Skip to navigation – Site map

The Colliding Worlds of Anthropology and Film-Ethnography

A Dynamic Continuum
Cristóbal Escobar


The article explores three consecutive periods in which the disciplines of anthropology and film ethnography collide. The first moment examines the common practice of Bronislaw Malinowski and Robert Flaherty’s Nanook of the North. I argue that Flaherty’s film illustrates the general fieldwork schema proposed by Malinowski to document the world of the Other-native. The second period connects the writings of Marcel Mauss and his influence in Jean Rouch’s cinéma-vérité. I state that Mauss’ radical sense of doubt about scientific pretentions of objectivity sustained Rouch’s cinematography with the general principle that reality is accessible only in partial form. Finally, the third period compares the anthropology of Eduardo Viveiros de Castro with the Sensory Ethnographic Lab’s film Leviathan. I argue that it is in both cases where bodily practices are being supported to account for more sensorial perceptions of the environment.

Top of page

Full text

Introduction: The communal Practice of Anthropology and Film Ethnography

1In 1977, while reviewing a film about East African people, P.T.W. Baxter stated that anthropology and film ethnography were incompatible, because “they fundamentally differ in methods and aims.” (In Taylor 1996: 64) On this occasion, as Lucien Taylor suggests in his article Iconophobia: How anthropology lost it at the movies, Baxter argued that each discipline seeks quite different aspects of truth and utilises different means of stitching scraps of culture together creatively. To Baxter, whereas anthropology is detached and open-minded, film is anything but: “Substituting a single glass lens for our two human eyes is imperious and monocular; its beauty is distorting; it tries to simplify and disarm, as well as to impose.” (1996: 64)

2A decade later, as Taylor continues to argue, Maurice Bloch not only declared that he is “not very interested” in ethnographic films, but more bellicosely that “he can hardly bear to watch them at all.” (1996: 64) Bloch states that if ethnographic films must be made at all, they should be made with a thesis component. For him, textuality itself, and textuality alone is the only means to legitimate a serious visual anthropological endeavour. Visuality, on the other hand, becomes merely ancillary, illustrative rather than constitutive of anthropological knowledge.

3In the same vein, the anthropologist Kirsten Hastrup has continued to defend the written primacy of the discipline to combat photographic and audiovisual representations of a given culture. In her article Anthropological visions: some notes on visual and textual authority, Hastrup places a whole series of oppositions between films and texts, seeking to qualify the difference between visual and textual power in anthropology. On the one hand, she argues that film is only capable of producing a “thin” description of happenings. Text, on the contrary, is able to fabricate a “thick” description of an event, which is already a happening invested with cultural significance. In her words: “While a thin description may capture forms, it cannot of itself convey implicit meaning. Forms are culturally meaningless when studied independently of local meaning relations and contemporary conventions of representation.” (1992: 10) The idea then is that while a happening is an objective occurrence viewed from afar, an event is embedded with first person subjectivity and narrated with perspective. Thus, only writing, as Hastrup suggests, can evoke the existential fabric of the place to someone who wasn’t there.

4Baxter, Bloch and Hastrup take films as an unquestionably lower epistemological production if compared to written texts. Moving images do not seem to be of cultural relevance in the practice of anthropology here. For them, the image and its soundtrack remain as an inferior manifestation of the idea; audio-vision is still an accessory attached to the delights of representation while text guarantees the meaning.

5Contrary to their belief, I do not see why film cannot be constitutive of anthropological knowledge. On the one hand, our vision, as cinema itself, constantly and literally frames the world; it comes naturally equipped with focus -and losing focus, depth of field, left and right edges, top and bottom limits. On the other hand, also our ears, as film’s acoustic surroundings, ensure the embodiment of visual perception, because we only see in one framed and flat direction, whereas hearing is always three-dimensional. Borrowing Marshall McLuhan’s expression, film becomes an enlargement of our own physical ability to see and hear the world.

6Rather than highlighting discordances between the written and the audiovisual event, I propose to take the worlds of anthropology and film ethnography as an interdisciplinary field that shares a community of practice, what Wenger et al. define as “a group of people who share a concern, a set of problems, or a passion about a topic, and who deepen their knowledge and expertise in this area by interacting on an ongoing basis.” (In Pink 2006: 4) In this line, I state that the interplay has been historically dynamic and continuous. Parallel preoccupations have governed the practice, concerns and viewpoints of anthropology and film ethnography. They share a common body of knowledge that frames, in different periods of time and space, similar assumptions about a given Other. The dynamic continuum thus perceived highlights three consecutive periods across anthropology and film ethnography, starting from the early decades of the twentieth century and concluding with current ethnographic practices. The first moment deals with Robert Flaherty’s Nanook of the North (USA, 1922) and the work of Bronislaw Malinowski who, among other anthropologists, was transforming the discipline from the nineteenth century’s natural science outlook to the twentieth century’s humanistic attitude. I argue that Flaherty’s film illustrates, in many respects, the general fieldwork schema proposed by Malinowski to record the life of native communities more accurately. It was in this period when Malinowski and Flaherty established what would become the conventional subject and the methodological schema for anthropology and ethnographic filmmaking.

