Friday, February 5, 2016

Bruno Latour’s Artistic Practices: Writing, Products and Influence

As an influential public intellectual Bruno Latour has held prominent teaching positions throughout his career – he is currently at the Paris Institute of Political Studies –, has published over a dozen books since the late seventies – including Laboratory Life (1979), Science in Action (1987), We Have Never Been Modern (1991), Reassembling the Social (2005) and An Inquiry Into Modes of Existence (2013) –, and has conceptualized some important concepts in the social sciences, most notably Actor-Network Theory (ANT), in conjunction with John Law and Michel Callon. In recent years, he has also served as the curator for successful art exhibitions.
The primary research question for this paper is how does the French sociologist Latour and his theories put pressure on cinema and cinema studies? This question can be interpreted a few different ways. What relationship does Latour have with cinema, in terms of the directors that he has written about and has expressed an affinity towards? Has he been an influence on any specific directors in terms of his mentorship as a professor? Has he himself made any films, or acted in any? Does Latour’s cultural writing in general offer an idea of the aesthetic that he would value in terms of the cinematic? What about his own artistic and cultural practices? Do they have any cinema-related value? And finally, does ANT offer a useful concept for analyzing films?
This paper will try to answer those questions by discussing how Latour has intersected with cinema throughout his career. For example, he developed personal connections with some filmmakers (Paravel and Green) and his theories may have had a potential influence on them or vice versa. Latour himself has made a film, The Tarde Durkheim Debate (1903/2008), or at least conceptualized it and acted in it. Latour’s writing, curation and creation (for example, of Paris: Invisible City) has provided a model for his aesthetic theory as metaphors for his social theory. Furthermore, ANT offers an interesting perspective to the analyses of films, most notably the documentary.

Véréna Paravel and Eugène Green
In recent years, Véréna Paravel, a PhD in Anthropology and one of Latour’s former students at the École Nationale Supérieure des Mines, with Lucien Castaing-Taylor, the director of the Sensory Ethnography Lab (SEL) at Harvard University, made an experimental documentary on a fishing ship, Leviathan (2012). This was the most direct connection between Latour and cinema and film studies. Scott MacDonald in American Ethnographic Film and Personal Documentary wrote, “Leviathan is surprising – its immersion of its audience within the audio-visual surround... feels not only overwhelming, but quite new in the annals of modern theatrical cinema.” The film would also receive the cover of Cinema Scope magazine and numerous conferences and lectures on it at Society for Cinema and Media Studies (SCMS) and Visual Evidence. It marked something new and exciting for cinema.
This immersive documentary on a fishing ship in New Bedford, Massachusetts stands out for being the cutting edge of the radical emerging style of the SEL movement and its content and form display certain tenets of ANT.  MacDonald wrote, “these films exemplify the commitment of the SEL to a sense of culture as continuous transformation, interpenetration, and imbrication.” Latour’s theory of a sociology of associations, of circulation and movements, the animating of non-human actors, like marine life or a fishing ship, found in Leviathan is a compelling illustration.
Paravel even, in an interesting anecdote on the French radio-show Hors-champs, spoke to Laure Adler about a unique Skype conversation that she had with Latour. During it he gave her a tour of his apartment room through his computer, as an illustration of some of his ideas. One could even detect certain parallels between this gesture and that of Leviathan’s with its digital cameras that are always in motion and how the film evokes the fascination of the sensory world.
Another example of Latour publicly supporting a filmmaker was in January 2010 at the Centre Pompidou where Latour gave a series of conferences titled Selon/Salon Bruno Latour. There Latour met with other philosophers and artists to discuss eloquence and demonstration and how they come together through articulation and composition. In the description of the series, Latour highlights the oratory arts and how they allow for movement and a liberating potential. The arts are a form of articulation and they relate to knowledge and science. Latour wrote, “Each intellectual discipline learns how to articulate the world in its unique method, to multiply its knowledge, to differentiate itself from other disciplines, and, to facilitate its expressions and representations.” This articulation leads to a need for composition and to understand how to group its varying elements. The archaeologies of these conferences on the social sciences, philosophy and the arts share a more subjective empiricism. Among the conferences the early Einstein-Bergson debate was recreated; Latour, Donna Haraway and Isabelle Stengers spoke on the potential of cyborgs; and surprisingly the filmmaker Eugène Green discussed the Baroque and L'âge de l'éloquence
 Latour in his introduction spoke about how Eugène Green inspires him. The focus of the conference is on the eloquence of Green’s dialogue most notably in his theater work, but also in his novels and films. Green’s method, he claims, has its roots in the Baroque period, between the Renaissance and an emerging Rational Age, at the intersection of science and religion. Latour wrote,

