(Reprinted Chapter 4 in A Crack in the Mirror: Reflexive Perspectives in Anthropology, Jay Ruby, editor. Philadelphia: University of Pennsylvania Press (1982). Original pagination preserved for citations purposes.)
My new conception of the film is based upon the idea that the intellectual and emotional processes which so far have been conceived of as existing independently of each other-art versus science-and forming an antithesis heretofore never united, can be brought together to form a synthesis on the basis of CINEDIALECTIC, a process that only the cinema can achieve. A spectator can be made to feel-and-think what he sees on the screen. The scientific formula can be given the emotional quality of a poem. And whether my ideas on this matter are right or wrong, I am at present working in this direction (from a speech given by Sergei Eisenstein at the Sorbonne -quoted by Brody 1930).
The sentiment expressed in this quotation exemplifies the issue I wish to explore in this essay. Eisenstein wanted to develop a cinematic form that would imbue content with a Marxist ideology. He was not the first or last Marxist to hypothesize that the medium had this potential. From Dziga Vertov to Jean-Luc Godard, many filmmakers have tried to create a revolutionary cinema. (2) In an analogous manner, anthropologists since 1896 (e.g., Regnault) have advocated a visual anthropology (cf. Ruby 1975 for a more detailed discussion of the relationship between a revolutionary and an anthropological cinema). Neither has met with much success.
(1) Some of the ideas presented in this paper have appeared in a more extended form in earlier publications. See Ruby 1976, 1977, 1980.
(2 ) Cinema is used here to stand for all the socio-cultural processes and events surrounding the production and consumption of film.
In this essay I would like to convert Eisenstein's conception from one of unwieldy magnitude to a manageable problem. I will therefore seek to explore the following seemingly paradoxical question: Why is it that anthropologists were among the first social scientists to examine the potential of the motion picture and yet a visual anthropology has played only a minor role in the development of anthropological thinking?
It seems to me that there are two possible explanations for the paradox: (1) There is something "inherently" unscientific about pictorial media, that is, film is, by its very nature, an art form; or (2) Our culturally conditioned assumptions have prevented us from exploring the potential of these media.
While it may be commonplace to talk about the "art" of the film, to restrict film to being exclusively an art form, that is, to insist that audiences pay primary attention to the syntactics of a film as a sign event, is premature because there is no evidence to support the position. To make an analogy, it is equivalent to restricting writing to poetry. On the other hand, most people do have aesthetic expectations when they view a film-regardless of its stated intent-and these expectations do pose a problem for the anthropological filmmaker.
Furthermore, our cultural assumptions create additional problems. These assumptions are situated within the folk models of art and science that pervade our culture and serve as the basis for the two most common film theories-Realist and Formative (cf. Andrews 1976 for a more detailed discussion of film theory). Given the alternatives of regarding film either as an expression of emotion, feeling, and art (The Formative) or as a surrogate for reality and the most accurate means for data collection (The Realist), most anthropologists have chosen to use the printed word to communicate their ideas. Hence the lack of an anthropological cinema and the virtual absence of professional filmmakers who are also professional anthropologists.
These culturally conditioned interpretive strategies for making sense out of a film-one strategy leading us to make inferences about film as art, as aesthetic object, and as fictionalized fantasy designed to amuse us, and a second causing us to deal with film as a document of reality that should be unbiased, objective and truthful-are at odds with what might be called a fundamental purpose of an anthropological communication, namely, to make scientific/humanistic statements about culture.
The strategies have their origins in the basic Western idea of the dichotomy. This dichotomy is formulated in pairs like art and science; mind and body; thought and feeling; cognition and emotion; and objectivity and subjectivity. It produced the opposition of realist versus formative in film theory and leads most people to assess films as being understandable either as fiction, that is, made-up fantasy, or as documentary, that is, real and truthful.
Stated in the terminology suggested by Worth and Gross (1974), films are interpreted either as natural sign events (3) that are assumed to exist with meanings assigned on the basis of attribution, or as symbolic articulations created so that meaning can be inferred from them.
