If there is error here, it is the paradigm that is errant, not the individual, and the
relevant
paradigm is the paradigm of sci- ence as currently constituted and as passed through the medical school filter.
The Public Work of Rhetoric_nodrm
8 An enormous variety of scholars have shown that the
120 Celeste M. Condit
technical sphere, like the public sphere, is at least partially shaped by rhetori- cal factors. 9 It might similarly, therefore, be susceptible to intervention through discursive efforts guided by rhetorical theory.
Even when addressing topics raised by science and technology, however, most rhetoricians have directed their discourse at other rhetoricians or at a presumed reading public inhabiting the public sphere. With regard to the topic of race and genetics, there are, for example, numerous analyses that show the problematic metaphors and categorizations of genetics,10 or reveal the rhe- torical strategies by which the linkage between "race" and "genetics" is con- structed in scientific discourses or reportage on science. 11 While there are some exceptions to the tendency to focus on technical communication as it plays out in the public sphere, the heavy publication slant suggests that it is likely that more activist interventions are also directed at the public sphere rather than involving engagement within the technical (or scientific) sphere. 12 To try to enter the scientists' own conversations, to argue with them within the ven- ues and rubrics of science, is a somewhat different enterprise. To broach the feasibility of the practice and some of the challenges, I recount four different episodes in which I engaged geneticists in scientific venues on the subject of the relationships between "race" and "genetics. " I describe some of the major reasons for the varying levels of success I had in each different situation. I conclude that, if my own experience is at all representative, then the funda- mental variables influencing persuasiveness in the scientific venue are remark- ably similar to those in the public sphere: you persuade people if you talk their language, but it is more difficult to persuade someone the more his or her highly interested worldviews are at stake.
My Coming to Terms with Race and Genetics:
The Human Genetic Variation Consortium
Until I moved to New Orleans, I reacted to race in most of the ways the aver- age white professional American today reacts to race. I wanted to be "liberal" and "tolerant," not a "racist," but these were mere platitudes with no depth or breadth of meaning. When I moved to New Orleans, however, I moved to an area called "the black triangle" because I did not want to commute an hour each day to teach at Tulane University, because I was a teacher who could not afford to live in a "white" area, and because my liberal platitudes told me that there was no reason to be prejudiced and avoid a "black" area of town. Liv- ing there gave me experiential lessons about the meanings of racialization in America, and it also led me to modify my research agenda to focus on race. Periodically I have become overwhelmed and despondent and "stopped work- ing on race. " But race in contemporary America is not something you can ever really walk away from.
Consequently, though I had "stopped working on race" half a dozen years earlier and turned to studying genetics, in 1998 I found myself submitting a
The Process of Remaking Race as Genetic 121
proposal to the National Institutes of Health (NIH) to study how laypeople, especially African Americans and Whites, understood the relationship between race and genetics. When that proposal was eventually funded, I was included in a group called the Human Genetic Variation Consortium. This was a group of researchers working in various ways on social, ethical, or legal issues re- lated to race and genetics. Most notably, the group discussed the "Haplotype Map" project (described below) and advised some of that project's advisers. At the close of the term of my cohort of members, a group of its members published a position paper in the Journal of the American Medical Association on the lack of relevance of genetic research to the amelioration of health dis- parities. 13
The consortium meetings during the first year were devoted to familiariz- ing ourselves with the new trend in genetics toward taking account of race in genetics and health studies. We were told that there were two major forces generating this trend. The first was a need for geneticists to account for popu- lation substructure in their work. Through several expensive and embarrass- ingly errant studies, geneticists had discovered that differences among the geographic ancestry of control and experimental populations could produce spurious findings that "a gene" was related to a particular health condition. 14 The simplest way of correcting that problem appeared to be to provide rela- tively rigorous controls on the boundaries among the "racial" groups that com- posed research populations.
The second concern was more compelling to the humanists and social sci- entists in the consortium. It was the issue of health disparities. There can be no denial that there are clear health disparities between groups that are under- stood as "racial" groups in the United States. Compared to people who self- identify as "White," people who self-identify as Black or African American, American Indian, or Hispanic American die younger and suffer more and worse disease across many common diseases, including cardiovascular dis- eases, diabetes, and several kinds of cancer (the situation is more complex for Asian Americans due to different historical and economic positionings of dif- ferent nationalized subpopulations. )15 Many doctors and researchers, includ- ing minority physicians, have come to believe that a part of these disparities is due to differences in biology, specifically in genes.
This rationale for attending to the possibility that "race" is "genetic" was compelling to many members of the consortium, because it meant that dis- missing biological accounts of race was not a move with positive, or even merely neutral effects, but might be positively harmful to minority groups. For example, ignoring race results in the recruitment of predominantly white pop- ulations to research studies, and it produces medicine that is tailored to white bodies and therefore may be less helpful to the bodies of members of other groups. If there is substantial biological variation among the bodies of ethnic or racial groups, then to ignore that variation is to promote "the invisibility
122 Celeste M. Condit
of whiteness," which is simply the contemporary route to privileging white people. 16 It is to mistake whiteness as the universal version of humanity.
The members of the Human Genetic Variation Consortium exhibited a range of reactions to these presentations. My own reaction was to be deeply troubled and indecisive. On the one hand, I was profoundly skeptical of the idea that race was biological. As a rhetorician, I understood that words made categories, and that the structure of material reality in the world did not sim- ply and neutrally require a particular set of labels. I wrote, and continue to write, rhetorical analyses designed to show the constructedness of race with regard to human genetic variation. 17 On the other hand, I had come to take seriously the fact that race could not be simply ignored, denied, or swept under the carpet. Health disparities were undeniable, and minority researchers and physicians I respected insisted that they saw important differences between average members of different groups. While it was clearly true that the "aver- age" member of a group was an imprecise construct, it might also be true that denying such average differences would merely reinforce a medical system that served white people well and other people not so well. Moreover, I had something of a consistency problem, as my own research required me to explicitly invite people to participate based on something like their "race. " If I were to be able to provide a sufficiently robust input from people other than white southerners, then I had to explicitly invite people into my research proj- ect based on their identity as "African Americans. " If this was a good, even essential, practice for the research I thought I should pursue, then how could I insist that it was not a good thing for people doing other types of research?
I took this profound uncertainty as an initial question for my research process. I decided to trust the community of people who would be most affected by a race-based medical research endeavor, African Americans. So, in conjunction with a fabulous group of graduate students and coresearchers, I designed and conducted focus groups to find out what ordinary people thought of these possibilities. 18 The answer African American participants in this research gave was overwhelmingly opposed to race-based medical ap- proaches. White participants tended to oppose them as well, though not as vehemently. Having learned to distrust my own white-based instincts on issues of race, I determined to trust these people's judgments. The effort to amplify their voice became the basis of my first attempt at engagement with the sci- entific community.
Engagement #1: Publish Data About the Challenges and Costs of Race-Based Medicine in Scientific Journals on Genetic Medicine
My goal was to engage medical geneticists who were being encouraged to par- ticipate in a paradigm of race-based medicine. It seemed to me that the flow of medical science was such that it was better to prevent the formation of an errant consensus by these experts rather than to contest that consensus once
The Process of Remaking Race as Genetic 123
it reached the public sphere, so I determined it best to engage in the process by which the medical consensus on race and genetic medicine was being for- mulated. All I had to offer by way of argument, however, was the fact that a substantial group of laypeople did not like the idea. How could that fact be made of relevance to the medical community?
Medical research in genetics, like all medical research, is largely bereft of attention to many of the issues of how a particular medical regime will be implemented in practice. Medical research is conducted in highly selective locations under near-ideal conditions. When a medical procedure or drug is approved, it is diffused for use under very different conditions. This, I thought, was exactly the problem for a race-based medicine. In an ideal world, where everyone was enlightened, using a social grouping such as "race" as shorthand for mild tendencies for human biology to vary might be acceptable. In the real world, as my participants clearly indicated, it just would not work. Since race-based medicine was being promoted primarily as an economic necessity-- due to the presumed expense of doing individualized genetic testing--showing the economic and other barriers to implementation seemed like a reasonable line of argument. 19
So I and my research team wrote "Attitudinal Barriers to Delivery of Race- Targeted Pharmacogenomics among Informed Lay Persons," and we were suc- cessful in publishing it in Genetics in Medicine. 20 The article and its companion article were written and published within the generic norms of a scientific article. 21 Its sections are "methods," "data," and "discussion. " It uses quanti- tative presentation of findings and employs the objectivating procedure of intercoder reliability assessments. As Kenneth Burke has noted, you persuade a person insofar as you speak his or her language, and because we wanted to be persuasive to an audience of scientists, we presented our research in the language and conventions of science (even though we were mostly a group of rhetoricians). 22
The contents of the article indicate that the effectiveness of a race-based medical approach would have to be assessed based on a variety of costs and challenges not presently included in the arguments in favor of its implemen- tation. These additional costs included the inability of doctors to assign peo- ple to discrete and accurate ancestries, the lack of knowledge of the biology of one's grandparents especially among African Americans, and very high levels of distrust of such medicines by their users, which is likely to lead to low com- pliance with prescriptions as well as a potential backlash against prescribers, pharmaceutical companies, and medical research institutions if ever such a drug turns out to have serious harmful side effects (which seems probable over the long run, especially given the contemporaneous tragedy of Vioxx and related cox-2 inhibitors). In other words, we gave the medical experts scien- tific evidence that race-based medicine was unlikely to be cheap or particu- larly effective.
