Last week I wrote two rather lengthy post (Part 1, Part 2) looking at shot scale as an element of film style and its connection to meaning by analysing sequences from three different films. Two of these films were live-action (The Lord of the Rings: The Two Towers and The Hunger Games: Mockingjay – Part 1) and one was computer animated (How to Train Your Dragon 2), but I used the same terms in describing all three, referring to things like camera movement despite the fact that in CGI animation there is no physical camera to move. 3D animation software allows for a virtual camera to be placed in a virtual space, but that does not shake the suspicion of using terms that derive from a physical apparatus to describe a film made without using that apparatus, and it also doesn’t account for 2D drawn animation, in which there may be a real camera but it is one that captures still images rather than an existing reality.
There are a number of reasons why I chose to talk about the sequence from Dragon the same way I talked about the sequences from the other two films.
Simple rhetorical convenience is one motivator. In my third year of cinema studies I took a pair of courses focusing on the history of animation and the theoretical approaches used to examine it, but I quickly realized that the terminology we normally use to talk about film seems inappropriate when applied to animation, since a great deal of that terminology is centered on the camera. Can we really talk about camera movement (pans, tilts, tracks) when there isn’t a camera? When I asked my professor how I should handle this lack of appropriate vocabulary, how I should talk about animation in general, he suggested qualifying everything with the idea of simulation. Instead of “the camera tilts up,” for example, one would write “the image moves up in the manner of a tilt.”
While this is a perfectly reasonable rhetorical method to suggest, and perfectly in keeping with ensuring the sort of distinctions between animation and live-action one might expect from animation studies, it’s a pain in the wrist. Qualifying everything with “as if it were” not only gets tedious, but also makes animation seem subordinate to live-action by suggesting it’s primarily concerned with imitation. Most of the time I’m fairly pedantic about what terms I use to describe film, but using the materially correct language here seems to cause more problems than it solves.
My other reasons are, for lack of a better term, polemic.
Animation is often marginalized in film studies, especially that branch of the discipline that concerns itself with studying the basic essence of the medium (what it is, what it does, how it does it), because a great deal of film theory is founded on the idea of cinema as a fundamentally realistic medium due to the ability of photography to capture and reproduce an extant reality more accurately than any other. As Tom Gunning argues, the “dominance of a photographic understanding” of the medium leads theorists to make “broad exclamations about the nature of cinema” that must always be qualified with “excluding, of course, animation.”
Responding to Stanley Cavell’s insistence that “projections of the real world” are at the heart of what cinema is, for example, Alexander Sesonske outlines the theoretical hurdles animation presents for this definition of the medium, since, he argues, the worlds animation offers to us never existed and only gain some measure of reality when projected: “Their projected world exists only now, at the moment of projection—and when we ask if there is any feature in which it differs from reality, the answer is, ‘Yes, every feature.’ Neither space nor time nor the laws of nature are the same.” For Cavell, Sesonske’s objections, based on the absence of a camera capturing an existing, physical world, simply offer an argument for why animation should not be considered cinema.
Traditional drawn animation is problematic for Sesonske and others because the origins of the image are so unlike those of live-action cinema. While the movement projected on screen in live-action and animation is created using the same basic process (the rapid succession of individual still images), in the former the movement and the figures performing it existed at some point in time before the film was projected, whereas in animation there was no such pre-existing action and no physical actors – the movements we see are entirely created. The knowledge of this basic difference and the freedom afforded to animators because of it, which allows them to stretch, squash, and morph their characters and environments in ways impossible in live-action cinema, effects how we experience what we watch.
The theoretical tendency to found a definition of cinema in the status of the image, its connection to an existing, physical reality, can largely be traced to the work of influential French theorist Andre Bazin, who argued for the basic realism of cinema due to the photographic image’s unique ontological qualities. For Bazin, photography had an almost mystical connection to the objects it captured, with the camera creating something beyond a simple representation of those objects, an image that possessed some of the reality of the original. Bazin’s use of terms like “fingerprint,” “mould,” and “impression” to describe this relationship between photography and the world led others to link his theories of cinematic realism to the idea of the “index,” instances when an existential bond exists between objects and their representations. A fingerprint, for example, is the result of physical contact – a finger touches a surface and makes a mark. This connection gives the print a different status than, for example, a drawing of a finger, in which no such bond exists between the object and its representation. Light rays bouncing off an object, entering a camera, and being fixed on a photographic medium, it is argued, represent an analogous process.
Because of this connection between the image and what it represents, photography tells us, unlike any other medium, that what it captures had really been there at some point in time. In animation, on the other hand, there is no such connection, since while animated images may gain an added sense of reality by infusing drawings with movement, they are still drawings, representations of people and things that never existed.
