From: Visualising Ethnography

An Interview with David Zeitlyn

The following interview was made in 2002 as part of Sarah Pink's Visualising Ethnography project.
This was online at from c 2002 until late 2012. It has been copied here on notification of the demise of the original site

spacer spacer spacer
spacer spacer spacer An Interview with David Zeitlyn spacer spacer spacer
spacer spacer spacer
In 2002 Stewart Coleman interviewed David Zeitlyn, of the University of Kent (Canterbury, UK).

Q - How did you first become interested in anthropology and, more specifically, how did you begin employing visual methods?

I got interested in anthropology through philosophy and science – visual anthropology has not been salient in my work until the last five years.
Like many anthropologists I was taking photographs and doing nothing with them, to all intents and purposes. Then a few years ago I was documenting this ritual [in Somiť, Cameroon] which I’d been documenting for years; it was the fifth time I’d seen the ritual and I was given access to the most private bits of the ritual that I’d never been allowed to see before, and the chief of the village turned round and me and asked how come I didn’t have a video camera. The ritual happens every two years, and so two years after that I turned up with my video camera saying “I’m just following orders”. So I shot video of all the ritual and that then posed me the question of what to do with it, and that led into the research project which I did with Mike Fischer. And then I saw an exhibition of African photography at the Barbican and that made me think that there are all these people taking photographs and so I’ve started talking to them and I’ve got a kind of research project that I’m trying to get started but I can’t get the start-up funding. So I’ve got a box that’s full of several thousand black and white negatives taken over the years by local Cameroonian photographers for a local audience. Now they’ve moved from 120 film to 35mm, everybody wants colour, nobody is interested in the old black and white, and these negatives are literally sitting there rotting. And there is this wonderful opportunity to document them now while a reasonable number of people are alive, but when I say these things I get accused of doing rescue anthropology, but it is rescue anthropology and I don’t think that rescue anthropology is essentially, inherently a bad thing. I’m not claiming that’s the only thing I want to do; it then becomes a wonderful instrument for doing all sorts of things about social memory, but if they’re not archived you can’t do it.

Q - Can you explain a bit more about what you hope this project will be?

Well one thing is to actually archive them and to make sure that the photographs are available in one form or another for Cameroonian scholars, which at the moment they aren’t and they never will be.
What I’m interested in doing - because I’ve got thousands of photographs spanning three different ethnic groups with three different kinship systems – I’m most interested actually in using it as an instrument to do old-fashioned anthropology looking at the relationship of kinship systems to social knowledge and social memory – who recognises who? So it’s a research instrument in that way.
I’ve now got a bit interested in other things that you can do at the same time. There’s an aesthetic side to it: what makes a good photograph? What are the local opinions of what makes a good photograph? The beautiful thing about doing this is that you can do all of them together, and everything will be better for doing that. But I haven’t got funding, so at the moment it sits here and nothing is happening because there’s no money.

Q - Do you hope to digitise them and get them onto a database?

Yes, it’s not cheap but it’s still cheaper and more flexible than making prints. It’s a lovely project and, well, I’m stubborn so I’m keeping on looking for funding.

Q - So, over these five years that your interest in visual anthropology has developed, has your understanding of the visual in ethnographic research and in representation changed at all?

