Books will soon be obsolete in the schools...It is possible to teach every branch of human knowledge with the motion picture. Thomas Edison (1913).
...and the applications are limited only by your imagination.
slogan of micro-computer advocates in the late 1970s.
Visual anthropology in the digital mirror: Computer-assisted visual anthropology
Books will soon be obsolete in the schools...It is possible to teach every branch of human knowledge with the motion picture. Thomas Edison (1913).
...and the applications are limited only by your imagination.
slogan of micro-computer advocates in the late 1970s.
Keywords: Visual Anthropology, Computer Applications.
It is difficult to identify 'distinctive' applications of computing for a particular humanities or social science discipline, and in particular for anthropology. Anthropology has a reputation, among anthropologists at least, for borrowing heavily from other disciplines. As with all disciplines that focus on people and their productions, anthropology is information rich and understanding poor. The computing methods used by particular specialist anthropologists tend to be similar to methods used in allied subjects in the humanities or social sciences.
Anthropology is a broad subject, with diverse sub-disciplines ranging from folklore to molecular anthropology. Where anthropology tends to be distinctive is its use of information derived from a method called 'participant observation' reported though ethnography. As a method this stereotypically consists of an extended period of living with 'research subjects' (who may be university professors Sheehan 1993 or other types of hunter-gatherer). You stay in a house, hut or tent just as 'they' do. You eat the same food, sit around the same fire, bar or tea stall and talk, get involved in personal relationships and personal obligations. You catch some of the same diseases and suffer from the cold or heat. Because of your origins as an 'outsider' and the surety that your future will not be theirs (or at least only in episodes), you cannot be a real member of the community and should not claim to be, but you do gain a great deal of experience. Others work "at home" where such membership does exist but the fact of education and the practice of research is alienating, so one ends in a somewhat similar situation to that of 'stranger' anthropologists.
This experience-rich method gives anthropologists a great deal of specific information they use to interpret what people say, what people do, and what people say they do. Although many sub-disciplines of anthropology cannot practice participant observation (e.g. archaeology and molecular anthropology), all branches of anthropology are influenced by the information collected by those who do. From this body of ethnographic information, emerge the problems that define anthropology, and thus the wider goals of all anthropologists.
Next we consider the use of computers in anthropology in general. This is a necessary prologue to consideration of our empirical findings since the results were obtained, in part, from the use of computers as a field tool. We have devoted much attention to the concern that the field techniques have created the problems we address rather than reveal them as an analytic problem. We discuss the specific reasons why the problems are not artefacts of the empirical methods used after a more general discussion of how computers and anthropology may be related.
If we are to identify a distinctive synergy of computers with anthropology it will most likely be in the collection, reporting and analysis of ethnographic data. We suggest that computer-based multimedia applied to ethnographic research is most likely to yield a distinctive mix that advances both to some extent. Computer-based multimedia has the potential to integrate visual and aural field material into ethnography, both of which have posed sufficient problems in the past to deter most anthropologists, despite broad agreement of the importance of such material. At the same time, the high demands of representing ethnographic material reveals important relationships between media units within computer-based multimedia that have not been widely reported or applied.
Within this context we will introduce some of the past problems with the use of computers in anthropology, and give a brief sketch of past computing applications. In particular, we concentrate on visual methods in anthropology, examining how computers address visual materials and how anthropologists can exploit computer-based capabilities. We conclude that most of these methods decrease the labour of working with visual materials, but do not essentially add new capability. New capability can be achieved only if we have a way to represent our knowledge about the material in the computer-representations of that material. Computer-based multimedia provides tools for doing this, contingent on our ability to represent our own knowledge formally. We identify different levels of knowledge that can be represented. We show that if we and others are to derive much value from it these levels must be incorporated into multimedia documents. These general points are illustrated by a case study in which our data suggest a radical revision of conventional understandings of ritual.
