Institute of Social and Cultural Anthropology

Interactive Multimedia and anthropology - a sceptical view

Marcus Banks, University of Oxford, June 1994


Note: This is a text of a talk given at a conference on 'Interactive Multimedia and Anthropology' at the University of Western England, 1 and 2 June 1994. The conference was organized by Richard Thorn of UWE and Andre' Singer and Dan Marks of Cafe' Productions. I deliberately overstated the anti-multimedia case in my talk to provoke discussion and comment, and I am aware that there are many oversimplifications and crude crystal-ball gazing in what follows.

In November 1994, Peter Biella - then of the University of Southern California - posted a response to this article. You can read this by clicking the link on the previous page. I would very much welcome further constructive criticism and feedback - my email address is given at the end of the article.

Interactive multimedia [IMM from now on] is a wide and diffuse term. For the purposes of what follows I shall take it to mean CD-ROM paradigmatically, with magnetic media storage (eg. HyperCard stacks) as a subset of this.

Let us start by considering the 'I' component of IMM - interactivity. You are of course, now experiencing a multi-media presentation, as you do at every lecture and seminar you attend. You are hearing me speak, watching me speak and move. If I wanted to be heavy-handed about this point you could also be looking at a slide or section of videotape, and studying a handout. As it is, your workshop programs will have to do. It is also an interactive multi-media presentation. You can stop me and ask questions, you can walk out, you can stand up and harangue me. Indeed, the interactivity works both ways. During the rest of my presentation I will speed up or slow down my rate of verbal delivery in response to gesture cues you give me, I may decide to skip a section of my paper if time is getting short, and so on. I can also depart entirely from my prepared text and sing you a song, or organize a group therapy session for us to discuss our repressed technophobia which we express in the classic denial pattern of enthusiastic endorsement. Similarly, you can interact with me by attempting to step outside the text and questioning me on what I had for breakfast this morning, or how much money I have in the bank.

But I'm stating the obvious, and we don't need to do any of these things...

The point is, human social interaction is critically bounded by context, and by generative rules. I won't sing you a song, and you won't suddenly ask me a question about my bank account, because that is not within the script for delivering a paper at an academic meeting. And if these things do occur, we all know that the normal script has been breached and that there will be some contextual explanation for this. In this light, the use of the term 'interactive' in the phrase 'interactive multi-media' has to be understood in a very particular way. It seems to me there is one intended meaning and one unintended or unrecognised meaning.

The former meaning is very narrow. With a highly limited stock of data, the computer controlling the IMM package understands extraordinarily specific and narrowly-focused choices of the human user. Computers, as you all know, are terribly stupid, or rather, terribly literal. They cannot, for example, understand natural language: the source data of most anthropologists most of the time. In fact, the computer is not interacting with the user at all in IMM. The user is interacting with the IMM creator or creators at second-hand and through an incredibly narrow communicative medium. Moreover, the interaction is essentially one-way. The IMM creator or creators can only guess at what the user wants to know and to provide the information in advance. The user must simply make do with it. The medium - the IMM package - takes a very long time to provide feedback to its creators, and then it does outside itself, outside the text. That is, if I as a user want information on a topic not contained within the package I must send the package's creator a letter, or telephone her, or simply let market forces send the message of dissatisfaction by refusing to buy any further IMM packages she creates.

Which brings me to the second meaning I see in the 'interaction' of IMM. This interaction is the interaction of human agents and their social actions and institutions within which the IMM package is located. By the time I slot my CD-IMM disk on the wonders of deep-sea diving, or the culture and customs of the So-and-so people into my computer, I have already completed numerous social interactions and transactions. I had to enter the market economy to get the job to provide the salary or research grant to buy the computer and the CD, I had to negotiate with my students to gain the free time to look at the CD and so on, and so on. And to write the CD the creator had to interact with social others - indeed, this meeting is itself part of the IMM interactivity process. Technology, as we all know, is socially-embedded. There is no socially 'neutral' technology, and we need to be aware of IMM's social embedding.

These are obvious points, but they are also anthropological points. Indeed, by focusing narrowly on the IMM package itself, as the computer industry and its followers do, we run the risk of falling into a trap that another anthropological medium has already strayed into.

