(Principles of Visual Anthropology, 2nd Edition, 1995)
Reality graphics in the Modern World
“Americans jolted by gruesome TV pictures” reads the morning paper in a report from Somalia. The battered face of US helicopter pilot Mike Durant looks fearfully out of a front-page picture made from a CNN videotape. And once again—only a year after the Rodney King episode of 1992 and the burning of central Los Angeles—the graphic evidence of film and video records seems likely to play a prominent part in our lives.
Looking back a bit further there were the audio tapes which precipitated President Richard Nixon’s downfall. And before them, providing irrefutable evidence of acts so far outside the range of normal human behavior that a sceptical later generation finds them increasingly difficult to believe, there were the film records made by the cameramen who entered Belsen and Auschwitz.
Each of these cases involves graphic facsimiles provided by mechanical record-keeping devices in which the role of human subjectivity is either negligible or irrelevant. All of them involve situations which are either difficult (Mike Durant’s) or impossible (Rodney King’s, the Nixon tapes, the death camp footage) to rehearse, repeat, direct or in any way control.
And all of them represent, in varying forms, the vital modern role of “reality graphics” in providing reliable information about states of affairs and matters of fact. Instead of reality graphics several synonyms would serve as well: objective graphics, video documents, electronic transcripts, biofacsimiles, and so on. What is denoted is the entire class of alinguistic mimetic facsimiles used for scientific research and public information—from the cloud chambers of nuclear physics, to satellite imaging, to medical endoscopy, to the scenes on the evening news. All of them indispensably assist both enquiry and observation today. Without them we would be handicapped in many ways.
The popular opposing view
This is not a popular view in social science. Anthropologists in particular urge us to regard the world not as the totality of facts or of things but of meanings. These are felt to have nothing to do with empirical observations, instrumental technology, or truth. “Thought lives on meanings, meanings are culture-bound. Ergo, life is subjectivity,” notes Ernest Gellner, summarizing the prevailing outlook. “Objective truth is to be replaced by hermeneutic truth” in a closed universe of self-admiring subjectivities (Gellner 1992: 24, 33, 35).
Besides, it’s argued, all the world’s a stage, Goffman rules, and today only simpletons could possibly be unaware that social life is full of deceit and dissimulation, of tricks and masks, or that wherever cameras are employed we find information and misinformation and disinformation inextricably combined.
There is of course something to this point of view. But to my mind it encourages a much too limited view of a large subject. I take it that even Hermann Eutic himself wouldn’t actually claim that we would know more, or would know it more reliably, if there had been no video evidence from Somalia showing Mike Durant; no endlessly reviewable tape of the beating of Rodney King; no recording of Richard Nixon’s voice; no cine-camera record of Belsen and Auschwitz.
I also assume that even the most besotted post-modernized devotee of interpretation would hardly have the gall to maintain that the unmediated subjectivity of Richard Nixon or David Irving (the latter being a man who maintains that the Holocaust never happened) would actually be preferable to the electromagnetic and photographic evidence we possess.
It is cases like these which make one realise that the account of culture as universal subjectivity and more-or-less motivated deception goes altogether too far. More than this, it demonstrates an ivory-towering blindness to life today. The entire fact-seeking and reality-testing side of modern culture is left out—that large part of it concerned with accuracy of description and adequacy of formal argument, and which has as its distinctive legacy the achievements of modern natural science.
Furthermore, the spill-over of its techniques into the public domain has had wide ramifications. In the last thirty years there have been huge advances in the ways citizens monitor their own activities and learn what is going on abroad. The images from Somalia and Moscow in 1993 were part of this. Real-time transmissions around the world never stop, from skirmishes in the Balkans to football matches in Paraguay, and each of these events may be taped, enlarged, reviewed in slow motion, or speeded up, all in the cause of seeing more clearly details of action and performance.
Yet to read theorists of ethnographic film you would hardly know this was happening. Intent on blurring the boundary between words and things, fiction and fact, the linguistic representations of narrative and the nonlinguistic (or a-linguistic) mechanized graphic records of events, they seem unaware of the deeper cognitive reasons for the continual expansion of facsimilizing in modern life. They still write as if what are called “documentaries” (those highly artificial artifacts) were the primary unit for analysis. In small groups at conventions they still gather in darkened rooms like Plato’s cave, admiring the shadows on the wall, seemingly unaware of the world outside.
From perception to transcription
Can we love the cinema and Plato too? asks Bill Nichols (1991: 187), before reluctantly answering “No.” But perhaps he’s been looking in the wrong part of the Agora. It was Aristotle, not Plato, who gave his mind to the question of likeness or “mimesis”. Just as it was Aristotle who first considered the need for mimetic accuracy in representation. And he went straight to the point. “A work of art is a likeness or reproduction of an original,” he wrote 2,300 years ago, “and not a symbolic representation of it” (Butcher 1951: 124).
This contains an important distinction, and it is a thousand pities that Daguerre and Fox Talbot and photography itself were not around at the time to help Aristotle develop his thoughts. They could have taken a photograph of his pupil Alexander, and its significance would have struck the sage at once. Almost certainly he would have felt impelled to revise the above passage as follows: “A photograph is a likeness or reproduction of an original, and not a symbolic representation of it.” For as S.H. Butcher noted in his 1894 commentary, Aristotle is saying that
A sign or symbol has no essential resemblance, no natural connection, with the thing signified. Thus spoken words are symbols of mental states, written words are symbols of spoken words; the connexion between them is conventional. On the other hand mental impressions are not signs or symbols, but copies of external reality, likenesses of the things themselves” (Butcher 1951: 125).