7The second moment is a transitional phase in the history of anthropology and film ethnography. I focus on the French school which, by the 1940s began to declare a radical sense of doubt about scientific pretentions of objectivity and the methodologies employed by previous anthropologists. Here I explore the connections between the writings of Marcel Mauss and his influence in Jean Rouch’s cinéma-vérité. For the former, it would be an illusion to state that anthropology could reveal the ultimate “truth” or “reality” of a given culture. Mauss constantly critiqued both the relative incomplexity of primitive thought in contrast to modern rationality and the pretensions of scientific objectivity as a transparent practice. This epistemology sustained Rouch’s cinematography with a couple of basic principles: that reality is accessible to knowledge only in partial form and that access to this reality involves poetry -the sensual- as much as science -the rational.

8Lastly, the third moment examines the sensory turn experienced across anthropology and film ethnography at the gates of the new millennium. In this section I introduce the anthropology of Eduardo Viveiros de Castro, a Brazilian social anthropologist who shows that what falls under the domain of “human” relations for Amazonian people is so broad -animals, plants, spirits are all endorsed with agency- that modern distinctions between nature and culture, or animals and humans are proclaimed to be useless. As a cinematic peer to this philosophy, I review the audiovisual works of the Sensory Ethnography Lab (SEL) of Harvard University. They support innovative combinations of aesthetics and ethnography that explore new bodily practices to account for more sensorial and embodied perceptions of the environment. The focus is on SEL’s critically acclaimed film Leviathan (Castaing-Taylor, Lucien and Véréna Paravel 2012), which demonstrates how in film-ethnographic practice the question of “the Other” has been framed with inadequate sophistication. Leviathan is based on SEL’s characteristic bodily approach to filmmaking, which adopts a highly sensorial form of personal expression. I claim that it is through its post-humanistic approach to mediated reality that the film meets the marked dimension of Amazonian people on multiple referents by portraying bodies as primary conductors of perspective.

Nanook of the North

Nanook of the North


A film by Robert Flaherty, USA 1922

Malinowski’s Argonauts and Flaherty’s Nanook

9As the British documentary movement was developing during the early decades of the twentieth century, the possibility that its two pioneer figures would one day work together seemed increasingly likely. It also seemed inevitable that they would clash. John Grierson’s belief in industrial progress and socially purposive attempts to depict workers as machines contrasted Robert Flaherty’s feeling for individual achievement and observational style. In fact, at the time when Grierson and Flaherty were working together on Industrial Britain (Flaherty 1931) Grierson declared that:

[Flaherty’s] flair for the old crafts and the old craftsmen was superb, and there will never be shooting of that kind to compare with; but he simply could not bend to the conception of those other species of craftsmanship which go with modern industry and modern organizations. (Grierson 1992: 91)

10Flaherty’s approach to film was primarily that of an explorer. With a flair for the old and the exotic, he used the camera to record unfamiliar territories. In his most notable film, Nanook of the North he wanted to depict the life of Inuit people in order to show in Europe how life was “for a typical Eskimo and his family” (Flaherty 1969: 216). Like the ethnographers of his time, Flaherty regarded native people as primitive versions of what was to become modern civilizations, a paradigm locked in the two-sided “hot and cold” societies that Levi-Strauss (1974) refers to in Structural Anthropology.

  • 1 Flaherty asked his subjects to revive a dangerous method of walrus-hunting that Inuit people had ab (...)
  • 2 Even when the Eskimos had a polygamist family unit, Flaherty depicted Nanook’s clan as monogamous t (...)
  • 3 In a long take, for example, the viewer watches and waits as Nanook sits above a hole he has cut in (...)

11Flaherty was a mining engineer and had lived among the Hudson Bay Eskimo for much of the decade before embarking on the film. He was convinced that living for a long time among his subjects would allow him to know them well enough in order to make a documentary faithful to their lives. However, Flaherty’s admiration towards Inuit community not only made himself surrender to the foreign lifestyle; Nanook of the North also portrays his harsh colonization of the Eskimo’s environment by domesticating all its strange domains. As suggested by William Rothman (1998) in The Filmmaker as Hunter: Robert Flaherty’s Nanook of the North, Flaherty did not seem to have any reservations about modifying reality, staging aspects of the seal hunt1, the Inuit’s family structure2, or telling us Nanook and his family are on the brink of starvation even though Flaherty is there with plenty of provisions. As such, the filmmaker sets out to convince the viewer that Nanook is a fine provider by demonstrating his prowess at hunting walrus, building igloos and harpooning seals.3

12During the same period, the anthropologist Bronislaw Malinowski was radically changing the identity of the discipline. Shifting from the methodological conventions of the nineteenth century, Malinowski - and his peers W.H.R. Rivers, Franz Boas and Alfred Kroeber - were transforming anthropology from its previous natural science outlook to a new humanistic enterprise. Malinowski’s aim, similar to Flaherty’s, was to understand and translate the “raw” societies to a common - western - language. Published in 1922, his Argonauts of the Western Pacific serves as an illustrative anthropological counter-part of Flaherty’s Nanook of the North. Malinowski’s impulse to reflect the natives’ way of life by decoding their cultural practices outside their area of origin is clearly seen in the distinction he makes between the view of actors -natives- and the interpretation of the analyst -anthropologist. In reference to the Kula ring in the Trobriand Islands, Malinowski declares:

