear only loosely related, however, I will begin with a brief synopsis of what has already come. Since all the topics hover over race, there are, no doubt, many ways to connect the major topics of the preceding chapters. Of course, I have written them with a particular set of questions in mind, so I present them here with an eye not only toward the content, but also the intended trajectory of the content of the first four chapters. So, I open this final substantive chapter with a bit about how I have arrived and where I intend to go. I began with a problem-a conceptual dichotomy, to be precise-that generates additional pernicious philosophical and practical (that is, real-world) consequences. Recall from the opening chapter that Gannett (2010) argues that race scholarship, especially among philosophers of science, involves "a dichotomous framing of alternatives" by which "race is either socially constructed or biological reality, a fiction like phlogiston or a genuine natural kind, a merely linguistic or a projectible predicate" (p. 364). Unfortunately, the conceptual rivalry 144 that ensues circumscribes our investigations of race concepts, especially with regard to finding a reasonable middle ground or advancing robust interactionist proposals. Those philosophical limitations, in turn, burden our practical aims. In particular, Gannett says, they preclude asking important questions about race, including-crucially-those that would inform our investigation of contexts in which biological and social causal factors interact to give rise to race. Intuitively, it seems patent that our working conceptions of race recruit both biological and social factors and that if we demand that race be either biological or social, we blind ourselves to interactive contexts. Even recognizing these intimately related problems, however, it is difficult to say how they could be addressed. Several of the authors considered in earlier chapters, for example, point to the difficulty offering a nonarbitrary (or non-"gerrymandered") interactionist description of race (Gannett, 2010; Hardimon, 2013; Kitcher, 1999). The problem remains largely unanswered, which is why I take Gannett's worry to be not merely local to biomedicine but representative of a broader set of problems in a host of contexts concerning philosophy of race. In Chapter 3, I showed that, at least among philosophers, a natural way to begin is by critically analyzing the concept ‘race' to reveal the necessary and sufficient conditions supporting its fundamental definition. Of course, unless philosophers could agree on a set of necessary and sufficient conditions, battling over definitions can serve only to reinforce the dichotomy, but Mallon shows that this is only the beginning of the problem for definitional approaches, arguing that the "semantic strategy" doesn't-can't, in fact-do the job we ask of it. Even if 145 successful, he says, a "project of semantic ascent" exaggerates metaphysical disputes, obscures metaphysical agreement, and does little to tilt the balance in favor of either conserving or eliminating ‘race' talk. In other words, the semantic strategy does not adequately address the guiding normative question, since, even if races are fictitious, we are not compelled by necessity to eliminate ‘race' talk-we might, for example, think that, even if "race" doesn't refer, its conservation is necessary to correct social ills. So, this classical approach is not-either in its rationalist or empirical manifestation-sufficient to advance our goals. As a result, our original problem, the dichotomy, is sustained by the apparent absence of techniques to efficaciously attack it. Fortunately, as I showed in Chapters 4 and 5, Mallon, along with Kelly and Machery, offers an alternative approach. The evolutionary-cognitive program advocated by those scholars shifts the investigative focus from race's "logical core" to its (opaque) psychological foundations. In this vein, Chapter 4 centered on the role of innate psychology in categorization, Chapter 5 on implicit, though not necessarily innate, attitudes. The former appealed to a growing body of evidence that suggests certain important aspects of racial categorization are (both temporally and geographically) universal in human populations. The most important of them is that, in spite of a scholarly consensus against biological essentialism, humans tend to racialize one another, that is, to presume that races are distinguished by racial essences that all or most members of a race share. In Chapter 5, I shifted focus from innate racial classification to implicit racial evaluation, showing that, in 146 addition to explicit attitudes, humans harbor implicit attitudes about others based on race affiliation and, most importantly, that both sorts of racial attitudes impact our behavior.17 Crucially, each chapter-and, in fact, ECP and implicit bias research themselves-raises challenges for social constructivist accounts of race that they do not appear capable of meeting. As a result, it appears that I have started with one dichotomy only to finish with another. Since I can hardly attack one dichotomy with another, the first task of this final chapter is to establish the feasibility of telling a coherent and practically useful interactionist story about race using complementary constructionist and ECP pieces. To this end, Mallon and Kelly once again lay the groundwork, so in the first section, I detail their general discussion of social construction and segue to two early attempts-one from Mallon and Stephen Stich, the other from David Sloan Wilson-to map out ways to combine the insights of social construction and innate psychology to render a more complete picture of race. After that, I will be in position to describe my approach, psychological constructivism of race, and detail its relevance to understanding race and its capacity to address the many unresolved social issues that surround it. Breaking down this second apparent dichotomy, in other words, will help me to chip away the first, along with some of the other issues raised along the way. All of this is to say that I am finally in 17 To be clear, none of this is to say that these tendencies-classification and evaluation-derive from the same (or a related) source. Even conclusive proof of an innate mental mechanism that determines significant aspects of racial classification provides no proof that implicit racial bias is subserved by the same (or a similar) mechanism. In fact, implicit attitudes needn't be innate at all. Fortunately, my aims are in no way dependent on such a claim. My thesis proceeds on a much weaker commitment, namely, to the apparent fact (based on mounds of evidence) that the human inclinations to both classify and judge by race derive from opaque psychological foundations of which we are not consciously aware. 147 position to describe and defend my primary theses, namely, that race is psychologically constructed and that conceiving of race that way helps us to fill the normative gaps left by previous accounts. So, to be entirely clear, I proceed as follows. In the first section, I offer a brief preliminary explanation of what it means for race to be psychologically constructed. Then, I distinguish PCR from semantic and purely18 social constructivist accounts. This move necessitates augmenting my previous discussions of social constructivism with some additional, complementary details. In Section 3, I finish the account of the psychological construction of race, first by offering a detailed description, and, finally, by distinguish it from a pair of closely related accounts: Wilson's "evolutionary social construction" and Mallon and Kelly's "hybrid constructionism," a constructivist account that also proceeds according to the insights of ECP and implicit racial bias research. The latter distinction is crucial, of course, both because of its proximity to my own view and because of the dissertation's heavy reliance on work from Kelly and, especially, Mallon. In the final section, I argue that PCR conduces to our normative goals involving race and then discuss its practical application to three salient racialized contexts of social concern. First, though, an introductory word about what psychological constructivism is and what it is not. I say "purely" here because it will become clear later in the chapter that, like Mallon, Kelly, and Machery, and, as I show below, Wilson, I think there is plenty of room for social constructivism to contribute to race scholarship even after I establish my position. In fact, it will turn out to be a crucial factor in describing the most comprehensive conceptions of race. 18 148 6.2 What Psychological Constructivism About Race Is and What It Is Not Although I dedicate most of the first part of this section to delimiting PCR by indicating what it is not, establishing a point of reference for the position will make that task easier, so I begin with a succinct definition of "psychological constructivism of race," as I think of it. After the brief introduction of PCR, I veer back toward what it is not; in particular, I spend some time distinguishing it from those strategies-semanticism and social constructivism-that it is meant to either replace or complement. Having earlier criticized some of the alternatives, I will want to distance my position from them, to be sure, but the short detour also serves the larger purpose of setting up the a fuller description of PCR. Psychological constructivism, which follows the evolutionary-cognitive approach in taking innate mental mechanisms to be foundational to race, represents an attempt to provide the conceptual nexus through which race's disparate constituents can be more coherently and consistently linked. It is especially designed to handle contexts in which, as Gannett says, the dichotomizing biological and social factors interact and, hence, to address our pragmatic needs in those racialized contexts. Alternatively, it represents an attempt to provide a new method for structuring the ‘race' talk that has informed the conceptual and practical problems I have described in the first four chapters. The choice of descriptive terms such as "nexus" and "structure"-as opposed to "definition," for example-is important. As Chapter 2's discussion of Gannett's paper showed, obsessive focus on the "object-ness" or "kind-hood" of race is 149 counterproductive, and in contrast to the more philosophically venerable notion of a "natural kind," "nexus" suggests a "bond, link, or junction" and, importantly, "the state of being connected or linked," while "structure" entails an "arrangement and mutual relation of...constituent parts," especially in "determining its distinctive nature or character...frame, or make-up" (www.oed.com). A nexus is a point of union; a structure provides a framework or skeleton. Each set of concepts is important to the explanation of my ends. These terminological choices confer several conceptual advantages. Describing those is, of course, one of the primary purposes of the chapter, but it is worth pointing out that they also enjoy what could be called a "tonal" advantage in that the tone they set for discussion of race is one that, unlike its predecessors and rivals, lowers the metaphysical stakes of race debates. This is not to say that they are irrelevant to metaphysical debates, but rather that, for present purposes at least, it helps table worries about whether races are really real, or natural kinds. As I argue in more detail below, PCR describes the conditions under which conceptions of race are constructed and acted upon. As such, it is best to think of any products of PCR-that is, of any "psychological race constructs"-as context-sensitive: Psychologically constructed race is not a once-and-for-all sort of thing, and it is not to be considered (in the strong sense) universal, though many aspects of the structure PCR provides likely are (in the sense preferred by evolutionary psychologists, that is).19 Again, its purpose is to supply a nexus around which complex aspects of race can be structured, 19 That is, in the sense of populational or cultural, not individual, universality. 150 analyzed, discussed, and most importantly, put to use. Now, on to the long version. Following Gannett and Mallon, I have suggested that two venerable scholarly approaches, the semantic approach and social constructivism, are ineffective and incomplete, respectively, with regard to the challenges that surround ‘race' talk. So, the defense of PCR as an efficacious method of understanding race requires showing first that it is not, or at least not merely, a version of those. So, I begin by distinguishing it from those more common approaches to race. 6.2.1 PCR Is Not Semantic As the name itself suggests, PCR proceeds from a foundation of research supporting the evolutionary-cognitive approach, not the semantic one. Recall that the semantic approach involves the attempt to arrive at the proper normative conclusions (primarily) about ‘race' talk by consideration of one's metaphysical commitments regarding race as determined by one's definitional analysis and favored theory of reference. In its direct appeal to ECP, psychological constructivism eschews traditional philosophical analyses. Instead of proposing necessary and sufficient conditions for the concept ‘race', it proceeds from the empirically-backed psychological conditions that give rise to common or regular elements among various conceptions. PCR's shift from concept to conception is crucial because, as I have argued, conceptions are more intimately involved with behavior. Following that lead, PCR shifts discourse from the semantics of race to the foundational psychological conditions of racializing behavior, including, of 151 course, the behavior of categorization itself. In the first instance, then, PCR does not signal an attempt to define ‘race' once and for all, or even too rigidly for a particular time and place, but to provide some flexible but limiting conditions for understanding both how race is being used in a particular context and the effect of that particular use. Moreover, psychological constructivism is largely agnostic on the deep metaphysical reality of race. Instead of investing intellectual resources to determine whether race exists in the world apart from the human mind, PCR holds only that the actions of opaque psychology causally impact conceptual and normative dynamics. So, one benefit of my approach is that, heeding Gannett's admonition, it lowers the metaphysical stakes and, ipso facto, attenuates the dichotomy she describes. Notice that lowering these stakes neatly complements Mallon's criticism of the semantic strategy. That approach, remember, involves more than just determining the conditions necessary and sufficient for being a race. According to the semantic strategy, the definitional status of race is used in conjunction with a theory of reference to determine its metaphysical status, which in turn is supposed to inform our normative decisions. Mallon (2005), of course, shows that this strategy does not function as supposed, arguing that ‘race' is "normative, not metaphysical or semantic" (p. 525). To say that ‘race' is normative in Mallon's sense is not to say that what "is normative is in the world, but how, when, and where we decide what to about what is in the world" (Mallon, 2006, p. 550). As Mallon's arguments from Chapter 2 show, however, the 152 attempt to link these two questions via the semantic strategy "has...resulted in...a misplaced emphasis on metaphysical and semantic concerns" (Mallon, 2006, pp. 550-1). Since PCR derives from opaque psychology, whatever the shortcomings of psychological constructivism, the shortcomings of the semantic approach are not among them. That derivation also demands distinction, of course, but I table the explanation of the distinction between ECP and other ECP-based approaches, on the one hand, and PCR, on the other, in favor of the more immediate concern: Clearly, PCR is constructivist, so is it social constructivist? 6.2.2 PCR Is Not (Merely) Social Constructivism Well, no and yes. Psychological constructivism proceeds from different fundamental assumptions than those generally associated with social constructivism. While social constructivists understand race as the result of its historical and social predecessors in combination with contingent human decisions, PCR absorbs the insights of ECP and implicit attitude research and proceeds from the claim that opaque psychology provides a foundational framework for race that is later furnished with contextual details, such as those provided by the existence of more than one phenotypically discernible population in an area. These points demand elaboration, but in order to extend this line of argument, I need to make an additional pass at social constructivism, highlighting some of the features that distinguish it from PCR and others that are complementary to it. I have already discussed social constructivism several times throughout the dissertation. In the presentation of Gannett's concerns in Chapter 2, for 153 example, social constructivism was cast as the antirealist rival to the realist biological description of race. In the second chapter, I again discussed antirealist constructivism, but this time via Mallon's distinction between it and representational constructivism, with the former accounting for the popular sense in which a proposed theory or kind is cast as "merely" a construction, the latter for the sense in which race (for example) is described a product of historical social and cultural practices, conventions, and decisions. Finally, in Chapter 3, I detailed the obstacles raised by the evolutionary-cognitive program for social constructivist accounts of race. To move the ball forward, however, I need to say just a bit more about some relevant distinctions within social constructivism. The section is not redundant, however-many taxonomies of social construction with differing levels of generality are possible relative to one's explanatory needs and several can be relevant to a single problem domain. That is the case with race. So far I have invoked some fairly specific constructionist issues for particular purposes, but in the present context I need a broader focus, so here I present some more general claims about social constructivism in order to distinguish it from my position. Mallon, once again, proves useful. In his "Field Guide to Social Construction," Mallon (2007) distinguishes two "foci of constructionist work: one centered on our ways of thinking about, representing, or modeling the world, and the second centered on parts of the world itself" (p. 95). Broadly construed, the former focus "includes many different types of theories-for example folk and scientific theories-as well as theories held by an individual, by a cultural group and so forth" (Mallon, 2007, p. 95). In 154 general, theory construction proceeds via defenses of "particular views of what (other than the facts or data) determines the content of accepted theories," which become contentious when "theory constructionists emphasize determinants that they take to be the ‘wrong kinds of causes' or the ‘wrong kinds of reasons' for beliefs" (Mallon, 2007, p. 96). This single argumentative strategy appears in many forms, however, including ones that stress "background factors," "intuitive judgments of plausibility and decisions about usefulness"20 and, in the case of Charles Mills' work on race theory construction, "self-interest of the powerful" (Mallon, 2007, p. 96). So, as Mallon (2007) says, "While a great deal of constructionist work concerns the construction of theories...more provocative constructionist claims seem to concern not only theories but the objects that those theories are about" (p. 95). The relationship between theory and object construction is contentious. For example, Mallon (2007) says that, on a deflationary reading, many "constructionist claims that are apparently about objects can be reinterpreted as primarily about theories," but some "constructionists may want to resist this deflationary reading, suggesting that recognition of the social construction of our theories should lead us to embrace the social construction of the facts those theories purport to describe" (p. 97). In any case, "there is good reason to think that social constructionism has special purchase in producing particular sorts of 20 Mallon (2007) offers examples of the first two as well, associating the former with Thomas Laqueur (1990) and the latter with Andrew Pickering's celebrated 1984 book, Constructing Quarks. 