To make an attribution, one makes an assumption that a particular sign-event is natural and existential, that is, the sign-event just happened to be there. Meaning is then assigned to such a sign-event by attributing to it characteristics associated with similar sign-events or with sociocultural or psychological stereotypes of such events. For example, one is shown a photograph of an oak tree and responds to it by saying, "That's a beautiful tree." If asked why, and the response is, "It's beautiful because it's an oak tree, and oak trees are beautiful," then the viewer is attributing to the photograph what he or she knows and feels about oak trees. I he interpreter, in this case, is interpreting not a photograph but a tree.
To make an inference about a sign-event, one begins by making an assumption of intention. One assumes that someone created the sign-event; that it was organized, structured, put together on purpose. One can discuss the purpose but only after the assumption of intention is granted. Meaning is assigned on the basis of a variety of socio-cultural conventions assumed to apply to the particular structure one recognizes. Not only does one infer from existing conventions of structure about persons, places, and events in general, but one uses the particular structures that apply to particular media and modes of communication. To return to the photograph of the oak tree, if one says that the photograph is beautiful because the photographer back-lit the tree, and back-lit trees are supposed to be beautiful or "art," then one is making an inference. Note that the viewer making the inference from a photograph doesn't have to guess whether or not the photographer thought backlighting
(3) Sign-event is used here to mean an organized group of signs-signs that are syntactically related and clearly delineated or framed in a way that sets them apart from other sign-events.
was beautiful. He knows the convention and assumes an intentional backlighting.
It should be emphasized that this model attributes no intrinsic meaning to the signs themselves. They are polysemic. The perceiver assigns meaning either by assuming existence and therefore attributing meaning to the sign-event, or by assuming intention and inferring meaning from the sign-event.
For our purposes, this suggests that films are interpreted either as natural sign-events that are assumed to exist and have meaning assigned on the basis of attribution, or as symbolic articulations created so that meaning can be inferred from them.
The range of inferential and attributional paradigms currently employed in our culture to assign meaning to films appears to be limited. I have suggested elsewhere (Ruby 1976) that when we infer from a film, we employ the conventions of art and deal with a film as an aesthetic object, that is, the sign-events are regarded as having primarily a syntactical significance. When we attribute to films, we assume they are "documents of reality" interpreted in the same way we would interpret the reality the films are thought to mirror.
The film as an aesthetic object need not be discussed at length. It should be obvious that the canons of science and not art are assumed to govern an anthropological communication. Therefore, if the most prevalent inferential system available for the construction of meaning for film is based on aesthetic criteria, we must look elsewhere, since a paradigm of science is more appropriate for an anthropological film.
If we dismiss an inferential strategy that causes us to regard film as art, we are led to an attributional strategy that causes us to see film as a record of the real world. In this schema a viewer attributes to the film what he or she knows about the events, objects, and persons depicted. The film as document becomes transparent-a mere conveyer of the content. Many documentary and ethnographic filmmakers thus assume that meaning resides in the world, that events and people can speak for themselves (that is, communicate meaning through the film without the interpretive aid of the filmmaker), and that the role of the filmmaker is to unobtrusively record this reality.
Employing attributions to understand an anthropological film leads viewers to meanings antithetical to anthropology for two basic reasons: (1) the use of attributional systems is based on a theory of perception
counter to the idea that culture organizes experience; and (2) the folk models underlying attributions are ethnocentric.
The belief that film can be an unmediated record of the real world is based on the idea that cameras, not people, take pictures and the naive empiricist notion that the world is as it appears to be.
The former concept, already examined by Byers (1966), reflects a profound navieté about the physio-chemical process of picture taking. The latter and more fundamental supposition presupposes a theory of visual perception called phenomenal absolutism that has been discredited at least since the work by Segall, Campbell and Herskovits (1966) on the effect of culture on visual perception. It is not feasible to present a detailed refutation of these positions here. I will simply assert that the camera creates a photographic realism reflecting the culturally constructed reality of the picture-taker and is not a device that can somehow transcend the photographer's cultural limitations. We cannot capture reality on film, but we can construct a set of images consistent with our view of it.
The argument against making attributions has a second thrust namely, that the folk models available for the construction of these attributions are ethnocentric. Most anthropological studies deal with cultures foreign to the experience of the viewer. The characteristics commonly attributed to exotic people are based on either the folk model of the Noble Savage-the Rousseauian, invincible, ignorant, natural man-or the Primitive-underdeveloped, culturally deprived, illiterate beast who is sorely in need of the benefits of civilization. Both models are obviously inappropriate.