It must be remembered that what appears to us an extensive, complicated, and yet well-ordered institution is the outcome of so many doings and pursuits, carried on by savages, who have no laws or aims or charters definitely laid down. They have no knowledge of the total outline of any of their social structure (...) Not even the most intelligent native has any clear idea of the Kula as a big, organised social construction (...) The integration of all the details observed, the achievement of a sociological synthesis of all the various, relevant symptoms, is the task of the Ethnographer (...) The Ethnographer has to construct the picture of the big institution, very much as the physicist constructs his theory from the experimental data, which always have been within reach of everybody, but needed a consistent interpretation. (2001: 83-84)

13The passage in the Argonauts of the Western Pacific as much as the scenes in Nanook of the North show the common practice employed by Malinowski and Flaherty to represent the “truth” of the natives as “the anthropologist and the filmmaker saw it”. However, and taking into account the more humanistic side of their project, even when Flaherty transformed Nanook and his family into fictionalized actors, he nonetheless actively collaborated with them to a degree that is still rare today. Nanook of the North is still considered a seminal contribution for the film ethnographic tradition. It is a pioneering work that helped to establish the form of observational documentary by being shot entirely on location, with no actors. Besides, Flaherty also showed more interest in the lives of native people than probably any other western documentary filmmaker before him. As Ilisa Barbash and Lucien Taylor suggest in Cross-Cultural Filmmaking: A Handbook for Making Documentary and Ethnographic Films and Videos, Flaherty screened some of the footage for his subjects, eliciting their feedback and suggestions for future scenes that they could film. (Barbash and Taylor 1997) Such interactive performance was an inspiration to Jean Rouch, who, as we will see shortly, coined the concept of “shared anthropology”.

14In this vein, Flaherty’s concern to develop scenes in collaboration with his subjects was methodologically comparable to Malinowski’s anthropology based on the “documentation of concrete evidence and the imponderabilia of everyday life.” (In Marks 1995: 340) In his Argonauts of the Western Pacific, Malinowski stated a general scheme for anthropological fieldwork, one which should collect “characteristic narratives, typical utterances, items of folk-lore and magical formulae. [This goal was] to grasp the native’s point of view, his relation to life, to realize his vision of his world.” (1922: 24-25) As such, Flaherty shared with Malinowski, to a remarkable degree, the convention that cultural practice makes sense in terms of the system in which it occurs. Both presented the native’s daily scenes in a way that their internal logic will become apparent for the external eye. In Malinowski’s words: “Field work consists only and exclusively in the interpretation of the chaotic social reality.” (1922: 238).

15In retrospect, Nanook of the North as much as the Argonauts of the Western Pacific transforms the break between the west and the non-west into a rift between culture and nature –or harmony and chaos. In both cases there is a borderline between the “civilized land” and “the wilderness”, a frontier underlying divisions such as the metropolitan versus the rustic, the settler versus the native, the law of the book versus the law of the harpoon. In short, the analyst-explorer versus the hunter-native. Therefore, if Malinowski who constantly staged and affirmed a demarcation line between the west and the native’s land, invented modern ethnography, then we should not hesitate to affirm that Flaherty invented ethnographic film.

The Anthropological Teachings of Marcel Mauss and Jean Rouch’s cinéma vérité

A film by Jean Rouch, France 1967

16In the decades following the publication of Argonauts of the Western Pacific, Malinowski’s conventions of participant observation for ethnographic fieldwork became widespread in anthropology. Ethnographic analysis started to focus more on the meaning of particular actions rather than providing an overview of broad-scale societal patterns. Similarly, film footage had also begun to be considered a more important medium for scientific research. The attractiveness of ethnographic documentary within anthropology was now enhanced by the development of a film technology which towards the 1950’s allowed ethnographers to record footage where it had once been impossible. As such, the financial and technical efforts of Flaherty in making Nanook of the North were overcome by the advance of lightweight cameras and portable sync-sound equipment, thus allowing filmmakers to record social actions at a level of detail that any ethnographer could hardly match.

17Around the same period, connections between anthropology and film ethnography were also coming closer together in France. Appointed as the General Secretary of the Comité International du Film Ethnographique et Sociologique in 1952, Jean Rouch aimed to establish “links between the human sciences, and the cinematographic art, both from the point of view of the development of scientific research and for the expansion of the art of the motion picture.” (In Eaton 1979: 4) Considered as one of the pioneers of visual anthropology, Jean Rouch used the camera as a recording instrument to document everyday life of different regions in Africa. For him, visual anthropology was a highly observational practice that aimed not to perform a wide description of everything, but to record “a close identification of one technique or ritual.” (Rouch 1995: 62).