In his (1998) commentary on the Banks and Morphy (eds: 1994) collection Chris Wright isolated three issues (among others) - the 'visual' as an anthropological subject/ object, visual documentation in anthropology, and the relationship of visual form to the anthropological subject. He distinguishes between the first and second category, though in a rather diffident manner. Following the 'anthropology of ...' model (economic anthropology being the anthropology of economics, and so on), we would expect visual anthropology to indicate the study of visual form in social or cultural context, not the use of visual data. There is no other area of anthropology which is distinguished solely by the inclusion of a specific form of data. In differentiating visual anthropology from the use of visual data in anthropology Wright exceeds himself, suggesting that the use of visual data might be called 'illustrated' anthropology. Should those of us who still use fieldnotes be said to do 'annotated' anthropology? Or are all those who wear glasses or contact lenses automatically to be considered as visual anthropologists since they have technologically mediated visual access to much anthropological data? It should be remembered that for most of the last century, continuing to the present day, conventional ethnographic studies have typically incorporated little or no visual data.
The reasons for such exclusion are varied: ranging from the cost, and portability of the equipment, through the lack of analytic tools for dealing with visual data, to a disregard for detail. The contemporary situation is probably more related to the 'political economy' of anthropology than an indication of the inherent value of visual data. It is quite possible to incorporate visual information as evidence for a broader study. But this kind of data tends to be oriented towards detail, and increasingly anthropologists are less and less concerned with detail, and more and more concerned with possible problems that might emerge if they were to undertake detailed studies. So long as anthropologists are more rewarded for uncovering new 'big issues' regardless of how ill prepared they are to do so, than for establishing a cumulative programme of research, then the concern with matters close to individual participants and specific manifestations of human emotion, thought and action within a society will remain undervalued.
There are, of course, different kinds of problems with different kinds of evidence, but these problems do not necessarily stem from the sensory form the evidence takes. Rather the problems come with the context and mode of recording, with the evaluation of the relationship between the evidence and the subject, others with the identification and evaluation of patterns of similarity and difference. These problems are numerous and laborious to offset, as anyone who works with visual and other recordings is aware. Recording light or sound (or impressions) from a situation is not a wholesale or total recording of that situation, or even a model of the situation. It is a recording of some details related to it, and must be placed within an interpretive framework to possess value as evidence.
Using visual data requires precisely the same care as incorporating any other form of evidence, be it written, spoken or aural (and presumably other non-textual sensory channel recordings as these become available). Of course, there are real issues concerning the use of visual data within 'mainstream' anthropology. Some might even argue that this is because it isn't useful. Wright introduces the film of the 'Yanomani Ax Fight'1 as a good example of the relationship between visual ethnographic data and its analysis in anthropology. We agree. In the 'raw' footage of the film, we hear Chagnon making immediate comments, talking to people and relaying their instant commentaries on the fight. In the middle section Chagnon's analysis suggests that these initial commentaries were at best inaccurate, and, in the normal usage of the term, incorrect. Wright suggests this analysis might be imaginary (as a synonym for being 'incorrect'). However, it came about not from an obsession of Chagnon's part to look for kinship related factors, but from his attempts to investigate both his initial interpretations as an experienced observer and the initial interpretations of the indigenous observers around him. He was unable to support those interpretations in the analysis. But within anthropology, in part for sound analytic purposes, in part for aesthetic reasons, we are obliged to define all indigenous views (held by 'others') as 'correct' views, and one of the goals of our accounts is to allow for variation in viewpoints, to make a body of seemingly incorrect and inconsistent views correct and coherent within our framework in order to validate the framework, not the views.
Kinship enters the picture, not because it is a 'natural' relationship, but because it is a way of 'measuring' social relationships as these are defined culturally, in particular, relationships which vary according to perspective, from one individual to another. Kinship theory as developed by anthropologists is in many ways unique in science or social science, because it provides a means for synthesising an overall account of a set of relationships which are different relative to each involved participant. They are multi-perspectival, relativistic in the strict sense of the word, but not for the individual actors who are constrained by each other (see Ruby 1995 for further discussion).
After conducting their analysis Chagnon and Asch present the film a final time. Now the viewer is armed with a set of 'measurements' to make and patterns to match. What was apparently chaotic in the first instance now is ordered, albeit synthetically. Their account is not the only one, nor perhaps even the best, and is certainly partial, but it is not imaginary. It is rooted in the anthropological aesthetic, and we are able to share this new account, as are casual observers in the classroom. This differs from countless other accounts of similar events which fail to connect what we see to the production of the accounts about them Wright suggests that this film might be more about the anthropological process than it is about the Yanomami. We suggest it is about both, in a classical sense. An ethnographic account is not a 'real' description. It is synthetic and analytic, as is any scientific, journalistic, or artistic account. It is constructed using evidence and tools for procuring and evaluating evidence. If it is constructed well, we are able to use it for a purpose such as 'understanding' the Ax Fight. If it is constructed poorly, we cannot.
