UvA-DARE ( Digital Academic Repository ) In Love with My Footage : Desirous Undercurrents in the Making of an Essay Film on Candomblé

Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.


Introduction
D uring the editing process of my film The Possibility of Spirits (Van de Port 2016), 1 I had a conversation with the video-artist Koštana Banović. Koštana is an experienced filmmaker, with a rich oeuvre of films on religious practices in different parts of the world. I have been writing as an anthropologist for decades, and I am a relative newcomer to the practice of filmmaking. We discussed differences between writing and editing processes. I told her how much I enjoyed rummaging through my footage, for hours on end. How I tend to forget everything around me, once I enter the worlds on my screen. My enthusiasm must have been evident. She told me, smiling: "From the way you are talking I can tell you've fallen in love with your footage! I've seen that happening more often. Recent converts to this medium fall in love with the images they make." I immediately realized how well these words captured the kind of engagement I have with my Bahian footage. I am infatuated with the beauty of my images. I identify with their content in a deep and heartfelt manner. I want to hand myself over to them, to get lost in their plenitude. I feel an urge to be in their presence, to revisit them, over and over again; to sing their praise to whoever wants to hear it. I have had such feelings for lovers. Never for my anthropological writings.
My urge to immerse myself in my Bahian footage raises an issue about anthropological filmmaking that is well worth pondering. After decades of pitting "textual" and "visual" anthropologies against each other, there is increasing consensus that such comparisons are, intellectually speaking, a dead-end street (cf. Ferrarini 2017;MacDougall 2004). They introduce a logic of "othering" whereby one medium acquires its particular qualities by setting it off against perceived characteristics of the other, with all the erasures, distortions, and blind spots that othering tends to produce (cf. Law and Ruppert 2016). There are no "visual" and "textual" anthropologies; there are only hybrid anthropologies, mixed-media anthropologies, multi-channel-modes-of-communication anthropologies.
And yet, arguments such as these did not diminish a burning impulse to shut down the bright, shadowless, orderly environment of my word-processing software, and open up my computer screen for the sounds, colors, textures, and movements of the footage I shot in Bahia. Which left me to conclude that at some level-more specifically: at the level of my desire-differences between textual and audiovisual anthropologies are real.
This essay picks up on that observation. It takes this craving to be with (work with, think with) my footage as a starting point for an alternative line of inquiry as to what kind of knowledge practice a multimedia anthropology might be. The aim was not to arrive at an understanding of the "essential differences" between textual and visual anthropologies, but to focus on the desiderata that underlie the actual process of mediation: our desire to communicate "fieldwork findings"; the way we find ourselves seduced by the communicative promises of particular media; the inevitable frustrations and disappointments of such promises; and the fantasmatic replacements that are brought up to cover the perceived lack.
Obviously, declaring the notion of desire to be relevant to the media we choose to report on in our research is like opening a window onto the murky, inchoate undergrowth of academic knowledge practices of which desire is part. 2 It is to question the cherished image of the academic as the master of his or her project, in full methodological and epistemological control. Instead, it exposes the researcher as a figure chasing fantasies that screen off the inevitable failure of attempts to cast the plenum of existence in one or the other mediatic form. However, I immediately sense what the gains might be with this approach. First, I am no longer analyzing "the medium," but the actual process of mediation. Second, looking at desire introduces a perspective that highlights the "lack" that is present in every mode of mediation: it invites us to consider media as fantasy objects, whose communicative promises are always bigger than what they actually accomplish, and which thus trigger the desire for alternative media that might make up for the perceived lack. 3 Third, by looking at the concrete interactions between a medium and its user, I am obliged to take in the situatedness of the process of mediation-the fact that media respond to the communicative desires, frustrations, and fantasies that come up in a particular fieldwork context, where particular questions are asked, and particular insights obtained.
In what follows, I situate the production of The Possibility of Spirits in my ongoing quest-henceforth called "desire"-to articulate what John Law calls "the messiness of the world," the fact that "the world in general defies any attempt at overall orderly accounting" (2004,6). I discuss how I was seduced by the idea of a "delinguified" ethnography, as exemplified in the filmic work of the Harvard Sensory Ethnography Lab (SEL), in which images, sounds, and silences move the viewer to new modes of understanding other people's lifeworlds; how I gradually became frustrated by that effort; and how I found (fantasmatic) solace in the genre of the essay film. In conclusion, I assess how this approach might reconfigure ongoing debates about the merits of textual and visual modes of doing anthropology.