Is the logical conclusion, then, that it is impossible to generate anthropological meaning from a film? Are anthropologists wasting their time taking pictures? If that were my conclusion, I would have not written this essay. One solution to the quest for an anthropological cinema may be found in the general shift in cultural attitudes toward being publicly self-aware or reflexive; the recognition of the social construction of reality; and the associated changes in science, anthropology, film, and communication.
I have suggested elsewhere (Ruby 1980) that being reflexive in public has now become respectable. I believe there is a growing realization that the world is not what it appears to be, and that what you don't know will and often does hurt you. People now want to know who
made it and what's in it before they buy anything-aspirin, cars, television news, or education. We no longer trust the people who make things to be of good will. Ralph Nader, the consumer protection movement, laws requiring financial disclosures by political figures, truth in lending and truth in advertising laws are all part of this felt need. The naive empiricism that pervaded our society and dominated nineteenth-century social science is being eroded. We are moving away from the positivist notion that meaning resides in the world, and that human beings should strive to discover the inherent, immutable, and objectively true reality (Stent 1975). We are beginning to realize that human beings construct and impose meaning on the world. We create order. We don't discover it.
The consequences of these changes are far reaching and interrelated and can be seen in science in general and anthropology in particular. They are reflected in the public acknowledgment of the scientist's role in the process of scientific investigation; the limitations of science as an epistemology; and the need to explicate clearly the methods employed in any scientific inquiry and to see them as a process integral to the products of science.
A convenient marker for this change is the 1962 publication of Kuhn's The Structure of Scientific Revolutions. Kuhn recognized that scientific knowledge is the product of a particular paradigm, and that science changes through the process of discovery of the inadequacy of the old paradigm and the subsequent construction of a new one. Labrot has said (1977:7):
Science is not static. Its development is determined to a great extent by the body of science as it stands at any given moment. This determinism is not one of a natural progression to a greater and greater number of known facts built on those previously discovered. It is rather one in which the fundamental principles, the structures in a broad sense, determine the nature of search for the facts and finally, to some extent, the facts themselves. So science which describes the world, also determined the world which it describes.
My position in this essay is that the paradigm of Positivism is insufficient to deal with questions now being asked. We need a new paradigm-in science in general and in anthropology in particular-one that will allow us to examine the symbolic environments (culture) people have constructed and the symbolic system (anthropology) we have constructed.
Within anthropology the foundations for this new paradigm already exist. I will mention only two of the sources: Clifford Geertz and Margaret Mead. From Geertz (1973a) comes the notion of anthropology as an interpretive science and ethnography as "thick description" in which data and theory cannot be separated; theory instead is regarded as the origin of data generation. (In other words, one regards data not as a property of entities but rather as an artifact of the questions one is researching.) From Mead (1976:908) we obtain a resolution of the science versus humanities conflict through the development of human science capable of accommodating both quantitative and qualitative knowledge:
It is in the sciences of living things that we find the greatest confusion but also the clearest demonstrations of the ways in which the two kinds of observation -human beings by human beings and physical nature by human beings-meet. One group of students of living beings have attempted to adopt as far as possible the methods of the physical sciences through the use of controlled experiments, the deliberate limitation of the number of variables to be considered, and the construction of theories based on the findings arrived at by these means. I he other group, taking their cues from our human capacity to understand through the observation of natural situations, have developed their methods from a natural history approach in which the principal reliance is on the integrative powers of the observer of a complex, nonreplicable event and on the experiments that are provided by history and by animals living in a particular ecological setting.... I would argue that it is not by rejecting one or the other but by appropriately combining the several methods evolved from these different types of search for knowledge that we are most likely in the long run to achieve a kind of scientific activity that is dominated neither by the arrogance of physical scientists nor by the arrogance of humanists who claim that the activities which concerned them cannot meaningfully be subjected to scientific inquiry.