155 object, human kinds," including race (Mallon, 2007, 95).21 Moreover, the construction of both theories and objects revolves around a shared "cluster of explanatory views about what determines the constructed phenomena" (Mallon, 2007, p. 102). Some of the shared explanatory views supplied by Mallon (2007) are already familiar-"that a particular human feature is culturally and historically local rather than universal" (p. 98) -others less so-"that as the content of our cultural representations of the kind vary, so does the kind" (p. 99)-but in one way or another all oppose the primary commitments of ECP, especially the contribution of innate mental mechanisms. PCR, on the other hand, embraces those commitments. Innate psychology is, in other words, the foundation of the psychological construction of race. It will soon become clear that many social constructivist commitments can (and must) contribute to robust, mature psychological constructs, but that need not detain me here-recall that my aim was only to show that psychological constructivism is not merely social constructivism. The role given to opaque psychology is enough to secure that conclusion. So, having distinguished PCR from metaphysical, semantic, and social constructivist positions, I move to its fuller positive development, a part of which centers on distinguishing it from two more closely allied positions. 21 More specifically, Mallon (2007) goes on to say that whatever one thinks about the relationship between theory and object construction in general, "there is good reason to think that connecting theory construction and objects may have special purpose in the study of human kinds" (p. 97). 156 6.3 What PCR Is In the introduction to this chapter, I alluded to the appearance of a second potential conceptual dichotomy-one between social constructivist and evolutionary-cognitive theorists-which represents the chief obstacle to establishing and defending PCR. The threat is serious: Mallon (2007) refers to the debate that underpins the new dichotomy as "human nature wars" and says that the "global positions [on each side] are powerful in that they guide research programs in more specific domains, domains like morality, the emotions, sexual difference, racial classifications, and so forth" (p. 97). Since, at root, psychological constructivism is an attempt to stake out a middle ground between the rival camps, and since I have distinguished PCR from social constructivism, the first task is to explain away this apparent dichotomy. Ultimately, I will carve out an interactionist solution that involves complementary aspects of the two sides, but the explanatory differences between them (some of which have been previously discussed) are quite real, so let me begin with a bit about how the debate plays out before turning to the solution. 6.3.1 The Makings of the Second Dichotomy According to Mallon (2007), this high-profile battle in the human nature wars pits those "who insist on a central role for innate human biology and psychology in explaining human traits, including dispositions and behaviors" against those "who argue that culture and human decisions fundamentally shape 157 the human kinds to which we belong" (p. 97).22 In an earlier paper, Mallon and Stich outline three more specific differences-one empirical, one strategic, and one semantic-between the two sides. There is, first, an "empirical disagreement about the extent to which all normal humans share innate...mental mechanisms that constrain our psychology and our social interactions and intuitions" (Mallon & Stich, 2000, p. 135). Strategically, the sides differ on "the best way to make progress"; specifically, ECP theorists "urge that we focus on what people have in common, while social constructionists think that it is more important to attend to the many ways in which people differ" (Mallon & Stich, 2000, p. 135). The third important issue is by now quite familiar: "it is a semantic disagreement....What is at issue is the meaning and reference of many ordinary terms" (Mallon & Stich, 2000, p. 136). In one way or another, I have alluded to each of these issues, and the last is, of course, the issue concerning the semantic strategy covered in Chapter 3. There are additional general disagreements. For example, social constructionism investigates primarily "phenomena that are contingent upon human culture and human decisions-contingent upon theories, texts, conventions, practices, and conceptual schemes of particular individuals and groups of people in particular place and times" (Mallon, 2007, p. 94). In contrast, "defenders of the importance of human nature often claim that there is a broad range of human universals" (Mallon, 2007, p. 98), so on that account, "human minds have a rich, species-typical cognitive architecture composed of functionally 22 The parameters of the rivalry are familiar from earlier discussions. See my discussion of the "received view" that race is a social construction in Section 3.2. 158 distinct systems-‘mental organs'...that have been shaped by natural selection" (Mallon & Stich, 2000, p. 134). That is not, of course, to say that the products of evolution, or these particular products, are not in some sense contingent as well-human evolution could have turned out differently for myriad reasons. Here, the point is about what the trait in question is contingent upon. For social constructivists, it is primarily the activities of humans, while for ECP theorists it is primarily evolved mental mechanisms. So, a significant chunk of the human nature wars turn on the role of evolution in constructing human kinds. To be clear, however, as Wilson (2005) argues, "the controversy does not center on the basic fact of evolution but on certain consequences, such as the importance of natural selection and especially the relevance of evolution to human affairs" (p. 20). Nevertheless, social constructivist approaches are "united in their commitment to the idea that individuals and societies have enormous flexibility in what they can become, in contrast to the inflexibility and determinism attributed to evolutionary approaches to human behavior" (Wilson, 2005, p. 20). Indeed, "while no serious social constructionist would deny that our innate mental endowment imposes some constraints on what we can learn and what we can do, they believe that most of these constraints are weak and uninteresting" (Mallon, 2000, p. 134). This, of course, is why social constructivists attribute differences to "surrounding culture" that is "in turn explained by differences in history and local conditions" (Mallon, 2000, p. 134).23 23 As a useful point of reference, one can consider a distinction made by David Buller in Adapting Minds. Buller uses the lower case "evolutionary psychology" to designate "a field of 159 Yet another point of contention is that social constructivism about race tends to be "radically" antirealist and antinaturalist, and while ECP can remain antirealist about race, its connection to evolutionary and cognitive psychology suggests a naturalistic framework. This is not to say, of course, that it is naturalistic in the same sense as biological explanations of race. Indeed, like social constructivist work, "work on racial categorization in cognitive and evolutionary psychology has...begun with the falsity of biological theories of race" (Mallon, 2007, 103). Given these general disagreements, it is no wonder that ECP and "social constructionism are widely regarded as fundamentally irreconcilable approaches to the social sciences" (Mallon & Stich, 2000, p. 133). Even so, the approaches can be described in complementary terms. Ironically, the easiest way to show this is by enlisting a couple of more specific examples of their competing views. For that, I return to Mallon's 2012 paper that informed significant parts of Chapter 4. Recall that Mallon (2012) argues that social constructivist stories about race most often depend on the "Conceptual Break Hypothesis" (or CBH), which suggests a relatively recent fundamental change in the way humans represent "human groups we now call ‘races'" (p. 77). In particular, it says that the change in concept, meaning, or theory turns on the essentialization of race at that time. inquiry" which covers a broad set issues that are "united only by a commitment to articulating questions about human behavior and mentality" (Buller, 2005, 8). In contrast, the capitalized "Evolutionary Psychology" is reserved for a specific "paradigm" in the Kuhnian sense. So, a way to understand Wilson and Mallon, here, is to say that they, like Buller, think that evolutionary psychology enjoys broad acceptance, but Evolutionary Psychology is rejected by social constructivists. In saying that social constructivists think that evolutionary constraints are weak and uninteresting, however, Mallon may be arguing that they are unimpressed even with the field of study. The point of the comparison is merely to provide a context that clears up the points of contention between ECP and SC. 160 The lynchpin of CBH is what Mallon calls the "HERE" hypothesis-the view that "racial essentialism is a culturally specific and historically recent way of thinking about some human groups" (Mallon, 2012, p. 79). Of course, ECP promotes an origin story for essentialist thinking based on evolved, innate mental mechanisms, and given that "a range of emerging evidence supports the evolutionary-cognitive thesis that racial essentialism is the product of an innate, domain-specific, and species-typical mechanism that supports lineage essentialist reasoning about human groups," Mallon argues that the "HERE hypothesis is mistaken," thereby undermining the CBH (Mallon, 2012, p. 86). Even so, Mallon (2012) claims to "join with a great deal of constructionist academic work suggesting that the modern treatment of human groups involves some sort of break" (p. 86, emphasis added). In fact, even "allowing close examination of the senses in which social constructionist and evolutionary-cognitive approaches compete," Mallon (2012) aims to "provides a case study of a potential site of integration," arguing that "the central idea of constructionism-that human decision and human culture exert profound and often unnoticed influence...remains interesting and provocative within a broadly naturalist and realist framework" (p. 79). In other words, "even in this context of evolutionary-cognitive vindication, both constructionists and evolutionary-cognitive explanations have a role as part of the complete causal story" (Mallon, 2012, p. 86). As a result, he and Kelly defend a position they call "hybrid constructionism," in which they "pair social constructionism about racial difference with a partially nonconstructionist account of racial representations" 161 (Mallon & Kelly, 2012, p. 512). Mallon and Kelly are not alone in this view. Wilson (2005) also aims at "a more productive explanation of the middle ground" and tries "to show that the heart of social constructivism can be given an evolutionary formulation" (p. 20). In fact, he goes farther, arguing that "Social constructivists have more to gain from adopting an evolutionary perspective than by avoiding it, and [ECP theorists] need to incorporate large elements of social constructivism into their framework" (Wilson, 2005, p. 20). So, these scholars supply the motivation for an interactionist/complementarian story about race. Social constructionists are particularly resistant to such a move, primarily because they tend to associate nativism with determinism. The trick is to show how they might fit together. Mallon and Kelly's hybrid constructivism, along with Wilson's position, which he calls "evolutionary social construction" offer useful paradigms for comparison. 6.3.2 Is an Interactionist Thesis Possible? The first move is to show that variability and malleability of racial categories and attitudes are possible, even if subserved by an innate mental mechanism. To this end, I take an intuitive (rather than conceptual or formally argumentative) approach in this section, beginning with two examples that support the sort of interactionist thesis that Mallon, Kelly, Wilson, and I have in mind. First, citing related work from both Paul Ekman and Paul Griffiths, Mallon discusses differences in emotive facial expression response between Japanese and Americans. Eckman's studies show that "Japanese suppressed their facial 162 expressions in the presence of authority figures" (Griffiths, 1997, as cited in Mallon, 2007, p. 100). In this respect, the most likely explanation for the difference between Japanese and Americans is a cultural one. Although some aspects of facial expressions associated with emotional response are innate- members of all human populations smile and frown, for example-"they can nonetheless be modified by cultural and human decisions" (Mallon, 2007, p. 100). "Such is the power and versatility of the human capacity to shape our environment and ourselves that many traits that develop extremely robustly across a wide range of environmental perturbation are nonetheless under the control of our decisions and our culture" (Mallon, 2007, p. 100). This example suggests that the social constructivists' resistance to nativism that grows out of concern for inevitability of certain traits is ill-founded. While some aspects of innate characteristics and capacities are likely inevitable, humans exert significant control over the full-fledged characteristic. My own example of the interaction of innate and sociocultural elements appeals to Chomsky's view of natural language, which I have touched on before. Again, to say that a phenomenon is (partially or fundamentally) psychologically constructed to say that it is underpinned by psychological mechanisms that are augmented by some social practices and deliberate decisions. As such, the helpful analogue to race according to PCR is not biological theories of natural kinds or even species, but the Chomskyan view of language.24 24 In this context, whether Chomsky is right about the origins of language is inconsequential. I chose the example to add clarity to my discussion of race and because it is one that is likely familiar to many philosophically and linguistically oriented readers. For details on his linguistic theory, see Chomsky (1955). 163 In broad outline, Chomsky's view is that language is something for which humans are hard-wired; in some sense, humans simply innately have language. Whatever the details, the human brain contains a language module, an innate mechanism for generating natural languages.25 So, according to the Chomskyan paradigm, particular natural languages-Dutch, English, Arabic, Mandarin, etc.- arise from the interaction of innate language mechanism, on the one hand, and social and cultural stimuli, on the other. For one to communicate in a particular language, then, one must acquire and display competence in the ways of those around one. In this analogy, a particular language may look like a representational construct, but language is itself, whatever it is, arises out of psychological construction.26 So, despite some fairly obvious differences, the Chomskyan line on language helps clarify the psychological construction of race. In fact, Chomsky's view of language is reminiscent of Hirschfeld's arguing that race is what it is "by virtue of the way it is acquired...young children acquire racial knowledge via a predilection to attend to input relevant to community standards and convention," and they acquire racial knowledge "largely on their own" (Hirschfeld, 2012, 22). So Hirschfeld's evidence, like the examples above, supports the move toward the evolutionary-cognitive approach to race.27 Race, according to this view, does not exist merely because of particular history of 25 There is an obvious disanalogy here, namely, that Chomsky thinks the module that gives rise to language was designed for language, whereas (again) the module implicated in race likely is not a race module. The two objects of inquiry needn't be analogous in this way, however, in order for the comparison to shed light on psychological constructivism about race. 26 For a more recent scholarly treatment of Chomsky's linguistic theory, see I-Language: An Introduction to Linguistics as Cognitive Science (Isaac & Reiss, 2008). 27 Machery and Faucher (2004) are critical of Hirschfeld's conclusions, but the focus of their objections is the particular cognitive mechanisms that Hirschfeld defends, not the claim that cognitive mechanisms are fundamental to racialization. 