Just why Aristotle should have thought this distinction important I’m not sure, but it is entirely in keeping with his interest in the accurate observation of living things.
And it shows that the difference between verbalizing and mental picturing had been noticed at quite an early stage. Anyone who tries to give a detailed verbal description of what he sees—the cat on the mat, the flowers on the table, the computer on the desk, with their colors and positions and size—will quickly find how hard it is to translate the one into the other. In the first place, because so many bits and pieces of verbal code are required. In the second place, because you also need to know about sentence construction and narrative form. And lastly, because of the sheer instability of things. The cat rises and slowly walks away, and even with the help of tense it is hard to bend language round the curve of time.
The limitations of words
Adequate verbal description always presents cognitive difficulties. But it is not merely difficult, it is next to impossible in the case of action which is quick, fleeting, unique, and uncontrolled. Then there’s the additional problem of how to communicate this sort of thing accurately to others.
In order to accurately record and communicate the sort of ever-changing pictorial imagery in the mind something very much better than language had to be devised, and eventually in the nineteenth century it was. Daguerre and Fox Talbot, and later Thomas Edison, met both the technological challenge and the cognitive need.
But in science the large class of facsimiles I am calling reality graphics were not developed and refined in order to provide us with “symbolic representations”—least of all the representations of literature. Homer had done that well enough, and the works of the Greek dramatists are both timeless and exemplary. There is no useful sense in which language, literature or symbolism have developmentally “progressed” since classical times, nor are they likely to in the future. And the reason is plain.
Language stands at a fixed and irreducible distance from whatever it describes, and this distance or gap has always been a problem. The true significance of all the facsimilizing going on around us lies in the historic effort of modern instrumentation to try and close that gap, to supersede the fuzziness and uncertainty of linguistic codes, to provide better, more accurate and permanent “likenesses of the original”, in Aristotle’s phrase. Photography, along with the whole range of ingenious imaging techniques we use today, is a cultural response to our cognitive need for better descriptions of states of affairs and matters of fact.
But doesn’t this ignore the obvious communicative use of such images? Weren’t the photographs from Somalia of the downed helicopter pilot Mike Durant sent by the Somalis as conscious messages of triumph, of popular defiance, of ethnic pride? Weren’t they deliberately intended to humiliate and embarrass President Clinton? And weren’t they correctly received and interpreted in America as such?
They were indeed. Any photograph may be used for any purpose. And for the purpose of argument reality graphics of this kind function very powerfully. Yet it is precisely because of their independent and superior reliability as objective a-linguistic records that they so powerfully influence our opinions and beliefs. A second-hand verbal report by Somalis about a captured US soldier admitting that “killing innocent people is wrong” belongs simply and exclusively to the error-prone domain of journalistic communication, and at its worst may be only hearsay. A video of that pilot uttering those words is both a communication and a quite different thing.
Confusion has arisen because of a tendency to treat all communicative phenomena as embodying a universal semeiosis. But the enjoyment of reducing everything to the lowest common denominator—all the world’s a text, etc.—is merely a hindrance here. It is unhistorical, it precludes an evolutionary view of the matter, and it fails to discriminate between the generalized linguistic use of pictures as expressive signs and symbols, and the specific role of facsimilizing as an emergent nonlinguistic level of description.
Karl Popper on language
Karl Popper’s comments on language may be helpful here. In his essay “Epistemology without a knowing subject,” he writes that the lower and universal linguistic functions of self-expression and signalling (to be found in all animal languages and all linguistic phenomena) need to be distinguished from the higher functions of description and argument (in Homo sapiens). There is a connection here with my suggestion that the universal communicative functions of pictures (of all sorts) need to be distinguished from the unique descriptive function of reality graphics (a specific sort).
And what Popper says next about the reason for all the confusion is also revealing. “The two lower functions are always present.” he writes. “when the higher ones are present, so that it is always possible to ‘explain’ every linguistic phenomenon, in terms of the lower function, as an ‘expression’ or a ‘communication’” (Popper 1973: 120). In his following statement too Popper’s thought can be read as illuminating the cultural role of “reality transcriptions” as an extension of “linguistic description”—a role which sees them helping to clarify issues of truth or falsity in the cases of Durant, Nixon, David Irving, and Rodney King:
With the descriptive function of human language, the regulative idea of truth emerges, that is, of a description which fits the facts … The argumentative function of human language presupposes the descriptive function: arguments are, fundamentally, about descriptions: they criticise descriptions from the point of view of the regulative ideas of truth; content; and verisimilitude” (Popper 1973: 120).
Although it is an important consequence of this line of thought. we need not go into the emphasis Popper places on the higher functions of language as indispensable to cultural rationality. But one last sentence is worth quoting: “Without the exosomatic descriptive language—a language which, like a tool, develops outside the body—there can be no object for our critical discussion,” no objective foundation, that is, for determining matters of fact (Popper 1973: 120).
Today a whole range of modern devices for recording, imaging, storing and reproducing data augment the “exosomatic descriptive language” on which analysis and argument finally rest. In the past, verbal description and argument provided the sole basis for truth claims. It was all we had, and its overall cognitive role has always been confused by the fact that any use of language contains a rhetorical and expressive element. It remains of the greatest importance. But today it is everywhere reinforced with mechanically produced facsimiles which in many cases provide evidence of a more accurate kind. Because such evidence often “fits the facts” more exactly, it is important in establishing matters of fact.