Basing a definition of cinema on photography and realism, on a camera capturing an extant reality, necessarily excludes traditional drawn animation, but greater difficulties arise when the definition of animation is expanded to other forms beyond 2D drawn cels.
Stop-motion animation, like live-action cinema, involves the photography of real objects in real spaces. The characters in stop-motion are unable to move without the intervention of the animators, who physically manipulate them for each frame, but they nonetheless have physical existence like their live-action counterparts and are reproduced with the same level of realism live-action photography is capable of, even if a puppet or clay figure is what’s being reproduced. A real camera is used, and the animators must make decisions about lenses, focus, and light levels like a live-action director. Like drawn 2D animation, however, the result of this process is a series of still images that are only invested with the illusion of life and movement when run through a projector. The objects are physically real – they did and may still exist – but their movement is not, putting stop-motion animation in an uncertain position. It is similar to live-action in its basis in photography, but is still animation – animated live-action.
More problems arise when new technologies in computer animation are thrown into the mix. DreamWorks recently unveiled their Apollo system, which allows animators to make changes to the image and to render them in real time using only a tablet and a stylus. Watching computer animators use the system is like watching stop-motion animators manipulate a real figure, if, admittedly, at one remove. While the characters still have no physical existence, the process of manipulating a digital body in digital space is more akin to the process of stop-motion animation than to drawing and photographing individual frames. There is a figure waiting to be moved in space even if that figure and space are created digitally rather than with clay, plastic, wood, and paint. Something similar could be said about animated films created by hacking video games, since the filmmaker is making use of a pre-existing set of characters in a pre-existing space.
These examples do not prove that theorists are wrong to claim that there is something different about a photographed image compared to an animated one. Yet, because the attempt to seek the distinctness of animation often serves, as Paul Ward argues, as a “rhetorical manoeuvre, a staking of animation’s right to be studied in and of itself,” animation scholars have often made overly grand claims for their subject that range from lionizing animation as more imaginative and open to creativity than live-action, to arguing that its freedom from the burden of representing reality affords it a metaphysical depth live-action doesn’t possess.
Animation’s ability to endow the inanimate with movement, and its related ability to endow the inanimate with life, are common claims, but the resulting scholarship often ends up adopting the mystical tone evident in Bazin’s writing on the realism of photography, making animation seem like an act of technologically infused magic, a Frankenstein’s monster of cinema.
I find animation wonderful on any number of levels, but as Andrew Darly argues, it “doesn’t create life in any real or normal sense of the word,” and in any figurative sense it doesn’t do so any “more than a host of other media such as painting, theatre, cinema and literature.” As a result, the analogy often occludes more than it explains.
I do not deny that the ontological status of the image, the origin of the image and the effects that origin has on how we perceive it, is an important avenue of enquiry. There are questions about film that can only be answered by addressing the differences between animation and live-action that arise from how their images are made. I have written before about how the created status of the image in CGI animation, the fact that it is not based in reality yet possesses some of the richness of detail that’s characteristic of reality, triggers a very particular reaction in those moments when we are simultaneously struck by the richness of a CGI image and the knowledge that someone made it. There is a different sense of virtuosity and wonder in these moments than in anything that live-action can produce.
For the type of stylistic analysis I was engaging in, however, the origin of the image doesn’t particularly matter. Some filmmakers and film scholars use “camera distance” interchangeably with “shot scale” because the relative size of objects in the frame is dependent on the perspectival relations created by the proximity of those objects to the camera. But those relations can also be created through non-photographic means. As David Bordwell argues, the “cinematic image yields a perspectival projection, but that projection need not record a pre-existing world,” it can be achieved through any number of image-making techniques, including 3D and 2D animation. There may not be a physical camera present, but the image can still possess the same characteristics even if they had to be deliberately simulated by an animator.
The idea of simulation, however, brings us back to my original problem in third year cinema studies, it means still having to deal with the clumsy rhetoric of “as if there were a camera.” Bordwell’s description of the cinematic image as a depiction of space with certain perspectival characteristics suggests a way to retain the idea of the camera without getting into the terminological stickiness that arises in animation from there simply not being one.
Rather than thinking about the camera as a physical machine, it is more productive to think of it as a particular type of perspective with certain limitations and possibilities that may have initially arisen from that machine, but aren’t intrinsically connected to it.
Thinking about the camera in this way helps address the actual look and feel of most feature length animation, which tends to hew quite closely to some of the qualities of live-action cinema even though the physical origins of those qualities are absent. To take one example, depth of field, the distance in front of the lens within which objects remain in focus, is largely dependent on the focal length of the lens being used, with short focal length or wide-angle lenses capable of greater depths of field than long focal length or telephoto lenses. Because no physical lenses are involved in creating its images, animation could potentially always have a limitless depth of field. The animators would simply have to create images that have clear planes of focus stretching as far back as the composition allowed.