Through his Theatre of Sapience, founded in 1977, the metteur en scène Eugène Green already was searching to revive an art of the baroque theater through utilizing his declaratory means and proper visuals: pronunciation, accentuation, rhythm, frontal acting, candle lighting, and gestures: Everything was codified by very specific rules. Today as a director and writer, Eugène Green develops in his books and in his films, with the most recent one being La Religieuse portugaise (2009), a new reflection on eloquence and the incarnation of the parole. How to re-find parole by a return to the artifice of the elocution, framing and eloquence?

How to bring together rhetoric and demonstration, Latour asks? These two concepts, which are major preoccupations for Latour, are elaborated by Green through the importance of dialogue and cinematography in his work. This interest in language parallels that of Latour’s in his emphasis on descriptive language for ANT studies. One of the topics that Green brings up comes from his 2009 book Poétique du cinématographe. In it he distinguishes between ‘bougants’ (‘move-ies’) and ‘Cinématographe’. This division into two categories parallels that of Latour’s division between a sociology of the social and that of a sociology of associations. For Green ‘bougants’ represent the commercial, non-artistic movies and the ‘cinématographe’ its more artistic and spiritual potential. Green elaborates in regards to how he films objects, people, geography and architecture which, he posits, allows for their material self to truly emerge. This parallels some of Latour’s concepts of how non-human objects can also have agency.

The Tarde Durkheim Debate
Another example of how Latour intersects with cinema is his role in the recreation of the important social theory debate between Garbriel Tarde and Emile Durkheim. This reenactment of the 1903 debate has Latour in the role of Gabriel Tarde, Bruno Karsenti as Emile Durkheim and Dominique Reynié as the Dean. It was filmed in Paris in 2007 in a conference room, filled with an audience, with its vintage stage by Frédérique Ait-Touatti, research by Eduardo Vargas and recording by Martin Pavlov.
The film illustrates the importance of their theories for Latour and a potential interest in the medium. Émile Durkheim, one of the founders of a scientific sociology, aimed at defining generalizable social facts like, for example, through statistics the general suicide rates in a particular Catholic community. Gabriel Tarde, on the other hand, argued for an emphasis on the microanalysis of actors and networks as, “every thing is a society and that all things are societies.” Latour prefers Tarde’s theory of associations and through this recreation aims to recall and call into question one of the problems with the social sciences.
It is an interesting film as the actors are not necessarily aiming for physical or oratory verisimilitude, and the debate itself has been reconstituted (the original whole is no longer available), but what stands out are the competitive ideologies of the participants, the rationale behind their ideas, and the confrontation between the hardened ‘scientific’ reason and a more abstract ‘subjective’ empiricism. The difference is that of a broad understanding of how society functions through generalizable facts, against that of an attempt to find the truth-value of a situation by a close analysis of the actors within a particular site. Through Latour’s identification with Tarde he is aligning himself with an undervalued tradition in the social sciences. The underlying gesture is to not pass over this monumental event, but to better listen to Tarde as his ideas are revelatory and effective ways to analyze society.
This gesture for a philosopher to recreate an important theoretical work has a precursor with Michel Foucault who participated in the recreation of the facsimile document of Moi, Pierre Rivière, ayant égorgé ma mère, ma sœur et mon frère (1976) by René Allio. Where Latour’s writing on the Tarde-Durkheim debate can appear to be too historical (why is this relevant today?), through its current re-creation the historical event is brought to life and is given new relevance. The debate also demonstrates Latour’s eloquence, as per his discussion with Green, and turns Latour into a cinematic screen persona. With his recognizable long face and stern expression, big black glasses, sharp suit and tie; he is giving his body to cinema as a visual expression for his ideas. The Tarde Durkheim Debate (1903/2008) is a fitting cinematic memorial to Latour’s importance.

