Since what I will be dealing with in this paper will be concerned with communication, and with the signs and messages through which communication occurs, it might be fruitful to start by examining the title of this paper as a possible communicative event composed of a sequence of sign units (Note 1). Our first task ought to be a preliminary analysis of some of the signs I am using. (I am arbitrarily assuming at this point that I know that individual words are signs.) I will try later on to suggest that analysis of similar sign units might be necessary in the development of a semiotic of ethnographic film.
By a semiotic, as used in this paper, I mean roughly: a science, or a doctrine, of signs that will help to explain how human beings communicate with one another within a shared system, exchanging units that have common significance and a commonly understood set of rules of inference and implication. This is not a standard definition, since most uses of the term semiotic include signs and sign systems employed by nonhuman organisms and include signs that do not, or may not, have a communicative function or context. I would like for the moment to concentrate on signs used in human communication.
By ethnographic, I mean broadly the study, description, and presentation
of the customs and ways of people all over the world. By a semiotic of ethnographic film, I mean a method by means of which we can study the signs, and the rules of implication and inference, that we employ when we use those signs in films that are intended to describe and present the customs and ways of people all over the world.
An ethnographic film, then, is something that we employ for some particular purpose. It is a kind of film that its makers or viewers use for the study, description, or presentation of people and culture. Its users have by agreement attached some of the concepts of a discipline labeled "anthropology" to their use of film in this way and generally believe that there is some relationship between the goals of that discipline and "ethnographic film."
There can therefore be no way of describing a class of films as "ethnographic" by describing a film in and of itself. One can only describe this class of films by describing how they are used and assigning the term ethnographic to one class of descriptions. Just as one cannot say that any sign--the arbitrary word sign dog, for example--has a significance in and of itself, one cannot say that any film has a significance or meaning in and of itself. For any film, just as for the word dog, there must exist a common and shared significance among the members of a group who make implications and inferences or draw meaning from the use of a sign in a process of communication.
Just as words are signs of, and in, a speech event, so are visual images parts of an image-event. An image-event, however, is much more difficult in some ways to deal with than the speech events that occur in verbal language. We do not have a lexicon of images, as we do of words, by which we may check on a culturally agreed upon signification. To some extent, when we use visual images, we depend not on arbitrary sign meaning but on an expected common set of perceptual mechanisms and a common set of rules by which we perceive and organize the world. We are, however, learning that this set of rules by which we perceive and organize the world is as much dependent on our culture as on some set of built-in perceptual mechanisms. The Whorfian notion that the language we speak determines to some extent how we see the world around us must be considered quite literally when applied to film, and most particularly when applied to ethnographic film use. Although image-events appear in no lexicon that has been bound and stacked in libraries, differing cultures use, organize, and imply and infer meaning from image-events in differing ways.
If, then, no single film can be classed as ethnographic by looking at it, in and of itself, it follows that any film might become an
ethnographic film because of its purpose or its use. It would seem, therefore, that in order to know something about ethnographic films, we must examine not the films, primarily, but why they are made and how they are used.
What I am suggesting is the obvious fact that it is not enough for our purposes to study the film, or even a film code itself. In order to know about ethnographic films, we must study the code within some specific functional context. In our case, it will do us no good to study film qua film. We must begin to determine the relationship between a film code and its context not only within ethnographic research but within the culture or the subculture that both produces and uses a film.
To develop this point further, I would like to borrow a question formulated
first by Sapir in 1938 and called to our attention by Hymes in a paper entitled
"Why Linguistics Needs the Sociologist" (1967). It might become
interesting if we were to challenge our interest in ethnographic film by
asking why the filmmaker needs the ethnographer, or even why the ethnographer
needs the filmmaker. In one sense, of course, the relation between these
two disciplines can be of trivial intellectual interest, even though of
great practical desirability. It might be thought of as analogous to asking
why the anthropologist or the linguist needs a sound-tape-machine operator.
The anthropologist in the field might need a similar technician to perform
a visual recording service for him. "Make me a picture of that dance,"
he would say. Conversely, the filmmaker might want to make a film about
Eskimos and would expect the anthropologist to act as a sort of straw boss
in the field. "Ask those guys what they're doing, and do you think
it's O.K. for me to say on the soundtrack that Eskimos love their dogs?
