Anonymity, Ethnography, and Alice Goffman: Welcome, journalists

I think I’ve written and thrown away three separate posts on the Alice Goffman debate trying to find something to say that people will find interesting. I personally don’t find the case to be very interesting, or to speak to core issues of what ethnography is or should be. In my opinion, the takeaway is: Goffman wrote a remarkable book at a remarkably young age, like all books it has some problems, and it is bearing an absolutely incredible amount of scrutiny fairly well. She did hard fieldwork and had to make hard choices writing her ethnography, and some people disagree with those choices. But that’s not an interesting theoretical problem. That’s just life.

For me the most interesting part of it is watching journalists run up against issues of anonymity and academic ethics for the first time and try to come to terms with it. I think its important to push back against the idea that ‘ethnography’ is suspect as a form of knowledge because it anonymizes identities and makes them “all but impossible to fact check”. An article by Leon Neyfakh making basically this claim appeared, embarrassingly enough, on the same day as an article by Jesse Singal which did fact check ing them. Singal — clearly an anthropologist at heart — walked around the neighborhood where Goffman did fieldwork handing out donuts until he fell in with the people she writes about in the book. What is at stake in this debate? Ultimately, I think the discomfort that some feel with Goffman’s book is that it forces them to think about issues of generalization, particularity, and anonymity in a way they haven’t before. Welcome, journalists, to the world of social science theory and ethics! We have been dealing with these issues for some time!

The idea that ‘ethnography’ is less empirical, true, or relevant than ‘journalism’ or ‘chemistry’ because it automatically anonymizes its sources is incorrect. In fact, there is no such thing as a single genre called ‘ethnography’. We can see this clearly in Neyfakh’s piece — the experts he interviews profess a wide range of opinions about anonymizing the people you study with. In sum, opinions on anonymization vary widely within the academic community.

They vary across time as well. Anthropologists have a long history of naming names. Sometimes this is because we view our interviewees as teachers or partners. In other cases it because we viewed them as so removed from the world of anthropology that they would never meet our readers, and so what did it matter?

On the other hand, anthropologists have also written purely fictional accounts, as as the 1922 volume American Indian Life. Here, entire people are made up. The goal is to provide a sense of what typical Indian life was like, but to do so by telling a compelling (and made up) story.

And we’ve tried everything in between: composite characters based on a mashup of different people, transcribed autobiographies that are not fact-checked at all, theoretical critique in the form of epic poetry, dance based on initiations we’ve undergone and more.

So ‘ethnography’ is a fundamentally experimental genre, or perhaps a loose assemblage of genres, not a single thing that can be blamed for Goffman’s success or failure, let alone used to condemn a field or discipline. In fact, as Neyfakh’s article makes clear, many of the issues surrounding anonymization were foisted on ethnographers by the IRB process, and are thus exogenous to ethnography itself.

Neyfakh makes an important and valid point to protest that the power of Goffman’s book comes from the particular biographies of the lives she narrates. To argue that the importance of the book comes from the general patterns of social life it generates, but to sell the reader on these patterns with biographical particularities does constitute a bit of a bait and switch.

But at the end of the day, her decision to write in this way is not a result of the general professional standards of ethnography, but because of the particular nature of her field site and her personal choices a writer. I think it was a legitimate choice, and I think Goffman’s book is a good read with solid ethnography and important lessons about race and policing in the United States. But it was a difficult one, one that no doubt left Goffman unhappy because her respondents could be identified, but also one that left journalists unhappy because they couldn’t be identified enough (at least, not without donuts). Which just goes to show that the Goffman debate is at heart not about problems of ethnography as a discipline, but about the fact that hard choices are hard, there’s no easy way to make them, and someone will always have preferred you make a different one.


Alex Golub is an associate professor of anthropology at the University of Hawai‘i at Mānoa. His book Leviathans at The Gold Mine has been published by Duke University Press. You can contact him at

11 thoughts on “Anonymity, Ethnography, and Alice Goffman: Welcome, journalists

  1. Thank you Rex for having addressed the topic here on Savage Minds. You answered to the critiques relating the anonymity and the ethnography as a “genre”, but there are other critiques of Goffman that you haven’t mentioned and that I find very convincing. For example: have you read the article written by Christina Sharpe on The New Inquiry ( I would be very interested in your opinion on that.

  2. Really – many thanks for this. I am trying – for the third time – to write a book about a group of Bedouin women weavers that I have known for 25 years. Hard choices about anonymity and describing experiences get no easier over time. Your article came at an important moment when I am trying to make those choices, And I think you are right – it is about choices and knowing that some people will have wanted you to make different ones – but it is also about the inevitable vulnerability of whatever choice you make.