Academic Writing and Its Discontents: Articulating Candomblé's Mysticism
I started studying Candomblé in 2001 in Salvador da Bahia. Previous research in war-ridden Serbia had already forced me to question the conventions of academic text making, which, as Law brilliantly put it, seek to "distort reality into clarity" (2004,4). I criticized the incapacity of the academic text to accommodate the messiness of the world I had studied, its odd insistence that by sorting out and cleaning up Serbia's ever muddier realities I would arrive at a privileged understanding of that world (Van de Port 1998). Bahian Candomblé questioned academic knowledge practices in other, no less radical, ways.
From the very beginning of my fieldwork in Bahia, I had encountered a polite but stubborn refusal on the part of my interlocutors to discuss and explain the mysteries that constitute Candomblé religiosity. This resistance resonated with much that I had been reading about mystical traditions, which consider language to be the instrument of human world-making, and as such incapable of articulating the divine. 4 Divine truths are encountered in immediate bodily experiences, which are provoked by a great variety of body techniques. And although many religions appreciate the efforts of mystics to report on their experiences to the community of believers in poetry or prose, this is not the case in Candomblé. Here, people actively seek to keep their mysteries a mystery ( Figure 1).
Much of my written work on Candomblé is an attempt to take this nonconceptual way of knowing seriously (Van de Port 2006, 2015. Basically, what the people from Candomblé brought up for reflection is: what kind of knowledge is produced in the inter- view where the informant explains that "initiation lasts seven long years"? How could a five-word message, which communicates an idea in a split second, ever capture the embodied experience of "seven long years" in real time? People from Candomblé told me it cannot. Insisting that "deep knowledge" cannot be unchained from the passing of time and the experiencing body, they asked me to ponder the question of what gets lost in the endless abstractions that social science writing performs.

Mattijs van de
By the time I finished Ecstatic Encounters (2011), I found myself thinking that in the next project, I would move to a more literary, poetic mode of writing: less prone to abstraction, hospitable to ambiguity and silences, appealing to the senses as much as to the intellect. A joint project with Koštana Banović in Bahia triggered a lingering interest in film as a medium for anthropological research. 5 Koštana was making a film on Candomblé, and I had offered her some assistance, introducing her to different places and settings. I was immediately intrigued by the kind of interactions her camera brought about in Candomblé temples-animated and spirited exchanges about beauty, instead of the cumbersome conversations that my anthropological questionnaire (going after the "meaning of things") tended to produce. These exchanges were not, in any way, openings to a conversation, but more like a constant pointing out of beauty; a punctuating of time with exclamations such as "look how pretty!" or "isn't this beautiful?"; an ongoing invitation to simply join the appreciation of beautiful things, without further questioning or comments. Koštana did just that, as her camera recognized and registered the enormous investments that were made to decorate the temple, embellish the altar, spectacularize the ceremonial dresses, arrange the dishes with food in an aesthetically pleasing manner. Time and again, I saw people asking her to replay her recordings on the LCD screen of her camera, frequently followed by enthusiastic remarks that on the screen, things "looked even more beautiful!" 6 Koštana's camera-based research made me see just how central "making things beautiful" is in Candomblé religious practices. Far from a mere preparatory act to a proper "religious" event, beautification is the sine qua non of spiritual presence. It is a mobilization of the sublime to provoke the ineffable presence of the divine. By filming, and not asking questions or seeking explanations, Koštana honored the mysteries. Now anyone could see them on her videos, and be baffled (Van de Port 2011).
Just as I realized that, methodologically speaking, the camera presence offered a way out of the awkwardness of asking questions about things that cannot be discussed, I also realized that film offered possibilities to explore the aesthetic itself in ethnographical reporting. For, as MacDougall points out, "images and written texts not only tell us things differently, they tell us different things" (1998,257). Or, to paraphrase Anna Grimshaw, film might complement the explanatory text with an open space of suggestive possibilities between the experiential and propositional, between the perceptual and conceptual, between lived realities and images of the Candomblé lifeworld (2011,258). Following these insights, I soon managed to silence skeptical voices in my head muttering that "maybe in another life" I could become a filmmaker. I bought a camera and allowed myself to become a filming anthropologist.

