Forthcoming in Colormute: Race Talk Dilemmas in an American School, by Mica Pollock. Princeton University Press, 2004.

Introduction

“It is all rather complicated.” -Edmund Leach, Political Systems of Highland Burma (1954, 44)

This is a book about race talk – about people in one school and district struggling with the basic American choice of when and how to describe one another racially. People in America have long struggled in various ways with racial categories, arguably some of humanity’s most conflicted creations. American race categories have become a social truth without ever having had a legitimately biological basis: created to organize slavery, retooled with waves of immigration, and naturalized over centuries by law, policy, and science, race categories are now everywhere, alternately proud building blocks of our nation’s “diversity” and the shameful foundation of our most wrenching inequalities. [1] Over the centuries, as people of various tribes, nations, and religions have taken their places in the nation’s small taxonomy of so-called “races,” we have only sporadically thought to ask each other whether these “races” actually exist: most of the time, we have worried less about the reality of our race categories than about what to do with the racialized orders we have made. Unwittingly or quite knowingly, we have built systems of inequality around race categories; but we have also built identities, friendships and marriages around them. And Americans, now never certain when race is a good thing and when it is a bad thing (and never certain about the moral or political implications of using race labels to categorize human beings) keep struggling with a particularly daunting question. When should we talk as if race matters?

Americans confront the question of whether and how race should matter, as I argue in this book, every time we wonder whether to talk as if it does. As this book will demonstrate, we encounter, every day, the pitfalls inherent in this most basic act of racialization: using race labels to describe people. We wrestle, for example, with the act of placing infinitely diverse human beings into simple “racial” boxes; we then wrestle with the fact that these categories of “racial” difference are central to the most troubling power struggles we have. Ultimately, we wrestle with the paradoxical reality that in a world in which racial inequality already exists, both talking and not talking about people in racial terms seem alternately necessary to make things “fair.”[2] Accordingly, though people in the U.S. arguably use race labels more often than do many other citizens of the world, we also seem to worry about doing so more than most other people. Many of us exhibit particular worries about being “racist ” with our very language: one anthropologist has described the “fear of being labeled a racist” as “perhaps one of the most effective behavioral and verbal restraints in the United States today” (Van Den Berghe 1996).[3]

Given the amount of worrying that race-label use seems to require in America, it is perhaps unsurprising that many Americans have proposed we solve our “race problems” by talking as if race did not matter at all. We are, in fact, in the midst of major battles in the United States about the very future of race talk itself -- and these controversies are a key context for this book. As Steven Gregory has noted, “diverse segments of U.S. society” claim that race “has become a tiresome topic, and one whose ‘polite repression,’ as Toni Morrison puts it, ‘is understood to be a graceful, even generous, liberal gesture’” (1996, 23). Lani Guinier and Gerald Torres (2002) claim that anti-race-talk arguments can be heard both on the American “right” and from “liberal progressives,” both of whom tend to argue for “masking race in political discussion” (32). Spokespersons on the right, they argue, suggest that “when one notices race, one is implicitly manifesting racial enmity” (38), that “noticing race is in essence a throwback to racism” (39), and that “whoever mentions race first is the racist in the room” (308), while in turn, “liberal progressives” argue that “race is something that good people simply do not notice” (51), since “a frank engagement with race” would “only heighten social divisions” (32). Indeed, some public figures are now arguing loudly that even using race labels publicly is tantamount to reproducing racism itself. A public referendum currently being proposed in California, the “Racial Privacy Initiative,” argues for the elimination of all race labels from public records, declaring that “the state shall not classify any individual by race, ethnicity, color or national origin in the operation of public education, public contracting or public employment” (my emphasis). The referendum’s key proponent, U.C. regent Ward Connerly, explains bluntly to the press that “The state should be blind to color, just as it is to religion or sexual orientation.”

More implicit quests to delete race talk from American life also simmer in our struggles over public policy. In 1996, when this ethnography was in the making in the under-resourced, “low income minority” California school and district where I myself had taught, a majority of California voters marked the ballot for a state proposition vaguely entitled the “Equal Opportunity Initiative,” which set out to make illegal not only “race-based” affirmative actions in the state’s universities, but also every “race-based” educational program in the state.[4] This “Proposition 209,” which fueled a concurrent nationwide wave of similar litigation intended to outlaw the consideration of race in college admissions, K-12 student enrollment plans, and programs for academic enrichment or student outreach, did not outlaw California’s racial categories themselves, of course. It also did not erase racial categories from Californians’

minds. Rather, it simply outlawed mentioning in official documents that these categories existed: in practice, the policy was less about being colorblind than being actively colormute.

As Lawrence Blum (2002) notes of such so-called “colorblind” policy, “A policy that makes explicit reference to race, or racial identity, is taken to stand condemned by that fact alone” (91, emphasis mine). Indeed, “colorblindness” can often be more accurately described as a purposeful silencing of race words themselves. Proponents of Proposition 209 effectively ordered district and university people to actively refuse to talk in racial terms. Yet actively deleting racial labels from applications and enrollment plans certainly didn’t mean the disappearance of racial patterns in education. Policymakers, of course, could not stop Californians from viewing each other racially, nor outlaw race as a system of categorization structuring people’s social and economic lives. Nor could they even outlaw daily racial references from school hallways and classrooms and lunch tables. Instead, policymakers simply banned race words from the official policy analysis –they deleted the race labels that appeared in school applications, program descriptions, and brochures. As the mostly-white-and-Asian freshman enrollments at the UC schools after Proposition 209 quickly made clear, however, officially erasing race words had far from erased racial patterns at the state’s universities. Indeed, the insistence on being colormute had actually allowed racialized disparities in pre-college opportunity to proceed unhindered -- helping increase racial disparities in UC enrollment and hinting that deleting race words can actually help make race matter more.[5]

Colormute policy and practice had specific consequences for this book’s subjects. In the spring of 1996, around the time Proposition 209 was in full swing, I was finishing my first year of formal research for this book at “Columbus High School” in “California City,” where I had the previous year been a teacher. Over that summer, angry district officials replaced 90% of the Columbus staff in a reform called “reconstitution,” wiping out not just the faculty themselves but also all the reform programs -- career academies, small learning communities – the faculty had devised. As we will see, this tumultuous event also involved dilemmas of speaking racially--and race silence here too had consequences. For while the “reconstitution” reform stemmed from the city’s desegregation order -- a court action concerned on paper with achieving racially equal academic opportunities and outcomes for “African-Americans” and “Hispanics” in California City schools—for over a year of “probation” before reconstitution almost no one had even used the words “African-American” and “Hispanic” in any district- or school-level public conversations about school reform. And as 100 adults left Columbus as the result of a silently racialized policy and a new staff of 100 strangers assembled at Columbus, I watched a new set of well-meaning adults reproduce all the prior staff’s habits of deleting race words -- and articulate identical dilemmas of race talk and colormuteness that would come to seem common American property.