Do you think that would be O.K. for the Board of Education?"
Unfortunately, such things have occurred. I am not talking about that kind of need. To clarify the original question, and to make more precise the area which is of significance intellectually, I should rather ask why the person interested in studying ethnographic film needs the ethnographer or why the ethnographer needs a person interested in the study of film.
Let us, for a moment, look at how film has been used by anthropologists
in the past. I will not detail this history (see Ruby 1971), but would rather
talk about types or categories of films used in a specific intellectual
community--that of anthropology. Films may, of course, still be examined
in terms of their own structure, but
if the use of the film determines its labeling as "ethnographic," it is its use that must first be established.
An ethnographic film is a set of signs used to study the behavior of a people. Films can be used in two basic ways to achieve this purpose: as a recording of data about</> culture, or as data of a culture.
First, a film can be made or examined as a particular form of data retention, collection, and transmission. Such a film records the basic data of culture in and within the cognitive system and value structure of the data collector. It is similar in this way to the linguist's use of the tape recorder. He chooses his speaker according to a determined set of rules developed by his discipline and saves memory storage by having a machine collect a particular set of sounds for analysis. In a like manner, one can choose a "chunk" of behavior and have the camera (with or without synchronous sound) record it. We can use that piece of film to save memory storage, or we can use it to observe units of behavior that are not visible at normal speed. Normally we do not see in units of 1/24 or 1/1000 of a second. The movie camera can record these chunks of behavior in frames passing before a lens at any given speed. We, however, point the camera, determine the speed at which the film passes behind the lens, and determine what we will view, and at what speed we will view it.
Second, films can be thought of as a phenomenon of culture in their own right, reflecting the value systems, coding patterns, and cognitive processes of the maker. Here the ethnographer is interested in what has been called the "ethnography of communication," and studies not only the behavior recorded on the film, but primarily the behavior of the man who organizes image events on film. Such use is analogous to that of the anthropological linguist or the sociolinguist who wants to know what, why, and how specific people in specific contexts say or do not say specific things, and how their utterances are related to other aspects of their culture, and to the culture of peple who live differently. Here one is interested in what things are said, why, to whom, and in what form.
Let us take these two different funtions and try to see what we need to
know in order to use film in these ways.
If one is interested in film only as a data bank, recording the behavior of an informant, one need not concern oneself with how the informant would organize the film. One must, however, be clear about how the ethnographer himself organizes the behavior he records. The notion that the film is in some way an objective record capturing some elusive "truth" must be recognized as the nonsense it is.
78 --- Studying Visual Communication
I believe this notion, implicitly held and sometimes explicitly stated, has done more damage to ethnographic research than any other "contribution" made about the use of modern technology. It has done damage not only because it is wrong--I can conceive of some very wrong ideas having quite positive effects in a research situation--but because it has prevented researchers from examining their own methods of organizing and structuring the visual events that are recorded in the field and subsequently analyzed and described as the behavior of an informant or group.
The obvious problem here is that--as Margaret Mead and others have pointed out so often--we forget that other things are happening when the camera is pointing in one direction, or that other things are happening outside that frame. We also tend to forget that all films are edited by someone. Even "unedited" research films are edited. Someone placed the camera and started, stopped, and started the camera again. Someone decided to take a close-up or a long shot. Someone decided what events were worth recording and what events were of no significance and not worthy of being recorded. In effect, someone acted as a human being making research decisions in the field.
Anyone who has ever analyzed a film can testify to the almost insidious power of the medium. We tend to believe that what is on the screen is--and is thereby some act of higher truth. Once one has looked at a film of an action or an interaction a number of times (and in some forms of microanalysis, three hundred times is not a lot) one is hard pressed to remember that what one has seen is not what happened, but what the cameraman and camera put on his film.
There is, however, no point in taking the position that if film is not objective truth, there is no use to it. I am arguing that there is great value in visually recorded data of culture--so long as we know what it is that we recorded, so long as we are aware of how and by what rules we chose our subject matter, and so long as we are aware of how we organized the various units of film from which we will do our analysis.