124 Celeste M. Condit
The impact of this pair of articles was, so far as we can tell, virtually zero. Citation indexes showed only three citations of either article. A wide variety of sources that should have cited it because they were addressing costs of race- based medicine or feasibility did not. My first intervention was apparently a total failure.
Engagement #2: Going Live to Genetics Researchers on the Harmful Impact of "Race as Genetic"
The opportunity for a second intervention presented itself as a matter of both serendipity and foresightful preparation. One of the major arguments that humanists and social scientists use for arguing against geneticization of racial categories is that such geneticization makes race appear absolute and perma- nent. The technical term for this is that it tends to "reify" race. In contrast, a nonreified understanding of existing labels for groups sees these labels as leaky and imprecise and as a temporary product of historical and social confluences. Indeed, the historical evidence is quite clear that racial group labels and racial groupings vary considerably through time and space. 23 In contrast, assigning a scientific basis for a grouping is believed to suggest that such groupings are permanent and absolute, because we tend to think of scientific categories as universal and rigid. Therefore, a scientific grounding for race should produce more "racism" than would an understanding of race as a product of social and historical forces.
Is this presumption correct, however? Having become skittish about the link between genes and racism, I wondered whether we humanists were really right to claim that people who saw health as differentially distributed by race and who saw race as distributed by genes would be more racist than people who did not. If ignoring race was tantamount to universalizing whiteness, perhaps recognizing race was necessary to overcome racism. And if that were true, then perhaps people who thought of health disparities as a simple matter of biological difference would not therefore conclude that everything about racial groups (for example, class distribution or IQ scores) was a matter of genetics and was therefore fixed forever.
I continue to puzzle over this issue. However, one way of addressing the question, and of addressing it in a way that would have credibility with the scientifically based genetics community, is to run an experiment to see if peo- ple exposed to messages linking health, race, and genetics increase the racism level of their attitudes as composed to people who do not hear such messages. So this is what my research team did. The results of the experiment seemed to me shocking and frightening. We exposed people to a very tame message about health disparities, one that simply mentioned genetics in passing as one possible basis for health variation. People who heard this message increased their level of racism an average of almost a full point on a five-point racism scale. 24 We had expected no effect at all, because we had used a very mild
The Process of Remaking Race as Genetic 125
message. We were quite concerned about the ethics of increasing people's lev- els of racism through an experiment, given that we were not sure that the debriefing message we would give them at the end would actually return their levels to preexperimental levels, so we had decided to start below the level where we thought we would create change and inch our way upward to a modest level of effect. The effect of a 20 percent increase that occurred from our first modest level message is unusual in message-design research, so this very large effect convinced us that messages about race, genes, and health most certainly would cause "reification" of attitudes about race and would amplify racism.
So that was the foresight. Serendipitously, around this time, because of my work with the Human Genetic Variation Consortium, I was invited to make a presentation to the sixth International Meeting on Single Nucleotide Poly- morphism (SNP) and Complex Genome Analysis. 25 This is a meeting of a group of researchers who participate in projects designed to track down details of variation in the human genome in order to produce the basis for identifying genes that are associated with common diseases and identifying medicines that might address those genetic variations. I presented the results of the mes- sage experiment in my fifteen-minute presentation close to the beginning of the conference.
The impact of the talk was, I believe, relatively substantial--among those present, for the duration of the conference. The audience was attentive, and later speakers evidenced awkwardness, discomfort, and attempts at circumspec- tion in labeling groups in their presentations. Moreover, the award-winning poster for the conference was one that indicated that apparent clusters of hu- man genetic variation by continents was an artifact of sampling procedures that left large gaps in samples between continental groups rather than using a continuous sampling procedure. I heard the winner being told that "as soon as I heard that talk, I knew you were going to win. "
I might well be exaggerating my own influence here. At the least, my pres- entation was one straw on a growing pile on an invisible camel (though there were no other presentations of this sort at the conference). I have to admit to being heartened and hopeful. It is important, however, to note the audience that evinced this effect. Most of these researchers do work that is well "up- stream" from actual medical applications on live human beings. They might gather DNA from people, but after that they sift and search among single bases, comparing dozens of genomes on computers, looking for variations among millions of base pairs. They are highly mathematical scientists, doing statistical tests and creating statistical tests to assess what might actually count as an association in such large and complex data sets. Their main stake in the "race is genetic" debate is merely the ability to control for population struc- ture. Population structure can be controlled simply through geographic dis- tance. It does not need current racial labels.
126 Celeste M. Condit
This audience also consisted of a very smart and northeastern-biased group of scholars, trained at excellent schools. Not only do these scholars not want to appear to be racist, they really do not want to be racist, and they were trained in an understanding of racism that indicated calling out difference meant racism. They would probably have been completely at a loss if one tried to suggest that one problem of racism is the invisibility of whiteness. Therefore, my presentation did not cost the audience much in terms of their scientific goals, and it fit well with their understanding of what "not being a racist" meant. Moreover, short-term ripples do not necessarily make long-term impacts, as the third intervention was to make clear.
Engagement #3: Relabeling the HapMap
During the second year of my stint on the Human Genetic Variation Consor- tium, some personnel of the National Human Genome Research Institute's Ethical, Legal, and Social Implications (NHGRI/ELSI) unit presented the con- sortium with plans for the "Haplotype Map Project" (or HapMap). In short, the project was designed to gather samples of human DNA from different groups in order to produce a tool for more rapidly and efficiently looking for variations in human genes that are associated with diseases. Various NIH per- sonnel and researchers associated with the project were concerned about the public response to it. A few years earlier a "Human Diversity Project" had gained harsh worldwide negative publicity when it had sought to sample DNA from global cultures that are threatened with extinction. The ELSI researchers wanted the HapMap to be done in an ethically responsible fashion, and other project personnel either wanted that or, at the least, recognized the need to avoid generating another hostile international response. Some members of the Human Genetic Variation Consortium were on the ELSI consulting group for the project, but NIH officials also presented the project to us to get our reaction.
The discussion within the consortium was intense, extended, and challeng- ing. The HapMap is a technically complex project. Understanding it requires careful explanation of detailed scientific information. The consortium included people ranging from an almost knee-jerk antiscience, antigenetics, antirace position, to medical researchers who were opposed to the use of race, to con- flicted humanists, and to social scientists from several disciplines. Although I took notes, I make no claim to be able to represent objectively the process. It is my impression, however, that my role in the conversation was to insist that if the HapMap needed to sample across broad geographies merely in order to assure that a broad range of SNPs (genetic variations) be included, then their populations should not be labeled so as to imply that what was being sam- pled was "race. " In other words, instead of labeling one population "Africans" or "sub-Saharan Africans," they should be labeled as coming from the nar- rowest precise geographic locale--for example, Ibadan, Nigeria. I was quite
The Process of Remaking Race as Genetic 127
insistent that the HapMap should not reify the idea that the genetic variation that needed to be sampled was racial groupings, given the project scientists' claim that they were merely trying to obtain diversity, sampling people from geographical regions that were far from each other. As a rhetorician, I focused intensely on the word choices in the labels, operating from the belief widely shared in rhetorical studies that particular word choices matter.