18Rouch also coined for the first time the term cinéma vérité -or truthful cinema- to adopt a deeply observational style of filmmaking. His call was for a more participatory documentary style based on the assumption that ethnographic enquiry could never be objective. Arguably, his purpose was to radically renovate the documentary episteme and make a new type of reality emerge:

Objectivity consists in inserting what one knows into what one films, inserting oneself with a tool which will provoke the emergence of a certain reality (...) When I have a camera and a microphone I'm not my usual self, I'm in a strange state, in a cine-transe. This is the objectivity one can expect, being perfectly conscious that the camera is there and that people know it. From that moment we live in an audio-visual galaxy; a new truth emerges, cinéma-vérité, which has nothing to do with normal reality. (Rouch 1978: 55)

19Rouch was taught, among others, by Marcel Mauss, the founding figure of social anthropology. As Ruben Caixeta de Queiroz (2012) suggests in his article Between the sensible and the intelligible, Mauss had a great influence in almost every branch of French anthropology, going from those of a more intellectualist tendency -like George Bataille and Claude Levi-Strauss- to those more experimentalists and artistic in nature, such as Jean Rouch and Germaine Dieterlen. As a matter of fact, the two latter figures found in audiovisual recordings a new means of artistic expression of Maussian anthropology in order to problematize the hegemony of the West.

20In their audiovisual ethnographies on the Dogon and the Bambara in Africa during the 1940’s and the 1950’s, Rouch and Dieterlen intended to demonstrate the complex nature of ceremonies and rituals performed by the Ogotemmêli people. According to them, the events of the Dogon, based on an oral archaic tradition, performed functions akin to the major written texts of western metaphysics, religion and literature, but in a way that “was unique to the Ogotemmêli world’s vision.” (2012: 207) As Marcel Griaule also suggests, the precautions and respect shown by the filmmakers for the native knowledge was faithful to the principle that “the local cosmological system was so complex that it was on a par with western philosophy.” (1965: 1)

21Rouch’s conception of reality as multiple, subjective and diverse is clearly influenced by Mauss’ critique of the relative incomplexity of primitive –magical- thought in opposition to western –modern- rationality and its pretensions of scientific objectivity as a transparent practice. In his work Theory of Magic, Mauss located modern science precisely in the subjective foundations of magical thought. In magic, he argues, there are officers, representations and actions. The performers -who can be shamans, alchemists, doctors, and astrologers, must give great importance to knowledge and its concerns in understanding nature. This is because “it is only by systematic means to possess the world that the officer can become a true magician.” (2001: 176) On the other hand, magical actions are only accomplished if the whole community believes in the efficacy of the rite performed, so it is only via the consent of the public opinion that shamanic actions can acquire symbolic significance. In short, what really makes magical rites eminently effective is belief, and it is precisely by doing things -creating a reality - that magic can be recognizable as such.

  • 4 Similar reasoning is derived from Mauss’ pupil Claude Levi-Strauss. In his book Myth and Meaning, h (...)

22For Mauss then, the logic of mythic thought becomes as fully rigorous as that of the moderns. In his words: “magic served science as much as magicians served scholars.” (2001: 176)4 This also implies that, as an essentially subjective practice, it is in both universes that the old alchemist’s mainspring knowledge is power becomes a major concern. Hence, in the light of this paradigmatic shift, subjectivity –knowledge as belief- comes to mean that there is constantly something of the real or the imaginary that remains unknown and inaccessible to those who try to apprehend it. Such Maussian approach to reality has been explicitly declared by Rouch. In one interview he declares:

Now the human sciences are something very specific. As Marcel Mauss said, the observer inevitably has, by definition, a perturbing role. Clearly the fact of speaking to people perturbs yourself and the others. From the moment when you interview me, you are no longer the same and I am no longer the same. (Rouch 1981)

  • 5 Mauss was very enthusiastic about the use of the camera towards anthropological investigations. He (...)

23Many of Rouch’s films had scientific purposes. Through the camera, he intended to describe the material fabric of rites of foreign cultures more accurately. However, even in his most realistic films - notably the series Sigui from 1966 to 1973 - the access to the raw material has always involved poetry -the sensorial- as much as science -the rational. According to him, and following Mauss’ suggestion to introduce films into ethnographic research5, cinema allowed for an examination of the more prosaic aspects of social life, that is to say, its material side, the conjunction of the lived with experience.

24Therefore, if we follow Rouch’s definition of “shared anthropology” given before (as the close identification of one technique or ritual), then the access to reality is somehow incomplete if scientific knowledge does not involve the senses as much as the rational. This is for him, and as we will see shortly for SEL’s bodily praxis too, the point of differentiation between written and (audio) visual anthropology:

Good anthropology is not a wide description of everything, but a close identification of one technique or ritual. The rituals are supposed to be dramatic. They are creations of the people who want them to be interesting and exciting. (...) What you can't get in writing is the drama of the ritual. Writing can't have that effect. That is the whole point of visual anthropology. (Rouch 1978: 4)

25As such, his cinéma vérité tried to offer a new possibility to show people and places in as unmodified a state as possible. However, it would be an illusion to state that his ethnographies can reveal the “ultimate truth” of reality. In Rouch’s project we are not asked to be the observers of an event, but rather the observers of an observation of an event. As an essentially subjective practice, “pure ethnography” is no longer possible, because films are everything the filmmaker shows by his/her methods and by his/her points of view. Hence, in cinéma vérité, ethnography becomes a very specific signifying practice that expresses reality with reality, or more precisely, a language that duplicates reality.