Reduction exploits redundancy. But the aesthetics of a good synthetic analysis is the extent to which information can be removed in the course of the analysis, then restored by 'animating' the resulting account. This does not ensure a 'correct' analysis, but it does result in a pleasing one. The Ax Fight accomplishes a non-trivial example of this by presenting the same information twice, sandwiched by an analytic account. From the first we learn only a few rudimentary 'facts' about the Yanomami, from the last we can animate the analysis and overlay it on the film and 'see' the film through new eyes which combine vision and knowledge. If the film is 'overused' it is because it is perhaps the clearest example to the lay person (or student) of the power of anthropological analysis to clarify rather than mystify.
Wright endorses Grimshaw's warning about accepting visual recording as mere technique without considering its broader implications within anthropology. We are not quite sure what is meant here. The only reason for the existence of a 'technique' within a discipline is to contribute to that discipline - precisely through the implications of its use. If it has none, it is not useful. If we do not know what the implications are, it is literally useless. Photography and video are only useful as techniques if they help us do better ethnography and/or analysis. This goes far beyond 'illustrated' anthropology, which may improve the storytelling aspect of ethnography, but makes no comment on visual material as primary or secondary evidence.
There have been potential computer-based solutions to some of these problems for well over 15 years, in the form of computer-controlled playback machines, but these have not been widely applied in research. Computer-based representations provide the organisational resource of quick access to individual images, video sequences and audio sequences which can be indexed to textual references or references to objects in the same or other media. Multimedia documents use non-sequential links between different instances of a range of media. Although relatively few anthropologists as yet are using computers to incorporate visual records into their research, we suggest that this represents a genuine contribution to the discipline. Visual records have been poorly utilised by anthropologists, although they have shown great interest since the advent of photography, cine film, and more recently, video.
The first example of a significant (e.g. not a demonstration) computer-based collection of ethnographic data in the form of images and text is the collection of material relating to the Naga in the Naga Videodisc project developed by Professor Macfarlane's group at Cambridge (see e.g. Macfarlane & Gienke 1989, now partly available via the web at http:// www.digitalhimalaya.com/). This has been followed by Professor Stirling's Forty-five years in the Turkish Village project incorporates demographic data, household data, migrations data, fieldnotes and about 1200 photographs and 1000 stills and out-takes from videotape (see http://lucy.ukc.ac.uk/TVillage/notes.html). Also, the computer accessible version of the Chagnon and Asch film (cited above) has recently become available.
Computers provide a platform for visual records that overcomes most of the barriers to using these in the research process. In the past these barriers have included:
In addition to resolving these technical difficulties, which in principle can be overcome (and have been on occasion), computer-based visual records offer new capabilities, most notably the ability to layer and interrelate in a non-linear manner data research data, at all levels from the field to publication.
Means by which computer-based methods address these problems include:
There is great variation in the extent to which anthropologists incorporate images into their research. Some use visual records as a central part of their research. Others take a few photographs simply for the purpose of illustrating aspects of their notes and other data. These may find little use for a computer other than, perhaps, the creation of a special register or 'card index' of photographs and legends; giving their photographs names then referring to them in the notes or other data resources. There are many more possibilities. The technical details of when, where and how to acquire images in the form of digital photographs or video, and the basic methods of analysis and use are no different from more conventional media for representation. These issues are well covered in Collier and Collier (1986) and Jackson (1967), as well as briefly discussed by Blacking (1984). However, the range of analysis possible and the ease of access are more extensive when using computers. These begin with text/code-based computer-assisted classification using authority lists such as a taxonomic thesaurus of classification terms for subject domains. Others include systems to enable cross-references between fieldnotes and people, places and events depicted in the records. More sophisticated yet is the incorporation of the images into a computer representation which can be not only accessed and viewed on screen from a catalogue, but can serve as an interactive element for data entry relevant to the images and objects in the images. Images can be incorporated into most computer applications which operate on graphical operating systems, including word processors, database programs. Images can be resized and otherwise manipulated using a 'paint program' capable of editing colour images. For video tape, support for partial or even full transcription at frame level is available, either through computer-controlled players, or through direct representation on the computer. Indeed, full transcription may not be necessary in some cases because, under computer control, specific video frames or sequences can be displayed on the monitor as a response to database queries, along with notes and markings added by the researcher.