Other people made similar arguments. Moreover, the conduit we had for making an impact was extremely indirect. We had the ear of one NIH scien- tist and one NIH (ELSI) program director and the ear of those members of the Human Genetic Variation Consortium who would participate directly in the HapMap project. In other words, I have no idea whether my personal argu- ments had any impact at all. Nonetheless, there is some clear evidence of adoption of the labeling practices we were promoting. On the HapMap Web site, the three groups are not labeled as "Africans," "Caucasians," and "Asians," as was a fairly predictable labeling structure given previous conventions in the scientific literature. 26 Instead, the "recommended descriptors" are "Yoruba in Ibadan, Nigeria (abbreviation: YRI), Japanese in Tokyo, Japan (abbreviation: JPT), Han Chinese in Beijing, China (abbreviation: CHB), and CEPH (Utah residents with ancestry from northern and western Europe) (abbreviation: CEU). "27
This novel labeling, and the official Web site's explicit attention to the ra- tionale behind it, suggests that my intervention, or co-occurring interventions by others, had some impact. It was, however, a negligible impact, because in the major article presenting the HapMap project in Nature, almost all the data combines the groups JPT and CHB and pits them against the other two groups. 28 This combining of the groups, presented without any rationale, indicates that the authors are still thinking of and presenting these samples as representing "Asia. "29 Furthermore, in later articles that cite and apply these findings, the HapMap data is relabeled with conventional race labels and is explicitly taken to support the idea of continent-based racial group- ings. 30
This does not demonstrate that the labels researchers choose are irrelevant. It merely indicates that getting scientists in a key locale to adopt a particular (nonracial) label set does not necessarily have very large ripple effects. Absent a clear understanding of the novel labels, those who encounter them merely translate them back into more familiar categories. In addition to simple famili- arity and inertia, a major force behind the maintenance of the old labels arises from those who have a stake in maintaining those labels. It was this group that my fourth intervention engaged.
Engagement #4: The NIH Roundtable on Race
The NIH has relatively recently attended to the ethnic disparities in health and in science in a variety of ways, including structural requirements for
128 Celeste M. Condit
researchers to include diverse populations in their research, outreach and financial support for minority researchers, bureaucratic components devoted to minority issues, calls for research projects that are likely to generate infor- mation of relevance to nondominant sectors, and a variety of conferences and workshops. The NIH Roundtable on Race was convened as one such con- ference. It specifically focused on issues surrounding the implications of genet- ics for race and for race-based medicine. Invitees and presenters included both medical researchers pursuing race-based medical or other projects and social scientists and humanists, most of whom were in some way skeptical of those projects. Minority scholars were included in both groups. Many of the schol- ars from minority groups, but not all, were favorable to race-based research and practice.
At this conference, I presented the same data I had presented at the SNP conference, as well as descriptive data characterizing the mass media's cover- age of race-based genetics research and practice. 31 I would have to call the re- sponse of the audience cold, if not hostile. An older female minority scholar told me to "think about the parable of the sun and the wind, trying to get the person to take off their coat. You are the wind. " In the parable, the wind fails because it only makes the person pull the coat tighter. The sun succeeds by making the person so warm they take off their coat. In light of the experi- mental results I had gotten, I was at a loss to figure out how to make people be less racist by applauding their assumption that race was genetic, so in spite of this chastisement I more or less continued a sustained argument against that view in the interactive discussion periods. Nonetheless, I learned a good deal from the people there about their motives, their sincerity, and the source of what seem to me to be their rhetorical blind spots.
I take as an example a researcher named Esteban Gonza? lez Burchard, a well-published scholar as well as a medical researcher dedicated to improving the health of Puerto Ricans. He has special interest, even passion, for address- ing the high rates of serious asthma in the Puerto Rican population. He and I had civil, mutually respectful interchanges, but from my perspective Dr. Burchard is fundamentally wrong. At this conference, the warrant he articu- lated for maintaining race-based approaches and for insisting that race had a genetic basis was the fact that Puerto Ricans have a high asthma rate and a low response to the available medications (much lower than other groups). However, "Puerto Ricans" is not a race, even according to the standards of Dr. Burchard's own published research, and Puerto Ricans are not typical of "Latin Americans" or "Native Americans" or "African Americans" (or any other of the groups geneticists describe as "racial") with regard to their asthma suscep- tibilities. 32
Nonetheless, as we talked, I gained additional depth of understanding about the ways in which the increasingly prominent version of antiracism that attacks "color-blind" approaches and the "invisibility of whiteness" applies in
The Process of Remaking Race as Genetic 129
health disparities. Dr. Burchard's perspective worked within this vision of anti- racism. Additionally, however, my perception became that Dr. Burchard was unwilling to entertain environmental or cultural explanations for evident dif- ferences because such accounts in some sense place blame, or at least respon- sibility, upon Puerto Ricans for their health problem. In contrast, a genetic explanation places the cause of the health disparity beyond the personal, social, or even cultural control of the population at risk.
There are, of course, other ways to avoid "blaming the victim. " The most common is to attribute the cause to the health care system or to environmen- tal factors that are the result of economic forces and structural discrimination. But Dr. Burchard, like many other minority physicians, appears to be in a posi- tion that makes it difficult to blame the medical system for health disparities. Minority medical personnel are part of that system, and they are working very, very hard to serve their populations through it. They are disinclined to blame the systems that do indeed provide extensive positive services to their popu- lations, and certainly they should not blame themselves, even though they must serve at least in part as agents of that system, its faults included.
Their acceptance of the system's perspective is not merely due to the social status it provides them. It also seems to result from the adoption of the sub- ject position of "physician," which, as many scholars have suggested, entails seeing patients in certain lights--as bodies, not as products of social sys- tems. 33 Indeed, the medical orientation that all physicians are trained into-- even minority physicians--runs to an even more deep set of biases in this regard than has generally been recognized. 34 Physicians are trained into the scientific worldview. The contemporary version of this worldview understands human beings almost exclusively through animal models. No other animal is subject to the variety and complexity of social, economic, and symbolic forces to which human beings are subjected. Thus the animal models provide no reason to factor social structures into health outcomes.
Dr. Burchard has heard extensive, detailed critiques of "race" (at this round- table as well as at the "Defining Race" NIH In-House Conference in 2006), but these have moved him only to restate more adamantly what he already "knew," not to step back and reconsider the ways in which he was using the term "race. " I assert that this sense of certainty arises from at least three causes--his passionate commitment to helping Puerto Ricans, his highly con- temporary view that helping Puerto Ricans entails a stance that highlights difference rather than ignores it, and an understanding of what counts in sci- ence as a source of causation (that is, genes, not social structure).
I have used Dr. Burchard as an example here not to belittle him, for I have the greatest respect for his intelligence, his hard work, and his dedication to the health of Puerto Ricans.
If there is error here, it is the paradigm that is errant, not the individual, and the relevant paradigm is the paradigm of sci- ence as currently constituted and as passed through the medical school filter.
130 Celeste M. Condit
I have come to believe that there is nothing I could write or say that would convince someone who cared passionately about the health of a minority group, and who accepted the current discursive framework of medical research, that he or she should abjure the utility of "race" in the study of medical genetics.
Scorecard
I have undertaken four major engagements with medical geneticists in an effort to get them to eschew the claim that race is genetic. I did so with the hope that by engaging rhetorically in the scientific realm, I could help stop the formation of a scientific consensus that would then be translated into the public realm as the indisputable scientific fact that race is genetic. I harbored no illusions that my efforts, on their own, would be sufficient. The scientific realm is large, with many players, and my voice is just one, able to be heard in few spaces. It is therefore difficult to evaluate these efforts. Who knows how things would have gone if many more people had engaged in a similar way at many more scientific venues? Nonetheless, a tenuous effort at evalua- tion is required for making any recommendations to others.
It appears to me that of the four engagements, two had some success. The engagement with the SNP consortium was as successful as I could have wished. Those who heard the argument appeared to make on-the-spot choices about what constituted "good science" that were, I believe, different than if I had not spoken. Individual speakers seemed to exhibit some uncertainty about their own labeling choices after listening to my presentation. It seems hard to ask for more from a single fifteen-minute speech. The engagement with the HapMap project achieved some visibility for a different way of labeling genetic variation data, and did so in a large, highly visible project, even if that visi- ble alternative was overwritten elsewhere by the older vocabularies we were trying to replace. One might hope that sustained efforts from multiple voices might ultimately be sufficient for normalizing the new ways of labeling.
These two positive results appear to be enormously encouraging for a pro- gram of rhetorical engagement in science, but that optimistic conclusion would overlook the fact that the audiences for these somewhat-successful engage- ments were scientists with little at stake in the racial labels. Both the HapMap and the SNP consortium truly could achieve their initial objectives with differ- ent (non-race-based) vocabularies and assumptions. Race was relatively tan- gential to the core of their research objectives, so they were willing to enact what they saw as good intentions to avoid what they were already convinced were racist ways of speaking. I simply provided a stimulus to breaking with inertia.
