Anthropology, Genetic Diversity, and Ethics 
A workshop at the Center for Twentieth Century Studies 
University of Wisconsin-Milwaukee  
Morris Foster
[Participant Information]

Before I begin, I should say that much of what I'm going to be talking about has been, the ideas have been developed in collaboration with Richard Sharp, from the National Institute of Environmental Health Sciences, who's here, and others have been, as a result of the collaboration with William Freeman, who is the director of research at the Indian Health Services. [tape ends, resumes]  The problem with this procedure is that everyone in this room is probably going to agree or disagree with different things, so everyone is going to end up disagreeing with me and agreeing with Rich, and so... 

I wanted to start by trying to differentiate more between kinds of, different kinds of genetic research, and different kinds of risks, or at least collective risks.  Because I think that in much of our discussion, in much of the literature, one of the problems is that we don't differentiate; we simply talk about genetic research in general, and risk in general.  We see, or I see, two different kinds of genetic research in general.  One is research into disease susceptibility, and that certainly is the kind of,  in NIH and the NIEHS, and I focus on.  The other kind of genetic research is into population history, and that's, we had yesterday the presentation on ancient DNA.  These have different objects of research: disease susceptibility research is researching the disease, that is the ultimate goal of that research; whereas population history research makes the population itself the object of the research.  And I think that this then means that different protections are possible, or not possible, depending on which of those kinds of research you're talking about.  Disease susceptibility research lends itself much more to population anonymity, whereas population history research, by its very nature, cannot be anonymous.  Disease susceptibility research usually, although not always, means that a smaller amount of genetic information is being focused on.  At least, that's the goal of the researchers, to narrow down the genetic information to that which is causing or leading to susceptibility.  Population history research, however, quite often deals with very, very broad swathes of genetic information, and those articles put out much more, and thus entail much more risk. 

As two different kinds of risks, or at least different kinds of collective risks, the ones that we know best are those that are extrernal to the community or the population, and that are usually or that are somewhat general or common across populations.  So the literature refers to employment and insurance discrimination; it refers to broad kinds of stigmatization.  But it often neglects, then, population-specific risks, because those are population-specific, and we don't, outsiders in particular, are not always very good at anticipating them.  We talked about some of these yesterday, in talking about the risks that population history research might have for indigenous land claims, for claims under NAGPRA, even for maintaining their native status, at least in respect to larger politico-economic formations. 

There are also, however, intra-community risks, risks that are very culturally specific, and that result not so much from outsiders stigmatizing or discriminating against members of the population, but from, potentially, members of the population having adverse relations with one another as a result of doing research.  And this comes, I think, not so much from the publication of results as potentially from the recruitment of participants.  So just doing the research, just going in and recruiting participants, is potentially risky, and I think mainly in an intra-community rather than an external manner.  It could disrupt social equilibrium; it could cause stigmatization potentially to those who participate, as opposed to those who don't participate; or, as Hank suggested, where you have a strong group consent in favor of research, those who do not participate potentially could be stigmatized by refusing to participate.  We don't usually think about these intra-community risks, and I think that we probably should. 

Other kinds of intra-community risks are culturally specific and have to do with things like genetic findings contradicting or conflicting with the community's sense of its own history, or its own historical narrative.  And this is often very difficult for researchers to credit.  The typical attitude of researchers is, "Well, you know, the scientific results are objective, and whether they agree with it or not, that's their problem," and a very good example of this which always comes up at meetings like this, and which came up yesterday, is, for Native Americans, Beringia, the land bridge.  Every meeting like this I have been to, every community discussion I have had, Beringia is brought up.  And it took me a while to figure out, this is important to these people; this is not just a political piece of rhetoric.  This is a real concern.  And I have, Rich and I have written some papers and submitted them to journals, and it's very hard to get academic reviewers of these papers to get the idea that because members of these communities perceive those kinds of conclusions as potential harms, that that is something that should be taken into consideration in the usual human subjects review, in the usual analysis of risks and benefits. 

So I want to distinguish, then, between population history research and disease susceptibility research; between external risks and intra-community risks; between the risks from publication of findings and simply the risks from recruitment of participants.  Those are all areas that I think we conflate, that we don't differentiate among, and I think we should.  Another area of conflation is between single use of samples and multiple use.  Again, the standard in genetics is to obtain samples and then use them ad infinitum, particularly when they're cell lines.  And that of course means that the possibilities for risk, particularly for intra-community risks, are just open-ended, and there's no way to reasonably evaluate risk, at least at the point of collection, when you have an open-ended possibility of use.  Single use, however, where you collect samples for a particular research question that you can talk about both to the community and to individual participants, and at the end of that, you then dispose of the samples in some culturally appropriate way, whatever is negotiated with the community.  That can much more precisely specify the risks.  It also can much more precisely specify the benefits of such studies. 