It has, because what we ended up doing was taking clips of my video and showing them to people back in the village and using that as a non-verbal elicitation method to get what inevitably ended up being verbal accounts, but it threw up some really bizarre results which are controversial and we are still trying to make sense of them. In short, people could recognise clips with absolutely no problem, so it’s not a problem of people being able to interpret video or interpret photographs, that wasn’t what was going on. But they weren’t able to connect - in a way that we expected them to - clips with, if you like, a sequential overview of the ritual. We then spent a long time in the field trying different ways of presenting the same clips to try and help us sort out what was going on. In the end we’ve come to the conclusion that our model of the ritual presumes a fixed chronological sequence, and that doesn’t seem to be the way that Mambila actors think about the ritual, and that has profound claims for the understanding of ritual and other action; which is why I’m hesitating slightly because we are building quite a lot on – well, I was going to say a slender amount of data, but it’s not that it’s a very slender amount of data… I mean, the wonderful thing about the way in which we’ve gone about collecting it is that it’s all very concrete in the sense that we can explain why we’ve arrived at our answers. Because we have digitised the video and we were showing people the clips via a computer, we have logs of what they were looking at when, which we can tie to the sound recordings of the interviews when we were actually discussing the images. And they are perplexing, and we spent most of our time in the field trying to convince ourselves that our results weren’t artefacts of the way we were going about doing the question gathering. And I am satisfied about that, which poses a problem. Granted that they can do the ritual, they do do the ritual on a regular basis, granted that they recognise the clips, what was it about saying, “Which of those comes first?”. That’s what we ended up doing, showing people two clips and showing people one clip and asking, “What come before, what comes after?”. If you say to someone, “How do you do this ritual?”, people will give you a very sparse outline, and basically what we now think is that people have sparse chronological outlines but piece together the details not following a template. It’s not like schema theory. If you’re not familiar with schema theory it’s like – how do you do a play? You follow the script but there are all sorts of things that are not in the script. In a way I suppose, it goes back to that but we are weakening it massively and then saying that everything outside that incredibly weakened schema is cobbled together, is done interactionally, so you can’t take one person, they cannot act as a key informant. And I should stress that we were interviewing some of the key informants, such as the chief, such the people who are in charge of the ritual, and they were just as bad or just as good as everyone else, there was no difference between age, there was no difference between genders. Very weird stuff.

Q - And what about when you have shown them the whole video in the order that it was shot?

What people are commenting upon is “Oh look, there’s so-and-so, he’s now dead”, “Oh look, this woman is a brilliant dancer”, and then somebody falls over and everybody laughs. They’re more interested in that sort of absolute minutia, the real, real details. And everybody takes for granted how to do the ritual because everyone knows, but it’s weird that everyone knows but can’t say it. And I’ve actually since been back to the ritual once or twice and asked rather pointed questions right in the middle of the ritual, “What’s going to happen next?”, and some of them just tell me things which are…. I’m shying away from the word ‘wrong’, perhaps I should be bold: I say “What’s going to happen this afternoon?”, and they tell me something which didn’t happen. Now, okay, almost certainly this person is drunk because most people are drunk, but they manage to organise the ritual in such a way that the fact most of the participants have to get drunk as part of the ritual doesn’t matter… and somehow that gets left out of the standard accounts… I’m going back to high theory here.
So I’ve been using the visual stuff as a research tool; it’s not the ultimate aim of my interests, or hasn’t been until now.

Q - Would you encourage students to employ similar research methods, and what do you think are the advantages of gathering this kind of data over other kinds?

I am encouraging students to use visual methods. What’s nice is that it’s always driven by a particular project, so it’s not “I want to do visual anthropology”, it’s “I have this problem”, and then you can see some ways in which you can use visual data, or get visual data which is pertinent. I’ve become sensitive to how interesting it is because it’s non-verbal. In terms of how any verbal account is a particular structure, it can be a nice corrective.
I’ll tell you about one absolutely tantalising thing, which isn’t as straightforward as it seems. There’s a lovely project that was done in South Africa where a guy from Edinburgh called Rick Rohde gave disposable cameras to a whole set of people and then organised an exhibition of their photographs of their lives. I haven’t done that; I’d be interested in doing it. But I have given cameras as gifts to people and developed some of their films; and so having looked at their prints…. they're not very good. It’s all people’s elbows… the people I’m working with have no expertise. They are not used to using cameras… I’ve also lent a video camera to one of my main informants during the big ritual that I was talking about. Of course, I didn’t want to interfere, I didn’t want to tell him what to shoot. The result was a lot of ground, a lot of elbows, he was messing around with the zoom, he kept both his eyes open so he could see what he was shooting and wasn’t really concentrating on what the camera was focusing on. It’s not very useful. And so there’s a dilemma that you can’t give cameras to informants, at least not informants that are as inexperienced at using cameras as the ones that I’m working with, and expect to get useful information out. You have this horrible dilemma: you have to give them some training, but of course, by so doing you are interfering with their untrammelled vision, so there’s an interesting set of dilemmas there for a researcher.