In contrast, with audiences who had more at stake--the medical research- ers working with minority communities--there appeared to be no success at all. These individuals defined their practice precisely in terms of what they
The Process of Remaking Race as Genetic 131
understood to be "race. " Although other people working with minority com- munities can shift to culturally based concepts of ethnicity that are more his- torically fluid and more open to social forces as a causative factor, medical personnel are not free to do that because their practice, their medical train- ing, and their science are defined in terms of bodies. They hold to the concept of race because it gives attention to the people about whom they care, and they impose upon race a genetic definition because biology is the only causal force their training permits them to employ.
The implication of this conjuncture is one that is not wholly novel, but perhaps deserves restatement--that scientific projects are neither simply uni- versal nor merely ideological, but task focused. Rhetorical efforts directed at correcting errors in such projects face the greatest difficulty when they inter- fere with central tasks to which a scientific research stream has been put. This task-focus has daunting implications with regard to the linkage between sci- ence and racism. Whenever science engages tasks that might be enhanced by grouping humans, the contemporary version of science may be predisposed to promote racism, because the contemporary version of science can only present groups as fixed physical entities that are genetically based, and such a version of human groups will, at the least, give aid and comfort to racism. This does not mean that the scientific project or goal is, in itself, racist. The task of ameliorating health disparities makes that point clear. It does mean, however, that the inability of science to envision categorical systems for humans that take account of changing socioenvironmental inputs to human biologies, and to treat the impacts of those socioenvironmental inputs on bodies as seriously as it treats genes, makes the current, restricted version of science prone to reproducing support for racism.
Broader Recommendations
In pursuing these rhetorical efforts, my team and I employed many of the classical tools of rhetorical theory, and we especially employed many of the insights of rhetorical studies of science. We focused on definitional processes35 and attended to audience's motives,36 to identities,37 and to subcommunity boundary marking. 38 Most important, we were scrupulously aware of generic constraints. 39 Rhetoricians have an almost anaphylactic reaction to doing experiments or using numbers, but this is the currency of the scientific realm, and to participate in the scientific sphere may require some use of these lan- guages and methods. At the least, using such methods and vocabularies may facilitate a hearing. If I had spent my fifteen minutes at the SNP conference using historical examples or ideological analyses to tell the conferees that their use of the term "race" might have bad effects promoting racism, I do not think it would have had nearly the impact that presenting them experimental results did. They believe experiments, so using experiments is probably the kind of proof that will have the greatest impact. In a broader sense, this is to say that
132 Celeste M. Condit
even in the technical or scientific sphere, one should argue, as Wayne Brock- riede put it, like a lover, not like an enemy. 40
This may be particularly crucial in technical fields because the zeitgeist of scientific argument is less welcoming of open conflict than that of many pub- lic venues. Furthermore, as Douglass Ehninger pointed out, the most authentic form of argument is one where you put your own beliefs at risk. 41 Through- out this engagement, I was never absolutely certain that my own beliefs were right. I always entertained the possibility that to end health disparities required outing "whiteness" and that a genetic diagnosis for "race" might do this. Although the engagements I undertook did not convince me of that, being open to the arguments of scientists like Dr. Burchard led me to understand the dynamics of the use of "race" in medicine in ways I would not have otherwise fathomed.
A second recommendation addresses the question of choice of audiences. I believed that I wanted to persuade the medical researchers themselves not to use constructions that linked genetics to race, but this was the most diffi- cult audience to persuade, and I was not successful at all with them. Perhaps the direct route is not always the best route in science (as it often is not in other realms). A more useful rhetorical strategy may consist of building allies within the scientific establishment from credentialed but less task-involved groups. Such experts are called by reporters to comment on such issues; they lead grant review groups, train students, and hire colleagues. Sometimes one cannot persuade the people on the other side of the boulder to stop pushing, but one can get others to jam up efforts on their side a bit. I am currently engaging in other activities of that sort.
Entering the largely alien scientific sphere to intervene "upstream" of the appearance of scientific products in the public sphere is a daunting endeavor. Understanding the scientific vocabularies and statistics with the level of knowl- edge necessary to be a credible conversant takes a substantial amount of time and effort. However, once these basics of common communication with the target community are absorbed, the rhetorical dynamics resemble strongly those in the public sphere, as rhetoricians of science have previously sug- gested. This means that activists trained in rhetorical theory wanting to shape scientific constructions already have many tools to draw upon. I hope that more rhetoricians will take up this task, as technological and scientifically based factors will inevitably continue to shape our futures.
Notes
1. Wade, "Race Is Seen. "
2. Risch et al. , "Categorization of Humans," 1-12.
3. Gould, Mismeasure of Man; Marks, Human Biodiversity.
4. Hamilton, "Am I Not a Man? "
5. Bix, "Experiences and Voices"; Hasain, Rhetoric of Eugenics. 6. Tucker, "Burt's Separated Twins. "
The Process of Remaking Race as Genetic 133
7. Berger, "Nazi Science"; Frank, "Rhetorical History. "
8. Farrell and Goodnight, "Accidental Rhetoric"; Killingsworth, "From Environmen- tal Rhetoric"; Segal, Health and the Rhetoric.
9. Ceccaeralli, Shaping Science with Rhetoric; Fahnestock, Rhetorical Figures; Gross, Rhetoric of Science.
10. Hedgecoe, "Transforming Genes"; Rosner and Johnson, "Telling Stories"; van der Weele, "Images. "
11. Condit, "How Culture and Science"; Happe, "Rhetoric of Race"; Smart et al. , "Stan- dardization of Race. "
12. Peterson and Horton, "Rooted in the Soil. "
13. Sankar et al. , "Genetic Research. "
14. Condit, "How Culture and Science. "
15. Smedley, Stith, and Nelson, Unequal Treatment. I use capital letters to designate the
noun "White" as a racialized term, but do not capitalize it when it is an adjective or adverb. Failure to capitalize "White" and "Black" when other racial designations are capitalized (African American or Asian American) introduces an asymmetry that can be read as effacing the racial quality of whiteness.
16. Nakayama and Krizek, "Whiteness. "
17. Condit, "'Race' Is Not"; Condit, "How Culture and Science"; Condit, "Race and Genetics. "
18. Bevan et al. , "Informed Lay Preferences"; Condit et al. , "Attitudinal Barriers. " 19. Nguyen, Desta, and Flockhart, "Enhancing Race-Based Prescribing Precision. "
20. Condit et al. , "Attitudinal Barriers. "
21. Bevan et al. , "Informed Lay Preferences. "
22. Burke, Rhetoric of Motives.
23. Lee, Mountain, and Koenig, "Meanings of 'Race. '"
24. Condit et al. , "Role of 'Geneticis. '"
25. This meeting took place November 20-23, 2003, in Chantilly, Virginia.
26. See the HapMap Web site at www. HapMap. org.
27. See http://www. hapmap. org/citinghapmap. html. en for further information.
28. International HapMap Consortium, "Haplotype Map. "
29. This occurs in spite of the fact that the sole graphic presenting separate JPT data
seems to be quite different from the other groupings. Ibid. , Figure 6. 30. Montpetit et al. , "Evaluation. "
31. The latter data is now published. Lynch and Condit, "Genes and Race. " 32. Risch et al. , "Categorization of Humans. "
33. Joralemon and Cox, "Body Values. " Note especially page 32.
34. Segal, Health and the Rhetoric.
35. Lynch, "Making Room"; Taylor, Defining Science.
36. Reeves, "Owning a Virus"; Mitchell, "Sacrifice. "
37. Keller and Longino, Feminism and Science; Happe, "Rhetoric of Race. " 38. Sullivan, "Keeping the Rhetoric"; Taylor, Defining Science.
39. Ceccarelli, Shaping Science; Miller, "Genre as Social Action. "
40. Brockreide, "Arguers as Lovers. " 41. Ehninger, "Argument as Method. "
Works Cited
Berger, Robert L. "Nazi Science--The Dachau Hypothermia Experiments. " New England Journal of Medicine 322, no. 20 (1990): 1435-40.
134 Celeste M. Condit
Bevan, J. L. , J. A. Lynch, T. N. Dubriwny, T. M. Harris, P. J. Achter, A. L. Reeder, and C. M. Condit. "Informed Lay Preferences for Delivery of Racially Varied Pharmacoge- nomics. " Genetics in Medicine (2003): 393-99.
Bix, Amy Sue. "Experiences and Voices of Eugenics Field-Workers: 'Women's Work' in Biology. " Social Studies of Science 27 (1997): 625-68.
Brockriede, Wayne E. "Arguers as Lovers. " Philosophy & Rhetoric 5 (Winter 1972): 1-11. Burke, Kenneth. The Philosophy of Literary Form. Berkeley: University of California Press,
1973.
------. A Rhetoric of Motives. Berkeley: University of California Press, 1969.
