In one of my first posts on Savage Minds I discussed the convention of using a “man in the street” interview in journalism. Two recent Mark Liberman posts on Language Log raise more general questions about the use of interviews as data. In “Ritual questions, ritual answers,” he argues that

The journalists already know what the stories are. Their questions are not designed to discover any new facts or ideas, but rather to get quotes that will fit in to designated places in the frameworks of logic and rhetoric that they have already erected.

And in “Down with journalists!” he reinforces this argument with a funny example in which a French journalist finds himself the victim of this very practice.

We all know that this happens, just as we know that these quotes often server little more than a ritual function, but what can we do about it?

One option is to make the source data – the interviews themselves – available to download. In fact, such “grey literature” may eventually become available as part of AnthroSource, but it will not be easy. For one thing, there are confidentiality concerns. How do we make our data publicly available while still protecting our sources? It is possible to do – but it would create a huge burden on researchers. In essence, one might be punished for being a good researcher and collecting large amounts of data, because then you would have to carefully censure much more data to make sure it is safe for public consumption.

Making such data available online is not something that is without precedent in the field. Johannes Fabian’s book Remembering the Present: Painting and Popular History in Zaire is very much a collaboration with his informant, the artist Tshibumba Kanda Matulu. The full interviews are available online, as part of the Archives of Popular Swahili website (which, in turn, is part of the Language and Popular Culture in Africa website). Now, this is somewhat different in that Tshibumba Kanda Matulu is an internationally famous artist, who is anything but an anonymous informant, and the book is constructed in such a way that his discourse is allowed to challenge Fabian’s authorial authority.

Such a model may not work for all ethnographies and all situations, but it will be interesting to see what happens as more and more primary anthropological data becomes available online. Will anthropologists creatively re-mix each other’s data? Will informants salvage their message from the grand narratives of the anthropologists? Will computational methodologies and google allow for the work to be analyzed in new ways? Or will anthropologists resist to the bitter end in the name of protecting their informants … even when they might just be protecting their own reputations? Which isn’t to say that confidentiality isn’t a real concern – just that we should think twice before ducking for cover.

NOTE: My thinking on this topic dates back to an e-mail exchange I had with Mark Liberman last year about the topic of posting primary data online.

13 thoughts on “Interviews

  1. I think the antagonism to journalists that acedmics sometimes feel is a classic case of the narcissism of small differences. There are good journalists and there are bad journalists, just as there are good and bad ethnographers. I often feel that journalists are BETTER at their job than academics since they are more closely scrutinized — which is to say someone actually reads them. The solution of more transparency has its problems, as you suggest, but I just sort of think training people better in grad school and holding them to the standards of a real journalist rather than sports color reporting is another straightforward solution. Too often I read ethnographies from people which take the form of “You know, I hung out for a couple of years in Kiribati, and it kind of reminded me of this one thing Benjamin said…” were the research is used as a sort of Proustian evocation to hand a grandiose (and often unrigorous) theoretical argument on.

  2. Journalism and ethnographic research have much in common. Actually, I’ve always seen fieldwork as an extension of my activities as journalist: While I have one day for doing research as a journalist, I have one year as an anthropologist. In both cases, you have to lay open your sources as much as possible – otherwise you’re not trustworthy. As a reader I want to get insight in the research / data gathering process, want to know the basis which the author’s / journalist’s conclusions rest on.

    So, when writing my fieldwork report on young HipHop artists in Switzerland, I published all the interviews (ok, it’s not more than eleven) – both on paper – and later online – not as grey literature, but as central part of the paper (which made the paper readable for non-anthropologists as well). Before doing so, I double-checked and authorisized the interviews with my informants. But, as stated above, this may not work for all ethnographies and all situations.

    So let’s hope that anthropologists creatively re-mix our data, that our work will be analyzed in new ways. That’s what science is all about.

  3. Rex: I think you are totally *right on* in your reading of antagonism between journalists and certain academics. Interestingly, I think *a lot* of academics actually and deeply *envy* the much wider public that (good) journalists both address and sustain through their writing.

    I do shiver to think of how much *more boring* ethnography would be if tons of raw data were able to be hyperlinked in some way to monographs. To me the problem isn’t that our data in ethnography/monographs are ‘framed’ (edited, theorized, systematized), just that the frames we have been choosing or that attract us are so very often bad and/or boring ones (e.g., ‘reminded me of something Benjamin said’).