If anthropologists are to collect data on film and to be trained to do this, it is necessary for them to conceptualize clearly the mode and specific methods of analysis that they will use, rather than to concentrate on the technology required to get a clear image on film. If Navajo Indians (Worth and Adair 1972) can learn to capture a clear image on film in two days of instruction, we ought to assume that anthropologists studying filming technique can do as well with a week's instruction, and that professors should spend time teaching something which cannot be picked up from an instruction book or from a day's instruction.
79 --- A Semiotic of Ethnographic
Film
What I am suggesting is that a semiotic of ethnographic film concern itself, on this data-collection level, with the study of three things. First, a description of existing analytic methods that are used in analyzing behavior recorded on film. Second, a description of the alternative ways of organizing the film image itself, and third, the development of new methods of analysis tied to ethnographic and anthropological theory.
The first concern, the description of existing analytic methods, is relatively simple because they are so few, and those few are fairly well developed (Birdwhistell 1971; Lomax 1970; Hall 1968; Ekman and Friesen 1969). At present, however, there are very few people trained in even this small variety of analytic methods. Achieving this first goal is a problem of educational organization within a multi-disciplinary field combining parts of anthropology, communication, and film.
The second concern, a description of alternative ways of organizing film, is a more difficult problem. There are at present few film schools in this country, few schools of communication, and almost no anthropology departments that I know of that are seriously investigating the ways human beings with differing cultures organize visual experiences when they record them on film.
The third concern, that of developing new methods of analysis tied to anthropological problems and theories, seems to me to be the most crucial. If we could make headway here, we would be able to integrate the study of ethnographic film into anthropology as part of a discipline.
Let me delay for a moment a fuller explanation of this point. I intend to come back to it after describing briefly the second of the two ways I mentioned earlier in which film can be used in the study of culture.
This second way in which film can be used is relatively new in some dimensions,
but not in others. Here I am referring to the study of film as a phenomenon
of culture itself, rather than its study as an anthropologist's recorded
data of that culture. Here we would analyze films for much the same reasons
that we analyze verbal language, methods of farming, child-rearing practices,
and so on.
Using films this way, or calling films that we study for this purpose "ethnographic," demands not only some of the same skills and conceptual clarity that were discussed earlier, but also the integration of some fundamentally new problems into existing theory, both in communication and in anthropology.
Looking at a culture through the films made in that culture poses some interesting questions. Wolfenstein (1953), Weakland (1966),
80 --- Studying Visual
Communication
Mead and Metraux (1953), Gerbner (1972), and many others have engaged in what Mead has called "the study of culture at a distance." They have made many assumptions which, so far as I know, have not yet been sufficiently clarified. Basic to much of this early work was the assumption that certain films--those made for mass audiences, primarily--reflect the "daydreams" of the culture that produced them. These films are then sampled and analyzed to ascertain basic values and interests of the culture in question. In many cases, such work yields fruitful insights. Yet some of the assumptions are much too simplistic. It is true that, analogously, the linguist taking an interest in problems of competence only can examine the utterances produced by such diverse groups as political figures, nightclub comics, or street-corner venders. But as long as the theoretical structure behind such analysis does not require one to differentiate among these speakers, and seeks only to deal with that part of language that they all share, it does not matter. If, however, the analysis has as one aim to compare value systems, "national daydreams," or traits in a body of films, and does not include an analysis of how the films were made, under what cultural rules, by what groups, for what purpose, the conclusions are bound to be difficult to substantiate.
For example, it is clear now (Guback 1969) that films made in the West (United States, England, France, Italy, Spain) are not so much national expressions as supernational products, designed, financed, and produced to appeal to as wide a group of viewers in as many countries as possible, by a multinational or supernational organization. It is almost impossible to single out a recent British film, for example, that was not financed and approved by American interests with multinational audiences in mind. The same holds true of the so-called Iron Curtain countries, whose films are financed by the Soviet Union and produced for much more than internal consump- tion.