The structural integrity of events
Our starting point, then, is that the basic materials for analysis are alinguistic. Next one sees that this material comes in isolable units—units we shall call events. These may be as short as a wink or as long as a marathon. They may be as trivial as a talk-show or as grave as a survivor’s report of a death camp. They may be as simple as a pitcher’s toss or as complicated in their framing as a play within a play within a play.
Whatever and wherever, when matters of fact matter, the reality graphics recording such events are bound by some concept of completeness and coherence. Bound, that is, by the structural integrity of events.
It might be noticed that the last 30 years have seen huge changes. At a New York film school towards the end of the 1950s a favorite class exercise was to give the student a scratched and valueless 16mm. print of an old documentary and tell him to reconstruct it. The challenge was to show one’s cleverness by treating each scene as an independent semantic unit, constructing new sentences and new meanings, writing a new commentary and fabricating a whole new interpretation which stood the old film on its head.
It demonstrated that where there’s wilfulness there’s a way, and was generally in accord with John Grierson’s definition of documentary as “the creative treatment of actuality”. The original people in the original scenes and the intention of the original work didn’t matter: these were grist for the mill. “Give it the treatment,” was the instructor’s advice—the creative treatment of actuality. And we did.
But cinema verité put an end to all that. And soon cameramen filming speech and music discovered the joys and frustrations of dealing with complicated internal structures of both sound and sense. Editorial fragmentation went out; unity and continuity came in. A manipulative eagerness to impose the merely extrinsic meanings of filmmakers became less and less acceptable, and a respect for intrinsic meanings took its place.
Above all, an indifference to the true dimensions of behavioral forms was no longer permissible. Replacing it came a respect for the natural topography of events, along with the assumption that for facsimiles of such events to be fully intelligible that shape had to be preserved.
This requirement applies to the minor as well as the momentous. It is as important to parents who record scenes of bathing the baby as it is to historians of the Holocaust watching the long unbroken record of Abraham Bomba, in Shoah, talking to Claude Lanzmann about Treblinka. And in the field of mass communications exactly the same considerations apply wherever matters of fact most matter—in international politics and their modem sublimation, sports.
When the helicopter pilot Durant replies to his Somali interrogators that “innocent people being killed is not good,” we know that any sign of a cut or deletion between the “is” and the “not good”—any flip-overs or image break-up or loss of lip-synch—would be grounds for suspicion; grounds for suspecting that a news editor was making Durant say something other than what he in fact said. Grounds, that is, for questioning the descriptive adequacy of the transcript of the event.
Sports fans are particularly demanding. In Formula One racing, lenses are attached to the cockpits of cars, and one gets immediate objective verification of whether or not Ayrton Senna was the culpable party in a “shunt”—whatever the subjectively volunteered extenuating circumstances of worn tires or driver fatigue. In football, vast sweeping views of the field bring audiences a whole range of complicated plays, and if signal drop-out occurs and a continuous and coherent presentation is disrupted, fragmented, and made unintelligible, viewers get very upset.
Their irritation reflects an intuitive understanding of the connection between cognition, continuity, and the phenomenal nature of the world. And this can be seen in both symbolic and non-symbolic representational forms. Where tense bends language round the curve of time (what was becomes what is and then what may be), the sinuously agile facsimilizing of cameras and lenses follows the temporal shape of events.
Where do photographic “stills” fit into all this? Mostly they don’t; for the reliable depiction of either identity or behavior time frames of 1/250th sec. are not enough. Casting directors might once upon a time have made do with portfolios of glossy still shots arriving in the mail from aspiring actors. Now they require videos of performance. In India matrimonial arrangements negotiated at a distance depended on carefully posed stills of prospective brides and bridegrooms, but as the MacDougalls’ recent film Photo wallahs shows, painful revelations often ensued (MacDougall 1992). Will videos soon displace stills as more reliable evidence of the wares on offer? Will there be a demand for more fully portrayed identities and events? It’s inevitable. In life, marriage is one of those times when matters of fact matter most.
Aristotelian unities
The need to pay attention to natural forms of life and action has a long history. In The poetics Aristotle’s conscious concern was with artistic mimesis. He wanted to define criteria for literary representation. But a naturally empirical turn of mind inclined him toward rules for storytelling which were essentially rules of intelligibility—rules designed in cognitive terms—and this in turn led to his famous declaration that “episodic plots are the worst”, and that the best narratives portray “one action and the whole of it” (Butcher 1951: 35, 37).
He warned that “if any part is displaced or deleted” the whole meaning will be “disturbed and dislocated” (ibid: 35). A meaning which is dislocated has been changed, perhaps irremediably. All of which indicates that from the very start of critical thinking about narrative, when cognitive matters mattered, displacement, deletion, disturbance and dislocation were seen as injurious.
Critics and dramaturgical theorists have always had trouble with this; and rightly so. When taken too seriously the principle of “the unity of action” amounts to a disabling restriction on the imaginative creativity we expect in all great drama. Where Aristotle’s rules more naturally belong is in empirical science, not art. And for documentary filmmakers they usefully set out general conditions of intelligibility for both the recording and understanding of social events.
The criteria of naturalistic coherence in time and space derive from our cognitive need somehow to seize phenomena in flux, making them available for contemplation. Respecting these criteria in the realm of reality graphics—respecting the structural integrity of events and employing lenses and framing procedures which retain bounded and isolable features of scenes intact—became defining criteria for the descriptive adequacy of the results.