Camus, Albert. "The Myth of Sisyphus. " In The Myth of Sisyphus and Other Essays. Trans-
lated by Justin O'Brien, 119-23.
120 Celeste M. Condit
technical sphere, like the public sphere, is at least partially shaped by rhetori- cal factors. 9 It might similarly, therefore, be susceptible to intervention through discursive efforts guided by rhetorical theory.
Even when addressing topics raised by science and technology, however, most rhetoricians have directed their discourse at other rhetoricians or at a presumed reading public inhabiting the public sphere. With regard to the topic of race and genetics, there are, for example, numerous analyses that show the problematic metaphors and categorizations of genetics,10 or reveal the rhe- torical strategies by which the linkage between "race" and "genetics" is con- structed in scientific discourses or reportage on science. 11 While there are some exceptions to the tendency to focus on technical communication as it plays out in the public sphere, the heavy publication slant suggests that it is likely that more activist interventions are also directed at the public sphere rather than involving engagement within the technical (or scientific) sphere. 12 To try to enter the scientists' own conversations, to argue with them within the ven- ues and rubrics of science, is a somewhat different enterprise. To broach the feasibility of the practice and some of the challenges, I recount four different episodes in which I engaged geneticists in scientific venues on the subject of the relationships between "race" and "genetics. " I describe some of the major reasons for the varying levels of success I had in each different situation. I conclude that, if my own experience is at all representative, then the funda- mental variables influencing persuasiveness in the scientific venue are remark- ably similar to those in the public sphere: you persuade people if you talk their language, but it is more difficult to persuade someone the more his or her highly interested worldviews are at stake.
My Coming to Terms with Race and Genetics:
The Human Genetic Variation Consortium
Until I moved to New Orleans, I reacted to race in most of the ways the aver- age white professional American today reacts to race. I wanted to be "liberal" and "tolerant," not a "racist," but these were mere platitudes with no depth or breadth of meaning. When I moved to New Orleans, however, I moved to an area called "the black triangle" because I did not want to commute an hour each day to teach at Tulane University, because I was a teacher who could not afford to live in a "white" area, and because my liberal platitudes told me that there was no reason to be prejudiced and avoid a "black" area of town. Liv- ing there gave me experiential lessons about the meanings of racialization in America, and it also led me to modify my research agenda to focus on race. Periodically I have become overwhelmed and despondent and "stopped work- ing on race. " But race in contemporary America is not something you can ever really walk away from.
Consequently, though I had "stopped working on race" half a dozen years earlier and turned to studying genetics, in 1998 I found myself submitting a
The Process of Remaking Race as Genetic 121
proposal to the National Institutes of Health (NIH) to study how laypeople, especially African Americans and Whites, understood the relationship between race and genetics. When that proposal was eventually funded, I was included in a group called the Human Genetic Variation Consortium. This was a group of researchers working in various ways on social, ethical, or legal issues re- lated to race and genetics. Most notably, the group discussed the "Haplotype Map" project (described below) and advised some of that project's advisers. At the close of the term of my cohort of members, a group of its members published a position paper in the Journal of the American Medical Association on the lack of relevance of genetic research to the amelioration of health dis- parities. 13
The consortium meetings during the first year were devoted to familiariz- ing ourselves with the new trend in genetics toward taking account of race in genetics and health studies. We were told that there were two major forces generating this trend. The first was a need for geneticists to account for popu- lation substructure in their work. Through several expensive and embarrass- ingly errant studies, geneticists had discovered that differences among the geographic ancestry of control and experimental populations could produce spurious findings that "a gene" was related to a particular health condition. 14 The simplest way of correcting that problem appeared to be to provide rela- tively rigorous controls on the boundaries among the "racial" groups that com- posed research populations.
The second concern was more compelling to the humanists and social sci- entists in the consortium. It was the issue of health disparities. There can be no denial that there are clear health disparities between groups that are under- stood as "racial" groups in the United States. Compared to people who self- identify as "White," people who self-identify as Black or African American, American Indian, or Hispanic American die younger and suffer more and worse disease across many common diseases, including cardiovascular dis- eases, diabetes, and several kinds of cancer (the situation is more complex for Asian Americans due to different historical and economic positionings of dif- ferent nationalized subpopulations. )15 Many doctors and researchers, includ- ing minority physicians, have come to believe that a part of these disparities is due to differences in biology, specifically in genes.
This rationale for attending to the possibility that "race" is "genetic" was compelling to many members of the consortium, because it meant that dis- missing biological accounts of race was not a move with positive, or even merely neutral effects, but might be positively harmful to minority groups. For example, ignoring race results in the recruitment of predominantly white pop- ulations to research studies, and it produces medicine that is tailored to white bodies and therefore may be less helpful to the bodies of members of other groups. If there is substantial biological variation among the bodies of ethnic or racial groups, then to ignore that variation is to promote "the invisibility
122 Celeste M. Condit
of whiteness," which is simply the contemporary route to privileging white people. 16 It is to mistake whiteness as the universal version of humanity.
The members of the Human Genetic Variation Consortium exhibited a range of reactions to these presentations. My own reaction was to be deeply troubled and indecisive. On the one hand, I was profoundly skeptical of the idea that race was biological. As a rhetorician, I understood that words made categories, and that the structure of material reality in the world did not sim- ply and neutrally require a particular set of labels. I wrote, and continue to write, rhetorical analyses designed to show the constructedness of race with regard to human genetic variation. 17 On the other hand, I had come to take seriously the fact that race could not be simply ignored, denied, or swept under the carpet. Health disparities were undeniable, and minority researchers and physicians I respected insisted that they saw important differences between average members of different groups. While it was clearly true that the "aver- age" member of a group was an imprecise construct, it might also be true that denying such average differences would merely reinforce a medical system that served white people well and other people not so well. Moreover, I had something of a consistency problem, as my own research required me to explicitly invite people to participate based on something like their "race. " If I were to be able to provide a sufficiently robust input from people other than white southerners, then I had to explicitly invite people into my research proj- ect based on their identity as "African Americans. " If this was a good, even essential, practice for the research I thought I should pursue, then how could I insist that it was not a good thing for people doing other types of research?
I took this profound uncertainty as an initial question for my research process. I decided to trust the community of people who would be most affected by a race-based medical research endeavor, African Americans. So, in conjunction with a fabulous group of graduate students and coresearchers, I designed and conducted focus groups to find out what ordinary people thought of these possibilities. 18 The answer African American participants in this research gave was overwhelmingly opposed to race-based medical ap- proaches. White participants tended to oppose them as well, though not as vehemently. Having learned to distrust my own white-based instincts on issues of race, I determined to trust these people's judgments. The effort to amplify their voice became the basis of my first attempt at engagement with the sci- entific community.
Engagement #1: Publish Data About the Challenges and Costs of Race-Based Medicine in Scientific Journals on Genetic Medicine
My goal was to engage medical geneticists who were being encouraged to par- ticipate in a paradigm of race-based medicine. It seemed to me that the flow of medical science was such that it was better to prevent the formation of an errant consensus by these experts rather than to contest that consensus once
The Process of Remaking Race as Genetic 123
it reached the public sphere, so I determined it best to engage in the process by which the medical consensus on race and genetic medicine was being for- mulated. All I had to offer by way of argument, however, was the fact that a substantial group of laypeople did not like the idea. How could that fact be made of relevance to the medical community?
Medical research in genetics, like all medical research, is largely bereft of attention to many of the issues of how a particular medical regime will be implemented in practice. Medical research is conducted in highly selective locations under near-ideal conditions. When a medical procedure or drug is approved, it is diffused for use under very different conditions. This, I thought, was exactly the problem for a race-based medicine. In an ideal world, where everyone was enlightened, using a social grouping such as "race" as shorthand for mild tendencies for human biology to vary might be acceptable. In the real world, as my participants clearly indicated, it just would not work. Since race-based medicine was being promoted primarily as an economic necessity-- due to the presumed expense of doing individualized genetic testing--showing the economic and other barriers to implementation seemed like a reasonable line of argument. 19
So I and my research team wrote "Attitudinal Barriers to Delivery of Race- Targeted Pharmacogenomics among Informed Lay Persons," and we were suc- cessful in publishing it in Genetics in Medicine. 20 The article and its companion article were written and published within the generic norms of a scientific article. 21 Its sections are "methods," "data," and "discussion. " It uses quanti- tative presentation of findings and employs the objectivating procedure of intercoder reliability assessments. As Kenneth Burke has noted, you persuade a person insofar as you speak his or her language, and because we wanted to be persuasive to an audience of scientists, we presented our research in the language and conventions of science (even though we were mostly a group of rhetoricians). 22
The contents of the article indicate that the effectiveness of a race-based medical approach would have to be assessed based on a variety of costs and challenges not presently included in the arguments in favor of its implemen- tation. These additional costs included the inability of doctors to assign peo- ple to discrete and accurate ancestries, the lack of knowledge of the biology of one's grandparents especially among African Americans, and very high levels of distrust of such medicines by their users, which is likely to lead to low com- pliance with prescriptions as well as a potential backlash against prescribers, pharmaceutical companies, and medical research institutions if ever such a drug turns out to have serious harmful side effects (which seems probable over the long run, especially given the contemporaneous tragedy of Vioxx and related cox-2 inhibitors). In other words, we gave the medical experts scien- tific evidence that race-based medicine was unlikely to be cheap or particu- larly effective.