Fantasies of a Wordless Anthropology
The making of The Possibility of Spirits was driven by a simple question: what on earth is a spirit? That question harkened back to several discussions in the study of religion that I was interested in. First, it linked up with Bruno Latour's admonition that researchers of religion should stop imposing their own understandings of the possible and the impossible onto the realities of others. "Follow the natives, no matter what metaphysical imbroglios they lead us into," he suggested (2007,48). Second, it referred to Stephan Palmié's observation that academics "cannot seem to resist transcribing spirits, gods or the work of witchcraft into codes that satisfy [their] deeply held beliefs that stories in which they figure are really about something else: category mistakes, faulty reasoning, forms of ideological misrecognition, projections of mental states, and so on…" (2002,3). Third, the question took in Birgit Meyer's critical revision of mentalist approaches to religion. Interested in how "the genesis of extraordinary presence occurs" (2012,29), Meyer suggests that one should study notions of the transcendent as grounded in experiences that are immanent to a material world, indicating how the possibility of spirits might be researched empirically.
As this guiding question for the film had only emerged after my fieldwork in Bahia, I had hours and hours of footage, reflecting my rather serendipitous filming activities, loosely organized around the religious calendar of Pai Ro's temple, the activities in the workplace of an old blacksmith who makes iron statues and other artifacts used in Candomblé ceremony (Figure 2); and a half-conceived plan to realize a multiscreen video installation together with Koštana Banović on permeable boundaries.
The process of editing started with selecting the footage I kept returning to when opening my files: a number of shots that spoke to me-or rather: called me out-and that I wanted to be in the film "no matter what." Two little boys watching me setting up my camera to film an altar of Candomblé in the woods, who then turn the tables by asking me if I ever made "a good film about werewolves." A lengthy close-up shot of Pai Ro being possessed by a spirit called Oxóssi, sitting on his throne, receiving the loving embraces of the members of the temple community. The landscapes around Santiago de Iguape. My friend Lucas and I jointly wringing a white sheet all the way until the air creates a bubble from which foam starts to emerge. The fact that I was drawn to these images, that they "spoke to me," was sufficient reason to put them on the timeline of my editing software, and I followed Koštana's advice "to simply take it from there" by put-ting different shots next to each other, and then to see what happens.
The search for the "right" footage-or the "right" juxtaposition of scenes-was akin to what John Law might call "a technique of deliberate imprecision" (2004, 3). The quality of being "right" would denote an interesting approximation to an idea, a suggestive tension, an invitation to the imagination of the viewer. Shots of antennas on rooftops, for instance, might speak to the calling of spirits. Shots from the circle dance xiré, juxtaposed with the tumbling laundry in a dryer, might situate Candomblé into the everyday world of a modern city. My editing, then, worked toward a "blunting of a meaning too clear, too violent" (Barthes and Heath 1978, 55). I sought to create space for what Roland Barthes called obtuse meanings-"the supplement that my intellection cannot succeed in absorbing" (ibid.).
Obviously, this highly intuitive play with images was informed (and to some extent legitimized) by my ongoing readings in visual anthropology, and my indulgence in ethnographic films. I read how other scholars had sought to use the medium of film to report on spirits and spirit mediumship (Mollona 2014;Russell 1999). I was especially inspired by the work done by SEL filmmakers, and in particular Sweetgrass (Castaing-Taylor and Barbash 2009) and Leviathan (Castaing-Taylor and Paravel 2012), which self-consciously exhibit "a retreat from language, and a withdrawal from the verbal and the denotive in favor of the visual and the connotative" (Barbash 2001, 370; see also Grimshaw 2011, 256). Concerning this approach, Lucien Castaing-Taylor has argued against the "linguification" of documentary film, explaining that he seeks to exhibit "a deficiency of authorial intelligence" and show "an ambiguity that is at the heart of human experience itself" (1996,86). Pulling away the documentary form from its "affinity with broadcast journalism, with the lecture format and with talking heads to discuss the world rather than experiencing it" (Castaing-Taylor, in Chang 2013), these visual anthropologists pushed the idea of a sensuous cinema, offering audiences the experience of being immersed in the worlds that they are presented with. 7 Given my attempts to acknowledge Candomblé's experiential mode of knowing, I felt these films offered the escape route from analysis and explanation I was looking for. Many of the reviews of Leviathan read like ever so many pleas to open anthropology to mystical modes of knowing. Hunter Snyder reports how he, as a member of the audience, became an "embodied camera" as "sensory depictions make for an internal emotional reality that takes over the body from the inside" (2013,178). Leviathan, he states, records "the illogicality of life itself" (ibid.). Matthey Battles similarly expresses the lure of the ecstatic mode of knowing that Leviathan proposes. "In Leviathan, a trawler is not a worksite to be explained or an environmental disaster to be descried, but a buzzing, blooming world emerging at the rheumy membrane where sea meets steal" (Battles 2014, 479). In a special issue of Visual Anthropology Review devoted to the film, Lisa Stevenson and Eduardo Kohn, having allowed themselves "to be made over by Leviathan" (2015, 50), report that "thanks to certain tools (such as multiple waterproof GoPro cameras strapped to the heads of fishermen, attached to the sides of the boat, submerged into the ocean's depth), we become resolutely, adamantly part of the thickness, the density, and the turgidity of a world in which it is very difficult to find our land legs" (ibid.).