This book, which uses everyday race talk controversies from Columbus and California City as primary data, is an attempt to map the contours of six basic dilemmas of racial description that tie Americans up in communal knots, and which we must attempt to better understand. For these traps of discourse, I want to argue, are extremely consequential. Having witnessed three full years of struggles over talking and not talking in racial terms at Columbus -- one as a teacher there in 1994-95, two as an anthropologist there from 1995-97 -- I have come to argue explicitly what policy debates across the United States are currently implying. Race talk matters.[6] All Americans, every day, are reinforcing racial distinctions and racialized thinking by using race labels; but we are also reinforcing racial inequality by refusing to use them. By using race words carelessly and particularly by deleting race words, I am convinced, both policymakers and laypeople in America help reproduce the very racial inequalities that plague us. It is thus crucial that we learn to navigate together the American dilemmas of race talk and colormuteness rather than be at their mercy; and that is the overarching purpose for this book.

Let me immediately explain my use here of an American “we.”[7] Different racialized communities in the U.S., of course, have very different experiences with racial description (Americans “of color” are described in racial terms far more often than are “whites,” for example, while “white” people, the racialized category into which I myself fall, experience disproportionate anxiety over using race labels even as we experience their application least of all).[8] Yet Americans are a giant speech community when it comes to some basic dilemmas of race talk: for we share not only our basic system of racial/ethnic categorization, but also the fundamental American question of when, how, and whether to take race “into account” in American life. We also share, as I want to demonstrate over the course of the book, the racialized inequalities we most struggle to discuss.[9]

That this book focuses on schooling talk is no accident, for public struggles over race have long centered on this particular shared arena of national practice. From nineteenth-century laws denying basic literacy to slaves, through decades of twentieth-century battles over mixing “the races” in desegregated schools, to contemporary multiracial debates on “colorblind” college admissions or curricular “multiculturalism,” our recurring debates over how race does and should matter in the U.S. have routinely circled back to address American schools.[10] Schools are key institutions where Americans “make each other racial” (see Olsen 1997): not only are schools central places for forming racial “identities,” but they are key places where we rank, sort, order, and differently equip our children along “racial” lines even as we hope for schooling to be the great societal equalizer.[11] School race talk, I argue, is thus one key version of American race talk: for the way we talk in school both reflects and helps shape our most basic racial orders.

Labeling (or not labeling) each other with race words, of course, is just one everyday way that Americans make each other racial -- and make race matter. Sociolinguists and linguistic anthropologists have studied many other ways we reproduce “racial” difference through our everyday talk, such as through the patterned use of particular languages, dialects, styles, or vocabulary.[12] Going beyond talk, we make ourselves and each other racial through the hairstyles we sport, our gestures, and the friends we display; through the music we listen and dance to, the people we sit down next to, the organizations we belong to, the resources we distribute, and the neighborhoods we choose to live in or not to live in. Race is also reconstructed when people make meaning of the genetically-insignificant physical characteristics, like skin color or nose shape or eye contour or hair texture, that we have used as markers of “racial” difference since pseudo-science codified this use centuries ago. Racial orders are built daily through movements of the body, through statistics and numbers, through glances across rooms to friends.[13] Racial orders in school are also built through the distribution of dollars, through the “tracking” of racialized bodies to designated schools and classrooms, through the false expectations that differential abilities reside in racialized minds, through an “institutional choreography” (Fine 1997) of everyday actions incessantly funneling opportunities to some students and not others.

In contrast to gestures, dollars, or knowing looks, the use of racial labels seems a bizarrely explicit way of making people racial. Race language is indeed itself a powerfully simple force: we become race-group members, or we must negotiate and resist so becoming, every time we are referred to in racial terms, and talking racially does prompt listeners to see the world anew in racialized ways.[14] This is no new claim: scholars have long viewed words as consequential actions that create the world rather than simply describe it.[15] Indeed, Americans, as this book will demonstrate, seem to know quite well that race words, in their bluntness, are extremely powerful actions. We seem somewhat less aware that our very resistance to using race words has major consequences as well.

We struggle over using race words, I argue, in part because the simplicity of racial descriptions so often seems to belie the complexity of human diversity. Imagine for yourself showing up for the first time at this book’s infinitely complex field site: “Columbus High School” in “California City.”

Trying to describe Columbus

Entering Columbus at the end of a typical day and glancing around the building, you might notice that there are some adults “of color” at Columbus, including its principal; but you might label the majority of the adults you see “white” without much thought. As Columbus students pour out the doorway, however, they might immediately appear to you stunningly diverse, a population that seemingly embodies the country’s breathtaking demographic complexity. Many Columbus students (or their parents) have immigrated to California City from various linguistically distinct islands in the Philippines, from numerous Central American and South American countries, from a list of Cantonese- and Mandarin- speaking regions of China, from both Samoas, and from a huge grab-bag of other places, such as Vietnam and Tonga. Recently or some decades ago, the parents of many African-American students at Columbus migrated across the country from the American South. There are just a handful of students who are called “white” at Columbus, most with grandparents or great-grandparents hailing from Ireland, Italy, Germany, or other places in Europe. Talking briefly to a few students, you might learn that some students have lived across the street from Columbus their entire lives, some moved to the city as young children, and some immigrated from another part of the globe just yesterday.

Watching the students emerge into the mid-afternoon sunlight, you might also realize that depending on how you are looking at these young people, you could frame them as a largely “of-color” unit, as divisible into a small list of presumed “ethnic” or “national” origins, or as a sea of faces on a seamless continuum of bronze and brown and beige. You might also begin to suspect that an accurate account of diversity at Columbus must take into consideration far more than what the classifying eye can comprehend. If someone were to hand you a sheet of the data that the California City Unified School District keeps on basic Columbus student demographics, for example, you would notice that students across Columbus are also astonishingly diverse linguistically: one third of the student body is in the process of learning English for the first time. Listening to the other two-thirds of Columbus students chattering in the hallway, further, you might notice that some shift flawlessly between three languages, while many seem to be speaking several dialects of one. Looking more at the district data in your hands, you might also notice that Columbus students seem diverse both economically and academically: the district gives a particularly low-income subset of Columbus students (40 percent) either free or reduced-price lunch (that is, 40 percent of Columbus students are willing to publicly claim such assistance), and the district also classifies 60 percent of Columbus students as “Educationally Disadvantaged Youth” -- students that are both low-income and under the 40th percentile on a statewide standardized test.

Go to some classrooms during the day and try talking to and observing students, though, and you will learn that describing Columbus’s academic and economic diversity is actually not so easy either. Some students write flawlessly, while others can barely read; a few can do calculus, while many others still struggle with basic fractions. Hearing more about students’ outside lives, further, you might find that while some students sleep on spare beds in “the projects” or in foster homes, others live in aunts’ extended family apartments, and still others wake up in stuccoed single-family, two-parent houses; some Columbus parents live on welfare, some clean hotels, others work as university librarians with Masters degrees. And talking at length to any student at Columbus, finally, you might find yourself challenged to describe the “diversity” of any individual. At Columbus, self-proclaimed “mixed” parentage is common enough that “what are you mixed with?” is a matter-of-fact student question. Indeed, ask any Columbus student “what” she “is,” and you may find that she offers different answers at different times of the day, week, or year.