164 social practices of certain sort. Instead, it appears to be grounded in a predilection to sort people into groups based on racialized qualities. In sum, despite resistance from social constructionists to evolutionary psychology, these real-world examples illustrate that at least some phenomena are best understood as combinations of complementary psychological and social forces. To this end, in fact, Wilson and Mallon make stronger claims. Wilson (2005) argues that social constructivism can be preserved, even strengthened by incorporating an evolutionary-cognitive foundation. Elaborate "innateness...does not exclude and indeed makes possible the potential for openended change...In short, the way forward for social constructivism is to become more sophisticated about evolution, not to deny its relevance to human affairs" (p. 28). Mallon (2007) expresses similar sentiments, arguing that "a weak social dependence constructionism need not be opposed to innateness" (p. 99). Citing the aforementioned work on emotions and facial expressions, Mallon (2007) argues that "even these ‘biologically determined' motor responses can be shaped by cultural reinforcement" (p. 100). In sum, these examples "show that it is possible for a trait to be both socially dependent and innate," which is to say that "social dependence is not incompatible with the biological or psychological claim that some important traits are innate. We may well be able to alter innate characteristics if we have the (individual or collective) will" (Mallon, 2007, p. 101). So, to find middle ground, social constructionists are likely going to have to broaden their views of what informs (at least some) constructions. Social constructionists are not the only ones who will need to adapt, 165 however. While innate psychology provides a framework for racial cognition, it does not tell the whole story. This much should be intuitive, given that evolutionary psychologists regard the mental mechanism responsible for racialization as an exaptation: if the module is an adaptation for something other than race-for example, biological reasoning-then one should not expect that it would account for race in toto. Wilson (2005), in fact, makes a point of drawing attention to limited nature of evolutionary psychology as well as social constructivism. "Evolutionary psychology in its current form...must take back some of what has been rejected as part of the ‘standard social science model', in particular open-ended, non-genetic evolutionary processes that adapt individuals and groups to their current environments" (Wilson, 2005, p. 15). Kelly et al. (2010) are less forceful on this point but similarly allow for dual contributions to the study of race, noting that evolutionary psychologists do not deny that "socialization plays some role, they simply insist that it is not the whole story" (p. 440). Finally, Mallon (2012) delivers another version of the message: "in effect, the content of our racial representations is determined by a range of forces that include both domain specific predispositions to believe...but also cultural and social forces" (p. 86). So, neither research program gives the complete picture. It appears, then, that innate psychology without social construction is empty, and social construction without innate psychology blind. A middle ground appears necessary, but so far little has been staked out, particularly with regard to race. Because the "debates have become so polarized," the "middle ground becomes a no man's land into which no one dares 166 to venture" (Wilson, 2005, p. 1). As a result, despite mounting empirical evidence and useful complementarian analogues, the "relationship of such accounts (growing out of cognitive and evolutionary psychology) to more traditional constructionist work in the social sciences and humanities has barely been considered" with regard to race (Mallon, 2012, p. 86). But, as Wilson has it, much of the debate that prevents complementarian theses is ado about nothing. The "middle ground that we have been discussing has remained unoccupied because of perceived implications, not just intellectual difficulty. Intellectually it is fully possible to achieve a theory of evolution...that serves as a resource for individual and societal change" (Wilson, 2005, p. 15, emphasis added). ECP and social constructivism are not mutually exclusive; in fact, "those of us who use the term evolutionary psychology broadly think of it not as a counterweight but as a framework for explaining all aspects of psychology from an evolutionary perspective" (Wilson, 2005, p. 15).28 The "framework" metaphor is, of course, familiar. I have already discussed psychological constructivism's providing a framework for race, and now the time has arrived that I can more fully develop the point. 6.3.3 PCR as Framework and Nexus Recall that Gannett's dichotomy problem arises from the tendency among philosophers of science to adopt "an essentialist approach to natural kinds" (Gannett, 2010, p. 371). The problem is that races, just like genuine biological categories, are nonessentialist-racial groups do not share biological racial 28 Again, see Buller's distinction described in footnote 7 above. 167 essences. Still, racialism won't go away, and given its recalcitrance, Gannett advises that "a better argumentative strategy recognizes that ‘race', as it is socially constructed, is essentialist" (Gannett, 2010, p. 371). The move to relocate the essentialist element of race is one that I adopt, but not quite in the same way as Gannett, who in the passage just cited goes on to say that "With society, not science, as the starting point, the question becomes whether or not social construction conforms to biology" (Gannett, 2010, p. 371, emphasis added). Unfortunately, this is just an instance of the semantic strategy, which I have already shown (via Mallon) to be inefficacious. So, I offer the following emendation: The first important point is to recognize that race, as it is psychologically constructed, is essentialist, and with psychology, and neither (classificatory biological) science nor society as the starting point, the question becomes how the skeleton provided by an innate mental module is fleshed with social phenomena to form robust, albeit context-dependent, conceptions of race. So, PCR begins with an innate human tendency toward racial classification, but to say only that humans tend to sort one another according to some typically racial standard would be insufficient. It is equally crucial to say that the classed groups are essentialized. An easy way to understand the importance of this point is to recall that in crafting his conceptual analysis of ‘race', Hardimon denies essentialism a place in its "logical core," the necessary and sufficient conditions common to all conceptions of race. The "logical core does not require that races have essences," he says, and for emphasis adds that the "ordinary concept of race is not essentialistic" (Hardimo, 2003, p. 449). But, 168 in the passage below, it's easy to spot the flaw in his justification: [The] logical core does not hold that human beings are divided ‘by nature' into a hierarchy of races…that each race is characterized by a fixed set of fundamental, ‘heritable', moral, intellectual, and cultural characteristics common to it…that each race has an essence that explains why it has the visible physical features it does, why it has the moral, intellectual, and cultural characteristics it does, or why the two are correlated. It does not hold that each individual member of a race necessarily shares the ‘essential' characteristics of his or her race. It does not-and this point should be underscored-require any intrinsic connection between skin color and humanly important traits such as intelligence or moral character. (Hardimon, 2003, p. 450) The problem is that what Hardimon (2003) says is not merely true, it is a truism. The view that essentialism is "in the object," that racialism is biological, is rejected as a part of the ontological consensus. Nevertheless, it is patent that people commonly and naturally essentialize race and that the essentialized conceptions of race are the ones most associated with social and political issues. So, while PCR agrees that race is not biologically essentialist, it takes the psychologically provided essentialist element as paramount. Even if racialism is not essential to the logical core of ‘race', it is essential to our pragmatic aims that we recognize the ubiquity and consequences of racialism. This is a point to which social constructivist Sally Haslanger appears to object. Haslanger's aims are similar to mine: "the task is to develop accounts of...race that will be effective tools in the fight against injustice" (Haslanger, 2000, p. 36). In fact, we agree even on the more specific concerns that guide her project, save one detail. Haslanger, like Hardimon, deflates her account by shedding essentialism. Consider the following footnote in which Haslanger is explicit about this point: 169 We need here a term for those physical features of individuals that mark them as members of a race. One might refer to them as ‘racial' features, but to avoid any suggestion of racial essences, I will use the term ‘color' to refer to the (contextually variable) physical ‘markers' of race, just as I use the term ‘sex' to refer to the (contextually variable) physical ‘markers' of gender. I mean to include in ‘color' more than just skin tone: common markers also include eye, nose, and lip shape, hair texture, physique, etc. Although the term ‘people of color' is used to refer to non-Whites, I want to allow that the markers of ‘Whiteness' count as ‘color'. (Haslanger, 2000, p. 53, fn. 7, emphasis added) Even Haslanger's way of expressing her conception of race suggests to me that "color" alone cannot constitute race, the formation of which demands an additional ingredient. To see this, consider a physical, but generally non-racial, characteristic: eye color (or Hardimon's choice, the Adam's apple). While it may not be true that no stereotypes exist with regard to, say, brown-eyed people, it certainly does not appear that we apply deep essential properties to all browneyed people, likely in part because brown eyes are shared by a large numbers of members from all recognized races. Both Hardimon's and Haslanger's minimalist conceptual accounts of race appear to treat all phenotypic characteristics as though they were like eye color, that is, as properties that demarcate different colors of people, rather than pointing to deeper psychological and moral characteristics typically associated with race. Their motivation, which is common to social constructivists, is easy to imagine. Racially essentializing persons and groups is clearly at the heart of many problems associated with race. But races, even if they exist, aren't essentialist, so by (re-)defining race without essentialism, social constructivists such as Hardimon and Haslanger remove the most harm-inducing element of common conceptions of race. 170 PCR shares that aim, but takes a different course: It accounts for racialism by providing a medium for racial discourse that correctly locates it in the human mind-that is, in the cognizing subject-rather than in the racialized being or group-that is, in the object so cognized. This line of argument becomes clearer if we consider an example. Psychological constructivism is committed to the view that group-typical physical characteristics are only that, until they are used to (fallaciously) predict deeper and, in the most extreme cases, essential racial characteristics. Hardimon is right that we could group people according to skin tone in rather innocuous ways-these people are brownish, these whitish, these yellowish, these reddish, etc.-but it does not appear that those groups constitute races until they are in some measure essentialized. That is why eye color is seldom counted as a racial characteristic-it is rarely used to suggest deeper meaning of the relevant kind. Similarly, if I say my friend's skin is darker than mine-or that another group generally has darker skin than the group with which I am associated-I doubt that that alone qualifies my claim as racial. The set of sunbathers, for example, has darker skin than I, regardless of their racial affiliation. Forming a race in the popular and pernicious sense requires something more than grouping according superficial characteristics. It takes a particular kind of story about those characteristics to make them racial, and PCR is founded on the view that racialist thinking best informs that story. Even allowing essentialism is not enough, however. Evidence of innate psychology's role in racial matters demands a thicker foundation that reflects the 171 fact that humans not only natively classify by and essentialize race, but also implicitly evaluate by race. In his discussion of points of potential integration of social and innate psychological factors, Mallon (2007) opines that "there is no reason that human kind constructionism about race cannot be paired with alternate accounts...of racial theories. Such account might hold that psychological predispositions contribute to the formation of racial social roles that have played an important role in racial oppression" (p. 103). In an earlier work, Mallon delivers a lengthier, but lucid description of the favored approach: Instead of focusing on defending an account of what race is or what racial concepts mean, we should ask: what kinds of conceptual apparatus do we need to discuss racial classification and racially associated phenomena in historical and contemporary life? We thus exchange the question of whether and how race exists for the project of developing an adequate metaphysical theory distinguishing as many accounts of race or racial phenomena as are needed to serve all our functional needs-including the various dimensions of racial identification, experience, appearance, and folk classification-so that their practical, social, and ethical significance can be discussed. Only in such a project of theoretical refinement are we likely to shed the persistent mistakes of ordinary racial thinking while continuing to refer to the world in ways that satisfy a multiplicity of theoretical needs. (Mallon, 2004, p. 668) That sort of supposition motivates the inclusion of racial evaluation, that is, of implicit racial bias in the psychological foundations of conceptions of race. The evidence leaves little doubt that the human predisposition to essentialize race contributes to racial oppression, but the more crucial concern involves what characteristics or tendencies are attributed to the supposed racial essence. Perhaps if racial stereotypes tended to be fairly benign, they could be left aside. And, in fact, Erin Beeghly uses her "descriptive view of stereotypes and stereotyping" to cogently argue that stereotyping is not intrinsically wrong" 172 (Beeghly, 2015, n.p.) Beeghly (2015) points out, for example, that even if the proposition "doctors wear White coats" counts as a stereotype, there "is no moral failing...no moral vice" when a "panicked father in an emergency room" grabs "the first person he sees in a White coat." So, instead of simply assuming that all cases of stereotyping are morally objectionable, Beeghly argues that "we might have to be pluralists about what's wrong with stereotyping. There may be no single wrong present in all the bad cases. Despite the gravity of such wrongs, an important fact would remain. When these wrongs do not accompany a case of stereotyping, stereotyping could be permissible and even good. (Beeghly, 2015, n.p.) One thing that could be wrong with racial stereotyping, in particular, involves the tendency to essentialize it, a point that Beeghly recognizes and allows (Beeghly, 2015). In light of what was said about race in Chapters 4 and 5, this consideration is particularly relevant to racial stereotyping, of course. Racial stereotyping, after all, tends to be deemed particularly and universally wrong. Indeed, as the Kirwan studies from Chapter 5 show, implicit racial bias (and, hence, racial stereotyping) is very likely causally relevant to oppressive, realworld disadvantages in a host of contexts. In fact, when it comes to race, even "positive" or "good" stereotypes can be harmful. An obvious problem is that positive stereotypes set standards that most members of the group will not reach. If we accept, even casually, the stereotype that Asians are good at math, then an Asian student who is poor at math is set up for criticism, even if he or she is has other robust talents. Anecdotally, I have been present for several pick-up basketball games in which 173 African American males who were chosen first by team captains are chastised when it turns out that they are not very good. The bigger problem with "positive" racial stereotypes, however, is that they make negative ones seem more plausible. Believing that "Blacks are naturally good athletes," for example, opens the door to believing "Blacks are naturally less intelligent." The point is that, very often, even "good" racial stereotypes aren't good. These claims, however, should not be taken as critical of Beeghly's position; rather, I think they promote Beeghly's analysis. If the previous chapters have shown anything, it is that we should assume very little about race, since our intuitions are often misguided and counterproductive. While it might very well turn out to be the case that racial stereotyping is usually (either directly or indirectly) harmful, Beeghly's thesis serves as further testament to the need to critically approach all aspects of race, even (perhaps especially) those that are most culturally entrenched. Returning to the original point, since PCR is designed to meet practical concerns, it is important to build implicit racial bias into its foundation.29 In sum, then, the foundational aspects of PCR-what I have been calling its "framework," "structure," and "skeleton"-are the innate psychological human tendencies to racially categorize ourselves and others, to imbue those categorizations with a presumed essence, and, for reasons that do not necessarily invoke nativism, to evaluate races and raced individuals via implicit biases. These ingredients, one could say, are cooked into the broth of racial cognition. One could object that not everyone who recognizes race or competently uses "race" will essentialize and/or exhibit bias toward those groups, but this anecdotal claim misses the point. The important point is that those sorts of conceptions are widespread and powerful enough that they explain a great deal that foments racial problems. 29 174 But this is only half the story. PCR does not stop at the attempt to place all the innate aspects of racial cognition under one heading. The point of PCR is to tell a fuller story about race by accommodating interaction between social constructionist and psychological approaches to race (in order, of course, that we might make headway in social and political issues). While many questions remain for PCR at its innate psychological foundation-which, if any, of the evolutionary stories give the best account of racial cognition, for example- accounting for the social contribution is, if anything, more difficult. Much of what I have to say about social contributions, then, is more speculative. What seems like a bug, however, is really a feature. Mallon's quote above helps explain why. Again, he says that we should shift our focus from defending accounts of what racial concepts mean to accounts of the "kinds of conceptual apparatus...we need to discuss racial classification and racially associated phenomena in historical and contemporary life" so that we can distinguish "as many accounts of race or racial phenomena as are needed to serve all our functional needs" (Mallon, 2004, p. 668). The ways in which we need to discuss racial phenomena, such as "racial identification, experience, appearance...so that their practical, social, and ethical significance can be discussed" will be highly contextdependent (Mallon, 2004, p. 668). Again, this thought is one I have promoted since the opening chapter's discussion of Gannett (2010), who complained that when focus centers on natural kinds, "the practical context in which the question initially arose...falls away," and who, therefore, argues that we should instead focus on "the very context-dependent ways in which biological and social factors 175 interact" (p. 375). So, the social considerations will depend largely on context, though we may find some relatively universal social results, such as the fact that White power structures tend to weigh disproportionately in outcomes for both Whites and non-Whites. A descriptive example of this process comes from Pieter Adriaens and Andreas De Block, who account for "male homosexuality as an evolutionary social construction" in order to aid progress "beyond the traditionally polarized debate between evolutionary psychologists and social constructionist" (Adriaens & De Block, 2006, p. 570). The relevant part of Adriaens and De Block's argument, which owes a debt to Wilson, is that "some sociohistorical conditions around 1700 may have quickened the transition from a mainly occasional kind of same-sex contact to…[an] exclusive kind we now call ‘homosexuality" (Adriaens & De Block, 2006, p. 583). For example, they provide evidence of a "sudden occurrence of homophobia around 1700," which may have been "an instrument used by marginalized individuals to (re)gain (more) power" (Adriaens & De Block, 2006, p. 582). As a result, one discovers a social "feedback loop" in which homosexuality and homophobia ironically feed one another.30 Just as with Chomsky's explanation of language, neither the details nor even the accuracy of Adriaens and De Block's view is at issue. The point of the example is to show only that beginning with an evolutionary-cognitive framework, Adriaens and De Block offer a reasonable and fresh view of how social and native factors interact 30 Here, Adriaens and De Block acknowledge a debt to Ian Hacking, who discusses feedback loops in "The looping effect of human kinds" and again in The Social Construction of What? 176 in a way that help us make sense of one important phenomenon surrounding the category ‘male homosexual'. Machery and Luc Faucher (2005) make a run at race that is similar to Adriaens and De Block's on male homosexuality (as well as the one I am currently describing). In an effort to "overcome the prevalent theoretical tribalism and inspire integrative theories of racialism," Machery and Faucher propose several "requisites for future theories of racialism," many of which could be relevant to the future development of PCR. For example, while underscoring the universal aspects of racialization, they acknowledge that racial categorization varies across cultures and argue that a "theory of racialism has to accommodate this diversity" (Machery & Faucher, 2005, p. 1030). So, PCR is designed to compensate for the explanatory gaps in social constructivist accounts, but not to replace it; rather, it complements the received view by filling those gaps and merging the explanations that social constructivism is well-equipped to supply. That alone does not distinguish PCR from similar projects, however. Wilson's evolutionary social construction and Mallon and Kelly's hybrid constructions are designed to do same, so I use the next two sections to distinguish my project from theirs. 