For a long time anthropologists appeared to have much in common with taxidermists inrecording and classifying phenomena such as dances, feasts, or meals – in short: the rituals of vanishing cultures. Their main concern was not to establish what was disappearing or how this could be understood, but how it could be classified in their museum. Similar to a taxidermist, the anthropologist engaged in dissecting vanishing rituals, remove their vital parts and preserve them in formaldehyde in order to present them to noted colleagues. As Rony affirms: “The metaphor of taxidermy – a form of representation which is infused with an acknowledgement of death, but also a desire “to be whole” – describes a plethora of technologies popular at the turn of the century used to represent the human body, including photography, film, and wax figures” (Rony 1996: 244).

Documentary filmmakers were – and what is more: they still are – even more notorious for their dissecting methods. As early as 1887, the brothers Lumière completed the Ashanti series. The film series features twelve short dances performed by colored women. Yet the fact that the series was recorded at the Lyon World Exhibition suggests an entirely different story, which is wrapped up in colonialism, imperialism and exploitation. “The irony – and this irony is at the heart of taxidermy – is that ‘reality’ filmed does not appear real. The filmmaker must use artifice to convey truth” (Ibid. 116).

Mainstream documentaries do not demonstrate reality: they are the result of a delicately obscured ‘taxidermy operation’. Selection, manipulation and other distorting processes, all part of the act of reconstruction, are carefully edited out. Although these manipulations may seem obvious, they have a painful and strikingly unjust result; the ‘object’ of interest, this human being is carved up and presented as a stereotyped distortion in a freak show. Moreover, the ‘viewer’ is unaware of the taxidermist at work and judges this ‘other’ as such. Narrative devices such as a detailed script, prefabricated quotes and characters, crosscutting – all seem misplaced in a documentary context. Yet, when I was asked to make a documentary about an immigrant’s family, working as a documentary maker for Belgian National Television, I not only received a detailed (and stereotypical) script, but also the timing of scenes and even quotes were specified. According to my editor, it was my job to fit a family into this predetermined matrix. From the perspective of the ‘viewer’, it seems that crucial information about the production process is obscured. As images are not critically contextualized the way written texts are – there are no footnotes, or bibliographical references, – the audience seems to depend on the status of the channel in order to evaluate the truthfulness of the images they see. The ‘viewer’, in consequence, has no point of reference in order to establish the program’s relation to reality. This, one might argue, is a fairly weak position of critique. More importantly, from the perspective of the people filmed, scripting seems absurd and most often even painfully stereotyped as they seem uninvited guests in their own script. There is no room for any participation or collaboration on the way they are represented. The script is as a mirror image of the producer’s reflections but does not have any relationship with the subject.

The problem is that mainstream documentary images are conceived in a conventional way, and therefore help to maintain a certain balance of power. Constructed and manipulated images, which represent clichés, stereotypes and established values that are part of a cultural hegemony emerge and are moreover presented by the TV channel as truth or reality. Gilles Deleuze points out that clichés rather than images typify our society (Deleuze 1983, 1985). Billions of people are surrounded and guided by clichés of identities and they model their lives upon them. Arjun Appadurai asserts “ordinary lives today are more often powered not by the givenness of things but by the possibilities that the media (either directly or indirectly) suggest are available” (Appadurai 1996: 54). The rich potentiality and importance of visuals in the construction of the self, on the one hand, and the formation of sodalities through those media, on the other, are consequently important challenges to anthropology (Ibid. 7). In fact, we have not moved on from the 19th century anthropology and taxidermy: images are cut and parts removed in order to show a reconstruction in formaldehyde, which supposedly refers to the truth. At what stage does the taxidermist appear and when does ‘the other’ leave the room? In general, documentary filmmaking prompts fundamental questions. First of all, why is the translation from reality understood as a representation? Secondly, how is this transformation manufactured? What does this process entail with regard to the ‘other’ that is filmed? Next, which information is obscured from the ‘viewer’ and what are the consequences? Finally, how can the ‘author’ prefigure the ‘viewer’ within the process of production in a way that (s)he has a critical position in the film? Acclaimed filmmaker Jean Rouch stated that every documentary is essentially a self-portrait (Rouch ed. by Feld 2003). How can one describe and question the ethics implied by the ways the ‘author’ handles the codes and values intertwined with the mediated context in his/her relationship towards the subject and the prefigured viewer? In sum, how can we understand the interactions between the ‘author’, the ‘other’ and the ‘viewer’ in (documentary) filmmaking during the production process?