If the apparently infinite variety of ways available to describe Columbus’s “diversity” now make the task of description itself seem impossible, fear not. Columbus students and adults will often readily make describing the school’s demographics exceedingly simple. One particular simplification of Columbus’s diversity shows up daily in conversation, as people place one another into a few simple categories they call “racial.”

There are, in fact, six lump-sum words that people at Columbus use to describe what they call the school’s main “races”: “black,” “Latino,” “Filipino,” “Chinese,” “Samoan,” and “white” (this last category includes mostly teachers). A student who tells me in one conversation that he is both “black” and “mixed with Puerto Rican” thus still writes for a class this poem describing Columbus with easy numbers:

“4 good teachers, with two bad ones a day

every 5 bad kids copping one great student

2 fights, 0 body breaking them up

6 different groups, and nobody cares about anything

over 1500 people different to the bone”[16]

In defining these “6 different groups,” Columbus students call “racial” even the groups scholars typically term “ethnic” or “national,” such as “Filipinos,” “Chinese,” “Latinos,” and “Samoans.” While some scholars would call this conflation of race, “ethnicity,” and “nationality” theoretically problematic, merging the three is a process key to daily social analysis at Columbus, just as it is for young people in many areas of the world.[17] The word “race” at Columbus, as elsewhere, indeed denotes “groups” imagined to be easily physically distinguishable, yet rarely do Columbus students suggest that they frame these six “racial” groups as populations that are somehow genetically or “biologically” distinct. Rather, calling these six groups “racial” indicates primarily that they are analogous parts in the school’s simplest taxonomy of “diversity” -- and importantly, often competitive parts in contests over social power.[18] While students occasionally change their nomenclature for categories, swapping “African-American” for “black” and “Mexican” for “Latino” (to the consternation of some “Latinos” of Guatemalan, Salvadoran, and Nicaraguan origin), students compress their diversity into six simple “racial” groups many times each day at Columbus. Adults at Columbus do the same.

This simple system of “racial” categorization, notably, is not limited to everyday life at Columbus. The California City Unified School District, for example, uses roughly these same six labels to keep records on what it calls Columbus’s main “racial/ethnic groups”: indeed, the district has been ordered by a federal court to distribute a set of nine such “groups” districtwide in proportional amounts. In the mid-1990s, district demographic records said Columbus enrolled “Filipinos” (roughly 30 percent), “Latinos” or “Hispanics” (roughly 30 percent), “African-Americans” or “blacks” (roughly 20 percent), “Chinese” (roughly 8 percent), “Other Non-Whites,” a bureaucratic category that included Columbus’s “Samoans” (roughly 8 percent), and “Other Whites” (roughly 5 percent). Columbus’s teaching staff was listed as roughly 55 percent “Other White,” 15 percent “African-American,” 10 percent “Filipino,” 10 percent “Latino,” 5 percent “Other Non-White,” and 2 percent “Chinese.”

With Columbus’s six main “racial” labels now in hand, you might with relief begin to describe the people pouring out of the building in their simplest racial terms. Yet you would have to take great care with what you were talking about, and to whom: for your “racial/ethnic” descriptions might well be met with uncomfortable silences or critical retorts. Descriptions of people at Columbus, you see, are only sometimes supposed to be racial. Stick around Columbus for a few days, and you will realize that to describe Columbus as people who spend every day there describe it, you will have to decide based on circumstance the most accurate or appropriate way to frame Columbus’s “diversity.” No one around you will know when you see various groups at Columbus, but the moments when you talk about these groups as groups will be analyzable acts -- and this fact may have you monitoring your speech rather carefully. The question this book asks is when -- in relation to which topics and in which social or institutional situations -- you might describe the people at Columbus as racial, and when you might resist doing so. Three years of talk collected at and around Columbus High School suggest that there would be some moments in which you worried quite explicitly about using race labels, other moments when you used race labels without thinking twice, and still other moments when you erased race terms from your talk quite purposefully -- and that all these actions would actually mimic others’ in an astonishingly precise choreography.

Using race labels: three main acts of racial description

Three main patterns of race-label use ran throughout the fabric of Columbus’s race talk, and they run throughout the fabric of this book as well: at different moments, speakers contested the use of racial labels, they used racial labels matter-of-factly, or they suppressed them altogether. First, speakers at Columbus and in California City often contested the use of racial labels quite heatedly. The inordinate complexity of Columbus’s very Californian demographics – its six “race groups,” and its multitude of self-consciously “mixed-race” students of color – actually accentuated the main pitfall of racial description anywhere in America. Racial descriptions of patterns, as well as racial descriptions of people, can always be wrong. And people at and around Columbus, struggling tremendously with when it was either accurate or appropriate to talk in racial terms, often worried as much explicitly. Students worried daily, for example, about accurately classifying themselves and others as members of the school’s six “races”; adults worried about accurately describing the racial demographics of patterns in schoolwide or classroom life. Students also occasionally apologized for comments about particular “races” at Columbus, or for even calling their teachers “white”; adults routinely questioned the appropriateness of speaking racially about school programs or school people, in conversations with the principal or the superintendent or their own peers.

In relation to some topics, though, such apparent anxiety over race-label use disappeared. Sometimes, everyone at Columbus talked quite matter-of-factly as if race mattered. Columbus people described classroom curriculum and public assemblies, for example, in straightforward racial terms: classrooms provided blunt units on “Latinos” or “Filipinos,” and people chattered happily about “black” students who read poems at a “black history month” assembly or “Samoan” students who danced at a “multicultural” event. It seemed similarly easy for students, teachers, and administrators alike to describe conflicts between students with racial labels: the “Latinos” beat up the “Filipinos,” people said matter-of-factly, or the “Samoans” beat up the “blacks.” Especially when discussing pleasant aspects of “diversity” or topics of school life ostensibly confined to students, Columbus people often talked nonchalantly as if they assumed race to be matter-of-factly relevant.[19]

In contrast, there were still other moments at and around Columbus when speaking in racial terms seemed either to indicate the existence of “racism,” or risk being “racist”-- and at these moments, people systematically suppressed race labels altogether from public talk. Adults, in particular, actively suppressed race labels when they were discussing inequitable patterns potentially implicating themselves. While adults spoke matter-of-factly in public about how race mattered to student-student peer relations, for example, they never spoke publicly at all about how race mattered to student-adult power relations: while “the Latinos fought the Samoans” was a possible public statement at Columbus, that is, “the white teachers are having trouble with the black students” was the sort of comment reserved only for private conversations. Similarly, while a teacher could announce nonchalantly at a faculty meeting a state writing contest targeted at “Chinese students” or a personal search for “Filipino literature,” no public discussion of school reform goals at Columbus—goals for which adults would be held accountable— labeled students racially at all. Similarly, achievement patterns, which intertwined the roles of students and adults particularly anxiously, caused adult speakers particular consternation: while adults at department meetings matter-of-factly described the racial demographics of curriculum, for example (e.g., “we need more black literature”), they never assessed the racial demographics of student academic performance (e.g., “we need more black students in honors English”). As important as the topic of conversation, finally, was the question of whom one was speaking to: while Columbus adults spoke privately in the hallway about racial patterns in school suspensions, they never discussed these racial patterns with one another in faculty meetings. And while district administrators presented racial suspension statistics in matter-of-fact charts to the court monitors overseeing the city’s desegregation plan, they deleted these very statistics from communiqués to be seen by defensive school faculty. Race labels could be used easily in school talk, it seemed, only in certain places at certain times --and when they popped up had everything to do with who was speaking where about what.[20]