It is important to realize that "films" are not all the same for ethnographic use, either. Certain films are made by individual mem- bers of a culture, but even here, as in "avant-garde" or "art" films, it is important to know which values and ideas are paramount for the filmmakers. Other films are made by members of a culture in a more ethnographic sense. But these have rarely been studied (Chalfen 1975). For example, "home movies" are made for showing to family and friends--for private, rather than public viewing--and probably reflect more of what we mean when we talk about how members of a culture organize their world through image-events than do the commercial infranational films. Recently, high school, college, and
81 --- A Semiotic of Ethnographic
Film
independent groups such as Indians, blacks, and Chicanos have been making films for their purposes in their ways.
Once one realizes that films are not made by a culture but by individuals or small groups within a culture or sharing some common culture, one is faced with a problem that has long preoccupied researchers interested in the complex relationship between the use and performance of verbal language, and culture.
In developing a semiotic of ethnographic film on the level of film as a datum of culture, the problem of determining how film signs are used by individuals according to rules becomes paramount.
Here we would be asking questions that have been asked about other modes of communication--speech, of course, but also about painting, carving, architecture, and so on. These might be considered questions of form or, using linguistic terms, questions of syntax, as well as content, or semantics. In analyzing ethnographic films in this way, we might want to ask such questions as: Does film have the properties of a language, and if so, what are they? Are there language communities of filmmakers and film viewers? Does film have the same function for all communities? and so on.
The decision to concentrate for the time being on developing a semiotic rather than a grammar was made because it seems unwise at this state of ignorance to prejudge whether film communication should be considered a language in a formal and serious sense. The notion of a semiotic allows for the discovery of rules of the linguistic type for film organization, but does not preclude other, less formal, less commonly understood, and perhaps different, patterns of use.
For ethnographers, the study of film as another way to investigate the cognitive organization of different peoples opens a vast new area. Since people have divided themselves into different linguistic groups, those studying culture through language have been forced to surmount the very difficult translation barrier. Descriptions of the attitudes, customs, and way of organizing experience rendered by the members of the studied culture are always filtered through the ethnographer's eyes, ears, and cognitive system. He must translate the myths, redescribe the customs, and talk secondhand about how the "native" sees his own world and his experiences in it.
Films, just like still photographs and paintings, can be made by the native people in question, and, what is more, it may be the case that we can understand them as the native does. Notice the qualification--I said, "It may be the case."
This seems to be the first job that those interested in a semiotic of ethnofilm must be prepared to tackle. What are the implications that the maker of a film puts into his film, and what are the inferences
82 --- Studying Visual Communication
that a viewer infers from it? Are there universal patterns of inference, of narrative forms, of image use, that are of common, shared signifi- cance across all cultures, or across certain groups of cultures? Are there certain structures of film communication which are related to other patterns of culture, such as verbal language, kinship patterns, and so on?
Not only do we have to know how "natives" organize their visual perceptions, but we must study the anthropologist's own sys- tem of organizing the films he makes. Does the language that he, the ethnographer, speaks make for a particular system of implication or inference? Does the anthropological theory that he is working with, which the film, as data, is meant to elucidate, influence the way in which he sees the world that he is trying to describe?
Leonard Bloomfield (1964), talking of White-Thunder's use of Menomini, implies that "Menomini is a language no one speaks tolerably" (Hymes 1967:636). About which Hymes comments
White-Thunder forces us to face the fact that for both the individual and the community, a language in some sense is what those who have it can do with it. . . . Differences in facility and adequacy may be encountered that are not accidental but integral to the language as it exists for those in question. [Ibid.]
We must know not only the differences in facility and adequacy among
different users of film, but if Navajos, Negro teenagers, or Trobriand Islanders
make films, what must we know about their methods of patterning these sign
units for us to use these films as data of culture? Using such films not
only gives us the possibility of seeing and analyzing what the other person
thinks important to show us--not only do we have the opportunity to determine
the patterns by which he organizes what he shows us, but, as with language
itself, we can see what he does not show and how he does not organize. In
other words, negative cases count, and nothing never happens.
When I talked earlier about the need to integrate a theory of ethnofilm
with theories in anthropology, I meant the kind of integration which would
be useful in tackling some of the problems briefly and cursorily mentioned
above.