Eduardo Viveiros de Castro’s Cannibal Metaphysics and Sensory Ethnography Lab’s Leviathan

A Film by Lucien Castaing-Taylor and Véréna Paravel, USA, 2012

26Cannibal Metaphysics, a recent ethnographic work on Amerindian people by Eduardo Viveiros de Castro, lands right at the beginning of the new millennium, a moment of epistemological earthquake for modern thought. The shift is what post structuralist theorists are calling “the ontological turn” in western history. Viveiros de Castro’s annotations on Amerindian cosmology have made significant contributions to a more recent theoretical programme underpinning the dichotomies of modernity to develop more sensorial understandings of possible worlds. As a cinematic peer to this philosophy, the film Leviathan, from the Sensory Ethnography Lab of Harvard University (SEL) explores new bodily practices to record material life from an optical reality that is not strictly centred on human form. As I will illustrate in what follows, it is in both cases where bodies become primary conductors of agency and perspective, hence, taking a new point of departure in regards to the twentieth century’s anthropological episteme.

27Under his theory on perspectivism, Viveiros de Castro puts forward a pluralist ontology that helps to discover what a point of view is for the bearer of the point of view. For him, the world presented by the indigenous in the Amazonas implies that in order to embrace a multiplicity of agencies one is required to live -and believe- in a place inhabited by a wide range of subjective agents. Araweté’s universe includes-but is not limited to- animals, plants, gods, objects, the dead, and the humans; all living in intense proximity and interrelatedness with one another. These people, like animals, objects and plants alike, are all equipped with the same general ensemble of perceptive and cognitive dispositions. For as Deleuze would say, in Amazonas there are no points of view on things, since things are themselves points of view.

28In a thought-world as different as the Amerindian, what really makes multiple perspectives equally valid is the common condition of species, that is to say, their shared status of personhood. According to the anthropologist, every relatable entity is conceived as having, whatever its bodily form, a soul of a human character, and that all beings thus perceive themselves as humans. Jaguars, for example, are thought to see themselves as humans, to see humans as human prey, and their own food as that of humans. In Viveiros de Castro’s view, all beings see and represent the world in the same way: “their worlds revolve around hunting, war, food, fishing, initiation rites, shamans, and spirits”. (2014: 71) But what changes among them is the world they see. For the anthropologist, it is the things other species see when they see them like we do what is different: “what we take for blood, jaguars see as beer; what humans perceive as a mud puddle becomes a grand ceremonial house when viewed by tapirs.” (2014: 71)

29Consequently, successfully negotiating one’s relations with other species requires adopting their perspectives. Amerindian shamans for example, in order to understand how other entities behave and see the environment, must abduct their agencies and become their corporeality, what Guimarâes Rosa calls “the who of things.” (In Viveiros de Castro 2014: 61) Since to “know” is to “personify” for the indigenous, then shamans are the ones who capture the point of view of what should be known, or rather, the one whom should be known.

30In his text Whose Cosmos, Which Cosmopolitics? Bruno Latour also recalls the native’s emphasis on the body as described. Latour revisits the old disputatio in Valladolid that Spaniards held in order to decide whether or not Amerindians had souls susceptible of being saved. The issue back then was to determine whether the natives had enough “soul” and “reason” to be taken as part of the human kingdom. What interests us today, however, is that despite the conquistadors’ controversy, also in the Amazonas there was a relevant claim to make. As stressed by Latour, their problem was not to decide whether Spaniards had souls, but rather if they had bodies, because for the Arawaté entities all have souls and their souls are all the same: “What makes them differ is that their bodies differ, and it is bodies what give souls their contradictory perspectives.” (2004: 452) Terms reversed, Amerindian views of the world are indeed multiple, heterogeneous and open to different interpretations, precisely because their scientific premises and procedures are determined not by metaphysical categories, but by corporeal experiments: bodies are there the structure of life’s perceptive and cognitive dispositions.

31This philosophy based on interspecies perspectivism arrives as an ontology -or rather, various ontologies- that reverses not only the western relation with its non-western Other, but more radically, all the terms of its long metaphysical dualism. What kind of lesson could film ethnography take from Amerindian’s views of the world? If anthropology and film ethnography have walked side by side along the twentieth century, shouldn’t the audiovisual medium also be transformed into a similar mode of existence?

32Post human cinema, alike Amerindian belief, presumes exactly a gaze which sees from an abnormal angle; that of a point of view from a nonhuman eye. It is a filmmaking style that tries to break with the systemic boundaries of cinema’s traditional distance by presenting an acentred system in which moving images become multiple referents; a structure where things vary in relation to one another. The audiovisual works produced by the SEL of Harvard University are in many respects an ethnographic mediation of such post-humanistic approach to visual culture.

  • 6 Sensory Ethnography Lab, accessed December 1, 2015,

33As stated on the website, SEL harnesses perspectives “drawn from the human sciences, the arts, and the humanities [aiming] to support innovative combinations of aesthetics and ethnography, with original nonfiction media practices that explore the bodily praxis and affective fabric of human existence.”6 Lucien Castaing-Taylor established the lab in 2006 as a collaboration space between the departments of anthropology and visual studies at Harvard University. Since then, the aim has been to instigate a wide range of audiovisual productions, ranging from video installations to sync-sound films by anthropologists-artists.