Field considerations for video as a research tool
Like all research methods but especially those which costs and technical limitations apply, the use of digital video in the field has to be critically assessed. For most researchers the principal application of displaying or making video in the field include:
It is primarily the first aspect that was explored in our recent case study (as described in accompanying papers). In all cases video clips can be manipulated by other computer tools, such as database management systems or inserted into word processing documents or multimedia authoring programs, as well as specialised programs for working with video material. With the system level support indicated below, there is support in each of these applications for making selections from the original clips, regardless of what application they are embedded within, and installing these new 'sub-clips' into the same or other application. This is done by reference, rather than making a copy, so the new sub-clip does not add appreciably to the storage requirements. A five minute video sequence can thus be broken into smaller and smaller sequences within an application such as a word processor database or spreadsheet, where these can be documented, while retaining the entire clip for reference.
As it is commonly used, computer-based multimedia refers to the presentation of documents which include a variety of visual, textual and aural information. As with most things in the computer world, multimedia is not an original development, but rather is based on existing non-computer models. We are not dealing with anything particularly new here. The ideas underlying hypertext documents are attributed to Vannevar Bush beginning in 1945, although some medieval scholars attempted to construct hypertext systems almost a millennia before (see Rouse and Rouse 1982). Computers may make many of these old ideas achievable in practice, rather than remaining of theoretical interest. This is a basis of the discomfort of many people, because although one is not obliged to be more concrete on a computer than any other technology-based medium such as writing on paper, the additional capacity to manipulate media on the computer seems to impel people towards the concrete.
There are currently several major models for computer-based multimedia. One might be called McCluhanist, in which a variety of media convey meanings perhaps only remotely related to the content of the media. This is commonly used in some advertisements or rock music videos. MDF's first exposure to the parent of this form of multi-media was in Prof. Krupa's 'Electronic Poetry Course' at the University of Texas in 1969. It was very difficult to enrol for the course, as it had a typical waiting list of perhaps 200, but one could walk in at lunch time and watch the show. He had set up an impressive array of movie projectors, slide projectors, tape recorders and overhead slides, all of which changed independently. To this background he would give his lecture. It is not clear what was learned about poetry in this class, but it was a striking display. Unfortunately, this is often the impression the computer version gives as well: "very flashy, but what is being communicated?"
A second model relates most closely to film; a sequenced presentation, commonly using visual stills, computer-generated graphics, text, sound and video. This form has been adopted widely in business for corporate presentations, entertainment or info-tainment, and educational purposes. There is some interaction by the reader (or user in computer-speak), for pacing the presentation and activating a few side-demonstrations in the form of small video clips or animations. As a finished document, this form corresponds most closely to film, in the sense that it is produced and sequenced in a more or less precise manner. The organisation is somewhat looser than that of film, in the sense that the presentation is often divided into units, and the units can activated (or 'visited') in any order, but the preferred sequence and the units have been been carefully crafted.
A third approach is essentially one modelled on the book. Frames, corresponding to pages, contain a mixture of image and text. There are innovations, such as the ability to activate new frames based on references within a frame (called hypertext or hypermedia links). One can move to a given 'page', consult an index, or perhaps search for a term. This approach should not be dismissed, because it is more 'powerful' than a book; it can incorporate information not currently available in print publications, such as digitised video and sound (e.g. Zeitlyn's audio appendices to a published transcript 1993a). And even more importantly, we can incorporate 'expertise' in the use of the material (e.g. help and guidance files, teaching notes etc).
A fourth type we call layered multimedia and it has the most promise for anthropological applications. It has some of features in common with books, but is designed in such a way as to provide different kinds of resources at different points in the text. These resources are organised into 'layers', which can be followed independently, or in combination with other layers. For example, in MDF's work on dance in the Cook Islands, he presents a textual analysis of the structures of male and female dancers in same- and opposite-sex interactions. In the text the definition of different terms are linked to drawings of abstracted dance positions. Selecting a drawing leads to the individual video sequences the abstractions are based upon. The reader of the document can follow the analysis down to the raw data as MDF assembled it (<http://lucy.ukc.ac.uk/tradition.html>). Another example is a cross-tabulation derived from a socioeconomic survey MDF undertook in Pakistan in 1982. The summary table is presented in the text. Selecting a cell in the table produces a list of the households in that cell. Selecting a household leads to the household members. Selecting a household member leads to a personal record for that person- albeit anonymised for public access (accessible via http://lucy.ukc.ac.uk/GreenNotes/).