Latour’s Art Writing and Paris: Invisible City
Some of Latour’s art and cultural writing offers interesting ways to bring his ideas to cinema. In Latour’s essay, Some Experiments in Art and Politics he discusses Tomas Saraceno’s Galaxies Forming along Filaments, Like Droplets along the strands of a Spider’s Web (2008) which he sees as a metaphor for social theory. The work, which was on display at the 2009 Venice Biennale, creates through organized wires an infrastructure similar to how Latour conceptualizes ‘networks’. Latour wrote,

What Saraceno’s work of art and engineering reveals is that multiplying the connections and assembling them closely enough will shift slowly from a network (which you can see through) to a sphere (difficult to see through). Beautifully simple and terribly efficient… Namely of explicating the material and artificial conditions for existence.

For Latour, this artwork offers a great amount of freedom for understanding connections as a thought experiment. Latour further elaborates on his artistic sensibility in the exhibition catalog essay From Realpolitik to Dingpolitik, or How to Make Things Public, where he addresses the need to think of the political beyond necessarily human and temporal categories to look at people in relation to objects to create a cohabitation between the two. There are a plethora of artworks, photography and installations from this Making Things Public exhibition at the ZKM Center for Art and Media Karlsruhe that attempt to illustrate this aesthetic. 
 Another good example of Latour’s engagement with artistic practices is his digital media online project, Paris: Invisible City, which is described as a ‘Sociological Web Opera’. It was first launched in 2004 as part of the Airs de Paris exhibition at the Centre Pompidou. In Reassembling the Social Latour wrote, “This somewhat austere book can be read in parallel with the much lighter… Paris ville invisible, which tries to cover much of the same ground through a succession of photographic essays.” There are four stages to this tour: Traversing, Proportioning, Distribution and Allowing. It begins at a department store on its rooftop panorama of Paris. Through this limited view Latour theorizes that there cannot be only one Paris (e.g. a bird’s eye view of the city, does not do justice to its local specificities or how it’s actually operated) but through many isolated representations of it, there can be a better understanding of it as a multiple, invisible (Latour’s word, referencing Italo Calvino) and virtual city.
Anne Friedberg’s conception of the virtual window is useful in relation Latour’s Paris: Invisible City. Friedberg description of the screen is that of both a surface and a frame. The screen becomes a reflective plane onto which an image is cast and the frame limits its view. This idea relates Paris: Invisible City towards the cinematic as, building upon Henri Bergson and Gilles Deleuze, Friedberg argues that computer screens in general have replaced previous incarnations of screens, from the architectural window to the cinema screen, and their specificity is the virtual quality of their representational images. Friedberg wrote,

The term ‘virtual’ serves to distinguish between any representation or appearance (whether optically, technologically, or artisanally produced) that appears “functionally or effectively but not formally’ of the same materiality as what it represents… Virtual images have a materiality and a reality but of a different kind, a second-order materiality, liminally immaterial.

The world that Latour recreated with Paris: Invisible City depends on the computer screen in a novel way, which reflects Friedberg’s conception of the virtual screen, as it finds new ways to describe a society and in its own way tell a story.