124 Celeste M. Condit
The impact of this pair of articles was, so far as we can tell, virtually zero. Citation indexes showed only three citations of either article. A wide variety of sources that should have cited it because they were addressing costs of race- based medicine or feasibility did not. My first intervention was apparently a total failure.
Engagement #2: Going Live to Genetics Researchers on the Harmful Impact of "Race as Genetic"
The opportunity for a second intervention presented itself as a matter of both serendipity and foresightful preparation. One of the major arguments that humanists and social scientists use for arguing against geneticization of racial categories is that such geneticization makes race appear absolute and perma- nent. The technical term for this is that it tends to "reify" race. In contrast, a nonreified understanding of existing labels for groups sees these labels as leaky and imprecise and as a temporary product of historical and social confluences. Indeed, the historical evidence is quite clear that racial group labels and racial groupings vary considerably through time and space. 23 In contrast, assigning a scientific basis for a grouping is believed to suggest that such groupings are permanent and absolute, because we tend to think of scientific categories as universal and rigid. Therefore, a scientific grounding for race should produce more "racism" than would an understanding of race as a product of social and historical forces.
Is this presumption correct, however? Having become skittish about the link between genes and racism, I wondered whether we humanists were really right to claim that people who saw health as differentially distributed by race and who saw race as distributed by genes would be more racist than people who did not. If ignoring race was tantamount to universalizing whiteness, perhaps recognizing race was necessary to overcome racism. And if that were true, then perhaps people who thought of health disparities as a simple matter of biological difference would not therefore conclude that everything about racial groups (for example, class distribution or IQ scores) was a matter of genetics and was therefore fixed forever.
I continue to puzzle over this issue. However, one way of addressing the question, and of addressing it in a way that would have credibility with the scientifically based genetics community, is to run an experiment to see if peo- ple exposed to messages linking health, race, and genetics increase the racism level of their attitudes as composed to people who do not hear such messages. So this is what my research team did. The results of the experiment seemed to me shocking and frightening. We exposed people to a very tame message about health disparities, one that simply mentioned genetics in passing as one possible basis for health variation. People who heard this message increased their level of racism an average of almost a full point on a five-point racism scale. 24 We had expected no effect at all, because we had used a very mild
The Process of Remaking Race as Genetic 125
message. We were quite concerned about the ethics of increasing people's lev- els of racism through an experiment, given that we were not sure that the debriefing message we would give them at the end would actually return their levels to preexperimental levels, so we had decided to start below the level where we thought we would create change and inch our way upward to a modest level of effect. The effect of a 20 percent increase that occurred from our first modest level message is unusual in message-design research, so this very large effect convinced us that messages about race, genes, and health most certainly would cause "reification" of attitudes about race and would amplify racism.
So that was the foresight. Serendipitously, around this time, because of my work with the Human Genetic Variation Consortium, I was invited to make a presentation to the sixth International Meeting on Single Nucleotide Poly- morphism (SNP) and Complex Genome Analysis. 25 This is a meeting of a group of researchers who participate in projects designed to track down details of variation in the human genome in order to produce the basis for identifying genes that are associated with common diseases and identifying medicines that might address those genetic variations. I presented the results of the mes- sage experiment in my fifteen-minute presentation close to the beginning of the conference.
The impact of the talk was, I believe, relatively substantial--among those present, for the duration of the conference. The audience was attentive, and later speakers evidenced awkwardness, discomfort, and attempts at circumspec- tion in labeling groups in their presentations. Moreover, the award-winning poster for the conference was one that indicated that apparent clusters of hu- man genetic variation by continents was an artifact of sampling procedures that left large gaps in samples between continental groups rather than using a continuous sampling procedure. I heard the winner being told that "as soon as I heard that talk, I knew you were going to win. "
I might well be exaggerating my own influence here. At the least, my pres- entation was one straw on a growing pile on an invisible camel (though there were no other presentations of this sort at the conference). I have to admit to being heartened and hopeful. It is important, however, to note the audience that evinced this effect. Most of these researchers do work that is well "up- stream" from actual medical applications on live human beings. They might gather DNA from people, but after that they sift and search among single bases, comparing dozens of genomes on computers, looking for variations among millions of base pairs. They are highly mathematical scientists, doing statistical tests and creating statistical tests to assess what might actually count as an association in such large and complex data sets. Their main stake in the "race is genetic" debate is merely the ability to control for population struc- ture. Population structure can be controlled simply through geographic dis- tance. It does not need current racial labels.
126 Celeste M. Condit
This audience also consisted of a very smart and northeastern-biased group of scholars, trained at excellent schools. Not only do these scholars not want to appear to be racist, they really do not want to be racist, and they were trained in an understanding of racism that indicated calling out difference meant racism. They would probably have been completely at a loss if one tried to suggest that one problem of racism is the invisibility of whiteness. Therefore, my presentation did not cost the audience much in terms of their scientific goals, and it fit well with their understanding of what "not being a racist" meant. Moreover, short-term ripples do not necessarily make long-term impacts, as the third intervention was to make clear.
Engagement #3: Relabeling the HapMap
During the second year of my stint on the Human Genetic Variation Consor- tium, some personnel of the National Human Genome Research Institute's Ethical, Legal, and Social Implications (NHGRI/ELSI) unit presented the con- sortium with plans for the "Haplotype Map Project" (or HapMap). In short, the project was designed to gather samples of human DNA from different groups in order to produce a tool for more rapidly and efficiently looking for variations in human genes that are associated with diseases. Various NIH per- sonnel and researchers associated with the project were concerned about the public response to it. A few years earlier a "Human Diversity Project" had gained harsh worldwide negative publicity when it had sought to sample DNA from global cultures that are threatened with extinction. The ELSI researchers wanted the HapMap to be done in an ethically responsible fashion, and other project personnel either wanted that or, at the least, recognized the need to avoid generating another hostile international response. Some members of the Human Genetic Variation Consortium were on the ELSI consulting group for the project, but NIH officials also presented the project to us to get our reaction.
The discussion within the consortium was intense, extended, and challeng- ing. The HapMap is a technically complex project. Understanding it requires careful explanation of detailed scientific information. The consortium included people ranging from an almost knee-jerk antiscience, antigenetics, antirace position, to medical researchers who were opposed to the use of race, to con- flicted humanists, and to social scientists from several disciplines. Although I took notes, I make no claim to be able to represent objectively the process. It is my impression, however, that my role in the conversation was to insist that if the HapMap needed to sample across broad geographies merely in order to assure that a broad range of SNPs (genetic variations) be included, then their populations should not be labeled so as to imply that what was being sam- pled was "race. " In other words, instead of labeling one population "Africans" or "sub-Saharan Africans," they should be labeled as coming from the nar- rowest precise geographic locale--for example, Ibadan, Nigeria. I was quite
The Process of Remaking Race as Genetic 127
insistent that the HapMap should not reify the idea that the genetic variation that needed to be sampled was racial groupings, given the project scientists' claim that they were merely trying to obtain diversity, sampling people from geographical regions that were far from each other. As a rhetorician, I focused intensely on the word choices in the labels, operating from the belief widely shared in rhetorical studies that particular word choices matter.