The book model and the layered model are examples of non-linear multimedia. That is, there is no necessary fixed order through the multimedia document (though there may be an order preferred or recommended by the author or teacher setting the text for a class).
The creation of multimedia documents has been associated with relatively large and expensive equipment, time consuming procedures, a remarkable amount of manual processing for a computer-based document, as well as non-portability, since often the only computer which can reproduce the document is the system on which it was created. Recent developments have dramatically changed this situation. It is now possible to record, digitise and process audio and video tape then author a multimedia document based on these materials using portable equipment weighing about 4 kg. The resulting document can be reproduced on any computer with appropriate software and a reasonable display. A standard for multimedia software and the means for dissemination has emerged in the form of the Hypertext Mark Up Language (HTML) and World Wide Web browsers such as Icab and Netscape (among many others). These make the use of multimedia possible both in a field context as a research tool, and as a means of dissemination to a range of researchers with less specialised hardware.
Ethnographic film has been criticised for being difficult to use in research. Indeed it is very difficult to find examples of ethnographic film being used in research at all (Loizos 1993). Banks (1994) has made similar criticisms of computer-based multimedia in anthropology. He suggests that multimedia databases may make it easier to exchange research material, but that there is little added by making the various interconnected (non-linear) links in the material. Biella (1994) counters this with a detailed analysis of non-linear structure, and how this structure constitutes a genuine encoding of knowledge within the document. Both focus on the 'freedom' of the multimedia user, and are principally disputing whether this freedom is of value or not. An aspect that both ignore is that this freedom is largely illusionary. The use of links in a multimedia document represents an increase in authorial control, not an advance in reader freedom (see Crane 1991). For example, it is difficult to imagine much more freedom than is afforded by a book! Although there is a linear order to the pages, the reader is free to look at the pages in any order, and parts of pages at will. Most books come with a device called an index which facilitates non-linear use. By processing the book through a concordance program, individual sentences can be utilised completely out of context (see discussion in Landow (ed.) 1992)
Multimedia documents afford the reader no more freedom than in books, and in many cases much less. Not every thing is linked, and only material that is linked can be accessed from a particular point in the document (although full-text searching provides some alternatives, but a misleading one since a poor choice of search term may mislead the user). Access to other parts of the document is controlled by the author. The manner in which the author produces the links in part reflects the author's knowledge of the content. Levels of knowledge might be represented from low to high as follows:
Each of these levels implies a greater control and knowledge about the material within. At present most multimedia is limited to the first two. These consist entirely of fixed direct links, or links that relate to classified material. Some forms of layered multimedia achieve level 3, such as the example relating table cells to households to individuals, all of which is related using a general syntactic description of the relationships. In such documents there are no hard links at all, but the information is retrieved systematically through the use of relationships between different types of information.
Because of the increase of authorial control in computer based multimedia, it is important that we are able to achieve at least level four. Otherwise the reduction in reader access will prevent them from using their normal skills in supplying semantic links (i.e. making cross-references to other relevant material). It is currently possible to insert such links in a hypertext document (although it has not been implemented as a multimedia model to our knowledge). The commonest example of this in conventional computing applications is the expert or knowledge-based system, where links between information are based on situational matching. A multimedia representation would involve tagging the textual, visual and aural segments with propositions that can interact with each other within a knowledge-based system to create relevant links. To accomplish this involves not more computer technology, but the development of more formal means of describing non-textual media.
The use of computers as a medium for preparing, storing and retrieving visual material greatly increases the potential for using visual material in anthropological research by reducing the mechanical difficulty of working with these materials. But this alone does not add any genuine new capability, it simply makes the existing one more attractive. New capabilities are only possible if we can impose sufficient structure on the material to reflect our increasing knowledge derived from working with it. With texts language enables us to express cumulative ideas, to impose organisation and structure. We can abstract, represent and communicate. Such techniques are less developed and much less formal for non-textual data. Computer-based multimedia provides a solution. Our interest in multimedia methods was derived from a general dissatisfaction with the limitations of working with aural and visual data manually, while recognising that it was essential that anthropologists consult non-textual data with rigour to supplement traditional ethnographic techniques. There are, however, very few methods for referring to still images, much less time-based media (such as sound recordings and movie/video recordings).