Benefits is something that I haven't talked about; I've talked about mainly risks, but benefits also can be both external and intra-community, and also are different for population history research versus disease susceptibility research.  Most of the benefits from population history research accrue to the researchers and their disciplines, although Dennis' example yesterday, I think, was a good one of a situation where some population history research actually was perceived, I believe, as beneficial, or at least as neutral, by members of the communities he was talking about.  I think that's a great example, I hope you publish something describing that process, not just the research, but the process too.  But for the most part, the risks of population history research far outweigh the benefits, from the perspective of the community; it's not from the perspective of the researchers necessarily.  Disease susceptibility research, on the other hand, at least has the promise of potential health benefits, although often very far down the line, and that at least is a better benefit to weigh.  But another evidence, another aspect of conflation in genetic research is that people who do population history research often suggest that disease benefits, that the health benefits, be taken into account against the risks of population history research.  They conflate, then, population history research with disease susceptibility research, and Jonathan Marks has pointed out in a letter that we have in the packet, how that happens, and why, at least in the case of the Divesity Project, the kinds of phenotypic information they are going to collect, or won't collect, make any disease or health benefits very unlikely.  I think this conflation of kinds of research, kinds of risks, kinds of benefits, has contributed greatly to the confusion in much of these ongoing discussions. 

Community consent or veto has become sort of the label that we have placed on discussions about how to work with diverse communities, and I think that's unfortunate, because a consent or veto is an outcome, it's not a process.  It's a goal, it's the goal of the researchers, but it doesn't go to the more fundamental question.  And the more fundamental question should really be, I think, identifying risks and ways to minimize those risks.  When you talk about consent or veto, when we talk about contacting communities, we often talk about it in a way that implies that we already know what the risks are.  And I think that's because bioethicists and geneticists generally think mainly of external risks that are common to all populations.  And what I've discovered, in working now with three different Native American populations, to do actual genetic studies, is that we don't know the risks going in, particularly not those that are of most concern to members of those communities, and those are the intra-community, culturally specific risks, of much more concern than insurance or employment discrimination.  So I think that we should think, talk about community review rather than community consent or veto.  And I think that the premise should be discovering what a community, or members of a community, perceive as risks, rather than trying to get some outcome of a consent or veto. 

If, then, we do a true risk-benefit analysis of the genetic research, once we have discovered the risks that are relevant to a particular community, most projects that communities perceive as risky would not be approved by IRBs anyway.  So you wouldn't even need to get to a point of having a community veto a project.  However, most, particularly most population history research projects, are not seriously evaluated in terms of risks and benefits; or if they are, they're treated as minimally risky.  If they were really subjected to the same stringent, or more stringent, risk-benefit analysis as are most disease susceptibility studies, I think we would see many fewer population history studies done.  Obviously, my conclusion is that in general, population history genetic research is more risky to communities than disease susceptibility research. 

Finally, the issue of individual versus community consent, or individual versus community rights.  Some of the literature has portrayed this as a sort of inevitable conflict between individual autonomy and community consensus for a consent.  Again, I think that's a confusion, a conflation, that is [unfortunate?].  This drives more often from a Western definition of individual autonomy, compared to a non-Western definition of community consensus or consent, where we are comparing the Western standard of individual informed consent with a community's own standard of collective decision-making.  Individual autonomy within a single community, I think, is always based on the collective frameworks and social structures of that community.  Thus, not only are those collective frameworks logically prior to individual autonomy, but individual autonomy, individual choice, derives from the collective consensus, the collective structures.  I think that means that in most cases, where you apply a community-specific notion of individual autonomy, you won't have a conflict between individual choice and community choice.  It's only when we create this cross-cultural situation of trying to apply a Western bioethical notion of individual autonomy that we get conflicts and contradictions.  Now that's not easily resolved, because as Hank pointed out, we are obligated to follow, legally and ethically, a Western bioethical notion of individual autonomy.  There's no way, at least legally, that we can get out of that.  Although, and then this goes into a debate you may have seen in the New England Journal of Medicine last year, I think last year or the year before, about AIDs trials or AIDS research in Africa, where American researchers did not use Western bioethical standards.  So that's a very difficult question.  But if you think of it just within the context of the communities being studied, it's not a conflict.  It's a conflict for the researcher, and for the researcher's institutional background, but not for the community members, and we shouldn't blow it up as such. 

I think that if we differentiate many of these things that I've talked about, we can actually resolve some of the seemingly intractable and enduring problems that genetic research has raised in working with socially identifiable populations.  I think that's probably the only way to move the debate forward; otherwise, we're just going to be having these meetings, where we come and disagree again and again.  Thank you.