Q - How did the visual approaches you were using tie in with the other methods you were using? What other methods did you use?

Basically that was straight participant observation in the ritual, and very weakly structured interviews with key players – some of those interviews would have been videoed but a lot of it predates me getting a video camera, most of it was tape recorded.

Q - What was the equipment you were using?

Originally Hi-8, and I’ve subsequently moved to using DV. The sound quality on the DV cameras is actually worse than the Hi-8. If you want a bit of practical advice, for God’s sake don’t economise by not having an external microphone. The sound quality on a lot of these is very poor – particularly the smaller ones as they are having to miniaturise the microphone and it picks up the motor noise more.

Q - Have you yet used any of this work – the images or the video footage – to represent your work in books, or multimedia presentations?

It’s work in progress. It hasn’t been an issue quite yet. We’ve got some on the web but not hugely publicised. We’re about to send an article to Visual Anthropology Review, which is setting out some of the background, we’ve got another paper we’re giving at a conference, and there’s a third paper which might end up getting sent to Current Anthropology, partly because of the way that they now have a linked website, so you do some kind of mixed-economy model of publishing: you publish on paper the conclusions and you can present the background material or data via the web.

Q - So this will be representation material as well as research material. Does that feel rather like publishing field notes?

No, it’s not like that because I’m not talking about putting the raw footage out. What we will do is make available some of the research tools that we developed in the field. So you will see very short clips, and you will be able to see the responses that showing those clips to people elicited. That’s the idea.

Q - Are there any other examples of visual work that you have done that people can see?

On the ERA website there’s “A Day in the Life”. I took 360-degree pans at three different locations in the village every hour on the hour for a day. I was there with a digital camera and I was wondering what I could do that would be a useful teaching tool and I came up with this idea of taking these 360 degree pans at three different locations over a 12 hour period. I’ve written a little bit about it but I’ve never had any feedback, so I don’t know what people make of it. It does demonstrate how profoundly boring life in the village is: not much happens, but in a sense that’s significant, it seems to me. It gives you some idea of what the Somiť looks like. But I’m intrigued: if it isn’t useful, why isn’t it useful? I’m very curious about what people get out of visual material.

There is also another project which we did, which was that Mike [Fischer] made a simulation of Mambila spider divination, which we ran on computer with a touch-screen, so again it was a completely visual interface which behaved in the way that Mambila divination behaves, and we showed this to lots of people in the village and they loved it, and they had absolutely no problem interacting with it whatsoever, and they just said, “Yes, this is how divination works”. And that is visual anthropology I would say, but it’s more like a diagram, it’s pretty schematic. That’s in ERA as well.

I’d like to ask you now about the new MA programme in Visual Anthropology that is starting at the University of Kent. How do you expect the new MA programme to differ from others on offer in the UK?

The people we’ve been most careful to distinguish ourselves from are Manchester, which is more ‘film anthropology’ than visual anthropology. I think you can do visual anthropology studying diagrams, and that is something that a lot of visual anthropology theory seems to miss. We are not teaching people how to make ethnographic films; we are trying to make sense of visual material, period. You could go and talk to the psychologists about face recognition; I would say that that could become part of visual anthropology.
The sorts of thing we hope to be able to do on the MA is to use multimedia not as an end in itself but in order to encompass whatever material the students want to work on. So sure, use video, sure use photography, sure use hand-drawn… whatever, but the focus is the anthropology rather than the medium.