Finally, let us ask where this leaves the visual in anthropology? One view of multimedia documentation is that it means we need to spend less time describing the visual or even its representation since the representations (photos, video, sound and other recordings, short explanatory texts and various notes and titbits) can easily be included in our 'texts'. The space saved, the energy that no longer has to be devoted to conveying in words a sense of what is seen, can be spent on analysis. That is on asking anthropological questions such as described in Wright's (1998) review discussed above: how to approach the 'visual' as an anthropological subject/ object, the status of visual documentation in anthropology, and the relationship of visual form to the anthropological subject. Where this leaves his greater question as to the relationship between aesthetic (including artistic and scientific) portrayals and anthropological accounts is not certain. For the time being we should do both and then examine the implications of their juxtaposition.
Multimedia methods have been subjected to considerable criticism for appearing 'transparent' while actually being 'constructed' by the analyst. Although grossly overstated, these criticisms deserve attention. Although we cannot 'remove' construction by choice of subject, we can vastly improve the contextualisation of a visual or aural corpus to an extent approaching that of a written corpus. This will not satisfy all critics, but it is a clear qualitative improvement.
Bateson, G., & Mead, M. 1942. Balinese character, a photographic analysis New York: The New York academy of sciences.
Banks 1994 Interactive Multimedia and anthropology - a sceptical view <http://WWW.rsl.ox.ac.uk/isca/marcus.banks.01.html>
Biella 1994 Codifications of Ethnography: Linear and Nonlinear <http://WWW.usc.edu/dept/elab/welcome/codifications.html>
Banks, M., & Morphy, H. (eds.) 1997. Rethinking Visual Anthropology. London and New Haven: Yale University Press.
Biella, P., Chagnon, N. A., & Seaman, G. 1997. Yanomami Interactive. The Ax Fight (Case Studies in Cultural Anthropology Multimedia Series) New York: Harcourt Brace.
Brennan,E. R. and Relethford, J. H. (1983). Temporal variation in the mating structure of Sanday, Orkney islands Ann hum Biol. 10: 265-79.
Blacking, John 1984. What Languages Do Musical Grammars Describe? In M. Baroni and L. Callegari, (eds.), Musical Grammars and Computer Analysis , pp. 363-70. Florence: Leo S. Olschki.
Clifford, J., & Marcus, G. E. 1986. Writing cultures. The poetics and politics of ethnography Berkeley: University of California Press.
Colby, K. 1967. Computer Simulation of Change in Personal Belief Systems. Behavioral Science 12(3), 248-253.
Coult, A.D. & R.R. Randolph. 1965. Computer methods for analyzing genealogical space. American Anthropologist 67, 21-9.
Colclough, Nevil and Bagg, Janet 1999 The Ascoli Project: A Puglian town and its hinterland. Electronic document available from <http://lucy.ukc.ac.uk/jb6/archives/ascidx.html>.
Collier, J. J. and M. C. 1986. Visual anthropology: photography as a research method Albuquerque: University of New Mexico Press.
Crane, G. 1991. Composing Culture. The authority of an electronic text. Current Anthropology 32(1), 293-311.
Davis, J. 1987. Zuwaya social organisation. In Facsimiles from the Libyan Sharis-ucdsa archives (ed.) A. Layish. Jerusalem: Hebrew University.
Fischer, M.D. 1986. Expert Systems and Anthropological Analysis. Bulletin of Information on Computing and Anthropology 4, 6-14.
Fischer, M.D. 1994. Applications in computing for social anthropologists (ASA Research methods in social anthropology). London: Routledge.
Fischer, M.D. and W.L. Lyon 1994. Tradition and Symbol in the Cook Islands, CSAC Ethnographics Gallery, (on the World Wide Web) http://lucy.ukc.ac.uk/tradition.html
Gilbert, J.P. & E.A. Hammel. 1966. Computer Simulation & the Analysis of Problems in Kinship and Social Structure. American Anthropologist 68, 71-93.
