[Part 3 in a series. You can read part 1 here and part 2 here.] In an earlier post, I contrasted Bazin‘s theory of film as “physical reality as such” with Georg Tholen’s formulation of the digital as “transmissibility as such.” While celluloid contains a chemical inscription that directly relates to the object it records, digital cinema is constituted by generic code–the pure difference between 1′s and 0′s. How should we conceptualize the middle space wherein text-specific cultural codes are grafted onto generic digital codes? In other words, at what level does actual meaning, or signification come into existence in the transition from “physical reality as such” to “transmissibility as such” in digital cinema?
Motion capture seems poised in a conceptual freeze frame within this transition from the physical to the transferable. Rather than recording images which are put through some recondite process of transduction into digital code, motion capture records discrete points of movement as data.
This technique was a significant part of the collage of digital effects behind Iron Man’s digital armor. From this great article:
For Stan Winston’s Shane Mahan and his suit design team, this required making a suit that could be worn in sections over the visual effects suit Downey wore. “The big challenge was trying to find ways to blend, cross-cut and inter- cut combinations of practical and CGI shots,”says Mahan. “It would be absolutely foolish for me to think that I could pull off every shot in the practical suit, so we created a combination for Robert consisting of the chest piece, helmet and arm sections combined with a full-body motion capture tracking marker suit underneath. It’s a great way to blend the practical with the computer-generated effects, enabling ILM to bridge any gaps between the physical pieces.
Iron Man’s armor itself contains this tension between the trace of physical pieces and constructed code. But motion capture seems to highlight a deeply troubling question about the ontology of the digital image, namely a question of degrees of indexicality. When the raw material taken up by the motion capture camera eye contains no indexical relationship to the physical appearance of the actor, but rather consists already of raw data abstracted form movement, is the “indexical contingency” (as Hansen refers to it in an earlier post) of the motion capture portions of the armor any more compromised than those portions that had existed in material reality on Robert Downey Jr.’s body?
The all or nothing critique of the digital as heralding the death of photography and film seems to lack a significant nuance. Thinking through the “degrees” of indexical contingency in the varieties of the digital image would entail a thorough engagement with Barthes’s “reality effect” (which perhaps finds its counterpoint in motion capture: a hidden language, unreadable between the process of recording and transmission, comes into the service of believably depicting the utterly unrealistic).
In an interview with animation supervisor Kenn MacDonald on the possibilities afforded by the motion capture technique for the recet Beowulf film, “This method of filmmaking gives him freedom and complete control. He doesn’t have to worry about lighting. The actors don’t have to hit marks. They don’t have to know where the camera is. It’s pure performance.”
What “pure performance” seems to imply for the actors is a complete liberation from the constraints of the cinematic apparatus. What “complete control” implies for the filmmakers is the possibility of ubiquitous and invisible manipulation. The actor performs freely on an empty stage while the process of image production and transmission takes place elsewhere, unseen. In this account, it seems as if the actor in the motion capture suit exists somewhere in between cultural and digital codes. She is an organism operating outside the technological apparatus whose body is nevertheless already encoded with data, and it is merely this data that the camera is interested in capturing instead of any indexical image.
Motion capture is at once a separation from or an invisiblity of the apparatus that enables transmissibility—pure performance—and a total saturation of the digital language that enables image production—complete control. The rhetoric of indexical contingency, of realism itself is inverted: the invisibility of the semiotic language (digital code) allows the depiction of the utterly unrealistic: Tony Stark testing his Mk-1 boot-jets and flipping upside down into a wall. In motion capture, the Byzantine process of digital transduction is condensed into the figure of the actor covered in data points: it is an encoding of the body prior to its encounter with the cinematic apparatus.
In the introduction to Siegfried Kracauer’s Theory of Film, Miriam Bratu Hansen writes, “Digital technologies such as computer enhancement, imaging, and editing have shifted the balance increasingly toward the postproduction phase, thus further diminishing the traces of photographic, indexical contingency in the final product” (vii). A powerful trend in post-war film criticism was the argument for the literal quality of the cinematic image, what Hansen here refers to as “indexical contingency.” As Erwin Panofsky described it in a 19** essay “Style and Medium in the Motion Pictures,” “the medium of the movies is physical reality as such. Cinema is committed to communicate only by way of what is real.” André Bazin similarly argued that the cinematic image “redeemed from sin” the false perspectivalism of Western painting. A film theorist deeply influenced by his Catholicism, Bazin writes, “The cinema is objectivity in time.”
While of course this unproblematic collapse of the signifier into the signified has since been tempered, as Hansen’s comment suggests, the mutability of digital images resurrects the discourse (rightly or otherwise) of our fundamentally intuitive understanding of cinema’s direct referentiality, if only as a “trace.” Roland Barthes’s diagnosis of the “reality effect” comes to mind, which posits “a break between the ancient mode of verisimilitude and modern realism,” a discourse “which accepts ‘speech-acts’ justified by their referent alone.” The “reality effect” is constituted by a language whose signified is the very absence of a signifier; it is a language whose aim is to simulate immediate presence. Now that the “indexical contingency” of the cinematic image is no longer a given, one would assume that the mutability of the digital would render the representational apparatus visible. In other words, whatever “traces” were left of the reality effect after theoretical attacks by poststructuralism now flare up because of their (ostensibly) final destruction by the technological attacks of digital code.