This is rarely the case, however, because in live-action cinema depth of field is often manipulated to help direct our attention, to keep us focused on parts of the image that are important to the narrative. If the director wants us to pay attention to something in the middle-ground, for example, the foreground will often be kept out of focus so that it doesn’t distract us.
Racking focus, refocusing the lens so that a different plane is clear, can similarly serve to direct our attention from one part of the image to another within a single shot. Shifting from background to foreground or foreground to background gives us two different perspectives on the same scene without cutting, which can be used to highlight important information or to create certain thematic effects. Animation sometimes employs depth of field in similar ways because doing so has been established as an effective means of directing attention that we are all accustomed to, in live-action and animation alike.
The goal isn’t necessarily to replicate live-action film, but rather to use the different effects created by how the live-action camera captures the world to achieve specific narrative or stylistic goals.
This is especially evident in the How to Train Your Dragon films, which have so much of the look of live-action cinema that drawing an “as if” distinction seems misguided at best and unfairly reductive at worst. During their production there was a concerted effort to give the films the feel of live-action photography, an attempt to ground their cinematography in the realism of a camera’s physical limitations. “One restriction we set for ourselves,” director Dean DeBlois said in an interview after the release of the second film, “is that each shot needs to look like there is an operator behind the camera.” The production team brought on Oscar-nominated live-action cinematographer Roger Deakins to consult on how that effect could be achieved. For Deakins placing some limits on what the camera can do is critical to keeping an audience engaged, since while “it’s often easy to do something that’s impossible in the real world,” in a CG animated film, since the animators are not bound by Sesonske’s “laws of nature,” it can “give your audience an excuse to disengage from your movie.”
This meant not only refraining from movements that would be impossible for real actors and the cameras capturing them, but also a particular use of light. Using natural sources like varying degrees of sunlight, and fire from candles, torches, or dragons, the films have intense patches of light and shadow, compared with other CG animated films’ tendency toward fully lit and fully visible compositions. And the light often behaves in the way it would if it were being shot by a physical camera. When Hiccup steps out into the open of Valka’s dragon sanctuary in the second film, for example, DeBlois momentarily overexposes the image as if the lens was suddenly hit with too much light.
I’m in no way suggesting that live-action and animated film are exactly equivalent. There are obvious differences that affect our experience of the films we watch that should be explored. But drawing those distinctions based on a notion of cinema that locates its essence in photography and realism can lead only to animation’s exclusion from film studies, preventing us from exploring those differences. Yet as Darley argues, “concentrated on the big and serious sounding questions” like theories of the ontological status of the animated image in order to rescue animation from obscurity can lead to the disappearance of actual instances of animation under a mass of clever extemporizing, leaving questions of style, narration, craft practices, historical changes, and any number of other areas of interest unanswered.
So, yes, I talk about animation in the same terms I talk about live-action when addressing certain issues of style. It makes no difference to me if there was no physical camera involved. Live-action and animation can offer similar ways of seeing, ways most easily described by using terms tied to that camera, even if it wasn’t really there.
David Bordwell, Figures Traced in Light: On Cinematic Staging (Berkeley: University of California Press, 2005).
Stanley Cavell, The World Viewed: Reflections on the Ontology of Film, Cambridge: Harvard University Press, 1979). Quoted in Suzanne Buchan, “Animation Spectatorship: The Quay Brothers’ Animated ‘Worlds’,” Entertext 4, no. 1 (2004): 97-125.
Andrew Darley, “Bones of Contention: Thoughts on the Study of Animation,” Animation: An Interdisciplinary Journal 2, no. 1 (2007): 63-76.
Tom Gunning, “Moving Away from the Index: Cinema and the Impression of Reality,” differences: A Journal of Feminist Cultural Studies 18, no. 1 (2007): 29-52.
Jim Hill, “How Cinematographer Roger Deakins Helped Blur the Line Between Live-action and Animated Features,” Huffington Post, May 13, 2013, http://www.huffingtonpost.com/jim-hill/roger-deakins_b_3268900.html
Sarv Kreindler, “Roger Deakins Brings His Signature Style to ‘How to Train Your Dragon 2’,” Connect 2 Media & Entertainment, June 6, 2014, http://www.c2meworld.com/creation/roger-deakins-brings-his-signature-style-to-how-to-train-your-dragon-2/
Paul Ward, “Animation Studies, Disciplinarity and Discursivity,” Reconstruction 3, no. 2 (2003).