These ideas are the foundation of an anthropology that is a humanistic and interpretive science of humankind, a science that accepts the inherently reflexive relationship between the producer, process, and product, a science founded on the idea that
Facts do not organize themselves into concepts and theories just by being looked at; indeed, except within the framework of concepts and theories, there are no scientific facts but only chaos. There is an inescapable a priori element in all scientific work. Questions must be asked before answers can be given. The questions are an expression of our interest in the world; they are at bottom valuations. Valuations are thus necessarily already involved at the stage when we observe facts and carry on theoretical analysis, and not only at the stage when we draw political inferences from facts and valuations (Myrdal 1969:ix-xvi).
Logically this point of view causes one to regard anthropology as "not only a general set of general statements about mankind, it is also the product of a particular culture with its history of ideas proper to itself; its formulations are culturally committed and in major part determined" (Krader 1968:885). When it is recognized that anthropologists ask research questions based on their overt theoretical positions and their less conscious cultural assumptions, and that when the questions are asked in a particular way, there is a logical way to generate data and an equally logical way to present the analytic descriptions called ethnographies, then the necessity of publicly disclosing the entire process becomes inescapable, and the phrase "reflexive anthropology" becomes redundant.
The implications of this general shift in consciousness for communication have already been partially discussed in terms of Worth and Gross' ideas about inference and attribution. They suggest that communication is "a social process, within a context in which signs are produced and transmitted, perceived and treated as messages from which meaning can be inferred" (1974:30). The implication is that all forms of human communication are motivated and ideologically based within the culturally conditioned expectations of what messages can occur in which contexts.
An application of this model to film produces some interesting results. If one regards film as communication, or more precisely, as an articulatory medium used for communicative purposes, it is possible to posit a range of filmic discourses-some of which are intentionally constructed to be regarded as art and others to be regarded as science. Given this perspective, the "film as art" model becomes transformed and contextualized as the aesthetic component of the communicative process, or, to put it in a semiotic framework, the syntactic elements of the sign-event.
If it is acknowledged that film has the potential for a variety of discourses, that it can have a variety of voices, the concepts of "film as art" and "film as documentary truth"-the foundations of Realist and Formative theory-can be seen not as a dichotomy but as having a dialectical relationship and, therefore, as a stage in the historical development of film. This false dichotomy of film as aesthetically satisfying experience versus film as the objective revelation of truth should be recognized as the product of a particular ideological structure called positivism.
Seeing these two approaches to film in a dialectical relationship makes it possible to construct the synthesis: a theory of film and the accompanying interpretive strategy that reflects the general change in self-awareness and the new paradigm of science discussed earlier.
The origin of the synthesis can be found in the re-emergence of the idea of film as a language. First proposed by Eisenstein in the 1930s, there has been a recent swell of interest in linguistic paradigms for interpreting film, probably the result of the growing popularity of structural linguistics and semiotics. In spite of the cine-structuralists' initial enthusiasm, semiotics is not producing the breakthroughs we had hoped for. If film is a language, it is unlike any other language known, since it does not respond to linguistic analysis.
Whether it is possible to construct a science of signs that is not so heavily dependent upon linguistic models-a semiotic that deals with all sign systems without making the automatic assumption of the primacy of language-remains unclear at this time. In any case, it is necessary to separate the idea of film as a communication system from the idea of film as a language. Language is only one variety of communication.
Film semiotics and cine-structuralism with all of their limitations do represent an attempt to break from the Realist/Formative dichotomy. Because this approach deals with film as a construction of culturally coded signs, it provides a basis for the development of the synthesis of Realist and Formative into a theory of film as communication, in which sign-events (films) can be organized to emphasize the syntactic (aesthetic), the semantic (informational), or the pragmatic (the call to action) elements.
Such a theoretical structure would allow for the construction of an ethnographic trompe l'oeil for film: the development of filmic codes and conventions to "frame" or contextualize the apparent realism of the cinema and cause audiences to "read" the images as anthropological articulations Once constructed, it will be possible to explore the consequences of transforming abstract thoughts, such as theories or models, into Images. The exploration can shed light on both the nature of anthropological thinking and the potential for images to communicate ideas.