    Somewhere else on SM, someone argued that there can be no ‘theoretical core’ in the human sciences given the nature of human being, and I am inclined to agree. No normal science has emerged in quite a long period of reflection on human beings, including reflection of a scientistic (and even scientific) sort. In the absence of a rigorous normal science sort of paradigm in our discipline, what we have are descriptions and analyses of varying degrees of interest and persuasiveness… just like journalism. To my mind, the heterogenous condition of theory about people and society (including culture(s) ) is precisely what makes much ethnographic writing not dissimilar from journalistic narrative, travel narrative, the novel, and more. What burdens so much anthropological writing, and what makes it so very often boring, is precisely the need to ‘distinguish’ itself via the mantle of a scientism that it will never truly attain, a need rooted in a narcissistic compulsion to stand apart from much-disparaged forms of discourse that resemble nothing so much as ethnography on a human scale.

    All the same, as detailed at and elsewhere, a lot of journalism today is driven not by the need for good writing, but more by the cult of celebrity… come to think of it… so is academia!

  4. Now half-coherent associations of Wolf Blitzer and Geraldo Rivera to Judith Butler and Derrida are floating through my head…

  5. Hmmmmm . . . Interesting topic. I would say that primary data takes so many different forms that this wouldn’t work for all projects. If I take my preferred brand of radical participation with the keeping of journals rather than “fieldnotes” in the narrowest sense and with informal dialogues rather than interviews, my raw data amounts to my daily journals in which observations, conversations and gut reactions are all intertwined. A lot of it made it to the ethnography almost intact so . . .

    But this begs the question again of what constitutes data. Having the data available allows a reader to see all kinds of stuff that was noticed and that didn’t make it into the written work. It also would allow one to detect what was *not* noticed . . . in a roundabout kinda way. Re-reading myself, this doesn’t seem to make sense. I know what I’m trying to say but it’s too late at night. I’ll get back to this tomorrow.

  6. Sol Tax was a big advocate of archiving and making available our notes, letters, and other relevant information from our fieldwork and other research — leading him to play a role in the establishment of the National Anthropology Archives at the Smithsonian, and indirectly making my research possible by convincing his students in the Fox Project to do contribute their materials to the archive. Just about everything you can imagine a field project producing, from field journals to letters to draft essays to conference proceedings to memos to publications to newspaper clipping files is represented in the Fox Project papers. Tax’s own papers are on deposit with the U Chi library, where he also provided funding for the preservation and cataloguing of the material — in which other researchers have reported finding shopping lists and dry cleaning receipts. A worthy effort would be to digitize and make freely available the entirety of the NAA — but as the Smith experiences severe funding cuts, year after year, this seems entirely unlikely. From a preservation standpoint, though, it makes great sense — research materials are regularly photocopied, each time degrading the quality of the originals, but digitizing would allow them to be scanned and, except for research that required hands-on inspection, essentially put away; photocopies could be printed from the digitized copies. I wonder if the AAA would be interested in adding these resources to AnthroSource?

  7. Nancy said, “Having the data available allows a reader to see all kinds of stuff that was noticed and that didn’t make it into the written work. It also would allow one to detect what was not noticed…”

    This is an interesting aspect you mentioned.
    I respect very much every academic dedicating their work to disclosure this way. It might be more “risky”, but I believe it may indeed mean new means for analysis and contextualization of anthropology itself, as Lorenz said.

  8. I think I really disagree that making more ‘primary anthropological data’ available more easily will lead to advances in anthropological analysis.

    On the one hand, I *absolutely* agree that fieldnotes and other documentary compendia should be archived for future historians and anthropologists to use, as well as for use by the peoples described in such documents. In my own work, I was able to use the brilliant journals and fieldnotes of the anthroplogist who worked in the upper Asaro valley (New Guinea) 40 years prior to my research (Philip Newman). That work was vital to helping me understand the transformations in social structure that I argue have taken place in highland New Guinea societies over the last several decades. However, I think I was able to use those documents productively because of my own field experience and my own research. They would be much harder to make productive for someone who doesn’t have a grounding in the site in question. The kind of immersion in a cultural world that ethnography requires can be attained through reading extensively through an archive. But would massive amounts of online ‘primary data’ (I question this term) precipitate that? There are so many ways in which interviews, notes, photos, kinship diagrams, and such are deeply perspectival, laden with all kinds of implicit presuppositions, etc. that I suspect the notion that making them freely available online would lead to productive ethnographic/ethnological analysis a bit naive. Of course, such availability I think is basically inevitable. I just don’t share the evident optimism that it would be much of a solution to the problem of writing persuasive analyses of social life (the problem identified in the original post regarding ritual questions and answers).