Some central considerations in this study of race talk

In 1969, anthropologist Frederick Barth advised colleagues to stop studying the cultural practices presumed internal to individual “ethnic groups” and start looking instead at how boundaries between multiple such groups were socially maintained. This book takes an analogous approach to racialization: I am interested here not as much in what it meant in some internal fashion to Jake to be black, what it meant to Felicia to be Filipina, what it meant to Ricardo to be Latino, or what it meant to Steve to be white, as in when, in the institution of schooling, people drew lines around Jake or Felicia and Luis and Steve that categorized each of them as race-group members -- and when Jake and Felicia and Luis and Steve drew such lines around themselves.[21] In privileging this most basic simplifying social practice of racial identification over the dynamic complexities of racial identity in this book (a distinction I explore further in Chapter 1), I build here on anthropological work that has looked closely at the basic practice of description itself—work treating categorizing and delineating “kinds of people,” as Charles Frake (1980) has put it, as a key piece of cultural practice.[22] I also build on historical explorations of how racial categories developed over time, which have been particularly good at showing us people -- including scholars of “anthropology,” this relatively young “science of the races” -- actively labeling people racially through law, policy, and science (indeed, these studies have reminded us that there was a time before racial categories existed).[23] Yet race is not something simply made in the past, of course, but something we can watch being made in the present. We continue to make race and to build racial orders, I argue here, each and every day in America, with the help of the very racialized language we use and refuse.

This study’s focus on race talk emerged gradually, over many months of struggling to understand racial practice at Columbus. I had actually embarked originally upon a more typical ethnographic investigation of how important “race” was to Columbus students’ “identities”; research questions about race and schooling (which, not incidentally, typically investigate one “race” at a time rather than framing “the races” as mutually constituted groups) regularly frame “race” as something students of color own, rather than as a shared set of racializing practices involving people of all ages and “races.”[24] Making race words themselves the unit of analysis displaced the study’s more traditional research focus on students of color as “racial” actors -- and in doing so, it revealed that all players inside and outside Columbus were actually producing racial orders together.

Despite the pain and chaos it caused two generations of Columbus people, reconstitution itself created a bizarre natural experiment that demonstrated that everyday race talk habits, too, were shared: for some basic patterns of talking and not talking racially were reproduced by a new community of 100 almost complete strangers. Privileged to conduct research at Columbus both before and after reconstitution, I have here been an odd sort of anthropologist: with one year of teaching and one year of observing already completed in the spring of 1996, I was far more “native” than most of the new adult “natives” (educators) who arrived the following fall. Yet these new Columbus adults revealed quickly that they were already native to a much larger speech community, for they, like their predecessors, also resisted speaking publicly of racial achievement patterns; they, too, preferred to talk about reforms for “all” students rather than for racialized groups of them; they, too, used race words easily to describe student relations but not relations between students and themselves. Continuing to enjoy a researcher’s privileged mobility to talk to players across the school building and school district, further (and with the time to read legal and district documents not readily accessible to school-level adults, who had little time for such investigation), I was now perfectly positioned to confirm that both district and school-level people talked in racial terms at predictable moments and conspicuously did not do so in others –and to learn to understand, over time, when people were conspicuously not talking about race.

To study race talk like this -- as a form of patterned cultural practice, with predictable scripts and silences -- required a special self-consciousness about ethnographic method. Scholars studying race far more often treat the talk they gather as simply quoted opinion to copy down: more rarely do researchers examine the everyday politics and patterns of talking racially.[25] I conducted a good number of interviews during this research that in the end went largely unused, having become convinced quickly that formally prompted race talk was always particularly packaged for a researcher and that the “informal logic of actual life” (Geertz 1973, 17) was best demonstrated by more naturalistic interaction with both students and adults. While I spent countless hours observing classrooms, too, much of the data presented in this book emerged in casual research conversations. Such casual conversations were already speech events central to Columbus daily life, for these conversations actually resembled those in which I had participated every day as a teacher: they were informal, impromptu discussions with students and adults in hallways, on outside benches, and in empty classrooms. During my years as a researcher at Columbus (1995-97), I spent almost every day embroiled in such discussions; as I was training to be an anthropologist, people understood that “hanging out” informally was my research. As I knew an omnipresent tape recorder would make these informal conversations stilted and awkward, I reconstructed these research conversations on paper immediately after they occurred. I did not need to recapture language at the level of grammatical detail required by most linguists, as I was interested primarily in the words and phrases that surrounded race labels in talk, the moments when race labels appeared and disappeared, and the apparent ease or anxiety with which people used them.

I thus participated in most of the school-level talk I present here (if I did not speak, I participated with my very presence), guided by the methodological mantra that I would allow others to bring race labels into any conversation first. Being “white” seemed to make me a more natural confidant of white adults, but being an ex-teacher sympathetic to and supportive of Columbus people’s daily struggles positioned me as an acknowledged comrade to Columbus players of all ages and “races,” both before and after reconstitution. I assisted administrators, teachers, and students throughout my years as a researcher at Columbus, and I remain friends with a subset of faculty and former students. Notes on my own teaching year in 1994-95 (which form the basis of much of Chapter 1) were taken from a personal journal I kept in the hopes of writing a (never finished) first-year teacher memoir; over time, this diary became an important window onto a teacher’s dilemmas.

The data eventually used for analysis, thus, was the talk of informal and public occasions when people at and around Columbus used racial labels--and the moments when they worried explicitly to me about doing so. Over time, as I realized that people were routinely talking to me about their concerns about race talk and silence, I learned to pay special attention to what linguists and linguistic anthropologists call the “metapragmatic” aspect of language -- people talking about talking. While Columbus people had “limited awareness” (Silverstein, 1981) of some of their racialized deletions and hesitations (no white adults pointed out how predictably they stuttered before saying the word “black,” for example), they were brutally aware of many of their struggles with race words. Columbus adults talked to me most agitatedly about the pitfalls of racial description, and our in-depth discussions about the troubles of using race labels became a vital data source.