I am obviously using the term trompe l'oeil metaphorically. According to Webster's Third International Dictionary, trompe l'oeil is the "deception of the eye, especially by a painting as (a) the intensification of the reality of component objects in an unnaturally arranged still life
through the use of minute detail and the careful rendition of tactile and tonal values . . ." (quoted in Mastai 1975:8-9). In a still life the painter attempts to create a tension between the aesthetic, or, as Webster puts it, "unnatural" arrangements of the parts of the painting-that is, the spatial relationships among the fruit and between the fruit and the bowl -and the painter's ability to realistically portray the fruit and the bowl. The painter strives to produce a temporary illusion that the apples are so real we could pick one up and eat it, while at the same time displaying his compositional skills. Bateson said it well, "Conjurers and painters of the trompe l'oeil school concentrate upon acquiring a virtuosity whose only reward is reached after the viewer detects that he has been deceived and is forced to smile or marvel at the skill of the deceiver. Hollywood filmmakers spend millions of dollars to increase the realism of a shadow" (1972:182).
The parallel with the anthropological filmmaker is striking. If we grant Susan Sontag's (1977) notion that the apparent realism of photographic reproduction is its greatest achievement and gravest danger, it can be argued that the anthropological filmmaker has to contextualize the realistic effect of film as merely an illusion by making overt the theoretical basis of the construction of the image. The tension between the indexical resemblance of the film sign-event to its referents, which causes people to attribute meaning to the film, and the ideological construction of the film, which causes people to infer meaning from the film, must be made overt, explicit, and unavoidable. It is essential that audiences understand the differences between the images we make of what people do and what people say they do, and what we interpret both to mean.
Audiences can have the pleasure of the illusion that they are participating in something they are actually watching so long as we make it very clear to them that they are seeing a representation we have constructed because we were motivated to present them with our view of the world. In an anthropological film we never see the world through the eyes of the native, but if we are lucky we can see the native through the eyes of the anthropologist. The beginnings of such a cinema can already be found in the films of the French anthropological filmmaker Jean Rouch, particularly in his African films-Jaguar, Petit a Petit, and Cocorico, Monsieur Poulet-in which anthropological interpretation is blended with folk explanations and fantasies in a way that defies the labels of fiction and documentary.
I have tried to suggest in this essay that the conditions are ripe for the development of a cinema of anthropology. It has become increasingly more common and acceptable to be self-aware and to be publicly concerned with integrating one's personal and social selves. We seem ready to accept the idea that science is a limited epistemology and not a religion. There IS a growing interest in the implications of reflexivity within anthropology. The boundaries between a positivist-based anthropology and humanistic and artistic expressive forms are being regarded as temporary historical stages rather than rigid barriers. We seem amenable to the idea that anthropology is at its best when we are telling the stories of our experiences to others. Since film allows us to tell stories with pictures, its potential becomes enhanced within a reflexive and narrative anthropology.
1976 The Major Film Theories. New York: Oxford University Press.
1930 A lecture on principles of the new Russian cinema by Sergi Eisenstein. Close Up 6(4).
1966 Cameras Don't Take Pictures. Columbia University Forum 9(1):27-31.
Segall, Marshall H., Campbell, Donald and Herskovits, Melville J.
1966 The influence of culture on visual perception Indianapolis: Bobbs-Merrill Co.,
1968 Person and collectivity: A problem in the dialectic of anthropoplogy. Transactions of the New York Academy of Secies. Series II, 30(6): 856-62.
Kuhn, Thomas S.
1962 The structure of scientific revolutions. Chicago: University of Chicaog Press.
1977 Two kinds of self-reflexiveness. Unpublished paper, Center for Humanities, University of Southern California.
Mastai, M. L. d'Ortange
1975 Illusion in art: Trompe l'oeil. New York: Abaris Books.
1976 Towards a human science. Science 191:903-9.
1969 Objectivity in social research. New York: Panetheon.
1975 Is an ethnographic film a filmic ethnography? Studies in the anthropology of visual communinication 2(2): 104-111.
1976 In a pic's eye: Interpretive strategies for deriving meaning and signification from photographs. Afterimage 2(9):5-7.
1977 The image mirrored: Reflexivity and the documentary film. The Journal of the University Film Association 29(4):3-13
1980 Exposing yourself: Reflexivity, film and anthropology. Semioticia 30(1-2): 153-79.
1977 On photography. New York: Farrar, Straus and Giroux.
1975 Limits to scientific understanding of man. Science 187: 1052-57.
Worth, Sol and
1974 Symbolic strategies. Journal of Communication 24(4):22-29.