    Moreover, *a lot* of what occurs in the writing and analysis of ethnographic narrative concerns arranging information in a way that protects the identities and interests of informants. Clearly, problems of confidentiality and anonymity are magnified in an online world.

    I detect an empiricist fetish in some of these posts around ‘primary data’ that surprises me.

  9. I think I was able to use those documents productively because of my own field experience and my own research. They would be much harder to make productive for someone who doesn’t have a grounding in the site in question.

    Sure, but who gets to decide who has the necessary experience and expertise? Is it necessary to keep this information out of the “wrong hands”? What about your informants themselves?

    I certainly don’t fetishize primary data – if anything I think my comments about the ritual aspects of interviews suggests the opposite view – but I also think that there might be something to be gained by relinquishing our desire to control this data and keep it to ourselves.

    Old data can become new again when used for different purposes: Just think of the possibility of studying field notes as a literary genre when you have a huge online database you can mine. Or the way a linguist might be able to learn about language use from interview recordings stored online. Etc.

    Besides, I believe that to worry about misunderstanding and misinterpretation is to assume that these are things that one can control to begin with – but our statements are always interable and subject to new uses and interpretations beyond our control. (Blame Rex for bringing up Derrida…)

  10. Kerim writes:

    “We all know that this happens, just as we know that these quotes often server little more than a ritual function, but what can we do about it? One option is to make the source data – the interviews themselves – available to download.”

    It sounds like you are here saying that distortions in the presentation of data can be corrected by the presentation of *more data*. This sort of empirical regression can be exercised ad infinitum. For example, transcripts are of course edited versions of encounters and carry their own brackets. Is a text enough? What about audio recording available online? What about video? What about video *plus* heart-rate monitors? What about multi-angle video plus key bodily indicators plus a diagram of the room/space the interview was conducted in? I am just not sure that ‘more data / more of the time’ is the answer to questions of analytic integrity.

    And you are quite right to question the proprietarial undertones implicit in a concern with ‘controlling data.’ My own emphasis is not on hording data, it is on what makes for good analysis and persuasive writing. Gobs and gobs and gobs of information/data does not necessarily equal convincing analysis, even when one wants to convince someone that convincing analysis is impossible. (As a student of James Boon, I know quite well the poly-voiced, multisemous, and re-readable nature of any cultural description [indeed, I am psychically paralyzed by the thought].) But anthropologists have *always* re-interpreted texts. Indeed, I would argue that the best re-reading is in fact enabled by robust analytic framing, and not by a cornucopia of raw data. For example, Marilyn Strathern’s devastating re-reading of Gil Herdt’s quasi-Freudian analysis of Sambia male initation (Gender of the Gift vis-a-vis Guardians of the Flutes) was, as Dame Strathern herself indicated, in many ways attributable to the thoroughly *analyzed* nature of Professor Herdt’s data, as presented in his monographs and essays.

  11. I’m not saying “more data” but more open access to the same data by different people. These are very different things. Nor do I imply that it will solve all the problems – just the one of informants feeling that they have been misquoted. Although I do go on to suggest other ways in which it might change the field (and many problems that might be associated with doing this.)

    If I was an informant I’m beginning to feel that I might wish to double check your interpretation of my speech 🙂

  12. Strongthomas — can you send a ref for “Marilyn Strathern’s devastating re-reading of Gil Herdt’s quasi-Freudian analysis of Sambia male initation (Gender of the Gift vis-a-vis Guardians of the Flutes)” ? sounds really interesting — I’ve never been able to bring myself to teach that Herdt account, though it makes it into many a syllabus and/or reader for intro to cult anthro in the States. Mostly this is because I embarrass easily when faced with a room full of 19 year olds. Partly it is because I’m not at all certain they will depart the classroom with any memorable take-home lessons except for the sex bits. But a smidgen of my unwillingness to teach it has been partially-formed intellectual misgivings that make Strathern’s critique, as you describe it, sound really intriguing.

Comments are closed.