Rotoscoping, Medium Theory, and Digital Cinema

Home , Doug Chiang, Kinetoscope, Ollie Johnston, Rotoscoping, Rotoshop, Zoetrope

UNIVERSITY OF CALGARY

Animating Reality Effects: Rotoscoping, Medium Theory, and Digital Cinema

Courtney David Brinsmead

A THESIS

SUBMITTED TO THE FACULTY OF GRADUATE STUDIES

IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE

DEGREE OF MASTER OF ARTS

PROGRAM OF COMMUNICATION STUDIES

CALGARY, ALBERTA

MAY 2008

ISBN: 978-0-494-44575-4

UNIVERSITY OF CALGARY

FACULTY OF GRADUATE STUDIES

The undersigned certify that they have read, and recommend to the Faculty of Graduate

Studies for acceptance, a thesis entitled "Animating Reality Effects: Rotoscoping,

Medium Theory, and Digital Cinema" submitted by Courtney David Brinsmead in partial fulfilment of the requirements of the degree of Master of Arts.

Supervisor, Dr. Brian Rusted, Communication and Culture

Dr. Jean-René Leblanc, Fine Arts

Dr. David Mitchell, Communication and Culture

Date

ii Abstract

Following up on Lev Manovich’s excellent question “what is digital cinema?” this thesis examines how the ontology of motion pictures and our relationship to them change after digitization. It posits that digital cinema is a hybrid of animation and live-action film, anticipated by the unique technology of the rotoscope, and continued with contemporary motion capture technologies. Using The Polar Express (2004) as an exemplary digital film, it demonstrates that film critics consensually characterise such films as eerie, strange, or uncanny. It attempts to explain this uncanniness, in part, using medium theory to dichotomize animation and live-action as handmade and automated techniques, respectively. It is hoped this can supplement existing groundwork to better characterise the changing nature of motion pictures in the digital era.

iii Acknowledgements

I wish to thank my family, friends, colleagues, and supervisor Dr. Brian Rusted for their support and feedback through what was an ultimately worthwhile, but not always easy process.

iv Table of Contents

Approval Page...... ii Abstract...... iii Table of Contents ...... v List of Figures and Illustrations ...... vi

INTRODUCTION: NASCENT DIGITAL FILM ...... 1

CHAPTER 1: LITERATURE REVIEW...... 11 Medium Theory...... 11 Embodied Vision and the Challenge to Visual Studies...... 20 Digital Cinema and the Post-Film Argument ...... 23 Towards a Definition of Animation ...... 33 The Rotoscope Phenomenon...... 40 The Conventional Approach...... 43 Newness and Novelty...... 45 The Human, the Machine, and the Uncanny Valley ...... 47

CHAPTER 2: METHODS...... 54 Proving the Object / The Object as Method ...... 54 The Medium is the Method: How to “do” Medium Theory...... 57 Methods Summary ...... 60

CHAPTER 3: ANALYSIS ...... 62 Part 1: The Polar Express Reviews: Technology Before Story...... 62 Part 2: Medium Theory into Practice ...... 70

CONCLUSION: REFORMULATING DIGITAL FILM ...... 87

REFERENCES...... 94

v List of Figures and Illustrations

Figure 2.1...... 48

Figure 3.1...... 84

Figure 3.2...... 90

vi 1

INTRODUCTION: NASCENT DIGITAL FILM

Peter Travers, film critic for Rolling Stone, condemns the The Polar Express

(2004) and its “gargantuan attempt to redefine animation.” He finds the characters creepy and quickly concludes, “The result is a failed and lifeless experiment in which everything goes wrong” (Nov 18, 2004). What is interesting about his brief, single- paragraph review is that he focuses entirely on the technology, on the way the film is made, rather than the story, characters, moral, etc. that usually comprise popular film criticism. Apparently, Travers’ dissatisfaction with the technology is enough to preclude any attempt to deal with the storytelling. He simply does not bother, but rather stalls pre- emptively at the starting line by bluntly denouncing, “the movie just doesn’t work.”

The technology in question is called motion-capture (mocap) or performance capture. Travers explains that, for the film, “Tom Hanks covered his face and body with reflective dots…[a]nd director Robert Zemeckis and his ‘performance capture’ team use those dots to create computer-generated images meant to reflect the illustrations in Van

Allsburg’s book [The Polar Express (1985)].” In fact, the dots have very little to do with how the images look (i.e. like illustrations) and everything to do with how they move.