Addiction to realistic modes of representation

This article explores questions concerning the construction of documentaries and their implication on representation. In search of an efficient way to unfold this questioning, I read Works and Lives. The Anthropologist as Author by the noted anthropologist Clifford Geertz written in 1988. Geertz advocates anthropology as a representational discourse. In his view this intermediary nature is text-based. Geertz proposes to shift the attention partly from fieldwork – the result of the anthropological enterprise – to the production of those ethnographic statements, the textual discourses. The reason why he advocates this shift is eloquently formulated as followed: “The advantage of shifting at least part of our attention from the fascinations of field work, which have held us so long in thrall, to those of writing is not only that this difficulty will become more clearly understood, but also that we shall learn to read with a more percipient eye. A hundred and fifteen years (if we date our profession, as conventionally, from Tylor) of asseverational prose and literary innocence is long enough” (Ibid. 24).

This coincides with voices heard in media and visual anthropology that express their annoyance with the systematic lack of competencies and codes in examining, dealing with and expressing through audiovisual media and their parameters. In contrast with linguistic education, there is scarcely any consistently and efficiently organized audiovisual education. Yet everybody uses audiovisual tools, is surrounded by images and imagined by them. At the core of such an environment lies a positivist belief in the representational qualities of the medium. This belief is anchored in a long tradition in Western visual culture, in which the search for representation of reality was one of its main objectives. According to Winston, Western culture should be comprehended by “a general cultural addiction to realistic modes of representation” (Winston 1996, 44), much in contrast with the amounts of efforts to construct, develop and design devices to perceive reality: “The technologies of seeing bring us ever closer to a sort of Borgesian map of reality – one that corresponds at all points with the external world – but as they do so, they do little to help understand their own historical and social realities. On the contrary, their basic illusionism disguises their artifice, their cultural formation and their ideological import” (Ibid. 118). Moreover, ‘our’ culture can be typified by a belief that symbolic systems have the ability to actually represent reality. Ginsburg asserts: “The lack of analysis of indigenous media as both cultural product and social process may also be due to our own culture’s enduring positivist belief that the camera provides a “window” on reality, a simple expansion of our powers of observation, as opposed to a creative tool in the service of a new signifying practice” (Ginsburg 1991: 93).

It is this positivist belief in the representational nature of texts and images, which allows me to use Geertz’s book. His aim is to make transparent foundational mechanisms in the elaboration of textual discourses so as to peel off some displaced authoritarian or naturalistic connotations. His main goal is to strip off some ‘pretensions’ of textual discourses, which obscure their construction so as to prevent the critical assessment of their authorship and rhetoric. As such, he is not giving in to a relativistic plea for the abolishment of authorship or for the questioning of the possibilities of meaning an sich, on the contrary: “..the burden of authorship cannot be evaded, however heavy it may have grown; there is no possibility of displacing it onto “method,” “language,” or (an especially popular maneuver at the moment) “the people themselves” redescribed (“appropriated” is probably the better term) as co-authors. (Ibid. 104) These pretensions, I want to argue, are also quite identifiable with assumptions associated with (mainstream) documentaries. I therefore take up Geertz’s formulations of several pretensions of textual discourses to confront these with documentary practices.