Other people made similar arguments. Moreover, the conduit we had for making an impact was extremely indirect. We had the ear of one NIH scien- tist and one NIH (ELSI) program director and the ear of those members of the Human Genetic Variation Consortium who would participate directly in the HapMap project. In other words, I have no idea whether my personal argu- ments had any impact at all. Nonetheless, there is some clear evidence of adoption of the labeling practices we were promoting. On the HapMap Web site, the three groups are not labeled as "Africans," "Caucasians," and "Asians," as was a fairly predictable labeling structure given previous conventions in the scientific literature. 26 Instead, the "recommended descriptors" are "Yoruba in Ibadan, Nigeria (abbreviation: YRI), Japanese in Tokyo, Japan (abbreviation: JPT), Han Chinese in Beijing, China (abbreviation: CHB), and CEPH (Utah residents with ancestry from northern and western Europe) (abbreviation: CEU). "27
This novel labeling, and the official Web site's explicit attention to the ra- tionale behind it, suggests that my intervention, or co-occurring interventions by others, had some impact. It was, however, a negligible impact, because in the major article presenting the HapMap project in Nature, almost all the data combines the groups JPT and CHB and pits them against the other two groups. 28 This combining of the groups, presented without any rationale, indicates that the authors are still thinking of and presenting these samples as representing "Asia. "29 Furthermore, in later articles that cite and apply these findings, the HapMap data is relabeled with conventional race labels and is explicitly taken to support the idea of continent-based racial group- ings. 30
This does not demonstrate that the labels researchers choose are irrelevant. It merely indicates that getting scientists in a key locale to adopt a particular (nonracial) label set does not necessarily have very large ripple effects. Absent a clear understanding of the novel labels, those who encounter them merely translate them back into more familiar categories. In addition to simple famili- arity and inertia, a major force behind the maintenance of the old labels arises from those who have a stake in maintaining those labels. It was this group that my fourth intervention engaged.
Engagement #4: The NIH Roundtable on Race
The NIH has relatively recently attended to the ethnic disparities in health and in science in a variety of ways, including structural requirements for
128 Celeste M. Condit
researchers to include diverse populations in their research, outreach and financial support for minority researchers, bureaucratic components devoted to minority issues, calls for research projects that are likely to generate infor- mation of relevance to nondominant sectors, and a variety of conferences and workshops. The NIH Roundtable on Race was convened as one such con- ference. It specifically focused on issues surrounding the implications of genet- ics for race and for race-based medicine. Invitees and presenters included both medical researchers pursuing race-based medical or other projects and social scientists and humanists, most of whom were in some way skeptical of those projects. Minority scholars were included in both groups. Many of the schol- ars from minority groups, but not all, were favorable to race-based research and practice.
At this conference, I presented the same data I had presented at the SNP conference, as well as descriptive data characterizing the mass media's cover- age of race-based genetics research and practice. 31 I would have to call the re- sponse of the audience cold, if not hostile. An older female minority scholar told me to "think about the parable of the sun and the wind, trying to get the person to take off their coat. You are the wind. " In the parable, the wind fails because it only makes the person pull the coat tighter. The sun succeeds by making the person so warm they take off their coat. In light of the experi- mental results I had gotten, I was at a loss to figure out how to make people be less racist by applauding their assumption that race was genetic, so in spite of this chastisement I more or less continued a sustained argument against that view in the interactive discussion periods. Nonetheless, I learned a good deal from the people there about their motives, their sincerity, and the source of what seem to me to be their rhetorical blind spots.
I take as an example a researcher named Esteban Gonza? lez Burchard, a well-published scholar as well as a medical researcher dedicated to improving the health of Puerto Ricans. He has special interest, even passion, for address- ing the high rates of serious asthma in the Puerto Rican population. He and I had civil, mutually respectful interchanges, but from my perspective Dr. Burchard is fundamentally wrong. At this conference, the warrant he articu- lated for maintaining race-based approaches and for insisting that race had a genetic basis was the fact that Puerto Ricans have a high asthma rate and a low response to the available medications (much lower than other groups). However, "Puerto Ricans" is not a race, even according to the standards of Dr. Burchard's own published research, and Puerto Ricans are not typical of "Latin Americans" or "Native Americans" or "African Americans" (or any other of the groups geneticists describe as "racial") with regard to their asthma suscep- tibilities. 32
Nonetheless, as we talked, I gained additional depth of understanding about the ways in which the increasingly prominent version of antiracism that attacks "color-blind" approaches and the "invisibility of whiteness" applies in
The Process of Remaking Race as Genetic 129
health disparities. Dr. Burchard's perspective worked within this vision of anti- racism. Additionally, however, my perception became that Dr. Burchard was unwilling to entertain environmental or cultural explanations for evident dif- ferences because such accounts in some sense place blame, or at least respon- sibility, upon Puerto Ricans for their health problem. In contrast, a genetic explanation places the cause of the health disparity beyond the personal, social, or even cultural control of the population at risk.
There are, of course, other ways to avoid "blaming the victim. " The most common is to attribute the cause to the health care system or to environmen- tal factors that are the result of economic forces and structural discrimination. But Dr. Burchard, like many other minority physicians, appears to be in a posi- tion that makes it difficult to blame the medical system for health disparities. Minority medical personnel are part of that system, and they are working very, very hard to serve their populations through it. They are disinclined to blame the systems that do indeed provide extensive positive services to their popu- lations, and certainly they should not blame themselves, even though they must serve at least in part as agents of that system, its faults included.
Their acceptance of the system's perspective is not merely due to the social status it provides them. It also seems to result from the adoption of the sub- ject position of "physician," which, as many scholars have suggested, entails seeing patients in certain lights--as bodies, not as products of social sys- tems. 33 Indeed, the medical orientation that all physicians are trained into-- even minority physicians--runs to an even more deep set of biases in this regard than has generally been recognized. 34 Physicians are trained into the scientific worldview. The contemporary version of this worldview understands human beings almost exclusively through animal models. No other animal is subject to the variety and complexity of social, economic, and symbolic forces to which human beings are subjected. Thus the animal models provide no reason to factor social structures into health outcomes.
Dr. Burchard has heard extensive, detailed critiques of "race" (at this round- table as well as at the "Defining Race" NIH In-House Conference in 2006), but these have moved him only to restate more adamantly what he already "knew," not to step back and reconsider the ways in which he was using the term "race. " I assert that this sense of certainty arises from at least three causes--his passionate commitment to helping Puerto Ricans, his highly con- temporary view that helping Puerto Ricans entails a stance that highlights difference rather than ignores it, and an understanding of what counts in sci- ence as a source of causation (that is, genes, not social structure).
I have used Dr. Burchard as an example here not to belittle him, for I have the greatest respect for his intelligence, his hard work, and his dedication to the health of Puerto Ricans.
If there is error here, it is the paradigm that is errant, not the individual, and the relevant paradigm is the paradigm of sci- ence as currently constituted and as passed through the medical school filter.
130 Celeste M. Condit
I have come to believe that there is nothing I could write or say that would convince someone who cared passionately about the health of a minority group, and who accepted the current discursive framework of medical research, that he or she should abjure the utility of "race" in the study of medical genetics.
Scorecard
I have undertaken four major engagements with medical geneticists in an effort to get them to eschew the claim that race is genetic. I did so with the hope that by engaging rhetorically in the scientific realm, I could help stop the formation of a scientific consensus that would then be translated into the public realm as the indisputable scientific fact that race is genetic. I harbored no illusions that my efforts, on their own, would be sufficient. The scientific realm is large, with many players, and my voice is just one, able to be heard in few spaces. It is therefore difficult to evaluate these efforts. Who knows how things would have gone if many more people had engaged in a similar way at many more scientific venues? Nonetheless, a tenuous effort at evalua- tion is required for making any recommendations to others.
It appears to me that of the four engagements, two had some success. The engagement with the SNP consortium was as successful as I could have wished. Those who heard the argument appeared to make on-the-spot choices about what constituted "good science" that were, I believe, different than if I had not spoken. Individual speakers seemed to exhibit some uncertainty about their own labeling choices after listening to my presentation. It seems hard to ask for more from a single fifteen-minute speech. The engagement with the HapMap project achieved some visibility for a different way of labeling genetic variation data, and did so in a large, highly visible project, even if that visi- ble alternative was overwritten elsewhere by the older vocabularies we were trying to replace. One might hope that sustained efforts from multiple voices might ultimately be sufficient for normalizing the new ways of labeling.
These two positive results appear to be enormously encouraging for a pro- gram of rhetorical engagement in science, but that optimistic conclusion would overlook the fact that the audiences for these somewhat-successful engage- ments were scientists with little at stake in the racial labels. Both the HapMap and the SNP consortium truly could achieve their initial objectives with differ- ent (non-race-based) vocabularies and assumptions. Race was relatively tan- gential to the core of their research objectives, so they were willing to enact what they saw as good intentions to avoid what they were already convinced were racist ways of speaking. I simply provided a stimulus to breaking with inertia.
In contrast, with audiences who had more at stake--the medical research- ers working with minority communities--there appeared to be no success at all. These individuals defined their practice precisely in terms of what they
The Process of Remaking Race as Genetic 131
understood to be "race. " Although other people working with minority com- munities can shift to culturally based concepts of ethnicity that are more his- torically fluid and more open to social forces as a causative factor, medical personnel are not free to do that because their practice, their medical train- ing, and their science are defined in terms of bodies. They hold to the concept of race because it gives attention to the people about whom they care, and they impose upon race a genetic definition because biology is the only causal force their training permits them to employ.