Finally, I also learned to go looking for race talk in multiple institutional locations. Seeking patterns in the use, contestation, and deletion of racial terms, I documented discourse from school board meetings, superintendent’s addresses, conversations between teachers held in classrooms, hallways, and happy hours, conversations between teachers, students, parents, and administrators in and out of classrooms, and conversations between students both in and out of school. I also gathered systematically the written artifacts of legal opinions, district and school-level statistics, district pronouncements and press releases, union newsletters, faculty newsletters and memos, student assignments, newspaper articles, and educational research itself. Over time, comparing private to public talk became essential to this analysis, as Columbus people often conspicuously de-raced their public talk of the very topics (achievement, discipline, opportunity) to which they privately muttered that race mattered most problematically.[26] In the end, though, six core dilemmas of race talk and silence seemed to pervade all levels of schooling: as they started to become evident everywhere I looked, they soon came to seem fundamental to American race talk. Each chapter in the book fleshes out one of them.

Dilemmas piled upon paradoxes: the organization of the book

The book begins with Columbus students challenging the very idea of “racial groups” (Chapter 1); it continues with adult and student debates over when race mattered to life at Columbus (Chapter 2). After expanding the analysis to include district and legal adults’ struggles with race talk and racial inequality (Chapters 3-5), it returns finally to Columbus to watch Columbus adults anxiously deleting one race label in particular, and in the process reproducing a racialized disparity despite themselves (Chapter 6). The book concludes (“Moving Forward”) by offering possible solutions to these dilemmas, arguing that we must become more proactively and critically conscious about race talk—that we must learn to discuss fruitfully not just our racial inequalities, but also the very question of when and how to use race talk strategically to address particular problems. A final section (“Practically Speaking”) addresses educators in particular.

Chapter 1, then, begins by exploring the most fundamental dilemma of U.S. race talk, one demonstrated daily by Columbus students engaged explicitly in the very process of racial classification: we don’t belong to simple race groups, but we do. Columbus students, many of whom proudly considered themselves multiracial or “mixed,” always challenged the accuracy of simple race labels to describe complex people when discussing racial classification itself; when discussing other topics, however, they regularly placed themselves and one another into a handful of lump-sum groups they called “racial.” These dynamics of racial identification were always intertwined with dynamics of power: while students’ talk about racial classification routinely called the very borders and reality of “race groups” into question, such contestation over group membership (“is he Samoan?” “What is ‘white’?”) vanished particularly in everyday talk about racial equality. In classrooms and at public events, even “mixed” students demanded the equal curricular representation of Columbus’s six “racial” groups, reifying these groups as things they should “learn about” in equal amounts; and adults and students speaking to one another typically proceeded as if people at Columbus belonged naturally to this simple, six-group “racial” taxonomy. Most school talk, in fact, takes no time to critique the boundaries or very notion of “race groups” -- for school people typically worry about racial inequality rather than the very idea of “race.” Accordingly, the sporadic student challenge to racial categorization remains submerged until the conclusion of the book, where I discuss the possibility of modeling Columbus students’ strategy of “race-bending”: that is, strategically interrogating the very notion of “racial” difference even while keeping race labels available for inequality analysis.

Chapter 2 begins to look at Columbus adults and students struggling not with whether clear-cut “race groups” existed at Columbus, then, but with when and how race mattered to everyday life there. In examining a particularly routine kind of everyday talk at Columbus -- talk about social relations -- the chapter explores a second fundamental dilemma of logic plaguing U.S. race talk: in American logic, race doesn’t matter, but it does. In the U.S., it seems, we want race sometimes not to matter and sometimes to matter very much; and daily talk at Columbus was accordingly routinely unclear about when race “really mattered.” Race labels waxed and waned in daily conversation, implying that race was only sometimes relevant; and when asked to sum up whether race mattered, people often denied bluntly that it did. Yet at less self-conscious moments, speakers used race labels matter-of-factly in public talk of student-student relations as if such relations were unequivocally racial, speaking so easily of student friendships and “riots” that the relevance of race to student relations came to seem almost normalized; and they deleted race labels from public talk of student-adult relations that were perhaps racial too problematically. The chapter concludes that the moments when we delete race labels from our talk are perhaps the moments in which race matters most dangerously. Figuring out how race matters thus involves attention both to moments when we talk easily “about race,” and when we resist talking about race at all -- and the rest of the book follows this prescription.

Chapter 3 next tells a legal and school reform story about Columbus and its district, in which race labels indeed vanished at the moments when they were most relevant: namely, in talk of reforms designed to achieve racial equality. Charting this tale historically and then over the turbulent spring of “reconstitution” at Columbus, the chapter demonstrates a third key dilemma of U.S. race talk: the de-raced words we use when discussing plans for achieving racial equality can actually keep us from discussing ways to make opportunities racially equal. Over several decades of desegregation in California City, I demonstrate, a specific 1960s concern for equalizing the educational opportunities of “black” students became submerged in a de-raced 1980s equality discourse of serving “all students.” While this discourse of “all students” contained a decades-long quest for racial equality, reform talk about serving “all” would itself replace discussion of improving educational opportunities for specific racialized groups. During the reconstitution battles of Columbus’s 1995-1996 school year, district representatives speaking of expected school reforms repeatedly submerged the court’s renewed concern for “African American” and “Hispanic” students within talk of reforms for “all students”; in turn, Columbus adults hoping to avoid reconstitution spoke only of serving “all students” in the presence of district personnel. After the district finally fired all Columbus adults in 1996, the new staff hired quickly reproduced an identical reform discourse of assisting “all” – completing a cycle in which no public analysis of assisting “African-Americans” and “Hispanics” in particular had occurred.

As Chapter 4 shows next, however, the absence of race labels from public reform discourse was the result of confusion over inequality analysis as much as an explicit resistance to speaking racially. Analyzing inequality, I demonstrate, poses another fundamental challenge for U.S. speakers, especially in particularly diverse places: at and around “low income,” multi-“minority” Columbus, speakers seemed perennially unclear about whether race groups like “African-Americans” and “Hispanics” actually had fewer opportunities to succeed in school than anybody else. Speakers contested all available descriptions of educational “disadvantage,” demonstrating deep uncertainty about who exactly was disadvantaged in comparison to whom; in trying to analyze opportunity citywide, speakers similarly shifted their analytic lenses both between “race” and “class” analyses and between an analysis framing all “minorities” as less advantaged than whites and one framing certain “minorities” as particularly “at risk.” Through such analytic motion, speakers ironically blurred all inequality analyses – and in the end, many abandoned altogether the task of determining how race mattered to educational opportunity. In our confusion over ascertaining the role race still plays in our complex multiracial and class-diverse inequalities, I warn, we often delete racial analyses of inequality prematurely. In doing so, we demonstrate another key dilemma of race talk: the more complex inequality seems to get, the more simplistic inequality analysis seems to become. The oversimplification of inequality analysis has real repercussions for students left to experience inequality without remedy: In the late 1990s in California City, policymakers dismantling the city’s desegregation order and deleting race from the city’s public inequality analysis (mirroring the statewide deletion of affirmative action remedies) would leave behind no sophisticated analytic framework for analyzing the distribution of educational opportunity at all.