Pretension number One

“There is text positivism: the notion that, if only Emawayish can be got to dictate or write down her poems as carefully as possible and they are translated as faithfully as possible, then the ethnographer’s role dissolves into that of an honest broker passing on the substance of things with only the most trivial of transaction costs” (Geertz 1989: 104).

The pseudo-positivistic claim of representational systems seems the source for the various assumptions on documentary images that need to be challenged. The positivist or naturalistic belief – the inference that what is being represented unquestionably refers to what has been experienced – seems even more dominant in audiovisual media than in textual discourses. Indeed, as visual media actually are able to represent an image of what can be perceived in reality, as the ‘real’ leaves recognizable and even mimetic traces in the audiovisual counterpart, positivist assumptions might appear much harder to battle. The idea persists that images represent without any censorship or manipulation whatsoever; images are supposed to have the ability to record the interviewees in their own words, with their own gestures and physical body language. Television formats such as “Life as it is” unmistakably tap into this assumed conviction so as to persuade the audience that the program reflects life ‘as it really happened’. Moreover, the recorded images are interpreted as the unmistakable evidence of the point of view of the interviewees, given the chain of anthropological documentaries such as the Smithsonian series and media libraries, such as the Albert Kahn museum in Paris in which pictures and recorded images are catalogued depicting vanishing communities and tribes. “What is presented as evidence remains evidence, whether the observing eye qualifies itself as being subjective or objective. At the core of such a rationale dwells, untouched, the Cartesian division between subject and object that perpetuates a dualistic insideversus- outside, mind-against-matter view of the world. Again, the emphasis is laid on the power of film to capture reality ‘out there’ for us ‘in here’. The moment of appropriation and of consumption is either simply ignored or carefully rendered invisible according to rules of good and bad documentary. The art of talking-to-saynothing goes hand-in-hand with the will to say, and to say only to confine something in a meaning. Truth has to be made vivid, interesting; it has to be ‘dramatized’ if it is to convince the audience of the evidence, whose ‘confidence’ in it allows truth to take shape” (Trinh 1990: 83).

Although Geertz’s book was written more than a decade ago, this positivist belief in the representational nature of texts (and of images) still remains quite unquestioned. I comprehend this by referring to the omnipotence of the audiovisual and its indexical qualities. Bill Nichols uses “indexical to refer to signs that bear a physical trace of what they refer to, such as fingerprint, X ray, or photograph” (Nichols 1994: ix). It is important to stress that the fact that something has been filmed, does not imply that it is real. This contrasts sharply with our understanding that something ‘real’ has actually been filmed. Because of the indexical quality, images might be wrongly interpreted as reality.

“Inevitably, the distinction between fact and fiction blurs when claims about reality get cast as narratives. We enter a zone where the world put before us lies between one not our own and one that very well might be, between a world we may recognize as a fragment of our own and one that may seem fabricated from such fragments, between indexical (authentic) signs of reality and cinematic (invented) interpretations of this reality” (Nichols 1994: ix).

Pretension number Two

“There is ethnographic ventriloquism: the claim to speak not just about another form of life but to speak from within it; to represent a depiction of how things look from “an Ethiopian (woman poet’s) point of view” as itself an Ethiopian (woman’s poet) depiction of how they look from such a view” (Geertz 1989: 104).