Q - How do you think visual methods compare to more traditional methods in their usefulness? Would you say they could only ever supplement other methods, or could they replace them?

I don’t think so really. I think anthropology is primarily about how people understand themselves, how they understand the world they live in, the world they construct through their lives. So we can never not talk to people, let me put it like that. So a song might be the focus of an anthropological discussion just as a photograph might, but it can’t be the be all and end all.

Q - Do you think the new technology, particularly digital photography and digital video will change the way that visual anthropologists work and that anthropologists work and the way that ethnography is presented?

There are several different questions there. Since photography got easy to do, post Second World War, lots of anthropologists took still photographs, and as I said, most of them did nothing with them. Now that video technology has become easy I think that an awful lot of anthropological researchers are more or less automatically taking video cameras. And I suspect they are going to do nothing with the video they shoot. It’s just there. It might be used to illustrate some fun thing you can do at a talk, you can put everybody to sleep by turning the lights out and showing them 10 minutes of a video, but it doesn’t serve an analytical purpose. You can make these things serve an analytical purpose, and it’s probably easier to do on a computer than not, and it’s sufficiently cheaper to do that it only becomes feasible on a computer, but conceptually I don’t think that digitisation makes any difference, and I want to keep a focus on the conceptual purposes and they still are a bit vague. You know, the day of visual anthropology is always fast approaching, it never quite arrives, and I’m sure exactly the same is being said about computers, but at least here we have always been very clear that we think that computers allow you to do old anthropology better, rather than suggesting that it’s changing anthropology profoundly. We see it as very much a tool in the service of anthropology, and we are proud of being old-fashioned anthropologists who believe in detailed empirical study of human society.

Q - What is the future of visual anthropology?

I’m going to answer that question with another question: what is the future of anthropology? People are continually announcing the death of anthropology. If anthropology is to survive then I think visual anthropology is firmly at the centre. I’m not sure that it should survive as a ghettoised separate sub-discipline; it’s a bit like feminist anthropology, which I think has quite rightly all but disappeared as a separate sub-discipline and now you just can’t do anthropology without being sensitive to gender issues. And the way I see it, I think that any anthropologist should be both using and thinking about things to do with visual data. This is easy for me to say because I’m not hugely concerned about making films as polished products, and there are a whole lot of technical skills you need to be able to do that which I simply lack. But granted that proviso, I want to bring more visual material into the mainstream, and one possible result of that is that stuff labelled ‘visual anthropology’ per se goes away. And I’m very concerned that JRAI , that American Anthropologist basically don’t publish things discussing the visual; it gets ghettoised into these specialist journals and I think that’s a bad thing.

Q - Is there any overall advice you would give to somebody about to embark on any kind of visual project?

That’s very general…. avoid Microsoft proprietary formats? Work to international standards (which of course means not Microsoft proprietary formats)? Certainly, if you are talking about digital material there is a conservation and archiving issue which simply wasn’t there with analogue or traditional photography. In other words, do you wish to be able to access your material in 10 years’ time? If you do, then don’t use Microsoft proprietary formats – period. AVI, which I think is their standard video format, they’ve changed their format so many times that AVI-encoded video produced four or five years ago is not readable by Media Player software, whereas QuickTime 1 is readable by QuickTime 5, so it seems better. So be careful right from the start to save backup copies in as generic a format as you can possibly do, and do not trust the computer support people who quite simply are not thinking about long-term access.
Sure, if you want to use a Microsoft product then do so, but be damn sure that you save out of it into one of the other formats, which they will export to. It’s just a question about the archiving. Anybody doing research has got to think that they might want to go back to their material at some point in the future, and it’s all too easy to lock yourself out.

Links to web sites referred to in this interview:

spacer spacer spacer
spacer spacer spacer spacer spacer spacer spacer
spacer spacer spacer