Those of you who have any point of contact with ethnographic film cannot fail to have noticed that it is rarely taken seriously by anthropologists. Although actively hostile opponents may be rare, there is a large number of anthropologists who regard it at best with indifference. There are a number of causes and effects to consider, but let me mention one of each: part of the indifference stems from the popular anthropological belief - however misguided - that those who devote large amounts of time to the production and use of ethnographic film do so because they have little to contribute to the main agenda of anthropological theory; an entailment of this - an effect - is that ethnographic film is almost never used in research, except by the anthropologist who produced it. For example, with one or two very recent exceptions, I have yet to encounter any anthropologist who cites another's ethnographic film as a source of data, let alone analysis, in a written publication. The use of film is largely confined to classroom teaching where it is seen as a relatively painless way to introduce students to basic ethnography. Part of the problem - if it is a problem - lies of course with anthropologists who are unwilling or unable to see in ethnographic film any greater potential. But part of the problem lies in ethnographic film as a medium (as well, of course, in the intentions of its creators). Ethnographic film is produced, used and enthused over largely for the qualities of the medium itself (it is thus no surprise that until Wilton Martinez' work, no anthropologist had stopped to consider whether the use of ethnographic film in the classroom was actually effective [1]). This preoccupation with the medium - Alfie Gell's 'enchantment of technology' [2] - has led to the major use of ethnographic film being in classroom teaching, not in research. From all the current signs it seems that IMM - like ethnographic film - will find its major use in undergraduate teaching rather than professional research.

Academics are notoriously bad teachers. In Britain at least they are not professionally trained to teach, they generally show very little interest in pedagogical theory or practice and - for the moment at least - promotion and overall status within one's discipline has very little to do with teaching quality.

It would appear that far more academic brows are being coated with chillier sweat over the recently begun HEFCE (Higher Education Funding Council, England) teaching quality exercise than over the research selectivity exercise that preceded it. The most common complaint by academics against the teaching quality exercise, I would imagine, is that it measures the wrong things and that it cannot measure and therefore will not consider the right things. Most academics, I would guess, would argue that the purpose of a university education is not so much to absorb facts about nuclear particles, or the customs of the So-and-so people, but for young people to learn how to think, or for mature people to learn to think in new ways. That is, the value of education lies outside the script of (say) the French literature curriculum, and within a larger script of enlightenment and maturity, which is itself embedded within still wider scripts inherited from the enlightenment and before. Unfortunately, these things are not measurable

IMM is above all else a medium of script limitation and bounding, which is presumably why the HEFCE is so keen on it, and why it will undoubtedly in future act as a quantifiable measure of teaching quality. In fact the rhetoric of IMM denies such bounding by claiming to offer constant choice. For example, it is currently being claimed that an IMM version of an ethnographic film will offer far more 'freedom' than the original film. At any point in the CD-ROM version of such a film the user should be able to click onto background text giving contextual details, or listen to a commentary on the film by one of its participants, or any one of a number of other things. But this is an endlessly recursive project. The user may require footnotes of the footnotes, further paths and diversions, beyond the existing paths and diversions. But the social context of IMM prevents this, and the more slick the IMM package, the more it bolsters its claims to completeness and wholeness. It calls on the twin rhetorics of 'freedom' and 'choice' to disguise its control and command of authority. Of course, the illusion of boundedness and of completeness exists in the traditional media of the academy - in books, and lectures. But we know it to be an illusion and we - I hope - encourage our students to see it as an illusion. Books and lectures exist in the wider context of academic production and in time maybe IMM will be seen within the appropriate context.

Yet some of the hype surrounding academic IMM either denies that context, or more worryingly seems set to make sure that its claims to completeness and boundedness obliterate that context. In particular, the endorsement of IMM in the form of 'courseware' by the HEFCE cannot be divorced from a wider political context concerning the future of higher education in Britain. As I'm sure you are all aware, the HEFCE recently invited bids for some very large sums of money (5 million pounds in 1992-3 alone) as part of the TLTP (Teaching and Learning Technology Programme). A smaller sum of money was made available a year or two ago in a joint venture with British Telecom. The stated aim of the TLTP programme is to create free-standing 'courseware' packages to teach students by remote control. The government has made it clear that it is (or was until recently) committed to a massive expansion in higher education without providing any additional funding for staff costs to teach the new students. Instead, existing academics are supposed to create content-independent 'courseware' packages into which they will distil their knowledge of a subject. As if in some Star Trek script, it seems inevitable that having stored their knowledge on disc or chip, the academics themselves will become redundant. Nor is this only the view from Britain; a message recently sent to the VISCOM mailing list reads (in part):

"I recently read an article in one of these how-to-use-new-technology-in-your-instruction newsletters that it takes approximately 100 person-hours of research and production time to put together 50 minutes of real time classroom multi-media instruction. I don't know about you, but I'd have a difficult time finding that kind of time in my life.