Chapter 5 proceeds by examining a kind of race analysis that did keep rearing its head in district and school-level discourse, even as analyses of educational opportunity were being erased: adults seemed mentally programmed to compare the school achievement of the available “racial groups.” District adults routinely informed the newspapers and the courts of districtwide racial achievement patterns in the mid-1990s, while Columbus adults repeatedly compared race groups in private conversations about how students achieved at Columbus; yet such talk about existing racial achievement patterns appeared in public only sporadically, and only for specific audiences. The seemingly omnipresent question about how race groups achieved, that is, was the very question district and school adults most often refused to articulate when speaking publicly to one another -- and when school and district speakers did mention racialized achievement patterns, they routinely blamed these patterns on actors other than themselves. Examining the wider dynamics of race and blame in achievement talk at and around Columbus, Chapter 5 argues that Americans actually expect school achievement to be racially ordered, yet we tend to name racial achievement patterns only when doing so does not seem to implicate ourselves personally. Paradoxically, the chapter thus reveals, the questions we ask most about race are the very questions we most suppress. I suggest that since both matter-of-fact talk blaming others for racial disparities and anxious silence about such disparities can serve to naturalize these very disparities, we might instead consider speaking proactively as if such disparities can be communally prevented and dismantled.

The final chapter addresses directly the most vexing paradox of racial description: Although talking in racial terms can make race matter, not talking in racial terms can make race matter too. We return to the Columbus corridors to explore this final race talk dilemma. Listening to Columbus adults discuss an everyday school “problem”—students wandering through the hallways cutting class—we examine the use and deletion of one specific race label at Columbus: “black.” In the hallways and in empty classrooms, adults privately muttered their perception that “black students” were overrepresented among the students wandering the halls. However, they admitted that even as they worried that this overrepresentation of “black students” among the “hall wanderers” “needed to be talked about”—since suppressing talk of “blacks” in Columbus’s hallways effectively allowed black students to wander the halls in disproportionate numbers -- they themselves self-consciously deleted the very word “black” from their public talk of the hall wandering “problem.” Intending to avoid the potential “racism” of describing the hall wanderers as disproportionately “black,” they explained, they omitted the word from public discourse quite purposefully. Yet few noticed an additional unintended consequence of these racialized deletions: actively suppressing the word “black” from public talk of school “problems” served daily to increase the perceived relevance of blackness to these problems. In whispering anxiously in private about the isolated disproportionate role of “blacks” in school “problems,” that is, adults deflected any analysis of their own role in producing the hallways’ racial demographics -- and they repeatedly framed “black” students themselves as an intrinsically “problematic” population. In knowingly saying nothing publicly about the overrepresentation of “blacks” in the hallway, further, Columbus adults effectively ignored black students in racial terms. In the end, such silence itself was a form of racialized action: for black students themselves remained both wandering disproportionately and quietly reviled. This book is, in the end, about how people anxiously remaining colormute risk institutionalizing the very racial patterns they abhor-- and I conclude with recommendations about talking more skillfully.

Columbus dilemmas as the dilemmas of us all

The description of “kinds of people” in the post-modern world, long acknowledged as a thorny analytic and social problem for anthropologists and other social scientists, is also a daily problem for “natives” themselves.[27] As both anthropologist and “native,” dealing with the dilemmas of race talk has been the central difficulty of writing this book. I want to make sure that readers see these dilemmas not as the fumblings and bumblings of strangers, but as dilemmas that belong to us all. Because this book focuses on one school’s struggles, my analysis could potentially backfire most on my teacher colleagues, since critiques of schooling seem to land most heavily on educators. I want to make clear that my position echoes that of George Spindler, the first champion of the anthropology of education, who stated in one of his earliest analyses that “It should also be clear that I have not been castigating teachers. They are the agents of their culture” (Spindler 1963, 156). All Americans, including this author, must fumble with race words often too clumsy to describe precise realities; we must fumble with the knowledge that both using and deleting race words can serve alternately to dismantle racial orders and to reinforce them. Most frustrating, we all must negotiate a world in which our very confusions over when to talk as if race matters help recreate a world in which it does.

A Note to Readers

Throughout the data segments presented in this book, I have occasionally put racial labels in boldface so that readers become accustomed to marking their appearance and disappearance. When I am arguing that certain apparently non-racial phrases are crucial to race talk, I have highlighted these phrases in bold face as well. I have also occasionally presented my own analysis in italics within some of the data I display, in an attempt to discuss race-label use as it occurs. Rather than have the reader scan a lengthy interaction and then have to return to it for plodding analysis, I make some of my analytic points as the reader goes through each data segment. Finally, I use racial labels myself to describe speakers throughout this book, a practice that may strike some readers as a blatant attempt to make race seem relevant to speakers’ words. Noting that speakers are “black,” for example, may set up their words to be heard as critical; describing teachers as “white” may set up their words to be heard as “racist” (to many U.S. ears, “white teacher” primes us to expect racism in a way “teacher” does not.) Yet it would seem a hypocritical silence, I think, to leave racial terms out of my own descriptions. For to me in this research, speakers almost always appeared racial.

---

[1] For discussion of the process through which pseudo-scientists reinforced the socially constructed “race” categories of “black,” “white,” and “Indian” (central to the institution of slavery) through massaging “biological” findings about genetically insignificant variations in human physical appearance, see Stanton 1972, Gould 1981, Stocking 1982, Banton 1998. Smedley (1999) provides a particularly useful discussion of how “race” was created as a North American system of “black”/“white”/“red”difference based fundamentally on categorizations and rankings of exterior appearances. For discussion of the extension of “race” categories over several centuries of immigration and colonial expansion to include the both-imposed-and-chosen racialized categories “Asian” and “Hispanic”/“Latino,” see Espiritu 1992, Almaguer 1994, Delgado and Stefancic 1998. The classic evidence against the existence of biological “races,” of course, is the fact that the handful of groups Americans call “race” groups are actually more genetically diverse within themselves than between themselves (see Montagu 1942).

[2] Minow (1990) provides an eloquent discussion of this tension in U.S. legal analysis; see also P. Williams (1997), and Appiah and Gutmann (1996).

[3] Van Den Berghe (1996) contrasts a general U.S. anxiety about race to Brazilians’ happier national discourse on race’s supposed irrelevance. Of course, the Brazilian mantra that race does not matter much in Brazil has been challenged by many scholars, who have argued that Brazil is an extremely race-conscious (and racially stratified) nation in which speakers simply resist conversations about racism. See, e.g., Twine (1998) and Sheriff (2000). Sheriff (2000) argues that this resistance to discussing racism is consistent across the Brazilian population’s racialized spectrum: “‘Ninguem gosta de falar [no one likes to talk about it],’ I was told over and over,” she writes (117).

[4] Proposition 209 also called for a halt to race-conscious outreach programs in public employment and contracting. The opening text of the Proposition, which eventually amended the California Constitution, read as follows: “The state shall not discriminate against, or grant preferential treatment to, any individual or group on the basis of race, sex, color, ethnicity, or national origin in the operation of public employment, public education, or public contracting.” For sociological examination of the proposition and its development, see Chavez 1998.