Karl Heider included this understanding in his recommendation that ethnographic films should portray “whole persons in whole acts” (Helder 1976: 107-109), and the cognitive rationale is not hard to see. Happenings are related as cause and effect, and to ask for the preservation of natural sequences is to ask for rules of facsimilizing in which “the logic of the situation” and “the logic of understanding” correspond.
Yet descriptive adequacy is only a first step. The objective record is only a foundation. Beyond it lie the perplexities of Goffmania, and upon entering this country one must ask the people whose images appear in these records who they are, and what they are doing, and why? Actors must explain their roles. When they do, will they confirm or deny their appearances?
Once filmmakers and editors have taken the original film documents and cut them about—using the shots of the cameraman as semantic units endlessly rearranged—will the intrinsic and heterogeneous meanings of the “footage” correspond to or contradict the extrinsic homogeneous meanings imposed by filmmakers in “films”? Once assimilated to this or that communicative form, will the men and women whose biofacsimiles appear in these films still recognize and confirm their identities and their acts?
Objective graphics, subjective confirmation
Trees are one thing; the rafters of a house are something else. And in most cases it would be difficult to draw any secure conclusions about the nature and form of the former from the latter. Of course, if one were able to interrogate individual pieces of lumber they might confirm their original identity despite all the cutting and sawing and industrious human use. But one should never lose sight of the fact that pieces of lumber begin as trees.
In much the same way it is a mistake to define the nature of reality graphics in terms of the artefacts called “documentary films”. For here once again we are dealing with constructions, in this case constructions in which the alinguistic products of facsimilizing are put to quasi-literary use. And the questions to be asked are much the same as the imaginary questions we might ask the planks and rafters: How much do the extrinsic purposes of the filmmaker allow the grain of original identities to be preserved? How much of the intrinsic meaning of the original event survives? What guarantee do we have that things really and truly are what they seem?
Suppose that helicopter pilot Mike Durant were to say in reponse to being shown his picture back in the U.S.A.—”That is not me”: suppose that he were to deny that he was the man in the truncated video clip. This plainly involves rather more than the repudiation by an annoyed and humiliated sitter of an artist’s unsatisfactory likeness in pen and ink. It is to deny the look of my eye while talking, the curve of my mouth as it speaks, the sound of my uneasy voice.
Such a repudiation has about it something of ontological suicide, and indicates the importance of the evidence of human actors themselves. As it happened, pilot Durant never did any such thing. And this example is only used to emphasise that if the built constructions of “ethnographic films” are to be persuasive, participants must confirm that even when the shots representing them have (so to speak) been cut, and sawn, and stained, what appears on the screen truly is what it purports to be. Here subjectivity plays an important role. It authenticates (or repudiates, as the case may be) the graphic reality portrayed.
I am not arguing here the general point that the self-understanding of agents is the only understanding relevant in social analysis: it is not. But in facsimiles of cultural phenomena the minds of agents are certainly the first things we should consult. A widespread feeling that this is so has deeply influenced ethnographic filmmaking in recent years. Underlying it is the tacit assumption that if asked, if given an opportunity to comment, the agents portrayed would themselves confirm and corroborate the meanings, the understandings and the interpretations which the communicative forms and formats of film convey.
One might indeed arrange ethnographic films along a continuum from those in which the intrinsic meanings of agents are central (the MacDougalls’ African work; Ian Dunlop’s later work in Australia) to those in which such meanings are largely irrelevant to the filmmaker’s extrinsic artistic considerations (Robert Gardner’s Forest of bliss, for instance). In this film about death and dying in Benares (a “silent” film in the sense that though dialogue is heard it is untranslated), a variety of implied or insinuated meanings were noted by Alexander Moore and Jonathan Parry (Moore 1988; Parry 1988).
But those meanings are all very ambiguous. One suspects that if some of the participants in Forest of bliss were shown the scenes in which they appear and asked “What are you doing?” the result would make a most diverting companion to the film itself. Surely in no other work claiming anthropological consideration is the disjunction between intrinsic meaning (the participant’s) and extrinsic insinuation (the filmmaker’s) so radical or so complete.
There is of course an infinite variety of intermediate types of ethnographic film, and an examination of selected cases throws light on what happens to reality graphics when they are used in various ways—anthropological, educational, or political. In each of the cases to be discussed the foundation of the work consists of reality transcripts having a greater or lesser degree of descriptive adequacy: in films about matters of fact that is the sine qua non. After that come the various ways in which subjectivity authenticates the record, witnesses speak, lawyers argue, and the wider processes by which reports become known as reliable begin.
Coniston Muster (1972) is a 30-minute film 1 made about Aboriginal stockmen (cowboys) on a central Australian cattle station (ranch). Its shooting ratio was about 20: 1. Much editorial work went into scene selection and sequence construction, and a variety of grammatical devices are employed. Action sequences are cut for kinetic effect. Extrinsic spoken narration is spare and only used at the beginning to set the scene. Silent episode intertitles are employed to punctuate and separate, to raise introductory questions, to point and comment.
All of which should make clear that the final result is very much a designed and packaged artefact for the marketplace. Yet the credibility of the result, if it is credible, depends largely on the descriptive adequacy of the original reality transcripts, the unedited footage, reinforced by the spontaneous commentary of an Aboriginal participant, Coniston Johnny, recorded as he watched the uncut film.