My understanding of work produced by the SEL allowed for the thought that it would actually be possible to put Candomblé's mysteries center stage, without explaining them away. Through my editing, I would have to expose the carnal, visceral, bodily dimension of Candomblé rituals; the rich material textures of this lifeworld; the demanding rhythms of the drumming; the emotionality of songs and prayers. I would work with lengthy shots that would give spectators the time to sink themselves into the images. I would privilege (extreme) close-ups, so as to draw spectators ever closer to the phenomenon of spirit possession, suggesting an intimate acquiescence with the spirit mediums, while simultaneously provoking sensations of bafflement and notknowing. I would avoid long shots, so as to prohibit all associations with the attempt to "map" this world, or provide a sense of "overview." In line with the idea to keep mystery center stage, I would work with the many shots I had where people would be standing "in the way" of a ritual action taking place, screening off the activity, offering only glimpses. Using shaky footage from scenes filmed with a handheld camera, I would seek to provoke the disorientation and dizziness of the early stages of possession by cinematographic means. I fancied I might sufficiently mess up those level-headed modes of academic knowing in which all that spirits can ever be is a form of false consciousness. And, who knows, I might actually produce the derailment and confusion that I deem to be a necessary precondition for academics to at least consider the possibility of spirits (Figure 3).

Fantasies Wear Off
It didn't really work. Sure, the footage was captivating enough to lead viewers into an elsewhere, where they might be moved out of habitual modes of understanding and get lost. But as I progressed in the editing, I was gripped by a number of questions that I could not answer.
The first set of questions concerned an unstated promise underlying films such as Sweetgrass and Leviathan, namely, that they give viewers access to the worlds they portray, and allow them to become part of their "thickness and density." Both films are extremely successful in providing for this experience. When you come out of Sweetgrass, you feel you have been in the Rocky Mountains. When you watch Leviathan, you are on a fishing trawler. However, the cinematographic production of the experience of being transported to an elsewhere doesn't allow for the revelation that "the scene" to which one is taken is an authorial assemblage: one that plays with the powers of the photographic reality-trace to seduce viewers to "take it for real," but a montage nonetheless. These films do little to remind me of the fact that the world on the screen is always marked by the camera presence and the performances such presence induces. I kept thinking: shouldn't the audience be alerted to the way it is being "played"? Wasn't the critical media awareness that the writing culture debate had provoked by asking "what is an ethnographic text?" equally relevant for the medium of film?