As the audiovisual positivistic claim is so dominant, it follows quite easily that a perspective from within a community can be effortlessly depicted. Quite in fashion lately are the numerous documentaries and television formats, in which the interviewees are invited to film themselves: to point the camera on themselves so as to guarantee the presupposed authenticity of the recorded images. In 2000 the BBC broadcasted a program called Video nation which was promoted for its truthfulness as all the interviewees had recorded themselves: they were given cameras to record their own lives ‘in their own way’. Yet what was strikingly obvious for a critical viewer, was that most of the audiovisual codes and parameters were in the hands of the series editors: the editing, and the choice of topics, obviously, but also the type of framing, the use of the tripod, sound and music, and the types of inserts. For instance, a sequence shows a parallel editing of three families. For one thing, the members of these families all talked about their housekeeping, definitely a choice by the editors. The framing of the image was conceived so as to contrast several families: one framing was extremely stable and neat, reflecting an old man’s tedious home, whereas another framing recorded a youngster of 16 who stereotypically lived in a sloppy room that needed to be cleaned. The framing was shaky, hand-held, the editing speeded up the frequency of the images and accentuated them by an up–tempo dance song, much in contrast with the silence background in the old man’s house. These codes were obviously chosen by the series editors so as to dramatize the program and accentuate the contrasts between the several characters. The result of these devices was a simplistic and very stereotypical depiction of these people, while letting the viewers mistakenly assume that the people on screen had had the total liberty over their representation.

“The relationship between mediator and medium, or the mediating activity, is either ignored –that is, assumed to be transparent, as value-free and as insentient as an instrument of reproduction ought to be –or else, it is treated most conveniently: by humanizing the gathering of evidence so as to further the status quo” (Trinh 1990: 84).

Pretension number Three

“There is dispersed authorship: the hope that ethnographic discourse can somehow be made “heteroglossial,” so that Emawayish can speak within it alongside the anthropologist in some direct, equal, and independent way; a There presence in a Here text” (Geertz 1989: 104).

Documentary film is, more than anything else, a matter of selection and intrusion. A crew consisting sometimes of five people, stampedes into a location and starts to rig up tripods, lights, cameras and microphones. Reality ‘as it is’ is disrupted, to say the least. As Stella Bruzzi confirms: “Documentaries are inevitably the result of the intrusion of the filmmaker onto the situation begin filmed, they are performative because they acknowledge the construction and artificiality of even the non-fiction film and propose, as the underpinning truth, the truth that emerges through the encounter between filmmakers, subjects and spectators” (Bruzzi 2006, 11). As a consequence of the selective nature of documentary making, and thus of the time-space linearity of film, narrative devices are developed to guarantee to the viewer the representational qualities of the film. Moreover, the use of textual discourses in the audiovisual system adds to this narrating and most often simplifying regime: voice-overs, interviews and other textual devices transform the image to a dramatized version of the reality experienced. “The producers (of the PBS series ‘Childhood’ AvD.) constantly search for dramatic material to illustrate intellectual points or to stand on its own. In the end, tensions get played out, more or less successfully, between the “magic” of documentary realism and the edification of expository explanation, between the programs as engaging televisual experience and the programs as scholarly knowledge, both tendencies mediated by the producers’ practical logic and the aesthetic ideologies of program production” (Dornfeld 2002: 257).

Raoul Ruiz uses the concept of a ‘central conflict theory’ to illuminate this idea. He defines it as an all-encompassing narrative and dramatic guideline that is ruled by conflict (Ruiz 1995: 14). He points out that “.. the criteria according to which most of the characters in today’s movies behave are drawn from one particular culture (that of the USA). In this culture, it is not only indispensable to make decisions but also to act on them, immediately (not so in China or Iraq). The immediate consequence of most decisions in this culture is some kind of conflict (untrue in other cultures). Different ways of thinking deny the direct causal connection between a decision and the conflict, which may result from it; they also deny that physical or verbal collision is the only possible form of conflict. Unfortunately, these other societies, which secretly maintain their traditional beliefs in these matters, have outwardly adopted Hollywood’s rhetorical behavior. So another consequence of the globalization of central conflict theory – a political one – is that, paradoxically, “the American way of life” has become a lure, a mask: unreal and exotic, it is the perfect illustration of the fallacy that Whitehead dubbed “misplaced concreteness”. Such synchronicity between the artistic theory and the political system of a dominant nation is rare in history; rarer still is its acceptance by most of the countries in the world” (Ibid. 21).