It is Johnny who identifies people and who personally answers the ever-present question, “What’s going on here?” In a scene which at one point includes himself he catches sight of an old friend struggling to lift some harness onto a truck. At this point he says, “There’s Paddy! Poor old bugger! [laughs] Loading them up. Helping Maurice load the truck.” Like most of his other comments these simple words go only a little beyond the self-explanatory nature of the event, so why are they appreciated by audiences?
Because that little is a crucial part of subjective authentication. Only a participant’s mild humor (“Poor old bugger!”) is either appropriate, understandable. or justified. It grows naturally out of the intersubjective experience of the world presented on the screen and is the expression of present consciousness reliving past events. At one level the image before him on the screen is merely the shadow of a shadow. At another it is closer to the Ding an sich than his own memory could ever hope to attain.
Johnny’s utterance is in part a communication with the world beyond the shadow, in part the murmured rumination of memory talking to itself. In either case his voice is self-evidently unrehearsed, undirected, and uncontrolled. During scenes in which a wild bull is released from a stockade in the early morning he begins to re-enact the desperate ride of the man who brings the animal back, and as he cries out in warning to the rider, past and present, objective graphic and subjective authentication, the viewer and the viewed, become for the moment phenomenologically fused.
Objective record, subjective disconfirmation
By way of contrast one might consider a particular sequence in Maasai women. This popular introduction to Maasai culture has been widely seen, and combines a certain amount of undirected observational material with interviews conducted by the anthropologist Melissa Llewelyn-Davies examining the situation of women. It is a useful film which I have often shown to classes in Sydney. But much of the information conveyed by the narration is both generalizing and normative—a feature hard to combine with the specificities of actual events.
Worse than that: because of a visible incongruence between the descriptive footage, on the one hand, and the information supplied by participants on the other, serious implausibilities arise. This is spectacularly so in the case of a sequence treating the subject of clitoridectomy.
It would be absurd to suggest that either the filming or the subsequent presentation on broadcast television of such an event was either practicable or desirable: it was not. But equally plainly it was open to the filmmakers to recognise this fact and not to pretend otherwise. As it is, the sequence treating this matter provides a clinical example of what happens when none of the criteria here being considered is satisfied—neither completeness and descriptive adequacy, nor subjective authentication.
The relevant scenes show a young girl being prepared for the operation, a few close-ups of her head being shaved, an interpolated interview from another time and place with an older woman, and then a number of post-operative scenes from which the patient is entirely excluded showing general collective celebration. Nothing of the surgical actuality is shown, and virtually nothing which even suggests the patient’s anxiety and tension before the event. As a result, a large burden of responsibility rests upon Llewelyn-Davies’ spoken commentary. This is perversely unequal to the task. It consists of a sustained attempt to abstract, romanticise, etherealise, and generally gloss over the painful bleeding centerpiece of the occasion.
The language of the narration is itself revealing. After all, what exactly is “female circumcision”? To most people it suggests an anatomical improbability, not the cutting off of the clitoral hood or the labia minora. Appealing figures of speech are employed poetically to distance and transfigure the occasion. The circumcision ceremony is described as “a girl’s farewell to childhood”, something “expected to transform a giggly girl into a mature and thoughtful woman” which is all in all “a bit like a white wedding”.
That the scenes dealing with the occasion are noticeably fragmentary and interrupted only increases one’s suspicions. In an abruptly inserted interview Llewelyn-Davies asks her principal female informant (and this interview fragment is used directly after the distressed face of the patient has been glimpsed), “Is the girl happy?” To which the Maasai woman appears to reply, “Very happy. Part of the ritual is to brew mead from nectar and honey … Her mother and father wear charms and they are happy because their daughter isn’t pregnant and they are drinking mead.”
Yet all too obviously the “happiness” referred to has nothing directly to do with the patient’s state of mind as she faces the knife—and from the look of the editing, nothing to do with her state of mind at all. It is a normative statement referring to how her father and mother regard such a ceremony overall, and its use at this point is decidedly misleading. Next we are told that “after the ceremony has taken place, the branches of a special tree are brought to mark the house where the girl is recovering” (one of the rare implicit references to the surgical nature of the event); and a procession of singing girls appears, adding both a musical and visual distraction from the matters of fact at hand.
Of course we know what Llewelyn-Davies means by her remark that “It’s a bit like a white wedding”. But with respect, analogies made on the lofty level of what all rites de passage have in common scarcely enable us to share, or even to glimpse, the experiential level of genital mutilation performed in a hut without benefit of antiseptic or anaesthesia. What would the participant herself have had to say about the event? Would the patient have endorsed the interpretative overlay? Would she have agreed with the older woman’s enthusiastically “functional” assertions? Or was she a victim of cultural circumstances that she would dearly like to see changed?
More to the point, perhaps, how complete were the original records of the event?—what uncut reality transcripts were made and available? Loizos tells us that the screams of the patient had been recorded, and that “a lively debate [took place] in the cutting room about how the issue should be handled”, one view being that “perhaps the girl’s screams should be heard, thus giving ‘symbolic’ expression to what was visually too horrific” (Loizos 1993: 123). In the event, however, nothing is shown, the screams are silenced, and the narrator pours a syrup of generalizing sociological interpretation over a whole episode. 1
But the relevant points are these: the original camera record is entirely inadequate. On top of this the natural shape of the event is broken up in a deliberately misleading way, being interrupted with highly tendentious “happiness” comments from an interview conducted at another time and place, followed by scenes of communal jubilation. Finally, no attempt is made to obtain the participant’s view of the event by exploring the intrinsic meanings available from the actor; indeed, quite the reverse. The “meaning” of the event is pretty much what you will find in Arnold Van Gennep. How extrinsic can you get?