A second set of questions concerned the idea of immersion itself. I did not doubt that I could offer academic audiences "an experiential encounter" with the world of Candomblé and that a refusal to offer "explanations" would keep the mystery a mystery. 8 Yet I kept wondering whether the responses of an academic audience to my images would be adequate to what is actually being shown on the screen. 9 Take the footage I had of the slaughtering of a sacrificial animal, a billy goat, for Pai Ro's caboclo-spirit Green Feather. These shots present the cutting of the animal's throat (and the spine-chilling howling of the dogs at the very moment of the killing); the pouring of large quantities of blood; the passage from life to death as the spasms of the dying animal slowly come to a halt; the white clothes of the adepts being splattered with blood, highlighting the intensity of the color red; the severing of the head and testicles of the goat. Undoubtedly, these scenes, introducing the powers of "primordial symbols" (Aretxaga 1995), are impactful as much for the participants in the ritual-who are visibly moved as the sacrifice takes place-as for any audience watching them. To state the obvious, the deep sense of being witness to an extraordinary happening is different for the people present in situ and for an academic audience in a cinema. For the former, the slaughtering of animals is also part and parcel of their everyday food practices. The killing in itself is not what moves them: what moves them is that the sacrifice opens a channel through which spirits and humans connect. For film audiences who may not be familiar with animal sacrifice, it is exactly the act of killing that triggers a sense of "shock and awe." To leave these differences unaddressed would maintain an understanding of experience as universal, rather than culturally informed. 10 Finally, I began to question a tendency in anthropology that celebrates indeterminacy, immediacy, fluidity, and affect as the "really real," and casts human world-making as a process that drains the magma of being of its vitality (cf. Van de Port 2015). I too subscribe to the idea that "being" is always in excess of the worlds of meaning imposed on it. In fact, the thought that "the world does not comply with our narrations of it" (Van de Port 2011) is something of a mantra in my own work. But that in itself does not qualify this blooming, buzzing world as life itself. World-makingnarration, mediation, the production of meaning-is as much part of this life as everything else. William Mazzarella's critical discussion of what he calls "vitalist" tendencies in the literature on affect articulates my discontent: …rather than positing the emergent as the only vital hope against the dead hand of mediation, why not consider the possibility that mediation is at once perhaps the most fundamental and productive principle of all social life precisely because it is necessarily incomplete, unstable and provisional. (Mazzarella 2009, 302, emphasis added) Indeed, as human beings we have no other option but to narrate (or as Mazzarella says: mediate) life and being. But instead of qualifying such narrations as "the dead hand of mediation" (ibid.), we ought to acknowledge that all that we ever study are the inescapable tensions that occur when being is subjected to meaning. These tensions define human existence-and anthropology as the study of human life.
It is stated in popular parlance, and theorized in Lacanian thought, that fantasies wear off. For a while, fantasies promise that a particular object-a lover, a change of career, a fine leather jacket, a new mode of communicating fieldwork findings-might give us what we are lacking, and the acquisition of that object may actually produce fleeting sensations of everything falling in place. When fantasies wear off, however, lack inevitably returns to the scene, and our desire to undo it drives us toward new fantasy objects. Thus, my acquaintance with immersive modes of filmmaking had triggered fantasies of being able to provide my audiences with a more full-bodied encounter with Candomblé: one where the lack of structural, semiotic, or cognitive anthropologies would disappear in the sensuous plenitude of moving images and sound. With the dilemmas sketched above, however, the lack in my newly envisioned mode of doing anthropology became ever more visible.
Luckily, however, the processes of mediation come with a learning curve. My search was not for the reinstallment of text as the privileged medium for anthropological research. The search was for a voice that could bring in the dilemmas sketched above. Clearly, this voice should not be the traditional voice-over narration of many documentary films, where a purportedly allknowing narrator claims expert knowledge about "the others" on the screen, and attributes the images with unequivocal meaning. Nor should this voice in any other way be "the master" of images and sounds, reducing film to a kind of "illustrated lecture" (MacDougall 2004, 223). The voice I was looking for would, as Trinh T. Minh-ha beautifully phrased it, not speak about its subject, but speak "nearby" it (in Chen 1992). It would, in other words, be one element out of the many that make up a film; co-present with moving images, sounds, and silences; part of the ever-changing configurations between these different elements as the film proceeds toward its ending; not only steering the interpretation of images toward possible meanings, but also being steered by images. My new fantasy object was the essay film, a genre that seeks to make the very tension between textual and visual modes of knowing present in a film.