The reflective dots are affixed to the actor, then tracked and recorded in three dimensions on a specially equipped sound stage. Next, the data, more-or-less a record of the actor’s performance, is piped into a prepared computer model resembling a virtual marionette, causing it to ape the movements of the actor. The gargantuan redefinition of animation

Travers mentions alludes to comes from the promotional material for the film. The production notes read, “An art form in its own right, Performance Capture effectively breaks new ground to offer images like nothing seen before” (¶ 30, 2004). Similar

2 rhetoric is used for Polar’s successor Beowulf (2007), with many overzealous film reviewers hailing it as “the future of filmmaking.” The technology is garnering a good deal of attention, even if it is not well understood.

Other critics have echoed Travers’ complaint calling the characters in Polar

“lifeless,” “eerie,” or “weird.” However, the somewhat esoteric technique does not necessarily have to spell doom for a film using it, as Travers’ review might suggest. In most popular and critical opinion it was used to good effect in the Lord of the Rings trilogy (2001 - 2003) and King Kong (2005), for instance. But this is not to say the technology, even when used well, is merely accepted and blithely taken as commonplace.

In describing such films, it is common enough for popular film reviews to speak of

“special effects wizardry” or “Hollywood magic” in place of a real understanding of what is beheld on screen. Words like wizardry and magic imply not only that audiences do not know the secret of how the trick is done, but also that what they see onscreen is unbelievable, difficult to describe, real but somehow unreal. Chicago Sun-Times critic

Roger Ebert exemplifies the point. “The characters in ‘The Polar Express’ don’t look real, but they don’t look unreal, either; they have a kind of simplified and underlined reality that makes them visually magnetic” (¶ 3, 2004). To my mind, Ebert does a great job of describing the indescribable. If the characters are not creepy, they remain

“visually magnetic,” which is more complimentary but still belongs to the same ineffable category of unfamiliar weirdness.

To pull these points together at once, I have sketched a striking, if somewhat silly, introduction: the future of filmmaking is weird. Of course, I do not take the studios at their word and do not accept mocap per se as the future of film. But there is good reason

3 to see it as part of a larger trend in filmmaking as cinema goes digital. And the matter of weirdness, however asinine it may seem at the moment, is not so easily subdued.

In his essay What is Digital Cinema? (2002), new media theorist Lev Manovich describes the floating feather in the opening sequence of Forrest Gump (1994). He writes,

To create the shot, the real feather was filmed against a blue background in different positions; this material was then animated and composited against shots of a landscape. The result: a new kind of realism, which can be described as ‘something which is intended to look exactly as if it could have happened, although it really could not.’ (pp. 409-10)

The last part is the most interesting and on its own makes for a curious statement. It recalls Ebert’s description above of that which is not real, but not unreal either. And the choice of phrasing (“the result”) begs that it be compared to Travers’ initial assessment,

“The result is a failed experiment.” The three references are all related, I think. Travers’ is dismissive, Ebert’s hopeful or interested, and Manovich’s more prescient. But

Manovich’s description is set apart because it refers not to mocap exclusively, but to the broader category of what he calls digital cinema. Indeed it may be that mocap is just one form of digital film, and Forrest Gump is another. And there may be reason to think that digital film has an unusual aesthetic. Manovich’s statement, at least, evokes a certain mystique. What does it mean to witness that “which is intended to look exactly as if it could have happened, although it really could not”?

Manovich’s quote also necessarily speaks of the relation between the real and the represented. The question of truth and what is “real” crops up repeatedly in digital image theories and is difficult to extricate from the discussion. Both he and Ebert refer to “a”

4 realism, highlighting the fact that many forms of realism are possible. The new realism of digital cinema and its attendant weirdness or visual magnetism, it will be shown, stem from a mixing of live-action and animation.

The stakes are higher than one might initially imagine. For one, the digital production process is now ubiquitous. It is no secret that there has been a significant rise in the use of digital special effects in cinema and television. More often than not, these effects can be described as a merging of live-action film and animation. And they are not limited to a handful of shots in a feature length film, but rather have become an integral part of big budget film production and are used throughout many films, as in the latest

Star Wars instalments, the Pirates of the Caribbean franchise, or Harry Potter film adaptations. Nearly all television ads or music videos with a considerable budget use digital effects. Indeed, the “new realism” Manovich describes is everywhere.