6.3.4 PCR Versus Evolutionary Social Construction Since I have referenced both Adriaens and De Block's and Wilson's own use of the phrase, I should say a bit about "evolutionary social construction," a title Wilson coined in the 2005 paper cited above. Wilson (2005), of course, argues that social construction can be given an evolutionary framework, using the aforementioned phrase to describe the strategy. Needless to say my 177 position, PCR, does something very similar, so similar, in fact, that I would not be distressed by someone who wanted to collapse PCR into evolutionary social construction. Even so, I want to argue that there is at least one reason, one with philosophical and practical consequences, to stick with my phrase. Although I am quite sympathetic to the evolutionary aspect of the project, I think that the cognitive aspect is, at least for now, more significant. To show this, I appeal to the distinction between proximate and ultimate causes, the difference being, of course, that proximate causes involve behaviors that are triggered by immediate stimuli, while ultimate causes involve explanations of a given behavior in terms of its evolutionary functional history. The distinction can be made clear with an example. I recently adopted a dog31 from the Humane Society, which received and cared for him after he was hit by a car when he was approximately 1 year old. Hunter is likely equal parts American Bulldog and American Staffordshire Terrier (or "pitbull"), and though he is in no way "aggressive," he is quite playful, often too playful. He has, in fact, sometimes instigated very minor and bloodless quarrels with other dogs, even though it is clearly a desire for play that motivates him. Those episodes have amounted to very little, but suppose he really were illtempered and violent. Out of concern for his and other's welfare, I might wonder why. Suppose the Humane Society workers told me that Hunter had been rescued from dog-fight breeders. They might say, he is ill-tempered and belligerent around other dogs because his trainers induced him to fight other His name is "Hunter Pence," after the 2014 World Series Champion San Francisco Giants' outfielder. 31 178 dogs, and now when he sees them, he acts on the impulse generated by that training. This would be the proximate cause of my dog's behavior. But suppose I respond that I've seen dogs engage in similar behaviors-say, using their front limbs and paws to control the head movements of another dog-and they couldn't all be former fighters. The workers might then respond that most or all dogs have a defense or fighting technique that comes out in play or battle, but that the universal behavior is best explained by evolution. The ancestors of modern dogs who developed effective fighting techniques survived and reproduced at a high rate, so now modern dogs sometimes exhibit those survival behaviors. In fact, those are the very behaviors that trainers manipulate to make fighting dogs. This is the ultimate cause of the behavior. By analogy, I might conceptually divorce proximate cognitive psychological causes of racialism from its ultimate evolutionary causes, and there might be advantages to doing so. The primary motivation for doing so would be that we might discover racializing behaviors in humans via tests such as the IAT, but remain completely in the dark about its evolutionary explanation. Given the urgency of racial issues, we might want to proceed with relatively independent investigations of the proximate and ultimate causes of the racializing behavior. This could confer both scholarly and practical benefits. In terms of scholarship, the partial division of labor between cognitive and evolutionary approaches allows the former to advance without being held up by the latter. In short, we don't want to hold up our investigation of the proximate aspects of racial cognition while we wait on the ultimate explanation. The following 179 anecdote shows that practical implications, however, the ones that apply outside of the academy, are likely more important. I was born in Alabama, and, although I am no longer a Southern Baptist, I was brought up in a rather large all-White32 Southern Baptist church in a rather small Alabama town. While many of the members of the church were "good folks," they did not generally hold what are considered progressive racial views. I would like nothing more than to affect change in the American South with regard to its racial attitudes, but notice that as an adherent of the evolutionary-cognitive program, I face not one but two major challenges-resistance to racial sensitivities, of course, but also resistance to evolutionary theory. If I want to make a difference in the south, I have to be able to talk to Baptists, so I cannot afford to make racial progress dependent on evolution. Wilson, in contrast, can, because he uses the term to support a thesis concerning literature and narrative, subjects whose finer points are largely confined to the academy. So, PCR maintains that we have certain universal racial beliefs and only secondarily inquires into their evolutionary roots. The phrase "evolutionary social construction" leads with a word that shuts some of those folks down. Again, I am entirely dedicated to understanding the evolutionary roots of racial cognition, and I fully understand that evolution theory is the only way to explain some aspects of racial cognition, so I am happy to eventually have PCR subsumed under Wilson's 32 People are often shocked to learn of the extent of segregation in my childhood, which spanned the late 1970s and early 1980s. My first grade school was segregated, for example, save for one Black student, whose surname was "White." The irony is somewhat amusing, but the fact that it alludes to-continued de facto segregation-is not. This anecdote serves as a reminder that while we need to move beyond the strategies of the 1960s civil rights movement, the movement's causes are still quite real. 180 favored research program. The evolutionary aspects, however, are likely not the most important ones for immediate social progress. Evolution should continue to inform our racial investigations, but it should not in the process hinder our racial progress. Since I have relied on Mallon so extensively throughout the dissertation, a likely more important distinction is between PCR and Mallon and Kelly's hybrid constructivism, a topic to which I now turn. 6.3.5 Distinguishing PCR From Hybrid Constructionism Mallon and Kelly set out to solve what they call social science's "race puzzle," which centers on the question, "If there is no biological basis for race...in virtue of what are racial categories a successful basis of informative, important social scientific generalizations?" (Mallon & Kelly, 2012, p. 507). The "standard answer"-that races "are social roles of some sort"-is not dismissed; in fact, Mallon and Kelly begin with the standard answer, which they take to be "basically correct," and then "suggest some ways in which such a solution might be extended and improved" (Mallon & Kelly, 2012, p. 508). The key to that move, and, ipso facto, hybrid constructionism (or HC), is "understanding the way racial social roles are psychologically constrained" (Mallon & Kelly, 2012, p. 508).33 The constraints of psychology on race are not trivial; in fact, to "say that a social role is psychologically constrained...is to claim that social roles are strongly 33 This move is analogous to one Wilson (2005) makes, in which he contrasts social constructivism that is "largely [but not entirely] unconstrained by human biology" and what he calls "anything goes" constructivism, and argues that the former is adaptable to the evolutionary approach (p. 21). 181 influenced, in some way or another, by the specific character of the psychological mechanism in play" (Mallon & Kelly, 2012, p. 523). The constraining mechanisms are, of course, the two familiar opaque psychological aspects of racial representations: the innate, domain-specific, species-typical mental module implicated in racialist classification and the aspects of racism that involve implicit and automatic evaluation (Mallon & Kelly, 2012, p. 508). So, just as it reflects important aspects of Wilson's evolutionary social constructionism, PCR reflects important aspects of hybrid constructionism, and, again, might ultimately be subsumed by it. As it stands, however, by conferring upon innate psychology and implicit bias roles foundational to the construction of race, PCR rests on a more specific commitment than HC. This claim can be secured by underscoring some ambiguity left in Mallon and Kelly's psychological "constraint requirement."34 I should first say that to call the constraint requirement ambiguous is not to criticize it. In fact, Mallon and Kelly appear to be intentionally ambiguous, committing only to the claim that social roles are constrained (or "strongly influenced") "in some way or another" by psychological mechanisms (Mallon & Kelly, 2012, p. 523). In short, they intentionally leave the door open to more specific proposals; they do not, however, take the door completely off the hinges. In other words, Mallon and Kelly do supply constraints on the constraint requirement. First, they suggest that psychological mechanisms "support and shape racial social roles" (Mallon & Kelly, 2012, p. 508, emphasis added). The choice of phrase is mine, not Mallon and Kelly's, but it should be uncontroversial and is used only for brevity. 34 182 Second, unlike traditional social constructivist accounts, HC pairs "social constructionism about racial difference with a partially nonconstructionist account of racial representation" (Mallon & Kelly, 2012, p. 523, emphasis added). By doing so, Mallon and Kelly can explain "the stability and distinctiveness of racial social roles" and "the evaluations associated with and the causal effects of those social roles," and, thereby, make the case that the "standard answer is best understood as psychologically constrained in a number of ways" (Mallon & Kelly, 2012, p. 523). Even with these specifications, however, I think that Mallon and Kelly's constraint requirement can be interpreted in at least two ways and that it is not entirely clear which, if either, they intend or prefer. In my view, the ambiguity rests on which (if either) aspect of the hybrid construction is primary or "given," that is, we can reasonably ask whether the social or psychological aspect is foundational. A brief explanation invoking a pair of simple examples should clear up what I mean. First, one could take the social facts-the "cultural understandings, social conventions, institutions, and common practices"-as primary and argue that the psychological mechanisms mold the social given. Here, one might initially imagine a block of marble (the given) chipped into shape by a sculptor. But, a better image is an above-ground swimming pool, which has walls that shape the water, constraining it in the sense that it can move all about but not over the wall. Either way, the point is that the psychological sets limits that the social phenomena bump into. Call this interpretation "constraint-ism." 183 Alternatively, one could take the psychological mechanisms as primary and argue that the social aspects arise from the accumulation of their effects. 35 Here, one might imagine, instead of a pool, a large water slide-not just the slide itself, but the entire mechanism. Metaphorically, the pumps "create" the water and the slide determines its trajectory. Accordingly, the second interpretation suggests that the psychological mechanisms are foundational in that they both construct the foundational aspects of "social roles" and guide their course. Call this option, "accretive constructivism." I favor this latter interpretation and have designed PCR to reflect that preference. In fact, I think that parts of Mallon and Kelly's discussion of the role of implicit bias in HC supports this interpretation. In a key section of their 2012 collaboration, they discuss one "prominent approach to understanding [persistent racial] disparities-one that emphasizes institutional racism" (Mallon & Kelly, 2012, p. 517). "Institutional approaches "focus primarily on unequal social level outcomes" and "typically favor a distinctive, anti-individualist kind of explanation" (Mallon & Kelly, 2012, p. 517). Accordingly, racial inequalities "are best explained not by appeal to the characteristics of individual people...but mainly by appeal to the institutions, social structures and policies of the society...and the policies that govern [it]" (Mallon & Kelly, 2012, p. 517). Since "racism...can exist and thrive even in the absence of racists," the contribution of the psychology of 35 Of course, a third option is that neither is primary and that they operate in tandem, perhaps according to a sort of looping effect. I resist that option here because it depends on greater understanding of each causal aspect of racial representations. Even if this third option eventually wins the day, then, it will probably have benefitted from the relatively independent investigations of each of the options described above. 184 individuals is, according to institutional accounts, "of minor explanatory value, if of any at all" (Mallon & Kelly, 2012, p. 517). Of course, Mallon and Kelly resist those claims, arguing that "approaches focusing on institutions need not be hostile to psychological research" (Mallon & Kelly, 2012, p. 518). What is interesting in this context is how they defend that claim. Mallon and Kelly begin with Ian Hanley-Lopez's "New Institutional" approach, which "expands the institutional approach so that it can take into account characteristics of individual actors" and "sets out to build a theory of racism that explains organizational activity that systematically harms minority groups even though the decision-making individuals lack any conscious discriminatory intent" (Haney-Lopez, 2000, as cited in Mallon & Kelly, 2012, p. 520). Of course, research on implicit racial bias provides an explanation for variance between explicit and implicit racial attitudes and so supplies a foundation for a view like Hanley-Lopez's. So, Mallon and Kelly accept some basic claims of the standard answer, but argue that it "leaves out an important piece of the puzzle: namely, the influence of specific features of individual psychologies on those classificatory dynamics, and the types of patterns those features can generate and sustain" (Mallon & Kelly, 2012, p .522). Although one could object that the effects of the constraining psychological mechanisms are "too small to actually shape population level regularities," they argue that even slight factors "will aggregate upward to affect the character of a population, especially when those factors are widespread" and that "collectively, the influence of implicit biases can scale up to shape the types of population 185 regularities that social sciences attempt to capture" (Mallon & Kelly, 2012, pp. 523, 519, emphases added). The highlighted phrases "aggregate upward" and "scale up to shape" suggest the accretive constructivist, rather than constraint-ist, interpretation, and, therefore, support PCR's psychological "ground up" rendering of racial phenomena. That interpretation is also supported by Hirschfeld's influential conclusions about the formation of racial cognition. Recall that Hirschfeld claims that the racial beliefs and attitudes of children develop independently of their parents, teachers, and peers. According to Hirschfeld (2008), "there is considerable evidence that even infants are capable of drawing distinctions between individuals that become basis of social category identity" (p. 38). In fact, he says, "the young child may play a special role in sustaining racial thinking. Rather than simply rehearsing adult racial beliefs, young children may be crucial to the way racial beliefs become a fixed part of adult cultural repertoire" (p. 42). Again, among the most important factors that children appear to supply is an intuitive essentialism, which I have argued is among the most important elements of race according to PCR. So, children don't blindly or passively acquire racial traditions, they "come to racial thinking because thinking racially is subserved by a cognitive susceptibility that makes race the sort of idea that is readily learned and stabilized in the culture of their elders" (Hirschfeld, 2008, p. 47). Quintana and McKown offer perhaps the best summary of the role of children in sustaining these phenomena, arguing that "between birth and adulthood, children become racialized beings…all of whom are to some degree beholden to the psychology of intergroup cognitions and 186 relations. We rapidly and automatically categorize and judge others on the basis of their group membership" (Quintana & McKown, 2008, p. 6). So, aspects of both the Mallon/Kelly and, especially, the Hirschfeld/Quintana/McKown lines appear to support a ground-up, accretive interpretation of the proposed psychological constraints of HC, according to which the psychological mechanism implicated in racial cognition provide a foundation out which social constructions of race are formed. PCR proceeds on that interpretation. Furthermore, in my view, the psychological ground-up orientation of PCR likely means commitment to a softer antisemanticism than Mallon promotes. Mallon