The implication of this conjuncture is one that is not wholly novel, but perhaps deserves restatement--that scientific projects are neither simply uni- versal nor merely ideological, but task focused. Rhetorical efforts directed at correcting errors in such projects face the greatest difficulty when they inter- fere with central tasks to which a scientific research stream has been put. This task-focus has daunting implications with regard to the linkage between sci- ence and racism. Whenever science engages tasks that might be enhanced by grouping humans, the contemporary version of science may be predisposed to promote racism, because the contemporary version of science can only present groups as fixed physical entities that are genetically based, and such a version of human groups will, at the least, give aid and comfort to racism. This does not mean that the scientific project or goal is, in itself, racist. The task of ameliorating health disparities makes that point clear. It does mean, however, that the inability of science to envision categorical systems for humans that take account of changing socioenvironmental inputs to human biologies, and to treat the impacts of those socioenvironmental inputs on bodies as seriously as it treats genes, makes the current, restricted version of science prone to reproducing support for racism.
Broader Recommendations
In pursuing these rhetorical efforts, my team and I employed many of the classical tools of rhetorical theory, and we especially employed many of the insights of rhetorical studies of science. We focused on definitional processes35 and attended to audience's motives,36 to identities,37 and to subcommunity boundary marking. 38 Most important, we were scrupulously aware of generic constraints. 39 Rhetoricians have an almost anaphylactic reaction to doing experiments or using numbers, but this is the currency of the scientific realm, and to participate in the scientific sphere may require some use of these lan- guages and methods. At the least, using such methods and vocabularies may facilitate a hearing. If I had spent my fifteen minutes at the SNP conference using historical examples or ideological analyses to tell the conferees that their use of the term "race" might have bad effects promoting racism, I do not think it would have had nearly the impact that presenting them experimental results did. They believe experiments, so using experiments is probably the kind of proof that will have the greatest impact. In a broader sense, this is to say that
132 Celeste M. Condit
even in the technical or scientific sphere, one should argue, as Wayne Brock- riede put it, like a lover, not like an enemy. 40
This may be particularly crucial in technical fields because the zeitgeist of scientific argument is less welcoming of open conflict than that of many pub- lic venues. Furthermore, as Douglass Ehninger pointed out, the most authentic form of argument is one where you put your own beliefs at risk. 41 Through- out this engagement, I was never absolutely certain that my own beliefs were right. I always entertained the possibility that to end health disparities required outing "whiteness" and that a genetic diagnosis for "race" might do this. Although the engagements I undertook did not convince me of that, being open to the arguments of scientists like Dr. Burchard led me to understand the dynamics of the use of "race" in medicine in ways I would not have otherwise fathomed.
A second recommendation addresses the question of choice of audiences. I believed that I wanted to persuade the medical researchers themselves not to use constructions that linked genetics to race, but this was the most diffi- cult audience to persuade, and I was not successful at all with them. Perhaps the direct route is not always the best route in science (as it often is not in other realms). A more useful rhetorical strategy may consist of building allies within the scientific establishment from credentialed but less task-involved groups. Such experts are called by reporters to comment on such issues; they lead grant review groups, train students, and hire colleagues. Sometimes one cannot persuade the people on the other side of the boulder to stop pushing, but one can get others to jam up efforts on their side a bit. I am currently engaging in other activities of that sort.
Entering the largely alien scientific sphere to intervene "upstream" of the appearance of scientific products in the public sphere is a daunting endeavor. Understanding the scientific vocabularies and statistics with the level of knowl- edge necessary to be a credible conversant takes a substantial amount of time and effort. However, once these basics of common communication with the target community are absorbed, the rhetorical dynamics resemble strongly those in the public sphere, as rhetoricians of science have previously sug- gested. This means that activists trained in rhetorical theory wanting to shape scientific constructions already have many tools to draw upon. I hope that more rhetoricians will take up this task, as technological and scientifically based factors will inevitably continue to shape our futures.
Notes
1. Wade, "Race Is Seen. "
2. Risch et al. , "Categorization of Humans," 1-12.
3. Gould, Mismeasure of Man; Marks, Human Biodiversity.
4. Hamilton, "Am I Not a Man? "
5. Bix, "Experiences and Voices"; Hasain, Rhetoric of Eugenics. 6. Tucker, "Burt's Separated Twins. "
The Process of Remaking Race as Genetic 133
7. Berger, "Nazi Science"; Frank, "Rhetorical History. "
8. Farrell and Goodnight, "Accidental Rhetoric"; Killingsworth, "From Environmen- tal Rhetoric"; Segal, Health and the Rhetoric.
9. Ceccaeralli, Shaping Science with Rhetoric; Fahnestock, Rhetorical Figures; Gross, Rhetoric of Science.
10. Hedgecoe, "Transforming Genes"; Rosner and Johnson, "Telling Stories"; van der Weele, "Images. "
11. Condit, "How Culture and Science"; Happe, "Rhetoric of Race"; Smart et al. , "Stan- dardization of Race. "
12. Peterson and Horton, "Rooted in the Soil. "
13. Sankar et al. , "Genetic Research. "
14. Condit, "How Culture and Science. "
15. Smedley, Stith, and Nelson, Unequal Treatment. I use capital letters to designate the
noun "White" as a racialized term, but do not capitalize it when it is an adjective or adverb. Failure to capitalize "White" and "Black" when other racial designations are capitalized (African American or Asian American) introduces an asymmetry that can be read as effacing the racial quality of whiteness.
16. Nakayama and Krizek, "Whiteness. "
17. Condit, "'Race' Is Not"; Condit, "How Culture and Science"; Condit, "Race and Genetics. "
18. Bevan et al. , "Informed Lay Preferences"; Condit et al. , "Attitudinal Barriers. " 19. Nguyen, Desta, and Flockhart, "Enhancing Race-Based Prescribing Precision. "
20. Condit et al. , "Attitudinal Barriers. "
21. Bevan et al. , "Informed Lay Preferences. "
22. Burke, Rhetoric of Motives.
23. Lee, Mountain, and Koenig, "Meanings of 'Race. '"
24. Condit et al. , "Role of 'Geneticis. '"
25. This meeting took place November 20-23, 2003, in Chantilly, Virginia.
26. See the HapMap Web site at www. HapMap. org.
27. See http://www. hapmap. org/citinghapmap. html. en for further information.
28. International HapMap Consortium, "Haplotype Map. "
29. This occurs in spite of the fact that the sole graphic presenting separate JPT data
seems to be quite different from the other groupings. Ibid. , Figure 6. 30. Montpetit et al. , "Evaluation. "
31. The latter data is now published. Lynch and Condit, "Genes and Race. " 32. Risch et al. , "Categorization of Humans. "
33. Joralemon and Cox, "Body Values. " Note especially page 32.
34. Segal, Health and the Rhetoric.
35. Lynch, "Making Room"; Taylor, Defining Science.
36. Reeves, "Owning a Virus"; Mitchell, "Sacrifice. "
37. Keller and Longino, Feminism and Science; Happe, "Rhetoric of Race. " 38. Sullivan, "Keeping the Rhetoric"; Taylor, Defining Science.
39. Ceccarelli, Shaping Science; Miller, "Genre as Social Action. "
40. Brockreide, "Arguers as Lovers. " 41. Ehninger, "Argument as Method. "
Works Cited
Berger, Robert L. "Nazi Science--The Dachau Hypothermia Experiments. " New England Journal of Medicine 322, no. 20 (1990): 1435-40.
134 Celeste M. Condit
Bevan, J. L. , J. A. Lynch, T. N. Dubriwny, T. M. Harris, P. J. Achter, A. L. Reeder, and C. M. Condit. "Informed Lay Preferences for Delivery of Racially Varied Pharmacoge- nomics. " Genetics in Medicine (2003): 393-99.
Bix, Amy Sue. "Experiences and Voices of Eugenics Field-Workers: 'Women's Work' in Biology. " Social Studies of Science 27 (1997): 625-68.
Brockriede, Wayne E. "Arguers as Lovers. " Philosophy & Rhetoric 5 (Winter 1972): 1-11. Burke, Kenneth. The Philosophy of Literary Form. Berkeley: University of California Press,
1973.
------. A Rhetoric of Motives. Berkeley: University of California Press, 1969.
Camus, Albert. "The Myth of Sisyphus. " In The Myth of Sisyphus and Other Essays. Trans-
lated by Justin O'Brien, 119-23.