ANT and Film Analysis
ANT as a model to analyze films has its benefits and drawbacks. Where it is probably correct to assume that Paravel had been influenced by Latour’s conception of ANT it is more difficult to broadly use it as a tool to analyze documentaries. Though one possibility would be to look at certain documentaries and then make a general taxonomy of how they animate some of ANT’s major tenets. But this can be limiting as well as in then how to interpret these scenes and their meaning? There are a few documentaries such as Silvered Water, Syrian Self-Portrait (2014), The Iron Ministry (2014) and 88:88 (2015) which are worth exploring as case studies for ANT’s usefulness for film analysis and to illustrate its potential.
Wiam Simav Bedirxan and Ossama Mohammed’s Silvered Water, Syrian Self-Portrait is a documentary on the atrocities of the Syrian civil war. It is a hybrid film made from the directors’ personal footage and found footage from the Internet of the atrocities plaguing Syria. The military regime that rules the country does not allow for filming (if caught filming the person is typically killed) and the violence of Silvered Water’s imagery is typically is not reproduced in Western reporting on Syria. In the documentary there is a scene of the ruins of an old building where a broken outdoor faucet is dripping. It is a lengthy scene as Bedirxan decides to focus on this one mundane activity. This broken faucet can take on what ANT describes as the agency of a non-human actor. ANT wants to distribute agency as broadly as possible. The social and historical trajectories of the country with its recent military violence take the specific form of this faucet as an actor in this scene. Even though the city is being destroyed there is still this micro-activity occurring. But the problematic aspect of focusing on this scene, solely to compare it to ANT, would be to take away from the overall project of the film with its message of urgency about the violence of the Syrian military on civilians and the destruction of Homs.  
 Another example is Isiah Medina’s experimental documentary 88:88. Through its portrait of a working class neighborhood in Winnipeg, Medina captures these interactions between nature, people, and community in a striking and unique way. If ANT posits that everyone and everything is profoundly relational then this experimental form of filming and editing can be seen as heightening the performative nature of these interactions and the networks connecting them. ANT presumes that a person’s identity is not prefigured by the moment of analysis (or filming) so in 88:88 brief shots of figures and lack of psychology have condensed the actants in enacting their relationships. These parallels offer some insight but perhaps a more helpful reference in understanding Medina’s film would be to compare it with works that it is most likely directly citing, such as Jean-Luc Godard’s Film Socialisme (2010) and Adieu au langage (2014) or other diary and structural films.
The same would apply to J.P. Sniadecki’s The Iron Ministry. Similar to the SEL Leviathan, The Iron Ministry is a condensed trip through the major Chinese rail-road system. It offers a fascinating glimpse to its busy activity, myriad of passengers and interviews. In one scene the camera is recording a young child criticizing American ideology late at night on the train. This scene recalls Latour’s focus on description without interpretation. But then what? The fact that parallels can be drawn between ANT and certain filming techniques used in specific documentaries does not confirm that it is necessarily a useful tool for the analysis of film. More research on the subject is still necessary.

ANT, Documentary and Media
There has also been scholarship on ANT and its relation to media and documentary which gives a better understanding of how other scholars have imagined this relationship. Here are two examples of scholarship on ANT’s relation to media and documentary: Ilana Gershon and Joshua Malitsky’s essay Actor-Network theory and documentary studies which discuss how science studies and ANT can inform documentary scholarship; and Nick Couldry’s Actor Network Theory and Media: Do They Connect And On What Terms? which elaborates on ANT’s relation to media in general. These two essays offer interesting methodologies on the subject.
Gershon and Malitsky’s essay is perhaps the better of the two to bring ANT to the analysis of documentary and its para-textual objects. Instead of asking how the film animates certain ANT tenets they attempt to use ANT to discuss the larger narratives surrounding the films, the films’ possible truth claims, and the broader social response to them. Gershon and Malitsky are reacting to the post-modernist critiques of ‘claiming the real’ and its recognition that truths are socially constructed. Instead they are claiming a similarity between ANT’s distrust of dichotomies in scientific practice to that of how documentaries construct truths. For this they propose to study the extra-textual information around the film as a method to interrogate its own truth claims.
Gershon and Malitsky, cite John Law, who “delineates how ANT is fundamentally a theory of relationality, the analytical task of figuring how these relationships condense in various people and objects.” ANT allows for techniques to reveal how truths are socially constructed and how interactions are transformed into representations. Gershon and Malitsky propose four conceptual consequences in bringing ANT to documentary: everyone and everything contributes to how interactions take place; not all actants are the same; ANT insists on the performative nature of relations and their forms; and actants are all network effects. This would lead to,