A lot more than this, to be sure. Instead of even the pretence of observational spontaneity one might for example have tried to organize the local community into acting out the whole episode, thus ensuring that nothing untoward occurred—no screams, no blood, no pain. In this ideal/typical functionalist version of the circumcision ceremony the desired social consequences toward which Llewelyn-Davies’ account is skewed—solidarity, unity, collective cohesion, happiness would then be even more systematically brought out, the numerous details of Maasai dress and decorum and speech and behavior being pointed toward this overriding goal. Real people would go through the motions of somewhat unreal acts. Real places would provide the backgrounds or “sets” for unreal scenarios.
There is of course a well-known tradition of political documentary which has always been made in this way (and Man of Aran was too). In this kind of work actuality is to be moulded as public opinion is moulded, and is always subordinated to overriding political goals. And if the set is a set-up, the men and women merely actors, the presentation of both self and setting a version of the Goffrnanian world where nothing is what it seems—what relevance have the criteria here set out? What can be done when the action is all too clearly rehearsable, directed, and controlled?
A witness testifies to events
Marina Goldovskaya’s remarkable recent film, The Solovki power: evidence and documents, suggests one possible approach. “Those who control the past control the future” was for seventy years the ruling principle of Soviet historiography. In accordance with this goal huge quantities of documents and evidence were either destroyed, or hidden, or suppressed, and systematic falsification was institutionalized.
But with Gorbachev the control of the past was relaxed, and among the things which escaped from the archives in 1986 was a film made about an early Soviet labor camp (the model for numberless others) built on the Solovetski Islands in the White Sea. This camp had been established in 1925 in an abandoned monastery. The treatment of prisoners was typically brutal. Escapees’ reports published in the West became an embarrassment. And so it was that in 1927 the GPU (later the KGB) commissioned a film to show the benevolent and uplifting nature of their prison regime.
“With an iron hand we shall drive mankind to happiness,” a Bolshevik poster had announced in 1918; and among those caught by the iron hand was 15-year-old Yefim Ligutin. Driven by a romantic yearning for faraway places and the sea he had run away from home. Unfortunately for him this was construed as the attempt of a would-be spy to enter foreign countries. He was sentenced to death, and when the sentence was commuted to imprisonment at Solovki he began the first of seventeen years in prison. Like a surprising number of other former inmates—scientists, engineers, writers, academics—this man was still available to be interviewed when Gorbachev freed Russia to inquire into its past, and in 1988 Marina Goldovskaya talked to them, showing the old propaganda film and inviting their comments on the mysterious world it both did and did not show.
This process can be seen as another variation on the theme of “objective record, subjective authentication”—except that the reality of the 1927 film was that of a “Potemkin Village”, and the role of living witnesses was one of retrieval: as each man and woman speaks it is less to confirm and authenticate than to disconfirm and repudiate the “creative treatment of actuality” the GPU had engineered. 2
“How nice and clean—and there’s linen too!” a former inmate says ironically about the scene. “A white tablecloth and even flowers. It must be a wrong close-up.” The remark is as spontaneous as it is revealing. The deception of the close-up derives from its role in information control. Beyond the frame of the close-up is the scene as a whole; beyond the scene, the play; and beyond that the play which contains a play. Yet within the tradition of situational encompassment this is something special. For to pull back and reveal the frame outside the frame is to reveal the high barbed-wire camp boundary within which this entire tragicomedy of political deceit is set.
In reality graphics treating matters of fact it is often tacitly assumed that what is contained within the narrower view typifies what is present in the wider field as welt, that the phenomenon pictured is continuous in space. In the kind of self-authenticating work of the MacDougalls in East Africa, we know that to pull back from the group of men gathered under the men’s tree—casually conversing, idling, working on strips of hide we will find across the wide plain beyond other men under other trees. Numerous clues lead us to trust the image.
By way of contrast, in the kind of work represented by the GPU’s propaganda piece we know that if we could pull back from the table with the white cloth and the flowers we would find only bare boards, dirt, and misery. Numerous clues lead us to distrust the image: the cramped view, the short takes, the posed people, the relentless voiceless smiles. We can be confident that tomorrow the cloth and the flowers will be gone forever—for what is presented is a space-time discontinuity so extreme that if one had been able to search the whole wide universe in the year this film was made, nowhere else could a camera crew be found in a concentration camp getting up to such theatrical tricks.
And one by one the former inmates confirm that this is so. “For God’s sake, what nonsense!” cries Yefim Ligutin, now 78, “My God, what rubbish! This never took place in Solovki. It’s just for show. Is that how they served food in the prisons? It’s a fake!” The scene shows a neatly dressed camp guard solicitously tasting the stew to be served to prisoners, followed by a title which claims that “Prisoners may request dry rations if they prefer them.” Throughout the Gulag dry rations were nauseatingly strong-smelling dried cod and a little bread and were universally detested. When Ligutin was offered this salted fish for the first time he was sickened, and struggled to get near an open window. Two guards knocked him down and kicked him in the kidneys.
“What I noticed then,” he says, “I have never forgotten: the shiny black boots of my tormentors.” And he laughs at the memory. Such scenes in Goldovskaya’s film draw on the intrinsic and unique experience of participants, those who have truly been there and had that sort of thing done to them. For this there is no substitute.