[5] Similarly, after the Hopwood decision in Texas ruled the consideration of race in law school admissions impermissible at the state university, admissions officers at private Rice University also decided to delete race words from admissions talk. According to the New York Times, “Almost overnight, the admissions officers at Rice stopped saying aloud the words ‘black,’ ‘African-American,’ ‘Latino,’ ‘Hispanic’ or even ‘minority’ in their deliberations. The next year, the proportion of black students admitted in the freshman class fell by half; the proportion of Hispanics fell by nearly a third.” (“Using Synonyms for Race, College Strives for Diversity,” December 8th, 2002.)

[6] Acknowledgements, of course, to Cornel West (1993).

[7] Scholarly or political talk of an American “we” is made suspect by many authors, who argue that any claim of “‘an overarching ‘we-ness’ across ethnic and racial difference” is itself a “colorblind” fantasy (Thompson 1999, 141).

[8] Frankenberg (1993) has argued that “race evasive discourse” is a “white” American habit, exhibited particularly by “whites” who, by insisting on essential human “sameness,” avoid all direct references to race as if such references are inherently “racist.” “From this point of view,” Frankenberg notes, “there are apparently only two options open to [whites]: either one does not have anything to say about race, or one is apt to be deemed “racist” simply by virtue of having something to say” (33).

[9] Gunnar Myrdal (1944), of course, famously described the “American dilemma” as the simultaneous existence of an ideology of equality and the practice of racial inequality (for further discussion, see “An American Dilemma Revisited,” Daedalus 1995). Despite the obvious problem that “American” marks all the Americas and not just the U.S., I utilize the term in this book because of its preeminence in U.S. ideological and political discourse, particularly about race and racism. See also Hochschild 1984, Hochschild and Scovronick 2003.

[10] For foundational scholarship on race and schooling, see Woodson, 1972. On slave literacy, see, for one, Genovese, 1974; on the history of desegregation (including the lesser-known history of battles for integrating “Mexican-Americans” and “Chinese”), see, e.g., Kluger 1975, Wollenberg 1976, Weinberg 1977, Kirp 1982, Anderson 1988, Orfield and Eaton 1996, Wells and Crain 1997, Donato 1997. For a basic outline of the history of (and debates over) affirmative action in college admissions, see, e.g., Takagi 1992, Curry 1996, Edley 1996, Chavez 1998. On debates over “multiculturalism,” see Taylor 1994, Banks 1995, Hu-Dehart 1996, Nieto 2000.

[11] To begin an examination of the central role of schooling in the reproduction of U.S. racialized categories, identities, and power structures, see, e.g., McCarthy and Crichlow 1993, Ladson-Billings and Tate 1995, Tatum 1997, Fine, Weis, and Powell 1997, Kincheloe et al 1998, Nieto 2000 (for discussion in a South African context, see Dolby 2001; for a Canadian example, see Yon 2000). Indeed, schools are institutions where people reproduce many such systems of difference and inequality: as others have argued, schools in the U.S. and elsewhere also help produce people who are “ethnic” or “indigenous” (Lukyx 1996), classed (Willis 1977, Eckert and McConnell-Ginet 1995, Varenne, Goldman and McDermott 1997), gendered (Thorne 1993, Luttrell 1996), citizens of nations (Levinson 1996), “abled”/“disabled” (Varenne and McDermott 1998, Mehan 1996) and members of various local status groups (Eckert 1989). For more on the racialized distribution of educational opportunity through schooling, see, e.g., Payne 1984, Oakes et al 1990, Kozol 1991, Carnoy 1994, Delpit 1995, Orfield and Eaton 1996, Secada et al 1998, Noguera 2000, 2001, Valencia 2002.

[12] For such scholarship on “race”-group linguistic patterns, see, e.g., Labov 1972, Kochman 1981, Bucholtz 1995, Smitherman 2000, Morgan 2002. Many of these scholars and others have long argued that this kind of “group”-specific talk matters particularly in school: see also, e.g., Cazden, John and Hymes 1972, Philips 1972, Erickson and Mohatt 1982, Heath 1983, Foster 1997, Baugh 1999. Linguistic anthropologists and sociolinguists have also been particularly interested in how people using specific words or languages create a sense of themselves as distinct populations (see, e.g., Gumperz 1982, Silverstein 1985, Gal 1989, Urciuoli 1996, Bailey 2000). For an important synthesis of anthropological theory on everyday ideologies about language and language use, see Schiefflin, Woolard, and Kroskrity 1998.

[13] For further theoretical discussion of these contemporary and historical methods of everyday racialization in the U.S. -- practices, that is, through which we place one another into groupings we call “racial” -- see Goldberg 1990, Roediger 1991, 1994, Omi and Winant 1994, Delgado 1995, Frankenberg 1995, Bucholtz 1995, Jackson 2001. On the production of racial difference via the employment of popular culture, see Roediger 1998, Perry 2002, Carter forthcoming; through housing and neighborhood “choice,” see Patillo-McCoy 1999, powell et al 2001; through social association, see Tatum 1997. Much influential theory on race-making also has emerged from scholars working in a European/diasporic context; see Fanon 1952/1990, Hebdige 1979, Hewitt 1986, S. Hall n.d., 1991, 1992, 1998; Mercer 1990, Gilroy 1991, 1993a, 1993b; Rampton 1995. “Race” grouping has been framed in scholarship as the production both of categories central to inequality systems and of community-specific practices. West (1992), for example, describes “blackness” in one formulation as 1) the experience of inhabiting a category that must weather the constant probability of subjugation by “whites,” and 2) the experience of enjoying the cultural practices shared by a community strengthened by resistance to such subjugation. Many authors have similarly named “whiteness” as both a category of privilege and a set of practices embodying that privilege (see Roediger 1994, Delgado and Stefancic 1997, Lipsitz 1998, Kincheloe et al 1998, Thompson 1999). Frankenburg (1993), for example, frames “whiteness” as “a location of structural advantage, of race privilege”; a “standpoint; a place from which white people look at ourselves, at others, and at society”; and “a set of cultural practices that are usually unmarked and unnamed” (1).

[14] Appiah (1996) calls identification this process of shaping lives and “life projects” around ascribed racialized labels (78).

[15] Words not only invite or deter actions from others, but act on people to in a sense create them. For classic analyses arguing generally that language shapes the world we live in, see Whorf 1940/2000, Althusser 1971, Foucault 1972, Wieder 1974. Scholars of “speech act” theory (Austin 1962, Searle 1969 ((for discussion, see also Hanks 1996)), and those working within the tradition of conversation analysis, have long been particularly interested in words as actions that produce real-world consequences. On conversation analysis, see, e.g., Cicourel 1970, Moerman 1988, Duranti and Goodwin 1992, Goodwin and Goodwin 1992, Tedlock and Mannheim 1995 (for useful syntheses of such research on discourse as social action, see also Briggs 1986, Mishler 1986). Mehan (1996) argues further that a “social constructionist” tradition of research has been particularly concerned with how institutional orders, including school orders, are “both generated in and revealed by the language of the institution’s participants” (243). “Because language is action,” Mehan writes, “different uses of language constitute the world differently” (262, emphasis mine). Critical Discourse Analysis scholarship (Fairclough 2001) has also argued particularly that language helps produce existing power relations even while displaying those relations; language, as Fairclough writes in summation, particularly “contributes to the domination of some people by others” (3). Critical theorists of race, finally (see, e.g., Delgado 1995, Delgado and Stefancic 1998), have looked particularly at how race labeling practices have organized U.S. history, while theorists who explore gender and sexuality (Thorne 1993, Spivak 1993, Hall and Bucholtz 1995, Ortner 1996, Butler 1997), have also explored the systemic and everyday social consequences of labeling.