A New Fantasy Object: The Essay Film
Considering the intellectual, theoretical bent of the essay film, the genre is remarkably absent from discussions on ethnographic film. The essay film grew out of attempts of experimental filmmakers who were interested in a "cinema of the word" and had been exploring how film could express thought (contrasting it with "show" and "mere entertainment"). The term was coined by German artist Hans Richter in his 1940 paper "The Film Essay." In the late 1940s, Alexandre Astruc famously proposed the development of "la caméra-stylo," which would allow for a "cinema that will gradually break free from the tyranny of what is visual, from the image for its own sake, from the immediate and concrete demands of the narrative, to become a means of writing just as flexible and subtle as written language" (1948). André Bazin used the concept of the "essay film" in 1958 in his musings on the work of Chris Marker, assessing that "the primary material in the essay film is intelligence… its immediate means of expression is language, and the image only intervenes in the third position, in reference to this verbal intelligence" (in Rascaroli 2008, 29). Later filmmakers whose work has been associated with the genre-among them Alain Resnais, Agnès Varda, Jean-Luc Godard, and Harun Farocki-opted for different configurations of intelligence, language, and image, or altogether rejected the kind of hierarchies introduced by Bazin. Yet an investment in exploring the cohabitation of verbal and pictorial "intelligences" in film remained a defining feature in their work. 11 In film studies, essay films are characterized by their interrogating mode of approaching the world, their "inquisitive attitude" (Rascaroli 2008, 33), their being a kind of "thinking out loud" (Corrigan 2011, 15). Laura Rascaroli states that, like written essays, they aim "to preserve something of the process of thinking" (Rascaroli 2008, 26). Another characteristic is that essay films are open to experimentation. Not only do they seek new ways to relate images, sounds, and texts; but they also allow for all kinds of materials to be included in the work: found footage, archival images, photographs, drawings, special effects, writings, musical scores, ambient sounds. Basically anything that helps to carry the development of an argument in the making is allowed. 12 Subjectivity is a third characteristic of the genre that is frequently mentioned. By definition, essay films introduce an authorial voice, which introduces a certain issue to ponder. This authorial voice, as Rascaroli phrases it, "approaches the subject matter not in order to present a factual report (the field of traditional documentary), but to offer an in-depth, personal and thought-provoking reflection" (Rascaroli 2008, 35).
Yet, since film operates simultaneously on multiple discursive levels-image, speech, titles, music-the literary essay's single, determining voice is dispersed in cinema's multi-channel stew. The manifestation or location of a film author's voice can shift from moment to moment or surface expressively via montage, camera movement and so on. (Arthur 2003, 58) Lastly, subjectivity in the essay film often implies an awareness of the act of mediation. The "inquisitive attitude" not only concerns the topic of the film, but also frequently takes in a questioning of the medium as a knowledge-producing instrument as well, exploring its enabling and disabling dimensions. In that sense, essay films are frequently auto-referential: they expose themselves as film. 13 It was this new "fantasy object," the essay film, that really allowed me to find a satisfactory form for The Possibility of Spirits. Giving up on the ideal of delinguifying film, I allowed myself to tell a story and make the most of "the surprises that could arise from encounters of the word and the image" (Pantenburg 2016, 33). Voice-over narration, which I had long rejected as an option for being a second-rank solution to problems that should be solved "cinematographically," turned out to be crucial. The voice I introduced, however, did not disambiguate images, but underscored their ambiguity. It is a voice that questions what appears on-screen, much like the Verfremdungseffekt in Brechtian theater, which seeks to keep the members of an audience critical and alert, and prohibits them from being carried away by the unfolding drama. In addition, I started to play with the different voices in the film: the voices of my interlocutors, contradicting themselves and each other in their attempts to articulate the notion of spirits; my own voice, as it appeared in the field recordings (giving directions to people in front of the camera, sometimes asking stupid questions); and my voice-over, which could provide critical comments on my voice in the field.
Clearly, I soon learned that "voice" has been amply discussed in visual anthropology (MacDougall 2004;Piault 2007;Ruby 1991). As Jay Ruby puts it in "Speaking for, Speaking about, Speaking with, or Speaking Alongside," these debates tend to concentrate on questions of authority and authorship (Ruby 1991, 50). The essay film adds to this discussion in interesting ways. Its searching, subjective, and inconclusive mode of thinking out loud invests an ethnographic film with an expressive layer of epistemological doubt. It uses voice as a tool to propose but not impose; to suggest but not state; to hint but not guide; to narrate while being critical of narration; to complicate, unsettle, and question. Highlighting the limits of our knowing (and the knowing of the Other, for that matter), the essay film seeks to induce curiosity and the joy of intellectual probing. As Trinh T. Minh-ha phrased it, when arguing against the use of language as a mere vehicle of ideas, an instrument for capturing or explaining cultural events: "every time one puts forth an image, a word, a sound or a silence, these are never instruments simply called upon to serve a story or a message. They have a set of meanings, a function, and a rhythm of their own within the world that each film builds anew. This can be viewed as being characteristic of the way poets use words and composers use sounds" (in Chen 1992, 85-86).