The ANT perspective makes the circulation of putative truth a question of how different actants contribute to shaping a network through specific interactions. What ANT scholars provide are techniques for sidestepping the ontological question of truth entirely and focusing instead on truth-value. In other words, what the ANT perspective offers are techniques for understanding how representations might be transformed into facts through the labour of specific networks.

For documentary the ANT perspective would involve bringing all the aspects of the documentary from production, distribution and reception to see how the documentaries themselves are actants, which each convey their own information. This would make the study of them more reflexive as analytical moves provide ways to think about how truths and facts are constructed. For Gershon and Malitsky, “That is, ANT provides a way of imagining documentary pre-production, production, post-production, distribution and exhibition practices as an integrated network for circulating knowledge.”
In his essay Actor Network theory and Media: Do They Connect and on what Terms? Nick Couldry describes ANT as an attempt to explain the social order. Couldry elaborates,

through the networks of connections between human agents, technologies and objects. Entities (whether human or non-human) within these networks acquire power through the number, extensiveness and stability of the connections routed through them, and through nothing else.

Couldry uses ANT to generate a theory of connectivity that brings together the social and nature, which includes the potential of media. Within these relations, it is the networks that set the agents in positions relative to other agents. Building on Roger Silverstone, Couldry argues that,

Networks (and therefore ANT) tells us something important about the embeddedness of social life in media and communications technologies, but they do not offer the basis for a completely new theorization of social order, nor even a new way of analyzing social action, in spite of claiming to do just that.

ANT is interested in humans and their entanglement with technology. Couldry argues that “ANT’s insistence on the necessary hybridity of what we call ‘social relations’ remains a valuable antidote to the self-effacing, naturalizing potential of media discourse and of much discourse in media studies.” Couldry citing Tarde elaborates on the increasing simultaneous conversations spread over a vast geography as one of the major important developments, which has grown exponentially since Tarde was discussing newspapers. Couldry instead of seeing technology, media or even the Internet as a faceless mechanical entity prefers ANT for being able to localize the specific relations entangled within it.
But Couldry also sees a limit to ANT and has his own critiques of it: ANT has a problematic relation to time since it neglects it. Couldry sees in this neglect of the long-term consequences of networks as overlooking of social power and the possibilities of resistance. He does not view ANT as successful when analyzing texts that are meant to be interpretative. As well Couldry argues that ANT has little to say about the processes that come after the establishment of networks.
Althought Couldry sees a relationship between ANT and media theory, he argues that it is both significant and uneasy. It is an antidote to the more functionalist versions of media theory but its problems of an insufficient attention to questions of time, power and interpretation are still serious. But it is a good base for more research around these questions.

Conclusion
Bruno Latour, even without directly being involved in film studies, has still managed to put pressure on the cinematic in interesting ways throughout his career. This influence is seen through his teaching relationship with Véréna Paravel, who would bring many of ANT’s tenets to the radical approach of the ethnographic documentary Leviathan, and also through his participation in the recreation The Tarde Durkheim Debate, which has him casted as Garbriel Tarde to eloquently debate some of the social theory ideas that contributed to shape him. His art writing, curation, and creations like the virtual Paris: Invisible City have theorized some key concepts in the aesthetics that he would privilege, which would enact his social theories by bringing together objects and humans with the goal of creating new relations. And finally actor-network theory allows for an interesting new approach to study the documentary, which still needs to be further developed. Latour’s intellectual work has provided cinema and film studies with many stimulants and his ideas still need to be further explored for a better understanding of what they have to offer.

No comments:

Post a Comment