According to Ruiz, this theory has turned into a predatory theory, a system of ideas that devours and enslaves any other idea that might restrain its activity (Ibid. 15). Yet there is no strict equivalence between stories of conflict and everyday life. People fight and compete, but competition alone cannot contain the totality of the event that involves this. Furthermore, he states that this theory yields a normative system. The products that comply with this norm have not only invaded the world but have also imposed their rules on most of the centers of audiovisual production across the planet, attempting to master the same logic of representation and practizing the same narrative logic (Ibid. 21). He claims: “The rules governing cinema (let’s say, Hollywood cinema) are identical to the simulation that is life today. This utopia reformulates the idea of salvation whose most perfect application is to be found in the theory of central conflict: the greater homage you render to narrative clarity or Energeia, the better your chances to be saved… In this permanent Olympiad, the citizens of the Ideal City are constantly pitched against each other in single combat” (Ibid. 29).

As a consequence of the ‘intrusive’ part of filmmaking, an exaggeration of performative behavior can be ascertained. When a camera enters a room, certain types of acting or staging are being stimulated: a sort of amplified form of common behavior can be noticed. Moreover, it is as though the camera itself leads to a situation where not only the person in front of the camera but also the people behind it acts in an almost programmed way. One of the students on our seminars on visual anthropology at the University of Ghent wrote a thesis on the ‘trap’ a camera could be. Even though he set out bursting with ‘good intentions’, lectures in visual anthropology and a good deal of common sense about what urban life might be, he found his own film on the black community in Brussels ‘trapped’ into a stereotype presenting blacks singing, dancing and sitting at the hairdresser. The audiovisual apparatus, the camera, the microphone and so on, often induces such a stereotypical behavior, and most of all when in the hands of amateurs or television professionals.

Yet as Geertz pointed out in regard to the construction of textual authorship and discourse, these essential elements of film are being dispersed or obscured. When, why, and how selection and intrusion has taken place is being camouflaged by means of an Ancient Greek view on drama directing the parameters to convey this drama as representation of an ‘authentic’ piece of reality. With a hand-held camera, an oftenblurred focus, and thus a deliberately ‘un-aesthetical’ style, the interviewees are followed in their whereabouts as well as possible, showing sometimes shaky images and less understandable conversations. “The documentary can easily thus become a ‘style’: it no longer constitutes a mode of production or an attitude toward life, but proves to be only an element of aesthetics (or anti-aesthetics), which at best, and without acknowledging it, it tends to be in any case when, within, its own factual limits, it reduces itself to a mere category, or a set of persuasive techniques. Many of these techniques have become so ‘natural’ to the language of broadcast television that they go ‘unnoticed’” (Trinh 1990: 88).

By submitting the flow of experiences to the structure of a classical drama, one confides in a certain appropriation and an ideology-laden use of images. The viewer cannot locate censorship or accountability. Form (the type of narrative, the scenario, the length of images, the frames, the angles,..) in and of itself thus carries a highly sophisticated ideological meaning. To ignore the mode of production of this form is to confine it in an ideological drama. Documentary filmmaking can therefore better be described as a site that constructs identities as opposed to representing them. In this sense, narratives dominate the reconstruction of the real. Furthermore, a documentary is deeply rooted in an economical framework, where decisions need to be taken for reasons of audience ratings, entertainment qualities, funding, etc.

Mainstream documentary images are moreover generally interpreted in a conventional way. These conventions are mainly based upon systems of belief of dominant cultural groups. Political relations are reflected in those interpretations. Although the representational system is essentially a system of open meanings, contextual interferences narrow the scope of interpretations into stereotypes. The codes of representation are generally obscure constructions by which cultural hegemony is maintained. “Electronic digital media at the end of the twentieth century have begun to alter many of our most precious assumptions about visual representation, as the image is no longer linked ontologically or indexically to something “out there” in the real world. Unlike the cinematic image, preserved on celluloid, the video image is made anew at every transmission, and digital image processing has opened up the possibility of infinite manipulation” (Russell 1999: 7).