Other teachers likely would face the same time limitation, so it would make more sense for a work group of people dedicated solely to that project to provide those new "instructional" materials that would take advantage of the new technology and would speak/visualize to a new generation of students weaned on MTV. Probably, these instructional materials would come packaged with "curriculum direction guidelines" as well. It will be more economical for new businesses to create these materials for a national higher education market. So, while it might appear rosy at first glance for us as instructors, having unlimited access to a smorgasbord of visual delights for our classes, in fact, just the opposite might be true. Because of time restraints and copyright/access fee limitations, decisions on classroom instructional materials and curriculum will be decided by faceless people somewhere else in cyberspace. These also won't be the people teaching the courses.

What will we become? Probably the best current model is the teacher's aide in public school. We will be the on-site administrators of the national, digital curriculum. Of course, we'd then be paid the equivalent of teacher's aides, with the same amount of respect.

Sound far-fetched? Put yourself in the position of the people with the pursestrings, i.e. legislators. No doubt, someone will make the numbers work so that it is far less expensive to pay the national curriculum consortium for access to the digital materials and then hire teacher's aides whenever possible, thereby cutting back on personnel costs for those over-paid and underworked higher education faculty, people who are more interested in research than teaching anyway." [Craig Denton, University of Utah, 26 April 1994]

This comment throws up another issue: Denton says that rather than academics being given the time, money and freedom to create their own IMM packages it will be 'more economical for new businesses to create these materials for a national higher education market'. The people who are really interested in IMM - if interest is measured in amounts of hard cash already invested - are the increasingly multinational telecommunications and entertainments industries. As far as they are concerned, IMM is about two things - interactive home-shopping, and pay-as-you-view video (or video on demand). It seems likely that companies currently involved with educational IMM such as Microsoft will come increasingly to provide the software platforms that enable home-shopping and video-on-demand (indeed, Bill Gates said as much in a recent interview on British television [3]). It therefore strikes me as probable that the look, feel, content and direction of commercially-produced educational IMM will be influenced by wider commercial interests. In Britain, we have already seen attempts by a fast-food company to get into the school classroom, and the day can't be far away when IMM 'courseware' is sponsored by and carries advertising for a whole range of products. (The only ray of sunshine in this scenario is that anthropology is such a small and under-funded subject anyway, that we may well escape the more blatant attempts to 'sell' the curriculum to business interests.)

In the context of vested commercial interests a particularly technological issue becomes apparent. Academics with a longer history of involvement in 'new' technology than me, have a number of similar stories to tell of obsolescence. As you are no doubt aware, today's shiny new toy is tomorrow's dinosaur. There are thus a number of anthropologists around with piles of computer punch cards neatly stacked in shoe boxes under their desks. With each year, possibly each month, that passes the chances of their being able to read these cards again, grows more and more remote. A few years ago, CDs were being touted as an indestructible and hence future-proof medium for music recordings. We now know that isn't true. We also know that the computer industry is a site of rampant change and obsolescence. Thus it seems more than likely that the exciting and innovative IMM systems of today will be obsolescent by tomorrow, and will require continuous inputs of cash and time to constantly upgrade. (This leaves aside as yet another problem the issue of cross-platform incompatibility). As anthropologists we already have numerous studies to hand outlining the effects of technological and productive innovation and introduction. The change from stone axes to steel, or from subsistence crops to cash crops, is not really so very far away from the shift from teachers and books to CD-ROMs, and as anthropologists we should not be naive enough to think that we will be immune from the changed social relations that such technological shifts and changes bring about.

I have three pieces of advice to offer. First, I would urge you all to think as anthropologists and read the existing literature by sociologists and anthropologists on the social relations of the new technologies [4], as well as the existing literature on technological change and innovation in our more traditional field research areas [5].

Second, I would urge you to forego work on educational IMM developments and concentrate instead on research applications. Academics as I have said are generally poor educationalists, and there seems no reason to believe that our IMM teaching packages will be any more carefully crafted or effective than our lectures and tutorials (for reasons of time and other pressures as much as for reasons of lack of skill). Personally speaking, most academically-produced higher educational IMM packages that I have seen strike me as slow, dull and naive. Like some ethnographic films they seem over-produced, concealing weak and shallow content with glossy presentation. More particularly, the much-vaunted strength of the hyper-textual links that IMM uses - that the user is liberated from linearity - seems unsuitable for analysis. If we agree that any path through a mass of audio, visual and textual data is as good as any other path, then the main intellectual purpose of our discipline is cast away. Most - probably all - intellectual analysis proceeds along a linear path, where pieces of data need to be assessed alongside each other in the right arrangement in order for the analysis to work. Abandoning linearity signals a return to Radcliffe-Brownian butterfly collecting: the arbitrary and decontextualised pursuit of comparison and connection for its own sake, or worse, the sheer observation of data for little more than immediate entertainment.