34According to SEL’s approach to reality, moving images are derived from the body. For its directors, bodily instincts have a primary function to dictate the action of what they film. Véréna Parave, co-author of Leviathan, declared during one interview that most of the time she does not have to see what she films, because “all what is shot comes from the sensual body.” (Parave and Castaing-Taylor, 2012) Such commitment to the corporeal also helps to restore part of the autonomy that viewers have lost in interpreting mediated reality. Leviathan for example, keeps authorial intervention to its minimum by adopting a highly observational and interactive style of filmmaking. It offers an aesthetic that favours long takes, close-ups, synchronous speech and a tempo faithful to the rhythms of real life –thus, discouraging cuts, re-enacting and interviewing. As explained by Castaing-Taylor, in this film the subjects “are less mutilated by the montage, and the spectators may garner meanings, or simply come away with sensations that are at odds with the makers.” (Parave and Castaing-Taylor 2012)

35Shot entirely aboard a fishing industrial trawler and recorded with tiny waterproof digital cameras placed in various locations, Leviathan narrates the experience of the filmmakers on the dark and dangerous waters of the North Atlantic Ocean off the Massachusetts coast. It approaches the material dimension of a fishing vessel devoid of any speaking subjects or intentional framing by attaching GoPro cameras to different locations around the boat: to the tactile and mobile body of labouring fishermen -in their heads, wrists and chests- to the not-yet-dead creatures on board, to the slippery floor, or placed on wooden sticks trying to reach flying seagulls up in the sky.

  • 7 The concept of deframing, or décadrage in French, has been originally employed by Pascal Bonizer to (...)

36In Leviathan the imposition of meaning is reduced to its minimum once the film’s points of entry become embodied in the multiple entities aboard the boat. It is by overcoming a more conventional and static notion of framing that the heterogeneity of camera angles gives another standard of measurement to what is being shot. Ohad Ledesman, apropos this notion of deframing7, makes an interesting observation when stating that the film is rather structured from the embodied perspective of the fly than that of the human eye: “Because of their spherical shape and protrusion from the fly’s head, the eyes give the fly an almost 360-degree view of the world. Thus a fly sees in a mosaic way and thousands of tiny images coalesce and together represent one visual image.” (In Wahlberg 2014)

37Leviathan’s obsession with the material world not only makes everything visceral in the film; it also makes the viewing-reading process rather difficult to digest. Without spoken words to guide us through, the viewer must struggle - as the fishermen and filmmakers must have - to know what is actually being shown and how to fit these images within the daily routine of the fishing boat. Right from its beginning, Leviathan reveals that we are immersed in the middle of a nearly black ocean, so that deprived of concrete visual orientation, spectators are in many ways more susceptible to the film’s acoustic effects than the imagery. As a work that could not easily be done through written ethnography, the audio plays a crucial role in this film.

38Designed by Ernst Karel and re-engineered by Jacob Ribicoff, the sound comes many times before -and after- the image. It makes the ear become accustomed to the din of the industrial process of harvesting the ocean once the engine of the fishing trawler is combined with the grasping noise produced by the encased GoPro cameras when submerged. Besides, taking into account that the auditory is a phenomenon related to waves -hence, also to movement and to this film in particular- Leviathan’s diegetic sound continually adds more vibration and motion to the already floating bodies on board. In a way, it is the ear that renders the image visible in Leviathan, because it is precisely the audio-event that then becomes the visual one. Anywhere the eye searches around the boat, the ear listens in on what is being searched: it covers, touches, and enfolds the spectator’s body.

39By way of conclusion, I propose to take the argument for embodied perception in Leviathan as an audiovisual counter-part to Viveiros de Castro’s theory on Amerindian cosmology. Bodies have become in both universes primary conductors of perspectives, attempting to bring points of view back to the phenomenologies of different species. In such way, the interspecies corporeal practice portrayed by the documentary creatively meets the marked dimension of Amerindians’ metaphysics on multiple referents. So, if the human has been traditionally seen as a “being-in-language” and the animal has been for long depicted as a “being-in-its-body”, then it is through SEL’s bodily praxis as much as contemporary anthropology where artistic and philosophical interrogations are exploring new strategies to leave behind the old dualism between the thinking cogito and the animal Other. Consequently, if the task was to make possible the experience of a place beyond its modern confines, then contemporary ethnographic practice depends exclusively on how much attention we pay to the concepts of difference, multiplicity, bodies and becomings. To borrow Latour’s expression, who is also acknowledged in the film’s credits, I hope to have shown how in SEL’s Leviathan as much as in the Amazonas ‘there are more ways to be other, and vastly more others, than the most tolerant soul alive can conceive.’ (2004: 453)