Digital cinema therefore, could be assumed to hold a productive relationship with the reality effect, in that it pushes the condition of the image as a signifier to the fore. Bazin’s faith in physical reality becomes skepticism in its digital reproducibility.
But this line of reasoning ignores several key facts about the specificity of this language, or rather the lack thereof. Digital cinema forces us to grapple with the fact that its smallest constituent level exists as the pure difference between 0’s and 1’s, a language that does not consist of linguistic codes or shared systems of meaning. Unlike the grooves of a phonograph record, for example, whose shapes have a direct correlation to the sounds they represent, digital code contains no semiotic specificity. It is a language that need not refer to moving images at all, that could exist as an instruction manual, an x-ray of a tooth, or a scan of a manuscript. The smallest constituent level of digital cinema is simply a neutral delivery system, what Georg Tholen calls “transmissibility as such.”
As Tholen articulates it, “Once ‘0’ and ‘1’ no longer represent something, but become markers of a system within which something appears, it makes possible not only the alternating oscillation of presence and absence but also ‘the universal medium of the electric current’ as a carrier that stays neutral to its message” (SAQ 101:3, 667).
We are faced with a significant problem then: while digital cinema provokes an awareness of the image as a representational language, it is only as a delivery system “neutral to its message” that the digital makes any sense. Nothing of this language remains to be seen, and the critique of the digital is itself rendered binary, a yes/no decision as to whether or not a represented action or object actually happened, was actually there. The problem is that this comes to resemble, once again, the “reality effect” wherein mimesis becomes invisible.
As we stare directly at the (cinematic) apparatus as it constructs a (CGI) biomechanical suit around the body of Tony Stark, what exactly do we see? Not, I think, a representational technique that gives the lie to the super hero genre’s realist aesthetic lifted from Moore/Miller’s 80s graphic novels (the desire to pull every SF element of the super hero genre down to the realm of the possible in the present day). The computer graphics are actually foregrounded here, but to what effect?
I know, after The Dark Knight, Iron Man feels like it came out about five years ago. But for the sake of my absence from blogging and the backlog of ideas I’ve been meaning to put up here, let’s look back.
To create the effect of the marvelous iron armor, a variety of technologies were used. From this article from Last Broadcast on “Tony’s workshop:”
For Stan Winston’s Shane Mahan and his suit design team, this required making a suit that could be worn in sections over the visual effects suit Downey wore. ‘The big challenge was trying to find ways to blend, cross-cut and inter- cut combinations of practical and CGI shots,’ says Mahan. ‘It would be absolutely foolish for me to think that I could pull off every shot in the practical suit, so we created a combination for Robert consisting of the chest piece, helmet and arm sections combined with a full-body motion capture tracking marker suit underneath. It’s a great way to blend the practical with the computer-generated effects, enabling ILM to bridge any gaps between the physical pieces.’
In this interview, Faverau says that part of the realism of the movie hinged on being able to show various parts of the suit being gradually constructed, culminating in Iron Man’s iconic red and gold Mark III armor. This trope in almost every superhero movie of the suit gradually being constructed or put on shows the gradual transformation of the human body into an icon, that is to say, into something that can be transmitted, disseminated, or broadcast.
This may take a few posts, but I’d like to consider the digital rendering of the superhero as symbolic of the inner workings of digital cinema itself. Iron Man, made up of various levels of physical reality, motion capture data points, and digital video encoding, is representative of digital cinema’s transformation of physical reality into transmissibility as such (1′s and 0′s rather than chemical inscription on celluloid). More to follow…
In an earlier post, I contrasted Bazin‘s theory of film as “physical reality as such” with Georg Tholen’s formulation of the digital as “transmissibility as such.” While celluloid contains a chemical inscription that directly relates to the object it records, digital cinema is constituted by generic code–the pure difference between 1′s and 0′s. How should we conceptualize the middle space wherein text-specific cultural codes are grafted onto generic digital codes? In other words, at what level does actual meaning, or signification come into existence in the transition from “physical reality as such” to “transmissibility as such” in digital cinema?
The all or nothing critique of the digital as heralding the death of photography and film seems to lack a significant nuance. Thinking through the “degrees” of indexical contingency in the varieties of the digital image would entail a thorough engagement with Barthes’s “reality effect” (which perhaps finds its counterpoint in motion capture: a hidden language, unreadable between the process of recording and transmission, comes into the service of believably depicting the utterly unrealistic).
What “pure performance” seems to imply for the actors is a complete liberation from the constraints of the cinematic apparatus. What “complete control” implies for the filmmakers is the possibility of ubiquitous and invisible manipulation. The actor performs freely on an empty stage while the process of image production and transmission takes place elsewhere, unseen. In this account, it seems as if the actor in the motion capture suit exists somewhere in between cultural and digital codes. She is an organism operating outside the technological apparatus whose body is nevertheless already encoded with data, and it is merely this data that the camera is interested in capturing instead of any indexical image.