was originally motivated to attack the semantic strategy to ensure that metaphysical and semantic disputes would not forestall normative progress. In Chapter 2, I offered an example of a way in which such disputes can result in neglect of important normative problems. My use of the arrow analogy suggests that definitions and ontological status are irrelevant to racial health disparities or unequal treatment in the criminal justice system-there is no use in suspending investigation of the Ferguson, Missouri Police Department's racist actions until we resolve what we mean by "race" and whether that word names something real. Even so, at some level of discourse, we will have to get straight on our terms. In fact, initially, Mallon's own arguments against the semantic approach are a bit more tempered. "The semantic strategy is problematic," he says. "Race theory ought not to rely on finding the correct theory of reference to determine the appropriate use of ‘race' talk" (Mallon, 2006, p. 528). It is only later, when he 187 engages Glasgow, that his stance hardens, and he suggests excluding semanticism from the discussion altogether. To see why I break from Mallon's strong stance, consider also that his initial aim was to show that the semantic strategy does not do any work for THE normative question of whether to conserve or eliminate ‘race' talk. The capitalization, "THE," is meant to suggest that the question whether or not to conserve is the primary, fundamental, guiding normative question in race theory. Mallon thinks that THE normative question should be addressed with arguments that do not depend on quibbles over definitions and are not suspended while we argue about what counts as real so that we can determine if race is real. To this point, I am on board with Mallon's antisemanticism. But, to say that the conservationism-eliminativism question is THE normative question is not to say that it is the only normative question. Once we decide whether to conserve (or eliminate) ‘race' talk, we will be faced with more specific, context-dependent normative questions. So, I suggest thinking of the normative racial realm as consisting of levels. At the highest, most general level, the question is simply whether to conserve ‘race' talk, and semantic approaches do not appear to do much work at that level. At subordinate levels, however, I suggest that we will have to get clear on our terms (albeit, I think Mallon is correct that even those decisions should not be held up by descriptivist-versuscausal-historical arguments over reference, for example). Moreover, Mallon sometimes appears to think likewise. In their description of hybrid constructionism, for example, Mallon and 188 Kelly discuss a series of studies about causes of academic underachievement among African American students. They report disagreement among the studies' conclusions, but pointing out inconsistency is not their purpose. Instead, they are making a point about the relative stability of racial conceptions, which flies in the face of social constructivist commitments to instability in classification and attitudes. In so doing, they argue that "Unless we take the predicate ‘Black' to have a common meaning purporting to pick out a single sort of person, it is unclear how a social role structured by such meaning could perform explanatory work stably across time and space" (Mallon & Kelly, 2012, p. 515, emphasis added). So, clearly at some (subordinate) levels, meanings do matter even to Mallon and Kelly. We will sometimes have to agree upon quite broad definitions, when, for example, we want to speak generally about decreasing oppression of non-Whites. Other times, we will have to be more specific, such as when we note that African Americans appear to respond better to BiDil than other racial groups. Finally, we will sometimes be forced to speak about race in surprisingly specific ways. In a recent paper, for example, Sean A. Valles argues that our racial classification is often too coarse-grained to be effective in medical contexts. He notes two examples: that the "Black immigrant population does not have the same hypertension risk as US-born African Americans. Similarly, Finnish descendants have a far lower rate of cystic fibrosis than other Caucasians" (Valles, 2012, p. 3). Often, these racial (sub)populations are simply ignored or, at best, clumsily placed in our broadest racial categories. But recognizing subsets of these groups leads to practical payoffs. It gives us a 189 better, more fine-grained understanding of the meaning of "race" relevant to the context, so Valles' grain of analysis strategy allows for more efficient use of limited resources. PCR, as I have described it, addresses our needs for contextdependent semantic agreement, again by offering a psychological (rather than, say, a "Hardimonian semantic-conditional") foundation that is adequately stable for broad aims, yet malleable enough to accommodate social vicissitudes that affect ‘race' talk. So, the model described and defended here depends neither on a traditional analysis of the concept ‘race' (at the general level) nor a settled-for-all time definition of the word "race"; rather, it offers a framework on which to build a range of definitions that are sensitive to the demands of a given context. PCR maintains only that humans act on their conceptions of race-what they conceive race to be generally or in a given context. What we want to know, in part, is what conceptions and attitudes about race lead individuals and groups to racialize the way that they do. It likely does not matter what, if anything, is common to all of those conceptions, and, in the main, PCR does not concern itself such questions. It does not ask, "How do we define ‘race' generally?" It asks, "How and why are we using it in this context? What are its material and social consequences?" and so on. Ultimately, it is aimed at social progress, the topic of the next and final section. 6.4 What Psychological Constructivism Does The primary aim of the dissertation has been to describe certain functional, flexible conception of race, and that project is, for present purposes, 190 complete. The point of building and defending PCR, however, is to supply a framework for a large range of racial conceptions that can be used to do normative work. That is, I have not aimed at something like a correct conceptual analysis of ‘race', but rather at offering an analysis that can be used to approach the normative problems neglected by other accounts. As Chapter 2's discussion of medical contexts suggests, my investigation should be meaningful and pragmatically useful-like Wilson, I am specifically trying to avoid "an exercise in idle diplomacy" between social constructivists and ECP theorists by offering a "serious attempt to find the common ground" (Wilson, 2005, p. 15). The ultimate goal of any study of race ought to involve real-world progress on social matters. To that end, the reintroduction of Gannett in this chapter has brought me full circle. I began with a desire to offer a medium for racial discourse "in the conservationist spirit" that allows us to address ignored and precluded questions, such as the ones Gannett describes, and also bring fresh hope to more traditionally recognized ones. So, I conclude the chapter by considering three brief (and unavoidably incomplete) suggestions for ways that PCR could promote our normative aims. The first is the one familiar from the earlier discussion of Gannett, namely, biomedical contexts in which racial disparities are best explained by the interaction of social and biological factors. The second involves what is, at the time of writing, the most salient and incendiary social issue regarding race, treatment of non-White minorities, especially African Americans, in the criminal justice system. For the last, I suggest that PCR offers a way to 191 respond to scientific racism. Of course, many of the details of PCR must still be sorted out. Among other things, we will need to investigate the processes by which social and innate psychological factors interact-by what process, for example, does a looping effect such as the one described by Adriaens and De Block proceed? Furthermore, as Machery and Faucher (among others) point out, PCR will benefit from greater cross-cultural study to determine to what extent its foundational facets can be described as genuinely universal. Since the conceptual apparatus remains incomplete, so does the normative investigation. Even so, I think I can use PCR to initiate some normative progress and offer the promise of more, so I now turn-admittedly cautiously-to the three normative areas of concern. 6.4.1 PCR and Biomedical Contexts Although Gannett introduces it as a philosophical-conceptual problem, the dichotomization of race into exclusively biological or exclusively social explanations is a problem of significant practical import. Thus, Gannett's ultimate focus is a normative one-the preclusion of a range of questions about race that involve biological and social interaction. The chief philosophical obstacle to addressing interactionist contexts appears to be the absence of a reasonable interactionist model. Recall, for example, that Gannett is dubious even of Kitcher's exclusively biological interactionist position, calling it "curiously gerrymandered." More to the point, Hardimon's tone borders on mockery when refers to the "erroneous belief that there is an amorphous thing race that is (somehow!) both social and biological" 192 (Hardimon, 2013, p. 6). I have allowed that Hardimon is right that no "thing" (by which he appears to mean something like a natural kind) is both social and biological, but Hardimon mislocates the interactionist foundation, just as he did the essentialist element of racialism, by reading it into the racialized object. According to PCR, the seat of interaction is in the cognizing subject, not the object so cognized, so here is where it is useful to switch PCR's metaphorical description from "framework" to "nexus." My claim is that the opaque psychological mechanisms that give rise to racial phenomena provide a nexus in which social and biological beliefs are combined in a common conception of race. This is not to say, however, that we simply have a false belief that should be eradicated, despite the resistance to interactionist models in this context. Instead, PCR accommodates such interaction. Recall that Gannett says that, generally, sociocultural factors impact the distribution of genetic variants (and, hence biology) and that that race is "socially constructed by enlisting biologic differences and investing these with sociocultural meaning," but this falls short as an account of the recalcitrance and power of the socializing of the biological. One might expect that if social constructivism accounted for the recruitment of biological truths to conceptions of race, we would be rid of the false biological pretenses of folk conceptions of race. Social constructivists, after all, generally hold that biology is irrelevant to race. If, in contrast, race is regarded as psychologically constructed, one can make better sense of the extent to which race is biologized by the folk. According to PCR, race is imbued with biological essentialism, likely as a 193 result of racial cognition's dependence on innate biological cognition. Since, as Gil-White (1999) argues, humans process "ethnic groups (and a few related other social categories) as if they were ‘species'" (p. 515), we should expect races to be conceived as real, complex biosocial classes. So conceived, we should then expect races to, in some manner, become real (or "real-like")-through selective intraracial breeding, for example, that could help account for some biological and medical commonalities among people of a shared race. This could be important for several reasons, one of which is the prospect of a better account the reification of racial categories. Furthermore, as indicated above, particular PCRs are informed by social, historical, and political factors in the ways championed by social constructivists. As a result, conceiving of race as psychologically constructed means allowing for interaction of biological and social factors. The case of BiDil, the heart medication targeted for African Americans, illustrates the point. Shortly after the FDA approved BiDil, controversy ensued. Some saw the new drug as a positive sign both for African Americans and for racial equality more generally. Others took it as yet another way to advance positions of scientific racism. For example, Gannett (2010) notes that New York Times science reporter Nicholas Wade asked, "Is there a biological basis for race? If there is not, as many social scientists and other argue, how can a drug like BiDil work so well in one race?" (p. 364). Gannett (2010) weighs in, pointing out once again that "this focus on the question of the biological reality vs. social construction of race leaves many other more interesting and socially and 194 politically important questions unasked" (p. 364). So, how does PCR apply to the case? The quote above suggests that Wade does not recognize the difficulty and nuance of the issue-he seeks a simple conclusion in a biological world that, as philosophers of biology such as Sandra Mitchell and Helen Longino suggest, is maddening complex and often demands proportionally complex explanations (Longino, 2013; Mitchell, 2003, 2009). Congestive heart failure can occur as the result of many factors, some biological (such as a genetic propensity for hypertension) and some social (such as diets that increase risk of hypertension). Apparently assuming a sort of biological realism, Wade's thinking mires him on one side of the issue. Wade digs in on the biological side, which gives him the illusion of a simple answer: BiDil works because race is biological. But even allowing that biology has some role in race, statistical biological traits encompass only a fraction of racial traits shared by African Americans, who also are more likely than other racialized groups of Americans to be poor and, hence, have the sorts of diets that promote heart disease. BiDil appears to be a treatment not for a biological condition, but a multidimensional medical condition in which biological and social factors interact. Wade's confusion is understandable, if (as his language suggests) he adopts a biological realist position on race. He should consider PCR instead. PCR alleviates the tension that Hardimon rightly points out, allowing specification of relevant boundaries without gerrymandering, and most importantly gives us a way to respond to people like Wade, who support similar 195 notions about race. Since PCR holds that race is not a thing-at least not in the robust, intuitive sense-we are not tasked with discovering a "thing" that is, per impossible, both social and biological. It is true that in many contexts we often treat race as a thing to be discovered and defined; however, its apparent reality in this respect is more likely the result of its reification based on human cognition and actions. The only way to account for the complexity of race-its odd combination of the biological and social-is along psychological constructivist lines. PCR, then, allows us to make sense of human race in crucial interactionist contexts without sacrificing the biological influence on race and without reducing race to biology. In sum, if we choose to conceive of race as psychologically constructed, we can offer a more robust way of analyzing the interaction of biological and social factors in specific contexts, which is to say PCR in no way precludes the asking of the crucial questions Gannett brings to light. Having said this, however, I do not mean to limit the interactionist element to the field of biomedicine. In my view, biological-social interactionist conceptions of race inform many contexts. PCR has implications beyond medicine, so I turn now to the topical issue of race in criminal justice contexts, including police interaction. 6.4.2 PCR and the U.S. Criminal Justice System As I showed in Chapters 4 and 5, opaque psychology better explains than explicit attitudes the persistent unequal treatment of non-Whites in a host of contexts. Among the most important factors is that humans naturally sort themselves into groups on the basis of not only shared traits, but also a deeper 196 shared essence underlying those traits. So, although it may not be obvious that an interactionist conception is relevant to criminal justice contexts, if one recognizes the biological essentialist element of racial cognition, the claim becomes clear. Recall also that we do not apply a merely generic essence to human groups; rather, as the implicit attitude literature suggests, we, again, naturally and unconsciously, imbue racial essence with particular traits, some of which are negative and oppressive. As a result, persons can be and often genuinely are explicitly nonracist and still harbor undetected implicit racial bias. So, since opaque psychology is foundational for PCR, I think it sheds light on persistent problems for racial minorities in legal contexts. One striking factor in these incidents is the tendency for the police officers involved to claim no racial bias (Kelly et al., 2010; Staats, 2013). Of course, consciously suppressed explicit bias remains a problem in the U.S.-see my citation of the U.S. Justice Department's conclusions on the Ferguson Police Department in the previous chapter, for example.36 There is little doubt, however, that at least some, if not most, of the officers involved in racialized incidents sincerely identify as nonracist. That is, of course, a good thing, but it also helps explain some of the inflammatory defensiveness of (especially White) officers 36 Also, I might relate-anecdotally-that while serving the U.S. Army I met many fellow soldiers who intended to transition from the military to positions as police and corrections officers. Many were well-intentioned-if one believes in the positive values of military service, those values are easily realized in police work as well. However, not all or even most but a significant number voluntarily reported other motives. One told me-no kidding-that he wanted to become a corrections officer because you get to "beat the shit out of niggers and spics," and, while not referencing his postmilitary career directly, another who aimed at a career in law enforcement had the habit of describing racial minorities, especially Blacks, as "just animals anyway." This is not intended as an indictment of all police and corrections officers, of course, but as a reminder that explicit racism is still rampant both institutionally (again, see the Ferguson case) and individually. 