40The collision between anthropology and film ethnography presented in this article can simply be read as an assemblage between the written and the audiovisual event. Filming wast taken here to be analogous to writing, that is, embedded with a first person subjectivity and narrated with perspective. In this way, by making possible the experience of a place coming into view is that film ethnography was endorsed with an ‘anthropological power’ open to both analytical meaning and multiple viewings. As such, the discursive relation between the disciplines was primarily addressed from their similar ethnographic set of problems and changing relations in regards to the notion of “the Other”. In Flaherty’s Nanook of the North and Malinowski’s Argonauts of the Western Pacific, shared methodological conventions guided the work of filmmaker and anthropologist to construct the picture of a foreign culture. For them, fieldwork consisted exclusively in the interpretation of the local reality to the extent of creating a native world explicated by a western “I”. Hence, Malinowski’s and Flaherty’s accounts demonstrate not only how one can easily fictionalise something -starting from a cultural, historical, political or psychological reality, but also how the interpretation of the observer can also be rendered as true, thus, inducing truth effects within a fictional discourse. For this reason, rather than a ‘becoming-Nanook’ or a ‘becoming-Argonaut’ -whose positions are strictly minoritarian, what I have put forward is the common gaze adopted by Malinowski -the anthropologist- and Flaherty -the filmmaker. Their shared humanistic practice is precisely what places in ethnographic proximity the text and the film event. Both showed more interest in the lives of native people than probably any other ethnographer before them. But as I have argued, they simultaneously blocked the native’s reality by exerting anthropological knowledge to express and translate the foreign world.

41A similar series of methodological assemblages make the differences between Jean Rouch’s cinéma vérité and Marcel Mauss’ anthropology quite indiscernible. They shared the conception of reality as a multiple and heterogeneous texture which led to their views of scientific statements as nothing but the subjective experience of the observer. “Pure ethnography” is no longer possible here. Rather, what they account for is a very specific signifying practice in which the expression of the material world is produced by the insertion of what one knows -methods and experience- into what one sees -points of view. Thereby, every act of knowledge becomes a universe with perturbing holes in it, because there is always something of “the real” that remains inaccessible to those who try to apprehend it. At last, Rouch’s and Mauss’ common zone of indiscernibility was also expressed in their examination of the more prosaic and material aspects of social life, that is, the involvement of the senses as much as reason in their ethnographic enterprise.

42In the case of Viveiros de Castro’s Cannibal Metaphysics and SEL’s film Leviathan, the notion of becoming was addressed more directly; becoming and multiplicity are here one and the same thing. Alliances are not exclusively performed by the previous methodological conventions that anthropologists and filmmakers had in common. Relations exist now within a much broader network of affects portrayed by the various entities dwelling in the environment. By adopting foreign agencies through corporeality, Amerindian people as much as SEL’s documentary have demonstrated the capacity of embodied perspective to find a zone of indistinction between the One and the Other.

43My hope then is that The Colliding Worlds of Anthropology and Film Ethnography will indeed be understood as a dynamic continuum. The purpose was to explore how ethnographic knowledge is always operating on a shifting basis, thus, conceiving also the conventions and experiences of the present as a reality that may throw the apparent assumptions about the disciplines into question tomorrow.

Top of page


Books and articles

Barbash, Ilisa and Lucien Taylor. 1997. Cross-Cultural Filmaking: A Handbook for Making Documentary and Ethnographic Films and Videos. Berkeley: University of California Press.

Caixeta de Queiroz, Ruben. 2012. Between the Sensible and the Intelligible: Anthropology and the Cinema of Marcel Mauss and Jean Rouch. Vibrant: Virtual Brazilian Anthropology 9(2): 184-211

Castaing-Taylor, Lucien and Parave, Véréna. 2012. NYFF Press Conference. Film Society of Lincoln Center. (accessed 29 July, 2017)

De Certeau, Michele. 1988. The Writing of History. New York: Columbia University Press.

Dieterlen, Germaine. 1990. Marcel Mauss et une école d’ethnographie. Journal des africanistes 60(1): 109-117.

Deleuze, Gilles. 2005. Cinema 1: The Movement Image. Continuum: London. [orig. 1983].

Eaton, Mike. 1979. Anthropology-Reality-Cinema: Films of Jean Rouch. London: BFI Publishing.

Flaherty, Robert. 1969. Nanook. In The Emergence of Film Art. Jacobs Lewis, ed.. New York: Hopkinson & Blake.

Griaule, Marcel. 1965. Conversations With Ogotemmeli: An Introduction to Dogon Religious Ideas. Oxford: Oxford University Press.

Hastrup, Kirsten. 1992. Anthropological Visions: Some Notes on Visual and Textual Authority. In Film as Ethnography. Peter Ian Crawford and David Turton, eds. Pp. 8-25. New York: Manchester University Press.

Latour, Bruno. 2004. Whose Cosmos, Which Cosmopolitics? Comments of the Peace Terms of Ulrich Beck. In Symposium: Talking Peace with Gods, Part 1, Common Knowledge. Duke University Press: 450-462.

Levi-Strauss, Claude. 1974. Structural Anthropology. United States: Basic Books

Malinowski, Bronislaw. 2001. Argonauts of the Western Pacific: An account of Native Enterprise and Adventure in the Archipelagos of Melanesian New Guinea. London: Routledge. [orig. 1922].

Marks, Dan. 1995. Ethnography and Ethnographic Film: From Flaherty to Asch and After. American Anthropologist 97 (2): 339-347.

Mauss, Marcel. 2001. A General Theory of Magic. New York: Routledge [orig. 1950].

McLuhan, Marshal. 2001. Understanding Media. London: Routledge. [orig. 1964].

Pink, Sarah. 2006. The Future of Visual Anthropology: Engaging the Senses. London & New York: Routledge.