But the most remarkable fusion of reality transcript and subjective recollection occurs at a later point. It is when the eyes of the 78-year-old camp survivor stare into the eyes of the 16-year-old he then was. He remembers when the film was being made. The crew had begun working in the camp and he recalls as a boy sitting and looking toward the camera. Now, after sixty years, he meets that boy again. As he does so he sees the facsimile of an enduring identity whose experiences, accumulated over 17 years in prison, would at last bear witness against those who could no longer control the past.
The orthodoxy of cinema theory
With monotonous repetition it is asserted that the objectivity of reality graphics is merely naive nonsense. Questioning the underlying assumptions of research films Luc de Heusch once wrote that the camera’s objectivity is “a myth” which is based “on a belief in its magic powers” (de Heusch 1962: 23).
What this means I’m not sure, but in the cases of Durant, of King, and of Nixon, it seems less than self-evident that magic gives these transcripts their persuasiveness. An even more sweeping claim was made years ago by Asen Balikci and Quentin Brown to the effect that “any subject that the camera photographs has been discovered already by the eye of the man behind the camera and hence the record acquires only the quality of illustration” (Balikci and Brown 1966: 27).
To which one can only respond that the cameramen Balikci and Brown are referring to differ radically from any I have known. On the evening news tonight there was a report about the Goroka festival in highland Papua New Guinea, one scene alone showing a quantity of uninterpreted data which it would take a well-informed anthropologist six months’ work to unpack.
Recent reports from Moscow presented wide-angle views from a high building showing crowds, military vehicles, trees and flowers in a park, and a line of distinctive edifices on the far horizon. If you were to give a highly magnified still of that scene to a dozen specialists in Russian affairs with a request for the interpretation of all identifiable military, botanical, demographic and architectural contents, six months might not be enough.
It is of course correct that in the normal course of events 99.9 percent of those data (add additional digits to taste) will never be interpreted. This is both true and irrelevant. After all, 99.9 percent of all sensory data in biological systems are uninterpreted, but so what? Sensory inputs of sight and sound last at their very best for the life of the organism—at worst they are erased within the hour. The cognitive difference between these two procedures (and the advantage of the mechanical over the biological) is that the cultural invention of facsimilizing makes data permanently available for review—whether or not this ever happens.
What then is the status of transcript data which are forever uninterpreted? Of “research footage” which is never seen? Of the ten miles of audio tape from Papua-New Guinea the linguist never gets around to analyzing? Of the video record from the monitor in the local bank before it is wiped out each month? It exists in a condition of latency—rather as latent photographic images exist before they are developed. Human consciousness is the chemical reagent which realises whatever potential is there, making it available for the next step in the construction of our knowledge of events.
A personal example comes to mind. Around 1971 the Australian Institute of Aboriginal Studies made a film, Larwari and Walkara, about Aboriginal ceremonies in the vicinity of Hooker Creek (now Lajamanu). As the cameraman on the project I worked with the anthropologist and liaison man Stephen Wild. Wild knew both the people and the language, whereas 1 knew neither, and knew even less about the rituals to take place. That they took place at all was largely due to our providing the food, transport, and encouragement needed to precipitate the events involved.
We also urged those responsible to do whatever they were going to do in as traditional a manner as possible. Beyond this we simply filmed what people chose to do, and my own role was little more than that of an uncomprehending observer of a three-day rite which both anthropologist and Aborigines believed to be important.
A complete and unedited copy of the original photographic transcript was subsequently given to the community, and a shorter edited film was also made. Then we all went off to do other things. Some years later however the data the footage contained were found to be relevant to an Aboriginal land claim, and statements made by individuals appearing in the film were mentioned in court. In this way the latent information of the original footage (an objective account of the event) was used as evidence on the basis of which legal argument determined certain matters of fact.
All in the cameraman’s head?
As to the remarkable claim that “it’s all in the cameraman’s head,” or that this or that subjective factor caused this or that subject to be filmed and the rest to be ignored, all I can say is that in this particular case the material used in evidence was wholly uninfluenced by the subjective interests of either the cameraman, the soundman, or the Chief Cook and Bottle-Washer who was also along on the trip.
Our very ignorance of the specific sentences on which argument would later rest ensured that all we did was to frame them, hoping that the result would satisfy minimal standards of descriptive adequacy. All such utterances were entirely unrehearsed, undirected, and uncontrolled—how could it be otherwise? In brief, the objective record we so blindly produced was the adventitious consequence of the disinterested documentation of events.
This film was indeed made by subjective persons with subjective purposes. Another tirelessly repeated absurdity claims that because photographic and video records are made for a purpose, the undeniable subjectivity of the intentional act subjectivizes all the consequences. In other words, that because my subjective purpose directed me to take a shot of a marketplace, all the data such a shot contains are equally the product and consequence of the initial subjective state.
But this is a non sequitur. The plenitude of data in reality graphics is an unintended consequence of purposive action—and like innumerable other unintended consequences it is absurd to pretend to “explain” them in terms of initial subjective conditions. The inventor of the automobile did not have Los Angeles “in mind”. Yet Los Angeles was certainly one of the objective consequences of his invention.
In the same way the superabundant data which register without any thought whatever on the part of numberless photographers in numberless shots every day are an unintended objective consequence of purposive action. Since in this case the specific product is not Los Angeles, but mere data in graphic form, it is appropriate to describe the result as an objective graphic. In the new world of facsimilizing such products are among the cognitive matrices of our lives.