Goody, J. R. 1976. Production & Reproduction Cambridge: Cambridge University Press.
Hammel, E. A., McDaniel, C. K., & Wachter, K. W. 1980. Vice in the Villefranchian: A Microsimulation Analysis of the Demographic Effects of Incest Prohibitions. In Genealogical Demography (ed. B. D. a. W. Morrill. New York: Academic Press.
Kippen, J. 1986. Computational Techniques in Musical Analysis. Bulletin of Information on Computing and Anthropology 4, 1-4.
Kunstadter, P. 1963. Demography & Prescribed Marriage. Am. J. Phys. Anth. 21, 511-519.
Landow, G.P. (ed.) 1992. Hypertext. The convergence of Contemporary Critical theory and Technology. Baltimore: John Hopkins University Press.
Levison, M., Ward, R. G., & Webb, J. W. 1973. The Settlement of Polynesia: a computer simulation Minneapolis: University of Minnesota Press.
Loizos, P. 1993. Innovation in ethnographic film: from innocence to self-consciousness, 1955-1985 Manchester: Manchester University Press.
Macfarlane, A. 1990. The Potential of Videodisc in Visual Anthropology. CVA Review Fall
Macfarlane, A., & Gienke, M. 1989. The Principles Used in Selecting, Editing and transferring Materials for an Archival Videodisc. Journal of Educational Television 15(3), 131-141.
Read, D. W., & Behrens, C. A. 1992. Computer Representation of Cultural Constructs: New Research Tools for the Study of Kinship Terminologies. In Computer Applications for Anthropologists (ed. M. S. Boone & J. J. Wood). Belmont: Wadsworth.
Rheubottom, D. B. 1988. 'Sisters first': betrothal order and age at marriage in fifteenth-century Ragusa. Journal of Family History 13(4), 359-376.
Rouse, R. H., & Rouse, M. A. 1982. Statim invenire. Schools, Preachers, and New Attitudes to the Page. In Renaissance and Renewal in the Twelth Century (eds.) R. L. Benson, & G. Constable. Oxford: Clarendon Press.
Ruby, J. 1995. Out of Sync: The Cinema of Tim Asch. Visual Anthropology Review 11, 19-37.
Also available online: http://astro.temple.edu/~ruby/ruby/sync.html
Ryan, N. S., 1995 The excavation archive as hyperdocument?. In J. Huggett and N. Ryan, (eds), Computer applications and quantitative methods in archaeology, number 600 in BAR International Series, pp 211-220, Oxford: Tempus Reparatum.
Schwimmer, B. 1996. Anthropology On the Internet - a Review and Evaluation Of Networked Resources. Current Anthropology 37(3), 561-568.
Sheehan, E.A. 1993. The academic as informant: methodological and theoretical issues in the ethnography of intellectuals. Human Organization 52, 252-259.
Spencer, J. 1989. Anthropology as a kind of writing. Man 24(1), 145-164.
Strathern, A., & Strathern, M. 1971. Self-decoration in Mount Hagen London: Duckworth and Co.
Werner, Oswald and Lauren Clark (1998). Ethnographic photographs converted to line drawings Field methods. 10:3 pp 54-6
Wilson, A. J., & Zeitlyn, D. 1995. The Distribution of Person Referring Terms in Natural Conversation. Research on Language and Social Interaction 28(1), 61-92.
Wright, C. (1998). The third subject: perspectives on visual anthropology Anthropology today. 14(4) 16-22.
Zeitlyn, D. 1993. Reconstructing Kinship or the pragmatics of kin talk. Man (n.s.) 28(2): 199-224. NB An electronic version of part of this article including digitised sound recordings is at <http://rsl.ox.ac.uk/isca/mambila/mambila.html>.
Zeitlyn, D., & Houtman, G. 1996. Editorial: Information technology and anthropology. Anthropology today 12(3), 1-2.
Zeitlyn, David mss.Talking Mambila Kinship.
1 Now available as an interactive CD: Biella, Chagnon, & Seaman, 1997
2 See Zeitlyn and Houtman 1996 and Schwimmer 1996 for summaries.
3 Keyword searches suffer because the same topic may be discussed using different terms (the problem of synonymy) and the same term may be used in the discussion of very different topics (the problem of homonymy).