Research is quite a different area: most of us are interested in our research to the point of obsession. Most of us have clear research goals and questions. IMM in research should be a tool for achieving intellectual ends, not an end in itself. But this may not mean a lot of fancy equipment and custom-designed software. Relational and flat-file databases are interactive in as much as they allow the data to be interrogated. They also allow a fair amount of hypothesis testing, by matching certain arrangements of data against other arrangements. It is also more likely that the potential of IMM systems to store vast amounts of data will make it easier to share research material with colleagues and for them to test the same hypotheses.

Finally, and on a technological note, I would advocate distributed computing environments over bounded packages for the storage and presentation of IMM data. Using free and freely available networking tools such as World-Wide Web client-server systems, allows one to create a continuously evolving and infinitely mutable database. Hypertextual links are soft, not hard, authorship is multiple and on-going, and new data are easily incorporated without having to forego earlier links or jeopardise existing ones. The 'ordered anarchy' of the Internet means that control over structure and content will probably never reside with a single interest group. Use W3 or similar to place material in hyperspace, create a few links, and let the users 'evolve' the text beyond your cramped imaginings. If you want this world, go for broke. Why buy MediaMaker (or similar authoring software) at around 600 pounds, when W3 client-server technology is (effectively) free?


[1] Martinez, Wilton (1990) 'Critical studies and visual anthropology: aberrant vs. anticipated readings of ethnographic film'. _CVA Review_ Spring: 34-47. [Back to text]

[2] Gell, Alfred (1992) 'The technology of enchantment and the enchantment of technology'. In_Anthropology, art and aesthetics_ (eds.) J. Coote, & A. Shelton. Oxford: Clarendon Press. [Back to text]

[3] The interview was recorded for the BBC2 computer magazine programme _The Net_ ; I forget the date of the actual programme but it was around mid- to late-May 1994. [Back to text]

[4] Fairly randomly one could cite the following as a starting point:

Ruth Finnegan, Graeme Salaman and Kenneth Thompson (1987) (eds.) _Information technology: social issues. A reader_. Sevenoaks: Hodder and Stoughton in association with the Open University

John Law (1991) (ed.) _A sociology of monsters: essays on power, technology and domination_. London: Routledge

Roger Silverstone and Eric Hirsch (1992) (eds.) _Consuming technologies: media and information in domestic spaces_. London: Routledge

Similarly, work on earlier communicative technologies should be of relevance, eg:

Elizabeth L. Eisenstein (1979) _The printing press as an agent of change_ (2 vols). Cambridge: Cambridge University Press.

Ithiel de Sola Pool (1977) (ed.) _The social impact of the telephone_. [MIT Bicentennial Studies, 1] Cambridge, MA: MIT Press.
[Back to text]

[5] Again, fairly randomly, one could cite:

Pertti Pelto (1973) _The snowmobile revolution: technology and social change in the arctic_ Menlo Park, CA: Cummings.

H Russell Bernard and Perrti Pelto (1987) (eds.) _Technology and social change_ (2nd. edn.) Prospect Heights, Ill: Waveland Press

R F Salisbury (1962) _From stone to steel. Economic consequences of a technological change in New Guinea_. Melbourne: Melbourne University Press.

Closer to the issue in hand, work on the 'empowerment' brought about by the introduction of satellite and video media to indigenous communities should be relevant:

Eric Michaels (1991) 'Aboriginal content: who's got it - who needs it?'. _Visual Anthropology_ 4.3-4: 277-300.

Eric Michaels (1991) 'A model of teleported texts (with reference to Aboriginal television)'. _Visual Anthropology_ 4.3-4: 301-23.

Terence Turner (1992) 'Defiant images: the Kayapo appropriation of video'. _Anthropology Today_ 8.6: 5-16.

James C.Faris (1993) 'A response to Terence Turner'. _Anthropology Today_ 9.1: 12-13.

Most works on economic anthropology and development anthropology discuss the consequences of shifts from subsistence to cash economies. [Back to text]

| Return to ISCA Home Page | Return to Anthropology and Multimedia |

Copyright Marcus Banks 1994

Responses welcome. Please email:

This page last updated (layout only): 10.11.98