  3. @Costanza, I feel like there’s a lot going on in that piece, but I have to confess that I can’t quite figure out what specific criticism of the book is being made. I get that Sharpe criticizes Penn for its relationship with the neighboring community and (possibly) not demanding/providing ethical guidelines from Goffman, and that there are a lot of issues with the book’s representational politics. But what would you say is her argument in a nutshell?

  4. That a reporter with a box of donuts can track down the real subjects in an afternoon despite all the efforts Goffman put into changing names and dates should not come as a surprise. I haven’t read the book, but I think anthropologists often delude themselves (and their informants) about the level of protection afforded by such measures.

  5. I think you’re right, Kerim. In some ways the discipline is long overdue for a conversation about pseudonyms—when they are needed, when not (and why), and how to make them effective rather than just expected.

  6. A good place to begin is old-fashioned journalistic ethics. Citations should be as specific as possible. Those who speak with us should be asked if what they say is on or off the record. If disclosure may be embarassing or dangerous to them, it should be they who decide whether or not they wish to be identified. The circumstances in which off-the-record statements are reported should be included in the report.

  7. I’m often surprised when informants are disappointed when I tell them that their names will not be included. Most of my topics are non-threatening, and even innocuous for my informants–and they often want their names attached. Journalists find this out too, I expect.

    What is clear to me is that the medical model of informed consent is inappropriate for the social sciences. The protections it offers are often limited to the bureaucratic need to guarantee the flow of continued federal money to the universities sponsoring the research, not the subjects themselves.

  8. Singal fact checked that at least some of the main characters in the book exist, yes. Neyfakh took issue with the events that AG witnessed with these people, which even Singal was only able to get partial confirmation of from his interviews. We are still left with a very powerful story about police checking names at a maternity ward of a (unnamed) hospital and arresting 3 boyfriends right in front of AG’s eyes. Is it important that this fact to be true? May be not if the larger message about excessive law enforcement is what matters. But, there is damage being done if boyfriends stop going to visit maternity wards. 11 year olds given 3 years of probation for riding in a stolen car? Again, may be not “true”, but we all know bad things happen in the criminal justice system, so again, even if this story isn’t true, it fits the larger narrative.

    The more powerful review of the re-reporting of her book is here:

  9. Is anyone else annoyed that in the kerfuffle about the Alice Goffman book, “ethnography” is represented almost exclusively as a method used by sociologists? I have seen very little that mentions anthropology or anthropologists as ethnography’s chief disciplinary practitioners (i.e., unlike sociology, where quantitative social science remains pre-eminent). Am I being petty, or is it significant that anthropology figures so marginally in this discussion?

  10. Daniel, it is only in the cloistered halls of anthropology that ethnography is taken to be something peculiarly anthropological. It was adopted as a method in urban sociology by the Chicago School in the 1930s, was eclipsed by quantitative methods (what Andrew Abbott calls the Standard Causal Model) in the 1950s and 60s, and has since enjoyed a renaissance. There are numerous qualitative anthropologists who do ethnography. “Ethnography” is now, moreover, a service routinely offered by marketing and consumer research suppliers around the world. Anthropologists may argue that short-term observational research combined with focus groups and in-depth interviews is not real ethnography — for the rest of the world such arguments are little more than academic bickering. In several articles (see, for example, Vol. 1, No. 1 of the free, online Journal of Business Anthropology, Marietta Baba has shown that an intimate tie between anthropology and ethnography is neither where anthropology began, as a form of armchair research using materials collected by others in the 19th century nor likely to be its future, given that most anthropologists will no longer enjoy the imperial privilege of prolonged participant-observation by scholars whose lives are expected to be dramatically different from those they study.

    This does not imply that anthropologists have nothing to say about the Alice Goffman case. It does suggest that the common stance of aggrieved purists defending their turf is unlikely to contribute much to the debate.

  11. I think Kerim is right that we often delude ourselves about how much it protects anyone when we change a few names and dates. I also agree with Tony–the protections that are offered only really serve the university itself. IRB forms and informed consent processes are filled with all kinds of legal jargon, but outside of the US how much weight do they really carry? Probably next to nothing. The other issue is that many IRB’s include a statement about having to show data to “legal authorities” in certain cases, which should tell us all just how few protections this offers for the people we work with. It’s all about covering the interests of the university bureaucracy–and not the subjects themselves (as Tony clearly points out).

Comments are closed.