[16] As seen here, I occasionally highlight key descriptive terms in boldface throughout this book. For explanation of this device, see “A Note to Readers,” p. 25.

[17] For further discussion of youth worldwide negotiating over (and eliding) meanings of “race,” “ethnicity,” and “nationality,” see, e.g., Gilroy 1993a, Sansone 1995, Amit-Talai and Wulff 1995, Sharma et al 1996, K. Hall 2002, Maira 2002.

[18] For more analyses of “race” as a system of distributing power to simplified groups (as opposed to the heritage-connoting, voluntary-group-association term “ethnicity”), see Omi and Winant 1986, Harrison 1995, Sanjek 1996.

[19] In contrast, Schofield (1995) describes a strikingly self-consciously “colorblind” school of the 1970s-1980s (with demographics listed as 50% “black,” 50% “white”) in which labels were almost completely absent from daily discourse. “In almost 200 hours of observations in classrooms, hallways, and teachers’ meetings. . .fewer than 25 direct references to race [for the author, direct uses of the words “black” and “white”] were made by school staff or students,” she writes (254).

[20] Moerman (1988) notes that “study of actual occasions has demonstrated that all terms for persons are said and understood by people with real interests. Those people actively orient to their immediate situation and type of occasion, to the setting, to when during the occasion the reference occurs, to the kinds of actors present, to social actions both ongoing and hinted at. All those factors -- and more -- influence the terms they use” (98). Gumperz and Hymes (1972) called the knowledge necessary to communicate appropriately in such various cultural contexts “communicative competence” (vii).

[21] For important work on “race” and identity development in the U.S., see, e.g., Cross 1991, Tatum 1992, Ward 2000.

[22] For Appiah (1996), “identification” in a more comprehensive sense is always intertwined with “identity,” since the very application of race labels helps shape how “an individual intentionally shapes her projects -- including her plans for her own life and her conception of the good -- by reference to available labels, available identities” (78). In so defining “identification,” Appiah builds on the work of Ian Hacking on “making up people”: Hacking writes that “numerous kinds of human beings and human acts come into being hand in hand with our invention of the categories labeling them” (Hacking 1992, cited in Appiah 1996, 78). I am concerned in this book primarily with the basic identificatory act of labeling itself, and with its effects in “shaping” shared social orders. For some nice examples of anthropological work that looks similarly closely at the processes and consequences of categorizing kinds of people (and negotiations over such categorization), see, e.g., Moerman 1968, Frake 1980, 1998, Bucholtz 1995, Mehan 1996, Bailey 2000, Jackson 2001. For some foundational work on categorization and labeling as a universal human process, see Levi Strauss 1963; for an excellent discussion of contemporary anthropology’s concerns with the very process of delineating “kinds of people” in an increasingly globalizing world, see Appadurai 1996.

[23] Through historical inquiry, we have watched not just public figures (see Jordan 1974 on Thomas Jefferson’s pivotal racializing role in U.S. history) but also 18^th- and 19th-century American and European scientists actually devise racial categories and struggle to rank them according to existing social perceptions (see, e.g., Gould 1981, Stocking 1982, Banton 1998, Baker 1998, Smedley 1999). Historical scholarship has also illuminated moments when people confronting the seamless continuum of human diversity struggled to keep social race categories intact, arguing passionately, for example, that some people were simply “black” and others simply “white” (see, e.g., Roediger 1991, Haney Lopez 1996, Davis 1997, Sollors 2000). Examining centuries of racial formation in America has also demonstrated how populations once categorized as members of tribal, religious, or national groups ended up “racial” (see Genovese 1974, Espiritu 1992, Omi and Winant 1994, Almaguer 1994, Ignatiev 1995, Sacks 1997).

As suggested here, anthropology itself played a key role in creating such “racial” categories in the first place; it has since taken on the responsibility of calling them into question. At the turn of the 20th century, skull-measuring anthropologists were still arguing vehemently that racial groups were biologically distinct populations that could be “scientifically” ranked on a scale of human worth; yet other anthropologists began to step in to debunk as pseudoscience the idea that race groups could be placed on any legitimate hierarchy (see, e.g., Boas 1895; for discussion, see Stocking 1982, Baker 1998). By midcentury, anthropologists were arguing loudly that “races” did not exist at all, pointing out that the world’s so-called “racial groups” were more genetically diverse internally than different from each other (for a seminal contribution, see Montagu 1942). Anthropologists now routinely frame racial categories as historical, cultural, and political creations rather than natural or “biological” realities. Current public education work by the American Anthropological Association emphasizes the genetic fiction of “race” (American Anthropological Association 2002). Our most nuanced spokespersons emphasize that after centuries of acting as if race categories exist, we have basically created a world in which they do (Harrison 1995, American Anthropological Association 1998). This argument will be explored further in Chapter 1.

[24] Thorne (1993) found similarly that the study of gender in childhood typically framed gender as a set of traits owned by boys and girls themselves (the typical question of childhood gender studies, she writes, is “are girls and boys different?”). Thorne shifted in her own fieldwork to ask, “How do children actively come together to help create, and sometimes challenge, gender structures and meanings?” (4). Gender, Thorne concluded, “is not something one passively ‘is’ or ‘has’”; rather, people “‘do gender’” (5). Appadurai (1996) articulates similar concerns about how the concept of “culture” is used in much theory and research. Even using “culture” as a noun, he argues, implies that “cultures” are easily delineated “things” owned by specific people. Rather, he notes, the social or “cultural” practices of “cultural groups” (and the maintenance of boundaries between such groups) are being negotiated constantly (12-13).

[25] In treating race talk as directly representative of inner thoughts and feelings, much research on race thus gets trapped in attempts to infer speakers’ “real” intentions (for comment, see Connolly and Troyna, 1998; see particularly Hammersley 1998). For an analysis of the many such pitfalls involved in research on everyday language, see Mertz 1992, Briggs 1986, Frake 1980. On the “ethnography of communication,” long known for explorations of patterns in the way “natives” talk, see the foundational work of Hymes (1962, 1964) and Gumperz and Hymes (1972).

[26] In this book, I call “private” any talk occurring in groups of two to around three or four people, while I call “public” the talk of more formal interactions involving additional participants (including written documents circulated to numerous readers).

[27] For concise discussion of anthropology’s contemporary problems with describing “kinds of people,” see Marcus and Fisher 1986, Clifford and Marcus 1986, Appadurai 1996, Gupta and Ferguson 1997.