Truth and consequences
Two images are relevant at this point—the gallows and the ivory tower. The first defines a situation of utmost seriousness, the kind which Dr. Johnson had in mind when he said that nothing so concentrates a man’s mind as the knowledge that he will soon be hanged. The second indicates a situation of legendary inconsequentiality, the sort of privileged unseriousness found only in the groves of academe.
Serious situations of life or death not only concentrate the mind, they put it under the kind of cognitive pressure which makes people rather more interested in distinguishing between truth and falsehood than they might otherwise be. Ivory towers, in contrast, are synonymous with the absence of any cognitive pressure at all. Matters of fact are always conjectural. Anything may be said; anything may be believed; beneath the “dreaming spires” the dreamers dream.
The appointment of cameramen to university posts in the last twenty years saw the academicization of a formerly honest trade. When this happened, putting a frame round the action gave way to framing arguments for other university folk. It became important to publish, and even more important to keep up with the latest fashions in anthropological thought. Each of these steps meant a move away from dealing with the non-semantic dimensions of culture, a retreat from fact-finding activities and exposure to serious cognitive pressure, and an increasing interest in the traditional preoccupations of arts faculties in ivory towers: that is, with words and meanings.
This has involved a more and more tenuous connection with the world in which reality graphics belong, and in which the only distinctive tradition of ethnographic filmmaking originated: the empirical tradition of direct behavioral observation. Where once an ethnographic filmmaker produced a film, and attached to it an appendix of published notes, today’s film may be merely an appendix to an essay. Sometimes an agonising selfconsciousness supervenes.
All of which closely resembles the situation in anthropology as a whole. Where fieldwork is more and more a thing of the past, where the study of culture has turned into the study of the “writing” of culture, where commentaries on commentaries proliferate and Sir Edmund Leach recommends that ethnographies should be read as novels, anthropology has become in some places little more than the onanistic subliterary chatter of people with nothing better to do.
Outside the universities, however, the cognitive needs of our culture still make themselves felt. There it is noticeable that where the stakes are high and when the consequences of getting things wrong are serious—when matters of fact matter—reality graphics are called on again and again. Hence the importance of the scenes of Mike Durant and Rodney King. Hence the significance attaching to the Nixon tapes and the footage from Auschwitz.
Where the consequences of error are momentous, the role of documentary footage in providing the evidential foundations for our knowledge of the world is most unlikely to diminish. Indeed, quite the reverse. Off campus, driven by our need for reliable information and broadly governed by the rules of intelligibility which Aristotle so presciently set out so long ago, reality facsimilizing shows every sign of continuing to expand.
Notes
1 For years Marxists berated functionalism for offering rosy interpretations of conflict and domination. There is a certain irony in the fact that someone like Llewelyn-Davies should make a film in which functionalistic talk about happiness and community well-being obliterates a young girl’s screams.
2 Named for one of Catherine the Great’s ministers, Potemkin, who inaugurated systematic theatrical deceptions for state visitors on their tours of the Russian countryside the chief object of such deception being Catherine herself. This tradition of political disinformation involved the building., painting, and populating of whole ersatz villages of happy peasants. Although fakery of this kind went on for over two hundred years in Russia, under Stalin it was carried to extremes never known before.
References – Literature
BALIKCI, ASEN, QUENTIN BROWN 1966 Ethnographic filming and the Netsilik Eskimos. Educational Services Incorporated Quarterly Report (Spring‑ Summer): 19‑33.
BUTCHER, S.H. 1951 [1907] Aristotle’s theory of poetry and fine art. New York: Dover Books.
GELLNER, ERNEST 1992 Postmodernism, reason and religion. London: Routledge & Kegan Paul.
HEIDER, KARL 1976 Ethnographic film. Austin: University of Texas Press.
HEUSCH, LUC DE 1962 The cinema and social science. Paris: UNESCO.
LOIZOS, PETER 1993 Innovation in ethnographic film. Manchester: Manchester University Press; Chicago: University of Chicago Press.
MACDOUGALL, DAVID 1992 Photo Hierarchicus: signs and mirrors in Indian photography. Visual Anthropology 5: 103‑129.
MOORE, ALEXANDER 1988 The limitations of imagist documentary. Society for Visual Anthropology Newsletter 4(2): 1‑3.
NICHOLS, BILL (WILLIAM JAMES) Representing reality. Bloomington: Indiana University Press.
PARRY, JONATHAN P. 1988 Comment on Robert Gardner’s Forest of Bliss. Society for Visual Anthropology Newsletter 4(2): 4‑7.
POPPER, KARL 1973 Objective knowledge. London: Oxford University Press.
References – Films
CURLING, CHRIS, MELISSA LLEWELYN‑DAVIES 1974 Maasai women Color, 53 minutes.
GARDNER, ROBERT 1986 Forest of bliss. Color, 90 minutes.
GOLDOVSKAYA, MARINA 1988 The Solovki power. B & W, 90 minutes.
LANZMANN, CLAUDE 1985 Shoah. Color, 9 hours, 30 minutes.
MACDOUGALL, DAVID, JUDITH MACDOUGALL 1972? Under the men’s tree. B & W, 20 minutes.
SANDALL, ROGER
1972 Coniston muster. Color, 25 minutes.
1974 Larwari and Walkara. Color, 45 minutes.