Instead of Conclusions: Food for Thought
The essay film collapses the oppositions that have been used to distinguish, delineate, praise, and criticize "textual" and "audiovisual" anthropologies. It explores the actual cohabitation of words and images in film. It seeks to make the most of the expressive potentials that rest in their neighborliness, their inextricable entanglement. Unsurprisingly perhaps, my encounter with the essay film triggered a new fantasy of completion: when textual and audiovisual anthropologies come together under one roof, one may make up for what the other lacks. I'm still under the spell of this fantasy. In conversations with other anthropologists, I find myself saying that it is a genre that might counter one of the strongest critiques of a visual anthropology-its lack of contribution to theory (cf. Pantenburg 2016). For if, as Annemarie Mol put it in her radical revision of the notion of theory, we do not theorize "to finally, once and for all, catch reality as it really is," but instead, to "make specific, surprising, so far unspoken events and situations visible, audible and sensible," and "to shift our understanding and to attune to reality differently" (Mol 2010, 255), then surely the essay film is a very promising genre. I am currently finishing a new essay film, 14 which not only explicitly addresses theoretical debates in anthropology, but also seeks to intervene in these debates with the pictorial, acoustical, temporal, and spatial affordances of film. Luckily, I find that ever more anthropologists are engaged in such experimentation (see Cox, Irving, and Wright 2016;Schneider and Pasqualino 2014).
What remains to be said in this essay is what a focus on the desiderata underlying the process of mediation brings to discussions about textual and visual anthropologies. I began by looking at anthropology as a practice of mediation in an attempt to communicate research findings. Dissatisfied with ongoing attempts to define essential differences between textual and visual media, and yet unable to deny my infatuation with my footage, I have sought to discuss the process of mediation in terms of the longings that drive it, introducing a vocabulary of desire, lack, and fantasy to analyze the choices I made in the making of The Possibility of Spirits.
What food for thought has been brought up by this exploration? First, it has underscored the importance of making discussions of mediation "case sensitive" (Van de Port and Mol 2013). I traded the desire to make grand statements and sweeping generalizations about one or the other medium for the specificity of the single case. What I made out of camera technology cannot be disentangled from the dilemmas I encountered in studying Candomblé. The editing of my film was never anything other than an attempt to resolve the issues of unspeakability that Candomblé gave me to ponder. In other words, the way I have come to understand camera-based research-what it is up to, what it can do-cannot be disconnected from the fact that I sought to resolve specific questions pertaining to a specific research setting. I suspect that had I studied dairy farming in Denmark, bankers in the city of London, or the role of transvestites at Sulawesi weddings, my understanding of what camera-based research is, and what it can do, would have been different (and I might not have fallen in love with my footage). Indeed, the camera would have been a different instrument, just as it will be a different instrument when I take it to new settings.
Second, looking at the desiderata underlying processes of mediation forced me to face the incompleteness of any attempt to mediate the world. The vocabulary of desire, lack, and fantasy reveals just how much we keep dreaming up totalizing accounts of all that we've learned in the field. As the editors of Beyond Text? ask, "What might be gained, we must ask, in taking seriously the ephemeral and fleeting senses that we encounter in the field but that do not achieve the requisite stability to enter into language, let alone theoretical interpretation?" (Cox, Irving, and Wright 2016, 5). Even if we have long given up on the rhetoric of "mastery" and acknowledged the hubris of holistic pretentions (Verrips 1988), what drives us to new modes of mediation is a sense of lack, of missing out on things that matter, of dimensions not yet touched upon, or even systematically ignored in a prevalent mode of mediation.
Within this case-specific, dynamic approach to the mediation of anthropological research findings, the notion of fantasy invites us to conceptualize media as "promising" us things, but not necessarily accomplishing what they promise. This is not to say that we should ignore the fact that different media present us with a different range of affordances, but-as the very notion of "affordance" denotes-these qualities are potentialities, and only come into being when they are being acted upon. The notion of fantasy helps us to follow these interactions between media and users. As media promise us things, they trigger fantasies: if only I lock myself up in my word-processing software, I will be able to bring order to my thoughts. If only I switch to my editing software, I might keep the sensorial fullness of Bahian lifeworlds up on its feet. If only I embrace the genre of the essay film, I can have the best of two worlds. Research along these lines would no longer be about what media can or cannot do in communicating research findings, but how their fantasmatic powers are enlisted in thinking up new anthropologies. and Arjang Omrani's Unless the Water is Safer than the Land. 14 Knots and Holes. An Essay Film on the Life of Nets, which is forthcoming in 2019.