197 when accused of racist motivations. Police officers are required to take an oath that binds them to serve "without favor or affection, malice or ill-will" (connect.lawofficer.com). So officers might reason that by virtue of willingness to take an oath of impartiality, they take an oath to be nonracist. Data show, however, that racial minorities are treated unequally by police in a variety of ways. Even in San Francisco, often considered among the most progressive cities in the United States, "Black adults...are much more likely to be arrested" (sanfrancisco.cbslocal.com). In addition, of "79 people in South Carolina who were fatally shot by police, 43%...were African American," even though African Americans comprise only 29% of the state's population," and despite accounting for only 13% of the U.S. population, nationally, African Americans were subject to 30% of police killings between April 2013 and March 2015 (Fischer-Baum & Bialik, 2015). African Americans also "fare worse in court" in a number of ways-they receive (on average) 25% higher bail fees, 12year longer sentences, and higher rates of capital punishment (Rachlinski et al., 2009, p. 1195). I concede that these data do not preclude social constructivist explanations. One could develop a story according to which contingent historical and cultural events along with intentional human decisions account of a long legacy of racist police and judicial mistreatment. But is that the best explanation? Consider the increasing social pressure against inegalitarian treatment. Can a judge or police officer risk social backlash that could affect both their 198 professional and personal lives?37 Intuitively, a better explanation is that, in many of these cases, implicit bias against essentialized racial groups plays a role. Moreover, the empirical data on implicit bias back that intuition. Recall that the Kirwan report shows that implicit bias is manifest in many ways in the criminal justice system, from the initial interaction with police-in which a suspect may be subject to weapons bias effect-to charge and trial to sentencing and appeal (Staats, 2013, p. 36). Moreover, recall that these independent influences can accumulate, causing an amplifier effect (Staats, 2013, p. 36). In short, essentialist thinking and implicit bias are relevantly associated with the phenomena and mesh with the intuition that most police and judges work to suppress bias when they discover it in themselves. As a result, "mere" social constructivist explanations appear inadequate to fully explain the phenomena. Give that, understanding race as PCR is advantageous in at least two ways. First, it better explains the persistence of unjust treatment of minorities in the justice system, given the apparent fact of social pressure against racism. Assuming that a majority of police officers and judges feel a duty toward egalitarian treatment, PCR explains why explicit attitudes and actions frequently diverge. Second, it helps societies better address the social ills tied to unequal police treatment. If implicit bias is active in these contexts, then instead of continuing to prescribe social nostrums, interventions and remedies can be designed to address race and racial attitudes that are psychologically 37 One could object that police officers are often exonerated in these cases, but there is no guarantee of exoneration, so explicit racist motivation would have to outweigh the social and legal risks. 199 constructed. So, at the risk of repetition, it bears mentioning that given the level and persistence of unequal treatment of minorities in the U.S. justice system, social constructivist explanations no longer appear utile. We need a new way of attacking these problems. PCR not only supplies one, but, as I have repeatedly argued, supplies one on the basis of the best available evidence on racial cognition. 6.4.3 PCR and Scientific Racism Although I have concentrated on some of her other arguments, Gannett (2010) also takes up the issue of scientific racism and its standing in race scholarship. Gannett (2010) reports that "the biological reality of race is a hot topic these days, for scientists and nonscientists, philosophers and nonphilosophers alike" (p. 381). Recall that Andreasen, Kitcher, and Pigliucci and Kaplan all attempt to rehabilitate or reinvent biological race along nonracialist, nonracist lines. I have also related that Collins, the former director of the Human Genome Project, concedes that "it is not strictly true that race or ethnicity has no biological connection" (Collins, 2004, p. S13). This news should not be assumed bad, however. Inasmuch as biology can inform the study of race, it should bring benefits, especially to biomedicine, so long as we keep in mind Gannett's claim that the connections between race and biology will remain "statistical not universal, interest-relative not mind independent, dynamic not static, [and] indeterminate not determinate" (Gannett, 2010, p. 383). Alarmingly, these points do not seem to be ones to which all biologically-minded race theorists ascribe. Not everyone forming connections 200 between biology and race, that is, does so in such meliorative terms. For example, Gannett (2010) reports that in Race: The Reality of Human Difference, Vincent Sarich and Frank Miele opine that "significant cognitive differences arising during the past 10,000 years among geographically human groups as cultures...makes it inevitable that racial groups will be disproportionately represented at the extremes of values for traits like being a criminal or having high-paying jobs" (pp. 381-2). The most well-known version of this line of argument comes from Richard Jensen, who pioneered modern scientific racism via his race-IQ studies, which in turn influenced the reasoning in the publicly recognized book The Bell Curve by Richard Herrnstein and Charles Murray. In crude terms, Jensen's main argument goes like this. Intelligence, as measured by IQ tests, has a heritability of approximately 80% in White populations. Furthermore, the average IQ score of Blacks is 15 points lower than the average IQ score of Whites. Since, IQ is highly heritable (at least in Whites), it is very probable that the IQ difference is largely due to genetic factors. From this argument, Jensen drew several conclusions with social implications. Among the most salient is that Compensatory education has been tried and apparently failed….Why has there been such uniform failure in compensatory education programs wherever they have been tried? What has gone wrong? In other fields, when bridges do not stand, when aircraft do not fly, when machine do not work, when treatments do not cure, despite all conscientious efforts on the part of many persons to make them do so, one begins to question the basic assumptions, principle, theories, and hypotheses that guide one's efforts. (Jensen, 1969, p. 3) 201 More recent works by Wade and Sesardic make similar claims. As James Tabery says, Sesardic's Making Sense of Heritability offers an "acrid, bitterly antagonistic contribution to the nature-nurture debate" (Tabery, 2009, n.p.); most of Sesardic's acerbity is aimed at philosophers of science he opposes, most of his worrisome conclusions at Black folks. Tabery notes that Sesardic commits himself to several controversial conclusions. He argues, for example, that given evidence of genetically caused differential IQ scores between Whites and Blacks, egalitarianism can survive only with modification (Tabery, 2009). But Sesardic does not stop at the conclusion that Blacks are inherently less intelligent than Whites; he argues that the average IQ of Blacks is likely overestimated. He argues, for example, that based on Bayesian analysis, if an American Black person and an American White person each score 100 on an IQ test, "then if we know nothing else about these two persons and if we want to get the best estimate of their true IQs on the basis of their measured scores, we should ascribe a lower IQ to the Black person than to the White person" (Sesardic, 2010, p. 225). This is simply a matter, he says, of regression to the mean, which "tells us if there are two different populations with different trait means and if the trait is normally distributed in both populations, then in any unbiased study that is not 100 percent reliable, measured individual scores should always be" corrected toward the mean. So, since IQ testing is not "100 percent reliable," and since the average White IQ score is 100, while Black average is 85, we correct by assuming that the Black subject who scores 100 actually has an IQ closer to the mean. In short, since "the Black mean is lower 202 than the White mean, Black true scores will have a ‘downward pull' compared to White true scores" (Sesardic, 2010, p. 225). According to Sesardic (2010), then, it's not just that Blacks are inherently less intelligent than members of other races. Their intellectual shortcomings exert dumbing gravitational power. While Sesardic never explicitly claims that race is a natural kind, as Andreasen and Kitcher, Pigliucci and Kaplan do, many of his conclusions rest on a de facto commitment to a natural kind view of race. PCR, I argue, places a new and formidable obstacle in path of scientific racists. To show this, let me begin with a roughly Kantian analogy. Hilary Putnam (1992) says that "Kant was the first really to see that describing the world is not simply to copy it. Kant saw that whenever human beings describe anything in the world, our description is shaped by our conceptual choices" (p. 28). Similarly, Mitchell (2003) says that since "Kant, most philosophers accept that every representation will be shaped, in part, by the concepts that humans bring to the task of describing the world" (p. 182). Even part of Gannett's quote from the previous page gets at this point. When we divide the world's population into races, we should consider those "cuts in nature to be interest-relative not mindindependent" (Gannett, 2010, p. 383, emphasis added). This point is one that scientific racists apparently ignore. That is, they treat race as though it were just copied from the world and neglect the innate conceptual, schematic, and classificatory contributions of human minds. Of course, in light of evidence from ECP, this simply won't do-we can't hope to understand race, unless allowance is given to the mind's interaction with 203 what is presented to it. In Kantian terms, it's as though-and I do not mean this literally, of course-scientific racists believe that they have access to the realm of racial noumena, forgetting that they experience only phenomena and, in so doing, they forget that minds interact with experience to form the classifications that guide inference. This way of framing the problem may seem pedantic or unnecessarily technical, so let me offer an example that helps secure the point. The problem for Jensen and like-minded scholars, of course, is that "all conscientious efforts" have not been exhausted. PCR, in fact, suggests that Jensen suffered from far too narrow conception of all that counts as "environmental." Most notably, through no fault of his own,38 he was not able to consider the impact of implicit bias on educational achievement, which the Kirwan study suggests is every bit as important to education as to law enforcement and to medicine. Again, "Implicit bias can permeate educational settings in several forms, all of which can yield disadvantageous consequences for students of color. Teacher expectations of student achievement, teacher perceptions of student behavior, and students' self-perceptions" are three of the most important factors (Staats, 2013, p. 30). This is not a problem for Jensen alone; it is shared by all of the scientific racists mentioned above. Each gives too much weight to genetic factors, and each appears to completely ignore the effects of implicit racial bias in reaching their very strong conclusions.39 By beginning with innate psychology, PCR not only avoids that problem, but Jensen's 1969 paper was published well in advance of implicit bias research on race. This is especially embarrassing for Wade and Sesardic, each of whom writes well after the advent of implicit bias research. 38 39 204 provides an additional problem for scientific racists. If they want to explain away environmental factors in race-relevant issues such as educational achievement and criminality,40 they have to explain away the psychological construction of race, and this would be a formidable task. And the deeper their commitments to biological race, the more their theories predict, the further are they removed from the critical insight that opaque psychology determines a lot of what constitutes our racial representations and attitudes. Moreover, introducing race as a psychological construct facilitates correction of the effects of implicit bias in the education context. Interventions with educators could, for example, focus not only on socially constructed aspects of race, but also on the opaque psychological mechanisms that underpin them. As such, establishing PCR facilitates fresh approaches to long-standing problems that strictly social interventions have yet to solve. 6.5 Conclusion I have used this final substantive chapter to describe the psychological construction of race and its meliorative potential for a number of practical contexts. Though it does support contextual definitions of race, PCR does not recapitulate the errors of the semantic strategy, and though it supports social construction, psychologically constructed race is not merely socially constructed 40 I should say I find the criminality point particularly objectionable, even before we consider implicit bias, because that argument depends on what we conceive of as criminal. To non-Whites, I imagine that given the socio-political status of the world's peoples, White actions must seem quite criminal. It seems patent that Whites are historically responsible for the most intense and frequent acts of criminality, violence, and aggression. Perhaps an interesting psychological study could focus on why White people appear to project these attributes onto other racialized groups. Perhaps that is also a matter of innate psychology. 205 race. In particular, PCR avoids the pitfalls of those traditional philosophical approaches by building insights of the evolutionary-cognitive program and implicit attitude research into its foundation. PCR is not alone in this move, however. At least two other approaches-Wilson's evolutionary social construction and Mallon and Kelly's hybrid constructionism-proceed according to those projects as well. Even so, I have shown that PCR, even if it resolves into one of those positions over the long-term, is distinguished from them in several crucial ways. Finally, given the full description of PCR, I argued that it answers several unmet needs of three salient contexts that have important consequences for racial disparities. CHAPTER 7 CONCLUSION Holmes (2015) calls postracial America a "fantasy," and nothing I've written here contradicts that claim. In fact, if race is psychologically constructed, then it is the case not only that we are not yet postracial, but that postracialism is close to impossible. That is not to say that human racial classifications and attitudes are static; they will change with social progress and scientific discovery. But, if races are PCRs, then some sort of racial classification is bound to persist until the human mind evolves away from the mechanisms that inform racial cognition. A review of what I have written here shows why. Again, I used Gannett (2010) because the problem she highlights is not merely interesting, but also representative. In other words, the dichotomization of race in biomedical contexts is not unlike its status in other contexts. Virtually everywhere race is relevant, it is constructed from disparate, sometimes intuitively contradictory parts. So, the point was not just to address her immediate concern, but also to describe a position versatile enough to inform racial discourse over a variety of contexts. If opaque psychology is an everpresent factor in racial cognition, then PCR is well-equipped to handle a host of racialized issues. It functions better, I have argued, than its rivals in that capacity. It is, unlike the semantic strategy, highly relevant to pressing normative 207 racial concerns. Since PCR's foundation is built of material supplied by the evolutionary-cognitive program and implicit bias research, it accommodates and explains facets of race that social constructivism cannot-it is not, for example, undermined by the "five lines of argument" presented by Kelly et al. (2010) against the favored explanations of the received view, and it can naturally account for dissociation between explicit and implicit racial bias. Although it clearly belongs to the family of complementarian approaches champions by Wilson, Mallon, and Kelly, it can be distinguished from them in useful ways. I have described it as a specifically accretive, rather than a broadly constraint-ist type of constructivism. That is to say, while it may be a species of hybrid (constraint-ist) constructivism, it is distinguished by the specific way in which in which the constraints it offers are understood. PCR's guiding commitment is that opaque psychological mechanisms are foundational to some social constructions, that is, that the individual psychology constructs, or plays a leading role, in the subsequent constructions. So conceived, PCR offers a versatile approach to recalcitrant social problems surrounding race. It facilitates effective racial discourse in a host of normatively problematic contexts, including health care, criminal justice, and education. That's good news, and given the prospect of goods news, I want to finish with a few positive reflections. It's so easy to become cynical when considering race and the social issues that surround it. In fact, it's far too easy to become far too cynical. But things likely have gotten and certainly can get better. When I was a kid in Alabama, people would sometimes ask, "Are ya'll racist," meaning, "do you and 208 your family subscribe to racism?" Among some of my peers, such a question would have been offensive, of course, but it was not considered so by the folks asking the question. They would ask in the manner that they might ask whether we were Democrat or Republican, or whether we supported University of Alabama or Auburn University football. Even given my strictly antiracist upbringing, the question did not seem unnatural. Today, I would respond to such a query with incredulity. My mind would race-"What could someone intend with that sort of question? I am bald; are they really asking if I am a White supremacist?" The point is that it's no longer okay, I assume even in Alabama, to ask whether someone or someone's family is racist, especially in a manner as casual as one used to ask whether we eat at the table or in front of the television. My students rarely recognize the word "miscegenation," and when I explain antimiscegenation laws, they're not offended, they're incredulous. It may not seem so at times in the United States, but in many ways, things are getting better. And that view doesn't follow merely from my anecdotal experiences. It wasn't just President Obama's election in 2008; several events coalesced to give the impression of a postracial United States. As Holmes (2015) notes, "data from 2008-9 showed that one in seven new marriages was between spouses of different racial or ethnic backgrounds" and an earlier New York Times report noted that "some people felt that ‘the blending of the races is a step toward transcending race, to a place where America is free of bigotry, prejudice and programs like affirmative action." And while many of the incendiary racial events 209 described earlier led to disillusionment, despair, and even some violent backlash, reaction to the Charleston shooting supplied hope and renewed concern. Shortly after the tragic event, White Christian groups gathered outside of the Emanuel African Methodist Episcopal Church to pray for it and its victims. Addressing the crowd, Reverend Brandon Bowers, a White man, said, "As a pastor in this city, a husband and a father to two boys and two girls, my heart broke in grief and disbelief….What the enemy intended for evil, God is using for good. We are here to pray for the healing that needs to come" (Eligon & Fausset, 2015, n.p.). Meanwhile, inside, Reverend Norvel Goff Sr., a Black man, claimed that the united gathering of Black and White Charlestonians "sends a message to every demon in hell and on earth. Some wanted to divide the race-Black and White and brown-but no weapon formed against us shall prosper" (Eligon & Fausset, 2015, n.p.). A sign outside read, "Holy City...Let Us Be the Example of Love That Conquers Evil" (Eligon & Fausset, 2015, n.p.). The unified reactions to the tragedy placed renewed pressure on Southern states to take down vestigial symbols of the Confederacy. The seeds of racism are in us, cooked into our broth. But so are the seeds of amelioration and union. We must renew our efforts to fight racism, and those efforts require new tools. But progress has been and can be made. A Black man became President of the United States, after all. REFERENCES Adams, K., Sabolinski, M., Worcel M., & Cohn J. N. (2004). Combination of isosorbide dinitrate and hydralazine in Blacks with heart failure. New England Journal of Medicine, 351(20), 2049-57. Adriaens, P. R., & de Block, A. (2006). The evolution of social construction: The case of male homosexuality. Perspectives in Biology and Medicine, 49(4), 570-585. Alcoff, L. M. (2003). Latino/as, Asian Americans, and the Black-White binary. The Journal of Ethics, vii, 5-27. Amodio, D. M., & Mendoza, S. A. (2010). Implicit intergroup bias: Cognitive, affective, and motivational underpinnings. In B. Gawronski & B. K. Payne (Eds.), Handbook of implicit social cognition (pp. 253-274). New York, NY: The Guilford Press. Andreasen, R. O. (1998). Race: Biological reality or social construct? Philosophy of Science, 67, S653-S666. Andreasen, R. O. (1998). A new perspective on the race debate. The British Journal for the Philosophy of Science, 49(2), 199-225. Andreasen, R. O. (2005). The meaning of ‘race': Folk conceptions and the new biology of race. Journal of Philosophy, 102(2), 94-106. Anzaldua, G. (2007). Borderlands/la frontera: The new Mestiza (3rd ed.). San Francisco, CA: Aunt Lute. Appiah, K. A. (1992). In my father's house: Africa in the philosophy of culture. New York, NY: Oxford University Press. Appiah, K. A. (2006). How to decide if races exists. Proceedings of the Aristotelian Society, 106(1), 365-382. Appiah, K. A., & Gutman, A. (1996). Colour conscious: The political morality of race. Princeton, NJ: Princeton University Press. Bamshad, M., Wooding, S., Salisbury, B. A., & Stephens, J. C. (2004). 