In studies involving human subjects who are potentially vulnerable in some way, do you see a difference in how researchers of different disciplines—say anthropologists versus geneticists— approach their ethical responsibilities?
The interesting thing about sociology, anthropology, and the humanities and social sciences in general, is that the history of their field is a core part of their canon. I don’t think that’s the case as often in the life sciences; they don’t get that history of the mistakes and the failures in their field.
I do think that makes a difference in terms of being a little bit more critical about the hypotheses you put forward. Not to say that there aren’t problems in biological anthropology—there are—but I think that in general, anthropologists are aware that when they’re dealing with race as a scientific object, they need to be more careful with the language that they use and how they’re linking race and genetics. They’re aware that in the history of social science, the study of race in the early 20th century and late 19th century was aimed at justifying race hierarchies. Population geneticists in general don’t have that history as part of their field.
I think it’s very important to bring the history of genetics and race and the ethics of research more into the center of training for population geneticists, and to make sure that as genetics comes to the fore in biological anthropology that the lessons of cultural anthropology and bioethics don’t get lost.
Right now, ethics tends to be an addon. This is how you get what happened at Berkeley with the freshman DNA testing. Those of us in the social science faculty who work on genetics issues at Berkeley tried to address the ethical concerns of that project. It was like talking to a stone wall, trying to get the scientists involved in the freshman DNA project to understand that no, you don’t just add an ethics panel at the end, after you’ve designed the whole program. Ethics don’t come at the end; they come at the beginning, when you’re conceptualizing. It’s an inherent part of doing science right, not something you add on at the end.
Of course, everybody—whether you’re coming out of a genetics department or an anthropology department—has to go through the institutional review board (IRB) at the university, so there’s a baseline. But what IRBs require is a bare minimum of the standards that you have to meet to conduct ethical research. IRB approval doesn’t constitute a thorough process.
Do you think that baseline can be made higher, or is it a general rule that IRBs only require the bare minimum?
IRBs vary from university to university, and some are much stricter than others. For example, the Arizona State University IRB is, after the Havasupai lawsuit, incredibly strict where tribes are concerned. If you’re going to do research with native populations, whether it’s biological research or even social science research, you have to get approval from the tribal council before the university will even look at your protocol. On the other hand, I’m doing a project at Berkeley where I’m interviewing both genetic scientists and tribal government people, and Berkeley didn’t look twice at my interview with indigenous people. I asked if they require some sort of documentation that I got approval from the tribe, and they said, “No, no, no, that’s not a problem.” So there are differences between IRBs as well as between disciplines.
What do you suppose accounts for these differences in the strictness of IRBs?
I’m not an expert on IRBs, but I can speak from personal experience—I have worked at both Arizona State and Berkeley, so I have seen the huge differences in IRBs. In short, the difference is that ASU has been sued. Before the Havasupai suit, ASU was lax as well.
I was at ASU in 2006 and 2007. As a social scientist, I was interviewing a range of people—native people, scientists, regulators—and the IRB was very strict about allowing me to talk to tribes. I had interviewees at five or six tribes, which meant I would have had to go through each one of those tribes to get approval for those interview questions. So, in order to get approval for my science piece, I backed out of the Native American community member questions.
This was also really interesting: I study the culture and politics of genetic science, and I think they should have been more strict and careful about my research questions for scientists. In my work, scientists are potentially vulnerable subjects. Now, I don’t actually think they are very vulnerable—I think they actually have a lot more cultural authority than I do in the broader world—but I’m a potential critic. While the native populations were seen as potentially vulnerable subjects, it didn’t seem to have crossed the IRB’s minds that scientists could be potentially vulnerable subjects, too.
It was the opposite at Berkeley, actually: they were much, much more concerned about my questions for scientists and protecting their confidentiality, and they seemed not at all concerned about my questions for indigenous people, at least from my perspective.
It seems that a researcher who belongs to the group being studied would be more concerned with their subjects’ values; but I wonder if that is not as obvious an outcome as it seems, since the researcher’s training also comes into play.
I think that’s right, it’s not just a natural outcome. Just because someone is, say, Navajo, and they’re researching the Navajo people, they would not necessarily do things very differently. If you have a PhD in genetics, you’ve been trained a certain way; you’ve been trained in a discipline that’s not accustomed to thinking about this kind of knowledge the way that Navajo people might.
For my next research project, I’m interviewing indigenous geneticists and their collaborators—people who are actually committed to working in indigenous communities over the long term. That seems to be the key: you need to have an understanding that in these communities, you cannot just go in, get your sample, and never come back. If you want to work with native communities, because of the kinds of suspicions they have of this type of science and the particular historical relationships they’ve had with biomedical research, you have to make a commitment to be in it for the long haul. With increasing development of tribal research review boards, tribes are quickly getting very savvy to the fact that they have the right to review research, the right to reject certain research projects, and they have the right to review publications before they go out to make sure their confidentiality is maintained.
I think that the researchers working in Indian country are increasingly committed to being there over the long term. Those relationships take a long time to build—it’s not easy to work with tribes at all, it’s very difficult. Or you see people who have just decided they don’t want to work with tribes, because they don’t want to have to go through a tribal research review board, they don’t want to let a tribal council or a tribal IRB have a say over whether they can publish something or not. I think that’s a good thing. I would rather see more researchers who are committed to working with native people over the course of their career and really spending the time to do it right and build those relationships; and the rest of them who don’t want to do that, that’s fine. Go do something else!
Whether a researcher is native or not, what matters is their commitment to being involved over the long haul. I think the chances are that a native person who wants to do that work in their own community is going to be more committed to being there long term, but I also know other genetic scientists who are really committed to working with native people over the long term and doing collaborative research.
We can certainly think of researchers being eager to define a group of people in biological terms. I wonder if there are groups who are just as eager to be defined, especially in terms of ancestry. It seems like it might be an American phenomenon.
It’s a very American phenomenon. You see that in the way that companies here all use the same four or five standard race categories that don’t have the same salience in other parts of the world. You do have a couple of companies in Canada and a half dozen in the UK and ancestry testing is popular there. But really the majority of people who do this stuff are Americans. I said this in a meeting where there were a bunch of genetic ancestry testing company guys, how tilted toward the U.S. their race categories were and one company founder said, “That’s not true! People globally are interested in this!” But it is not in fact a global phenomenon because whatever race categories you use, their relevance is specific to particular countries. The British companies, for example, use different ancestry categories— usually a greater number—than the four standard race categories that predominate in the American companies’ tests.
Deborah Bolnick wrote a very interesting article—she’s a biological anthropologist I work with in Texas—on the structure program, one of the main software programs used in AIMs (Ancestry Informative Markers) analyses which a lot of the ancestry companies do. It’s AIMs that they’re analyzing when they give you the results telling the different percentages of your ancestral background. She found that the program will basically break those markers down into whatever number of racial groups you put into the program. If you put in the standard five that make sense to us in the United States, it will spit back an analysis slotting the markers into one of those five racial categories; but if you put in twenty racial categories, it will break down markers at that level. There are some interesting ways in which our social ideas of race shape our technologies. It’s not as easy as our technologies just confirming what we already thought existed.