Rothman, William. 1998. The Filmmaker as Hunter: Robert Flaherty’s Nanook of the North. Detroit: Wayne State University Press.

Rouch, Jean. 1981. Entretien avec Jean Rouch. (Unpublished text. Ms.)

Rouch, Jean. 1995. The Camera and the Man In Principles of Visual Anthropology. Paul Hockings, ed. Pp. 79-98. Berlin & New York: Mouton de Gruyter.

Rouch, Jean 1978. Cine-Transe: The vision of Jean Rouch. Film Quarterly 31(3): 1-10.

Taylor, Lucien. 1996. Iconophobia. How Anthropology Lost It at the Movies. Transition 69(1): 64-88.

Udayan, Gupta and Judy, Janda. 1978. The Politics of Visual Anthropology, An Interview with Jean Rouch by Dan Georgiakas. Cineaste 8(4).

Viveiros de Castro, Eduardo. 2014. Cannibal Metaphysics. Minneapolis: Univocal. [orig. 2009].

Wahlberg, Malin. 2014. Leviathan: From Sensory Ethnography to Gallery Film. Necsus: European Journal of Media Studies. Autumn. (accessed 14 July, 2015)


Castaing-Taylor, Lucien and Véréna Paravel, dirs. 2012. Leviathan Harvard Sensory Ethnography Lab et al. USA (production). 87 min.

Flaherty, Robert, dir. 1922. Nanook of the North. Revillon Frères, USA (production). 78 min. (accessed 29 July, 2017)

Flaherty, Robert, dir. 1931. Industrial Britain. John Grierson and Robert Flaherty, Great Britain (production). 20,5 min. (accessed 29 July, 2017)

Rouch, Jean. 1966-1973. Sigui series. Centre National de la Recherche Scientifique (CNRS), Comité du film Éthnographique, France (production). 8 films.

Rouch, Jean. 1967. Jaguar. Pierre Braunberger, Les Films de la Pléiade, France (production). 88 min.

Top of page


1 Flaherty asked his subjects to revive a dangerous method of walrus-hunting that Inuit people had abandoned when rifles became common by 1920. For further discussion, see William Rothman 1998: 24.

2 Even when the Eskimos had a polygamist family unit, Flaherty depicted Nanook’s clan as monogamous to show his character in a more customary family structure according to the western tradition. For further discussion, see William Rothman, The Filmmaker as Hunter: Robert Flaherty’s Nanook of the North. Detroit: Wayne State University Press, 1998: 24. (1998: 24)

3 In a long take, for example, the viewer watches and waits as Nanook sits above a hole he has cut in the ice, waiting for a seal to appear for a breath of air. At that moment, Nanook will hurl his spear into the hole and bring food home to his family.

4 Similar reasoning is derived from Mauss’ pupil Claude Levi-Strauss. In his book Myth and Meaning, he argued that the activity of the mind in imposing form on content is fundamentally the same for archaic and modern thought. Levi-Strauss saw no differences in quality between scientific reasoning and the mythic thinking of native communities.

5 Mauss was very enthusiastic about the use of the camera towards anthropological investigations. He regarded it as an innovative technological practice to describe cultural habitus more accurately. In Rouch’s words: ‘Mauss recommended to ethnographers the use of film to record certain modes of behaviour. He had not been able, because of his experience, to detect the specific nature of a filmic orientation; for him the camera was a visual memory which could record the totality of a phenomenon." (1979:2)

6 Sensory Ethnography Lab, accessed December 1, 2015,

7 The concept of deframing, or décadrage in French, has been originally employed by Pascal Bonizer to argue that the cinematographic image can be divided ad infinitum. For further discussion, see Gilles Deleuze, Cinema 1: The movement image. London: Continuum, 2005: 16.

Top of page

List of illustrations

Title Nanook of the North
Caption Screenshot
Credits A film by Robert Flaherty, USA 1922
File image/jpeg, 68k
Title Jaguar
Caption Screenshot
Credits SOURCE:​films/​jaguar/​ right owner​fr/​catalogue-film-jaguar-DJAGU01.html
File image/jpeg, 212k
Title Leviathan
Caption Screenshot
Credits SOURCE:​theatrical/​leviathan_press.htm
File image/jpeg, 318k
Top of page


Electronic reference

Cristóbal Escobar, « The Colliding Worlds of Anthropology and Film-Ethnography », Anthrovision [Online], 5.1 | 2017, Online since 30 June 2017, connection on 20 April 2018. URL : ; DOI : 10.4000/anthrovision.2491

Top of page

About the author

Cristóbal Escobar

University of Melbourne, Screen Studies Department

Sociologist, Universidad Católica de Chile. Master of Moving Image, University of Melbourne. PhD Candidate Arts-Philosophy, University of Melbourne.

At present Cristóbal works as an editorial member for the Yearbook of Moving Image Studies (YoMIS, Germany) and collaborates with LaFuga film journal (Chile). His research focuses on the crossroads of film, philosophy and the social sciences.

Top of page


© Anthrovision

Top of page
  • Logo European Association of Social Anthropologists
  • Logo IMAF - Institut des mondes africains
  • Logo Max Planck Institute for the Study of Religious and Ethnic Diversity
  • OpenEdition Journals