211 Deconstructing the relationship between genetics and race. Nature Reviews Genetics, 5, 598-609. Banaji, M. R. (2001). Implicit attitudes can be measured. In H. L. Roediger III., J. S. Nairne, I. Neath, & A. Surprenant (Eds.), The nature of remembering: Essays in honor of Robert G. Crowder. Washington, DC: American Psychological Association. Banaji, M. R., Lemm, K. M., & Carpenter, S. J. (2001). Automatic and implicit processes in social cognition. In A. Tesser & N. Schwarts (Eds.), Blackwell handbook of social psychology: Intraindividual processes (pp. 134-158). Oxford, UK: Blackwell. Beaton, B. (2007). Racial science now: Histories of race and science in the age of personalized medicine. The Public Historian, 29(3), 157-162. Beeghly, E. (2015). What is a stereotype? What is stereotyping? Hypatia, 30(2), 675-691. Bharadwaj, A. (2008). Biosociality and biocrossings: Encounters with assisted conception and embryonic stem cells in India. In S. Gibbon & C. Novas (Eds.), Biosociality, genetics, and the social sciences: Making biologies and identities [Kindle Edition]. Retrieved from http://www.amazon.com. Bolnick, D. A. (2008). Individual ancestry inference and the reification of race as a biological phenomenon. In B. A. Koenig, S. J. Lee, & S. S. Richardson, (Eds.), Revisiting race in a genomic age (pp. 70-85). Rutgers, NJ: Rutgers University Press. Boxhill, B. (Ed.). (2001). Race and racism. New York, NY: Oxford University Press. Boyd, R. (1999). Homeostasis, species, and higher taxa. In R. A. Wilson (Ed.), Species: New interdisciplinary essays (pp. 141-186). Cambridge, MA: The MIT Press. Boyd, R. (1991). Realism, anti-foundationalism and the enthusiasm for natural kinds. Philosophical Studies, 61, 127-148. Boyd, R. (1989). What realism implies and what it does not. Dialectica, 43, 5-29. Buller, D. J. (2005). Adapting minds. Cambridge, MA: The MIT Press. Coats, T. P. (2007, February 1). Is Obama Black enough? Time. Retrieved from http://content.time.com/time/nation/article/0,8599,1584736,00.html 212 Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge, MA: The MIT Press. Chomsky, N. (2000). New horizons in the study of language and mind. Cambridge, UK: Cambridge University Press. Chomsky, N. (2002). On language and nature. Cambridge, MA: Cambridge University Press. Condit, C. M., Parrott, R. L., Harris, T. M., Lynch, J., & Dubriwny, T. (2004). The role of "genetics" in popular understandings of race in the United States. Public Understanding of Science, 13(3), 249-272. Condit, C. M., & Bates, B. (2005). How lay people respond to messages about genetics, health, and race. Clinical Genetics, 68(2), 97-105. Cooper, L. A., Roter, D. L., Carson, K. A., Beach, M. C., Sabin, J. A., Greenwald, A. G., & Inui, T. S. (2012). The associations of clinicians' implicit attitudes about race with medical visit communication and patient ratings of interpersonal care. American Journal of Public Health, 102(5), 979-987. Cosmides, L., Tooby, J., & Kurzban, R. (2003). Perceptions of race. Trends in Cognitive Science, 7(4), 173-179. Cox, K., Tayles, N. G., & Buckley, H. R. (2006). Forensic identification of "race": The issues in New Zealand. Current Anthropology, 47(5), 869-875. Cunningham, W. A., Johnson, M. K., Raye, C. L., Gatenby, J. C., Gore, J. C., & Banaji, M. R. (2004). Separable neural components in the processing of Black and White faces. Psychological Science, 15(12), 806-813. Daniel, G. R. (2006). Race and multiraciality in Brazil and the United States: Converging paths? University Park, PA: The Pennsylvania State University Press. Dawson, M. C., & Bobo, L. (2009). One year later and the myth of a postracial society. Du Bois Review, 6(2), 247-249. Degler, C. N. (1971). Neither Black nor White: Slavery and race relations in Brazil and the United States. New York, NY: MacMillan. Dingel, M. J., & B. A. Koenig (2008). Tracking race in addiction research. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 172-197). Rutgers, NJ: Rutgers University Press. 213 Doris, J. M. (Ed.). (2010). The moral psychology handbook. Oxford, UK: Oxford University Press. Downes, S. (2014). Evolutionary psychology. The Stanford encyclopedia of philosophy (Summer 2014). E. N. Zalta (Ed.). Retrieved from http://plato.stanford.edu/archives/sum2014/entries/evolutionarypsychology/ Du Bois, W. E. B. (1986). Writings. New York, NY: Penguin. Dupre, J. (2008). What genes are and why there are no genes for race. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 39-55). Rutgers, NJ: Rutgers University Press. Edwards, A. W. F. (2003). Human genetic diversity: Lewinton's fallacy. BioEssays, 25, 798-801. Ekman, P. (1992). Are there basic emotions? Psychological Review, 99(3), 550553. Epstein, S. (2010). Beyond inclusion, beyond difference: The biopolitics of health. In I. Whitmarsh & D. S. Jones (Eds.), What's the use of race? Modern governance and the biology of difference (pp. 63-87). Cambridge, MA: The MIT Press. Eversley, M., & James, M. (2014, December 3). No charges in NYC chokehold death; federal inquiry launched. USA TODAY. Retrieved from https://www.usatoday.com/story/news/nation/2014/12/03/chokeholdgrand-jury/19804577/ Fazio, R. H., Sanbonmatsu, D. M., Powell, M. C., & Kardes, F. R. (1986). On the automatic activation of attitudes. Journal of Personality and Social Psychology, 50, 229-238. Feldman, M. W., & Lewontin, R. (2008). Race, ancestry, and medicine. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 89-101). Rutgers, NJ: Rutgers University Press. Fitzsimmons, E. (2014, November 24). 12-year-old boy dies after police in Cleveland shoot him. New York Times. Retrieved from https://www.nytimes.com/2014/11/24/us/boy-12-dies-after-being-shot-bycleveland-police-officer.html Fodor, J. A. (1983). The modularity of mind. Cambridge, MA: The MIT Press. Friese, M., Smith C. T., Plischke, T., Bluemke M., & Nosek B. A. (2012). Do 214 implicit attitudes predict actual voting behavior particularly for undecided voters? PLoS One, 7(8), e44130. Fujimura, J. H., Rajagopalan, R., Ossorio, P. N., & Doksum, K. A. (2010). Race and ancestry: Operationalizing populations in human genetic variation studies. In I. Whitmarsh & D. S. Jones (Eds.), What's the use of race? Modern governance and the biology of difference (pp. 169-183). Cambridge, MA: The MIT Press. Fullwiley, D. (2008). The molecularization of race: U.S. health institutions, pharmacogenetics practice, and public science after the genome. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 149-171). Rutgers, NJ: Rutgers University Press. Gaertner, S. L., & McLaughlin, J. P. (1983). Racial stereotypes: Associations and ascriptions of positive and negative characteristics. Social Psychology Quarterly, 46(1), 23-30. Gannett, L. (2010). Questions asked and unasked: How by worrying less about the ‘really real' philosophers of science might better contribute to debates about genetics and race. Synthese, 177, 363-385. Gannett, L. (2004). The biological reification of race. British Journal for the Philosophy of Science, 55, 323-345. Gannett, L. (2003). Making populations: Bounding genes in space and time. Philosophy of Science, 70, 989-1001. Gibbon, S. (2008). Charity, breast cancer activism and the iconic figure of the BRCA carrier. In S. Gibbon & C. Novas (Eds.), Biosociality, genetics, and the social sciences: Making biologies and identities [Kindle Edition]. Retrieved from http://www.amazon.com Gibbon, S., & Novas, C. (2008). Introduction: Biosociality, genetics, and the social sciences. In S. Gibbon & C. Novas (Eds.), Biosociality, genetics, and the social sciences: Making biologies and identities [Kindle Edition]. Retrieved from http://www.amazon.com Gilroy, P. (2000). Against race: Imagining political culture beyond the color line. Cambridge, MA: Harvard University Press. Gil-White, F. J. (2001). Are ethnic groups biological ‘species' to the human brain? Current Anthropology, 42(4), 515-554. Gil-White, F. J. (2001). Sorting is not categorization: A critique of the claim that Brazilians have fuzzy racial categories. Psychological Review, 102(1), 215 219-249. Griffiths, P. E. (1997). What emotions really are. Chicago, IL: The University of Chicago Press. Glasgow, J. (2003). On the new biology of race. The Journal of Philosophy, 100(9), 456-474. Glasgow, J. (2009). In defense of a four-part theory: Replies to Hardimon, Haslanger, Mallon, and Zack. Symposia on Gender, Race and Philosophy 5(2), 1-19. Glasgow, J. (2009). A theory of race. New York, NY: Routledge. Greely, H. T. (2008). Genetic genealogy: Genetics meets the marketplace. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 215-234). Rutgers, NJ: Rutgers University Press. Green, A. R., Carney, D. R., Pallin, D. J., Ngo, L. H., Raymond, K. L., Iezzoni, L. I., & Banaji, M. R. (2007). Implicit bias among physicians and its prediction of Thrombolysis decisions for Black and White patients. Journal of General Internal Medicine, 22(9), 1231-1238. Greenwald, A. G., Banaji, M. R., & Nosek, B. A. (2015). Statistically small effects of the implicit association test can have societally large effects. Journal of Personal and Social Psychology, 108(4), 553-661. Greenwald, A., & Krieger, L. H. (2006). Implicit bias: Scientific foundations. California Law Review, 94(4), 945-986. Häggqvist, S. (2005). Kinds, projectibility and explanation. Croatian Journal of Philosophy, 5(13), 71-87. Hacking, I. (1999). The social construction of what? Cambridge, MA: Harvard University Press. Hacking, I. (1995). The looping effect of human kinds. Causal Cognition, 351383. Hacking, I. (2006). Genetics, biosocial groups & the future of identity. Daedalus, 135(4), 81-95. Hacking, I. (2007). Natural kinds: Rosy dawn, scholastic twilight. Royal Institute of Philosophy Supplement, 61, 203-239. Haslanger, S. (2000). Gender and race: (What) are they? (What) do we want 216 them to be? Noûs, 34(1), 31-55. Haslanger, S. (1995). Ontology and social construction. Philosophical Topics, 23(2), 95-125. Haslanger, S. (2008). A social constructionist analysis of race. In B. A. Koenig, S. J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 56-69). Rutgers, NJ: Rutgers University Press. Haslanger S. (2005) You mixed? Racial identity without racial biology. In S. Haslanger & C. Witt (Eds.), Adoption matters: philosophical and feminist essays (pp. 265-289). Ithaca, NY: Cornell University Press. Hawkins, C. B., & Nosek, B. A. (2012). Motivated independence? Implicit party identity predicts political judgments among self-proclaimed independents. Personality and Social Psychology Bulletin, 38(11), 437-452. He, X.-S., Ji, X., Hale, M. B., Cheung, R., Ahmed, A., Guo, Y., . . . Greenberg, H. B. (2006). Global transcriptional response to Interferon is a determinant of HCV treatment outcome and is modified by race. Hepatology, 44(2), 352359. Herrnstein, R. J., & Murray, C. A. (1996). The bell curve: Intelligence and class structure in American life. New York, NY: Simon & Schuster. Hinterberger, A. (2010). The genomics of difference and the politics of race in Canada. In I. Whitmarsh & D. S. Jones (Eds.), What's the use of race? Modern governance and the biology of difference (pp. 147-167). Cambridge, MA: The MIT Press. Hirschfeld, L. A. (1996). Is the acquisition of social categories based on domainspecific competence or on knowledge transfer? Race in the making: Cognition, culture, and the child's construction of human kinds (pp. 201233). Cambridge, MA: The MIT Press. Hirschfeld, L. A. (2001). On a folk theory of society: Children, evolution, and mental representations of social groups. Personality and Social Psychology Review, 5(2), 107-117. Hirschfeld, L. A. (2012). Seven myths of race and the young child. Du Bois Review: Social Science Research on Race, 9(1), 17-39. Holmes, A. (2015, July 15). America's ‘postracial' fantasy. New York Times. Retrieved from https://www.nytimes.com/2015/07/05/magazine/americaspostracialfantasy.html?emc=edit_tnt_20150701&nlid=66648065&tntemail0=y&_r=0 217 Hugenberg, K., & Bodenhausen, G. V. (2003). Facing prejudice: Implicit prejudice and the perception of facial threat. Psychological Science, 14(6), 640-643. James, M. (2012). Race. The Stanford encyclopedia of philosophy (Fall 2015). E. N. Zalta (Ed.). Retrieved from http://plato.stanford.edu/archives/fall2015/entries/race/ Jenks, A. C. (2010). What's the use of culture? Health disparities and the development of culturally competent health care. In I. Whitmarsh & D. S. Jones (Eds.), What's the use of race? Modern governance and the biology of difference (pp. 207-224). Cambridge, MA: The MIT Press. Kahn, J. (2008). Patenting race in a genomic age. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in genomic age (pp. 129-148). Rutgers, NJ: Rutgers University Press. Kahn, J. (2010). What's the use of race in presenting forensic DNA evidence in court? In I. Whitmarsh & D. S. Jones (Eds.), What's the use of race? Modern governance and the biology of difference (pp. 27-48). Cambridge, MA: The MIT Press. Kang, J. (2005). Trojan horses of race. Harvard Law Review, 118(5),1489-1593. Kang, J. (2010). Implicit bias and the pushback from the left. St. Louis University Law Journal, 54, 1139-1150. Kang, J. (2012). Communications law: Bits of bias. In J. D. Levinson & R. J. Smith (Eds.), Implicit racial bias across the law (pp. 132-145). Cambridge, UK: Cambridge University Press. Kang, J., Bennett, M., Carbado, D., Casey, P., Dasgupta, N., Faigman, D., . . . Mnookin, J. (2012). Implicit bias in the courtroom. UCLA Law Review, 59(5), 1124-1186. Kant, I. (1965). Critique of pure reason. (N. K. Smith, Trans.). New York, NY: St. Martin's Press. (Original work published 1929) Kaufman, J. S., & Cooper, R. S. (2010). Use of racial and ethnic identity in medical evaluations and treatments. In I. Whitmarsh & D. S. Jones (Eds.), What's the use of race? Modern governance and the biology of difference (pp. 187-206). Cambridge, MA: The MIT Press. Kelly, D., Machery, E., & Mallon, R. (2010). Race and racial cognition. In J. M. Doris (Ed.), The moral psychology handbook (pp. 432-471). Oxford, UK: Oxford University Press. 218 Kitcher, P. (2007). Does ‘race' have a future? Philosophy & Public Affairs, 35(4), 293-317. Konigsberg, L.W., Algee-Hewitt, B. F., & Steadman, D. W. (2009). Estimation and evidence in forensic anthropology: Sex and race. American Journal of Physical Anthropology, 139(1), 77-90. Krieger, N. (2010). The science and epidemiology of racism and health: Racial/ethnic categories, biological expressions of racism, and the embodiment of inequality-an ecosocial perspective. In I. Whitmarsh & D. S. Jones (Eds.), What's the use of race? Modern governance and the biology of difference (pp. 225-255). Cambridge, MA: The MIT Press. Kteily, N., & Cotterill, S. (2015). Is the defendant white or not? New York Times. Retreived from https://www.nytimes.com/2015/01/25/opinion/sunday/isthe-defendant-white-or-not.html Kuhn, T. S. (1996). The structure of scientific revolutions (3rd ed.). Chicago, IL: The University of Chicago Press. Kurzban, R., Tooby, J., & Cosmides, L. (2001). Can race be erased? Coalitional computation and social categorization. Proceedings of the National Academy of Sciences of the United States of America, 98(6), 1538715392. Lacquer, T. W. (1990). Making sex: Body and gender from the greeks to Freud. Cambridge, MA: Harvard University Press. Lamport-Stokes, M. (2014, February 1). NBA-Silver takes over as commissioner from Stern. New York Times. Retrieved from http://www.nytimes.com/reuters/2014/02/01/sports/basketball/01reutersnba-stern-silver.html Lapchik, R., Hippert, A., Rivera, S., & Robinson, J. (2013). The 2013 racial and gender report card: National Basketball Association. The Institute for Diversity and Ethics in Sports, University of Central Florida. Retrieved from http://www.tidesport.org/racial-and-gender-report-cards.html Larsen, T., Price, J., & Wolfers, J. (2008). Racial bias in the NBA: Implications in betting markets. Journal of Quantitative Analysis in Sports, 4(2). Retrieved from http://scholarsarchive.byu.edu/cgi/viewcontent.cgi?article=1191&context=f acpub Lee, S. S. (2008). Racial realism and the discourse of responsibility for health disparities in a genomic age. In B. A. Koenig, S.-J. Lee, & S. S. 219 Richardson (Eds.), Revisiting race in a genomic age (pp. 342-358). Rutgers, NJ: Rutgers University Press. Lehrman, S. (2008). Cops, sports, and schools. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 285-303). Rutgers, NJ: Rutgers University Press. Lieberman, M. D., Hariri, A., Jarcho, J. M., Eisenberger, N. I., & Bookheimer, S. Y. (2005). An fMRI investigation of race-related amygdala activity in African-American and Caucasian-American individuals. Nature Neuroscience, 8(6), 720-722. Lock, M. (2008). Biosociality and susceptibility genes: A cautionary tale. In S. Gibbon & C. Novas (Eds.), Biosociality, genetics, and the social sciences: Making biologies and identities [Kindle Edition]. Retrieved from http://www.amazon.com Longino, H. E. (1990). Science as social knowledge: Values and objectivity in scientific inquiry. Princeton, NJ: Princeton University Press. Longino, H. E. (2013). Studying human behavior: How scientists investigate aggression & sexuality. Chicago, IL: The University of Chicago Press. Lorusso, L., & Boniolo, G. (2007). Clustering humans: On biological boundaries. Studies in History and Philosophy of Science Part C: Studies in History and Philosophy of Biological and Biomedical Sciences, 39(1),163-170. Love, B., & Tosolt, B. (2010). Reality or rhetoric? Barack Obama and postracial America. Race, Gender & Class, 17(3/4), 19-37. Machery, E., & Faucher, L. (2005). Social construction and the concept of race. Philosophy of Science, 72, 1208-1219. Machery, E., & Faucher, L. (2005). Why do we think racially? In H. Cohen & C. Lefebvre (Eds.), Culture, evolution and cognition (pp. 1010-1031). Amsterdam, Netherlands: Elsevier. MacNamara, J. (1981). Physical objects: The relation between psychology and philosophy. Canadian Psychology, 22(2), 271-281. Mallon, R., & Kelly, D. (2012). Making race out of nothing. In H. Kincaid (Ed.), The Oxford handbook of philosophy of social science (pp. 507-532). Oxford, UK: Oxford University Press. Mallon, R. (2008). Naturalistic approaches to social construction. The Stanford Encyclopedia of Philosophy (Winter 2014). E. N. Zalta (Ed.). Retrieved 220 from http://plato.stanford.edu/archives/win2014/entries/social-constructionnaturalistic/ Mallon, R. (2007). A field guide to social construction. Philosophy Compass, 2(1), 93-108. Mallon, R., & Stich, S. M. (2000). The odd couple: The compatibility of social construction and evolutionary psychology. Philosophy of Science, 67, 133154. Mallon, R. (2010). Sources of racialism. Journal of Social Philosophy, 41(3), 272292. Mallon, R. (2007). Human categories beyond nonessentialism. The Journal of Political Philosophy, 15(2), 146-168. Mallon, R. (2009). Commentary on Joshua Glasgow's theory of race. Symposia on Gender, Race, and Philosophy, 5(2). Retrieved from http://web.mit.edu/sgrp/2009/no2/Mallon1009.pdf Mallon, R. (2012). Was race thinking invented in the modern West? Studies in History and Philosophy of Science, 44, 77-88. Mallon, R. (2006). Race: Normative, not metaphysical or semantic. Ethics, 116(3), 525-551. Mallon, R. (2004). Passing, traveling and reality: Social constructionism and the metaphysics of race. Noûs, 38(4), 644-673. Mallon, R. (2007). Human categories beyond nonessentialism. The Journal of Political Philosophy 15(2), 146-168. Marks, J. (2008). Race: Past, present and future. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 21-35). Rutgers, NJ: Rutgers University Press. McLaughlin, M. (2015, June 24). Former Baltimore police officer comes clean about corruption on force. Huffington Post. Retrieved from http://www.huffingtonpost.com/2015/06/24/baltimore-officertweets_n_7655204.html McCoy, K. (2014, December 4). Choke-hold cop sued in prior misconduct cases. USA TODAY. Retrieved from https://www.usatoday.com/story/news/nation/2014/12/04/choke-hold-coppantaleo-sued/19899461/ 221 McWhorter, J. (2008, December 30). Racism in America is over. Forbes. Retrieved from https://www.forbes.com/2008/12/30/end-of-racism-opedcx_jm_1230mcwhorter.html Millgram, E. (2015). The great endarkenment. New York, NY: Oxford University Press. Mills, C.W. (1996). Blackness visible: Essays on philosophy and race. Ithaca, NY: Cornell University Press. Mitchell, S. (2009). Unsimple truths: Science, complexity, and policy. Chicago, IL: University of Chicago Press. Mitchell, S. (2003). Biological complexity and integrative pluralism. Cambridge, UK: Cambridge University Press. Morning, A. (2008). Reconstructing race in science and society: Biology textbooks, 1952-2002. American Journal of Sociology, 114, S106-S137. Morrison, T. (1998, October 5). Talk of the town: Comment. The New Yorker. Retrieved from http://www.newyorker.com/magazine/1998/10/05/comment-6543 Moss, A. J., Hall, W. J., Cannom, D. S., Klein, H., Brown, M. W., Daubert, J. P., . . . Zareba, W. (2009). Cardiac-resynchronization therapy for the prevention of heart failure events. New England Journal of Medicine, 361,1329-1338. Moskowitz, G. B., Stone, J., & Childs, A. (2012). Implicit stereotyping and medical decisions: Unconscious stereotype activation in practitioners' thoughts about African Americans. American Journal of Public Health, 102(5), 996-1001. Mountain, J. L., & Risch, N. (2004). Assessing genetic contributions to phenotypic differences among ‘racial' and ‘ethnic' groups. Nature Genetics, 36, S48-S54. Nahman, M. (2008). Synechdochic ricochets: Biosocialities in a Jerusalem IVF clinic. In S. Gibbon & C. Novas (Eds.), Biosociality, genetics, and the social sciences: Making biologies and identities. [Kindle Edition]. Retrieved from http://www.amazon.com National Basketball Association. (2014, October 27). NBA Sets Record With 101 International Players From 37 Countries and Territories. Retrieved from http://www.nba.com/global/nba_sets_record_with_101_international_playe rs_37_countries_territories_2014_10_27.html 222 Neal, L. V. I., McCray, A. D., Webb-Johnson, G., & Bridgest, S. T. (2003). The effects of African American movement styles on teachers' perceptions and reactions. The Journal of Special Education, 37(1), 49-57. Nelson, A. (2008). The factness of diaspora. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 253-268). Rutgers, NJ: Rutgers University Press. Newman, A. (2014, December 3). The death of Eric Garner, and the events that followed. New York Times. Retrieved from https://www.nytimes.com/interactive/2014/12/04/nyregion/04garnertimeline.html Nosek, B. A. (2007). Implicit-explicit relations. Current Directions in Psychological Science, 16(2), 65-69. Nosek, B. A., Greenwald, A. G., & Banaji, M. R. (2007). The implicit association test at age 7: A methodological and conceptual review. In J. A. Bargh (Ed.), Automatic processes in social thinking and behavior (pp. 265-292). New York, NY: Psychology Press. Novas, C. (2008). Patients, profits and values: Myozyme as an exemplar of biosociality. In S. Gibbon & C. Novas (Eds.), Biosociality, genetics, and the social sciences: Making biologies and identities. [Kindle Edition]. Retrieved from http://www.amazon.com Omi, M., & Winant, H. (1986). Racial formation in the United States: From the 1960s to the 1990s. New York, NY: Routledge. Ossorio, P., & Duster, T. (2005). Race and genetics: Controversies in biomedical, behavioral and forensic sciences. American Psychologist, 60, 115-128. Ousley, S., Jantz, R., & Freid, D. (2009). Understanding race and human variation: Why forensic anthropologists are good at identifying race. American Journal of Physical Anthropology, 139, 68-76. Outlaw, L. T. (1996). On race and philosophy. New York, NY: Routledge. Outram, S. M., & George, T. H. (2010). Arguments against the use of racialized categories as genetic variables in biomedical research: What are they, and why are they being ignored? In I. Whitmarsh & D. S. Jones (Eds.), What's the use of race? Modern governance and the biology of difference (pp. 92-123). Cambridge, MA: The MIT Press. Palsson, G. (2008). Genomic anthropology: Coming in from the cold? Current Anthropology, 49(4), 545-568. 223 Parsons, C. A., Sulaeman, J., Yates, M. C., & Hamermesh, D. S. (2011). Strike three: Discrimination, incentives, and evaluation. American Economic Review, 101(4), 1410-1435. Payne, B. K. (2005). Conceptualizing control in social cognition. The role of automatic and controlled processes in misperceiving a weapon. Journal of Personality and Social Psychology, 89(4), 488-503. Payne, B. K. (2006). Weapon bias: Split-second decisions and unintended stereotyping. Current Directions in Psychological Science, 15(6), 287-291. Peirce, C. S. (1955). Philosophical writings of Peirce. New York, NY: Dover Publications. Phelps, E. A., O'Connor, K. J., Cunningham, W. A., Funayama, E. S., Gatenby, J. C., Gore, J. C., & Banaji, M. R. (2000). Performance on indirect measures of race evaluation predicts amygdala activation. Journal of Cognitive Neuroscience, 12(5), 729-738 Pickering, A. (1984). Constructing quarks: A sociological history of particle physics. Edinburgh, UK: Edinburgh University Press. Pigliucci, M., & Kaplan, J. (2003). On the concept of biological race and its applicability to humans. Philosophy of Science, 70, 1161-1172. Press, N. (2006). Social construction and medicalization: Behavioral genetics in context. In E. Parens, A. R. Chapman, & N. Press (Eds.), Wrestling with behavioral genetics: Science, ethics, and public conversation (pp. 131149). Baltimore, MD: Johns Hopkins University Press. Price, J., & Wolfers, J. (2010). Racial discrimination among NBA referees. The Quarterly Journal of Economics, 125(4), 1859-1887. Putnam, H. (1995). Pragmatism. Cambridge, MA: Blackwell. Quintana, S. M., & McKown, C. (2008). Handbook of race, racism, and the developing child. Hoboken, NJ: John Wiley and Sons, Inc. Rabinow, P. (2008). Concept work. In S. Gibbon & C. Novas (Eds.), Biosociality, genetics, and the social sciences: Making biologies and identities. [Kindle Edition]. Retrieved from http://www.amazon.com Rabinow, P. (1996). Artificiality and enlightenment: From sociobiology to biosociality. In J. X. Inda (Ed.), Essays on the anthropology of reason (pp. 91-111). Princeton, NJ: Princeton University Press. 224 The Economist Group Limited. (2012, January 28). Race in Brazil: Affirming a divide. Retrieved from http://www.economist.com/node/21543494 Rachlinski, J. J., Johnson, S. L., Wistrich, A. J., & Guthrie, C. (2009). Does unconscious racial bias affect trial judges? Cornell Law Faculty Publications, 3, 1195-1246. Rajan, K. S. (2008). Biocapital as an emergent form of life: Speculations on the figure of the experimental subject. In S. Gibbon & C. Novas (Eds.), Biosociality, genetics, and the social sciences: Making biologies and identities. [Kindle Edition]. Retrieved from http://www.amazon.com Ramachandran, S., Deshpande, O., Roseman, C. C., Rosenberg, N. A., Feldman, M. W., & Cavalli-Sforza, L. L. (2005). Support from the relationship of genetic and geographic distance in human populations for a serial founder effect origination in Africa. Proceedings of the National Academy of Sciences of the United States of America, 102(44), 1594215947. Rawls, John. (1989). Themes in Kant's moral philosophy. In E. Forster (Ed.), Kant's transcendental deductions (pp. 81-113). Redwood City, CA: Stanford University Press. Reardon, J. (2008). Race without salvation: Beyond the science/society divide in genomic studies of human diversity. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 304-318). Rutgers, NJ: Rutgers University Press. Risch, N., Burchard, E., Ziv, E., & Tang, H. (2002). Categorization of humans in biomedical research: Genes, race and disease. Genome Biology, 3(7), 112. Risch, N. (2006). Dissecting racial and ethnic difference. New England Journal of Medicine, 354, 408-411. Roberts, E. F. S. (2008). Biology, sociality and reproductive modernity in Ecuadorian in vitro fertilization: The particulars of place. In S. Gibbon & C. Novas (Eds.), Biosociality, genetics, and the social sciences: Making biologies and identities. [Kindle Edition]. Retrieved from http://www.amazon.com Robles, F. (2015, May 16). Racist police emails put Florida cases in doubt. New York Times. Retrieved from http://www.nytimes.com/2015/05/16/us/miamiprosecutor-reviews-cases-of-police-officers-who-sent-racistemails.html?mabReward=CTM&action=click&pgtype=Homepage®ion= CColumn&module=Recommendation&src=rechp&WT.nav=RecEngine 225 Root, M. (2000). How we divide the world. Philosophy of Science, 67(3), S628S639. Root, M. (2003). The use of race in medicine as a proxy for genetic differences. Philosophy of Science, 70, 1173-1183. Rosenberg, N. A., Pritchard, J. K., Weber, J. L., Cann, H. M., Kidd, K. K., Zhivotovsky, L. A., & Feldman, M. W. (2008). Genetic structure of human populations. Science, 298, 2381-2385. Rosenthal, R., & Jacobson, L. (1968). Pygmalion in the classroom. New York, NY: Rinehart & Winston. Rushton, J. P., & Jensen, A. R. (2005) Thirty years of research on race differences in cognitive ability. Psychology, Public Policy, and Law, 11, 235-294. Sabin, J. A., & Greenwald, A. G. (2012). The influence of implicit bias on treatment recommendations for 4 common pediatric conditions: Pain, urinary tract infection, attention deficit hyperactivity disorder, and asthma. American Journal of Public Health, 102(5), 988-995. Sabin, J. A., Nosek, B. A., Greenwald, A. G., & Rivara, F. P. (2009). Physicians' implicit and explicit attitudes about race by MD race, ethnicity, and gender. Journal of Health Care for the Poor and Underserved, 20(3), 896-913. Sankar, P., Cho, M. K., & Mountain, J. (2007). Race and ethnicity in genetic research. American Journal of Medical Genetics Part A, 143(9), 961-970. Santos, R. V., Fry, P. H., Monteiro, S., Maio, M. C., Rodrigues, J. C., BastosRodrigues, L., & Pena, S. D. (2009). Color, race, and genomic ancestry in Brazil: Dialogues between anthropology and genetics. Current Anthropology, 50(6), 787-819. Sarich, V., & Miele, F. (2004). Race: The reality of human differences. Boulder, CO: Westview Press. Saul, S. (2005, June 13). U.S. to review heart drug intended for one race. New York Times. Retrieved from http://www.nytimes.com/2005/06/13/business/us-to-review-heart-drugintended-for-one-race.html Schmidt, M. (2015, July 3). Charleston suspect was in contact with supremacists, officials say. New York Times. Retrieved from https://www.nytimes.com/2015/07/04/us/dylann-roof-was-in-contact-withsupremacists-officials-say.html 226 Schorr, D. (2008, January 28). A new ‘postracial' political era in America. National Public Radio. Retrieved from http://www.npr.org/templates/story/story.php?storyId=18489466 Sesardic, N. (2005). Making sense of heritability. Cambridge, UK: Cambridge University Press. Shear, M. D. (2015). Making a point, Obama invokes a painful slur. New York Times. Retrieved from http://www.nytimes.com/2015/06/23/us/obamaracism-marc-maron-podcast.html Shim, J. K. (2005). Constructing ‘race' across the science-lay divide: Racial formation in the epidemiology and experience of cardiovascular disease. Social Studies of Science, 35(3), 405-436. Shriver, M. D., & Kittles, R. A. (2008). Genetic ancestry and the search for personalized genetic histories. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 201-214). Rutgers, NJ: Rutgers University Press. Silverman, C. (2008). Brains, pedigrees, and promises. In S. Gibbon & C. Novas (eds.), Biosociality, genetics, and the social sciences: Making biologies and identities. [Kindle Edition]. Retrieved from http://www.amazon.com Simon, R. (2009, August 7). What happened to postracial America? Politico. Retrieved from http://www.politico.com/story/2009/08/what-happened-topost-racial-america-025890 Skinner, D. (2006). Racialized futures: Biologism and the changing politics of identity. Social Studies of Science, 36(3), 459-488. Staats, C. (2014). State of the science: Implicit bias review 2014. The Kirwan Institute for the Study of Race and Ethnicity, Ohio State University. Retrieved from http://kirwaninstitute.osu.edu/wpcontent/uploads/2014/03/2014-implicit-bias.pdf Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69(5), 797-811. Stevens, J. (2008). The feasibility of government oversight of NIH-funded population genetics research. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 320-341). Rutgers, NJ: Rutgers University Press. Steele, S. (2008, November 5). Obama's postracial promise. LA Times. 227 Retrieved from http://www.latimes.com/opinion/opinion-la/la-oe-steele52008nov05-story.html#page=1 Stone, J., & Moskowitz, G. B. (2011). Non-conscious bias in medical decision making: What can be done to reduce it? Medical Education, 45, 768-776. Tabery, J. (2009). Making sense of the nature-nurture debate. Biology & Philosophy, 24(5), 711-723. Tallbear, K. (2008). Native-American-DNA.com: In search of Native American race and tribe. In B. A. Koenig, S.-J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 235-252). Rutgers, NJ: Rutgers University Press. Taylor, P. (2005). Race: A philosophical introduction. Cambridge, UK: Polity Books. Taylor, P. (2011). Rehabilitating a biological notion of race? A response to Sesardic. Biology & Philosophy, 26, 469-473. Tenenbaum, H. R., & Ruck, M. D. (2007). Are teachers' expectations different for racial minority than for European American students? A meta-analysis. Journal of Educational Psychology, 99(2), 253-273. Tooby, J., & Cosmides, L. (1992). The psychological foundations of culture. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 19-136). New York, NY: Oxford University Press. Trawalter, S., Hoffman, K. M., & Waytz, A. (2012) Racial bias in perceptions of others' pain. PLoS One, 7(11), e48546. Tutton, R., Smart, A., Ashcroft, R., Martin, P., & George, T. H. (2010). From selfidentity to genotype: The past, present, and future of ethnic categories in postgenomic science. In I. Whitmarsh & D. S. Jones (Eds.), What's the use of race? Modern governance and the biology of difference (pp. 125146). Cambridge, MA: The MIT Press. United States Department of Justice, Civil Rights Division. (2015). Investigation of the Ferguson Police Department. Retrieved from: https://www.justice.gov/sites/default/files/opa/pressreleases/attachments/2015/03/04/ferguson_police_department_report.pdf van den Bergh, L., Denessen, E., Hornstra, L., Voeten, M., & Holland, R. W. (2010). The implicit prejudiced attitudes of teachers: Relations to teacher expectations and the ethnic achievement gap. American Educational 228 Research Journal, 47(2), 497-527. van Ryn, M., Burgess, D. J., Dovidio, J. F., Phelan, S. M., Somnath, S., Malat, J., . . . Perry, S. (2011). The impact of racism on clinician cognition, behavior, and clinical decision making. Du Bois Review: Social Science Research on Race, 8(1), 199-218. Vanman, E. J., Saltz, J. L., Nathan, L. R., & Warren, J. A. (2013). Racial discrimination by low-prejudiced Whites: Facial movements as implicit measures of attitudes related to behavior. Psychological Science, 15(11), 711-714. Wilson, D. S. (2005). Evolutionary social construction. In J. Gottschall & D. S. Wilson (Eds.), The literary animal: Evolution and the nature of narrative (pp. 20-37). Evanston, IL: Northwestern University Press. Wilson, M. (2005). Wandering significance: An essay on conceptual behavior. New York, NY: Oxford University Press. Winant, H. (2000). Race and race theory. Annual Review of Sociology, 26, 169185. Witherspoon, D. J., Wooding, S., Rogers, A. R., Marchani, E. E., Watkins, W. S., Batzer, M. A., & Jorde, L. B. (2007). Genetic similarities within and between human populations. Genetics, 176(1), 351-359. Wittgenstein, L. (1958). Philosophical investigations. London, UK: Prentice-Hall International. Workman, K., & Kannapell, A. (2015, June 18). The Charleston shooting: What happened. New York Times. Retreived from https://www.nytimes.com/2015/06/18/us/the-charleston-shooting-whathappened.html Zack, N. (1993). Race and mixed race. Philadelphia, PA: Temple University Press. Zack, N. (2002). Philosophy of science and race. New York, NY: Routledge. Zakharia, F., Basu, A., Absher, D., Assimes, T. L., Go, A. S., Hlatky, M. A., . . . Tang, H. (2009). Characterizing the admixed African ancestry of African Americans. Genome Biology, 10(12), 1-11.