Many years ago, I was involved in judging an interdisciplinary grant competition. At one point, there was an intense discussion about a proposal where part of the research involved ethnographic research that concerned illegal activity in a developing country. We were all convinced of the researcher’s skill and sensitivity and the topic itself was unquestionably important. We were also convinced that it was plausible and that the researcher could handle immediate issues of safety for the researcher and the people being studied. The disagreement was about whether the subjects could ever give “informed consent” to being studied in a project that might ultimately identify enough about how they conducted their activities to put them at risk no matter how carefully the researcher disguised the identities of the informants.
I had to acknowledge that there was potential risk. When I teach Ellen Hellman’s classic sociological study of African urban life, Rooiyard, I point out to the students that she learned (and disclosed) enough about how women carried out illegal brewing to potentially help authorities disrupt those activities, which is one of the reasons (though surely not the only one) for the degree of suspicion that Hellman herself says she was regarded with.
But I thought this shouldn’t be an issue for the group because I believed (and still believe) that the men being studied could make up their own minds about whether to participate and about the risks of disclosure. Several of the anthropologists on the panel disagreed strongly: they felt that there was no circumstance under which these non-Western men in this impoverished society could accurately assess the dangers of speaking further with this researcher (who already knew the men and had done work with them on other aspects of their social and cultural lives). The disparities in power and knowledge, they felt, made something like “informed consent” impossible. Quite explicitly, my colleagues were saying that even if the men in the study felt like it was ok to be studied, they were wrong.
Now this was long enough ago that on many campuses, Institutional Review Boards were only just getting around to asserting their authority over qualitative and humanistic research, so in many ways our committee was providing that kind of oversight in the absence of it existing on individual campuses. Over multiple years of participating, I only saw this kind of question come up three or four times, and this was the most “IRB-like” of all these conversations.
I was alarmed then and have remained alarmed at the potential for unintended consequences from this perspective. Much as we might like to blame those consequences on bureaucratic overreach or administrative managerialism, which today often functions as all-purpose get-out-of-jail-free card for faculty, the story at least starts with wholly good intentions and a generative critique of social power.
From a great many directions, academics began to understand about forty years ago that asymmetries of power and wealth didn’t simply disappear once someone said, “Hey, I’m just doing some research”. There were a great many critical differences between an ethnographic conversation between an American professor and an African villager on one hand and a police interrogation room on the other, but those differences didn’t mean that the former situation was a frictionless meeting between totally equal people who just decided to have a nice conversation about a topic of mutual interest.
The problem with proceeding from a more self-aware, self-reflexive sense of how power pervades all social relations and interactions to a sense that everyone with less power must be protected from everyone with more power is that this very rapidly becomes a form of racism or discrimination vastly more objectionable than the harm it alleges to prevent. What it leads to is a categorical assertion that entire groups of people are systematically less able to understand and assess their self-interest, less able to understand the consequences of their actions, less able to be trusted with their own agency as human beings. The difference between this view and the imperial and racist version of colonial subjects is small to nonexistent. Yes, there may be contexts like prisons or the aforementioned interrogation room where it takes specific attention to protect and recognize moments of real consent and communication, but it is important that we see those contexts as highly specific and bounded. There are moments where it is strategically, ethically, and even empirically important to defend universals, and this is one of them. Subjectivity has difference, but the rights and perogatives of modern personhood should be assumed to apply to everyone.
A good researcher, in my experience, knows when something’s been said in a conversation that it’s best not to translate into scholarship. Much as a good colleague knows when to keep a confidence that they weren’t directly asked to keep. We’re all sitting on things that were said to us in trust, sometimes by people who were trying to impress us or worried about what we might think, that we never use and often consciously try to forget that we heard. The problem occurs when this kind of sensitive, quintessentially situational judgment call gets translated into a rule, a committee, a structure, a dictum because we’re afraid of, and occasionally encounter, a bad researcher (or a good one who makes a bad judgment call).
I accepted my colleagues’ call in that long-ago conversation though I thought and still think they were wrong, because it was one project being evaluated in one discussion for one organization. I don’t accept it when I think I think the call is being made categorically, in whatever context. If you want an example of what can happen when that sort of view of human subjects settles in to stay and becomes a dictum, I think a distinction between an American doctor being judged capable of making informed consent to taking an experimental drug for ebola and a Sierra Leonean doctor being judged of not being capable of informed consent.