The mechanical amalgam of a film theory not concerned with ethnography with an anthropology not concerned with ethnofilm problems will not do. The kind of training in which anthropologists learn about Eisenstein's theory of editing and filmmakers learn some basic anthropology will produce perhaps filmmakers who know a little anthropology, and anthropologists who know a little about how
83 --- A Semiotic of Ethnographic
Film
to make films, but will not contribute much to the development of problems whose solution can be integrated into a scientific theory of culture.
Rather than attempting to list a set of problems appropriate to this kind of analysis, let me give one example of the kind of problem for which an adequate theory must be developed if there is to be such a thing as a semiotic of ethnofilm.
Whorf and others have advanced the notion that the language one speaks influences the way one perceives and organizes the world. Let us state as a hypothesis that the way one organizes units of a filmed event depends only on the way one organizes verbal descriptions of that event. (I do not think this is true, by the way, but it is a testable hypothesis.)
Obviously, if this were true, it would be extremely important to both anthropological theory and to film theory, yet neither group can at present deal with the complexities of the problem. It would, in order to test this hypothesis, be necessary to develop a descriptive system of film organization similar to grammar and syntax that can be compared with other relevant dimensions known to facilitate ethnographic description.
It would be necessary to examine much more closely how indeed the use of verbal language, in narrative, mythic, and everyday events, relates to what we normally call the grammatical or syntactic. At the same time, we would have to examine how film "narrates," "tells stories," and deals with everyday events, in a way that could relate that to the formal, structural, or syntactic codes of film use. The concept of "Language and Culture" as a label for the relationship of culture to modes of communication is clearly too narrow. When we make films, paint pictures, carve doorposts, dress, set our tables, and furnish our homes, as well as when we speak, we are using symbolic forms which are part of culture and which are all possibly related. A semiotic of ethnographic film is part of the study of the relationship between culture and communication itself.
Unfortunately, I have not had time to present the above arguments as fully as I would have liked to. Let me sum up, however, by restating some of them and indicating some of their perhaps controversial implications.
(1) There is no such thing as an "ethnographic film"; or, to put it another way, any film may or may not be an ethnographic film, depending on how it is used.
(2) Learning how to make films as films, with the emphasis on the technical or artistic aspects of the medium, is not relevant to anthropology. Learning how to study about films in relation to some
84 --- Studying Visual Communication
specific anthropological problem is relevant. At present, this kind of study is almost nonexistent.
(3) Whether those interested in analyzing culture through film use or define ethnofilm as their record of their observation or as the native's own way of seeing things, a similar body of knowledge about ethnofilm analysis is required. I am suggesting that this knowledge must include: a) a description of alternative ways of manipulating film; b) a description of alternative ways of analyzing film; and c) the development of a body of theory, or at least the identification of problems relevant to existing theory that the analysis of ethnofilm will be designed to solve.
(4) A semiotic of ethnofilm should be concerned with questions that have broad disciplinary interest and that stem from new problems generated by the congruence of film analysis and the analysis of culture. In other words, a semiotic of ethnofilm, if it is worth getting involved in, must become part of the science of culture and communication, and the questions and problems it deals with must be those that the larger discipline cannot help getting interested in ãnot only because films are fun, but because film studied in this way offers answers to questions that concern us as professionals.
(5) Fifth, but by no means least important, is the fact that training students in this new area is extremely difficult to do at the moment. Some universities have instituted ethnographic film services, some are experimenting with joint courses run by filmmakers and anthropologists. Various plans are afoot, but, in my opinion, we have just begun. The problems have hardly begun to be clarified, and training in this area is still not accepted as necessary for a degree in film, communication, or anthropology.
Although the interesting questions are the research questions, I would like to emphasize the pedagogical implications of such interests. If culture and communication are worth studying, the entire field must determine their relevance and the methods of training which will enable those entering the field to pursue these problems. Just as the study of language has not been left to the formal linguist alone, so the development of a semiotic of ethnofilm cannot be left only to the filmmaker or only to the ethnographer.
1. This paper appeared in the Program in Ethnographic Film (PIEF) Newsletter 3 (1972):8-12. An earlier version was presented at the 1968 annual meeting of the American Anthropological Association--Ed. Return.
©1996 Tobia Worth. All Rights Reserved.