But the line between live-action and animation is a blurry, permeable one.

Consider that the Gollum character from Lord of the Rings was created using motion capture technology to record the performance of actor Andy Serkis which was then fed into a computer graphic (CG) model (the character viewers see onscreen). Come Oscar time, there was considerable buzz over Serkis’s eligibility for an acting nomination.

Director Peter Jackson and the film’s producers lobbied to have him nominated. Was his contribution to the onscreen character worthy of such acclaim? Or were the other technicians and artists who shaped the humanoid character and put it all together more deserving? In the end, Serkis was deemed ineligible. The boundaries of this technology, it seems, are still unclear.

If it can be argued seriously that a virtual humanoid character, Gollum, might be a best actor, one wonders if the LOTR trilogy or even Forest Gump might qualify for animated film awards. Those four films all won visual effects Academy awards, but practically would not have been considered as animations. However, the debate has also been raised over the motion-captured film Beowulf (2007), where the filmmakers are vying to have it considered an animated film, for award purposes (“Beowulf raises,”

2007). This is essentially the opposite argument put forth by Serkis et al. If the move goes ahead, should the actors then be disqualified from performance awards since their characters must be, as it were, animated?

Whether or not Serkis or other mocap performers receive gold statues to place on their mantles at home is of little concern to the general public, but concurrent with these debates, there has been some stir over Beowulf’s PG-13 rating. Some of the fuss is over

Angelina Jolie’s virtual doppelganger’s nude appearance. Opening her review of the film, Jamie Portman of CanWest News Service writes, “It’s undeniably an accomplishment for Hollywood to get away with displaying Angelina Jolie in the nude without being hit with a ‘restricted’ rating” (2007). The underlying message is that the character Jolie plays is animated (albeit in a very photo-realistic rendering) and animated nudity does not constitute the “real” thing. Presumably, were the flesh and blood actress to appear unclothed, the R rating would be slapped on the packaging without question.

Add to this the opinion of Parent Previews, an organization that rates films based on their suitability for family viewing, that gives the violence of the film a D- rating (the absolute lowest possible on their scale), rebuking “this film ranks as possibly the most violent US rated PG-13 title ever released” (Gustafson, 2007). Again, the implication is that

6 animated violence is somehow less affecting than “real” (i.e. staged and fake, but live- action) violence.

In the Polar production notes, director Zemeckis muses about mocap technology,

“The good news is that anything is possible. The bad news is that anything is possible”

(¶ 79). He has in mind the artistic freedom afforded by digital filmmaking and the burden it brings of coping with too many choices. Everything including set design, costumes, lighting, texturing, camera moves, the performances themselves, etc. can be infinitely tweaked and everything must be thought of and planned in advance or it will not appear in the final picture. But his joke can have more sinister implications. At its extreme, such limitless control of images has sounded alarms, for instance, with those concerned with child exploitation and kiddie porn. If it is possible to animate a dragon, orc, or naked Jolie, it is equally possible to make images of undressed children. As

Zemeckis says, “anything is possible.” Of course, it has long been possible to draw or animate naked children, but the concern is heightened given the photorealism of CG images. The worry makes for interesting and important debate. With virtual child porn there is no actual, or at least specific, victim. However, the problem is galvanized by thinking evident in Jamie Portman’s review, mentioned above. The headline reads: “Jolie may as well be naked in ‘Beowulf’” (my emphasis). Jolie is not naked but her virtual avatar, which looks remarkably like her, is. A law article, “Image is

Everything”(Gibeaut, 2000) addresses the problem of virtual child porn, observing that since 1996 U.S. prosecutors now have “to prove the images were of actual children and not computer creations, a burden the government didn’t carry beforehand” (¶ 8). Clearly the debate raises important ethical issues, and one can think of countless other cases

7 where the inability to distinguish the photographic from the virtual can be exploited for personal gain. The film Wag the Dog (1997), for instance, shows faux wars and spurious acts of heroism being staged against green screens, and then broadcast to an unwitting audience.

The benefit of this situation may be that it reinvigorates debates and understanding about truth and representation. However, while technology has made static photograph alterations undetectable, moving film is not quite there. It still looks a little off, weird, or even eerie. Many think convincing virtual motion pictures are just around the corner. Tom Hanks, for one, fears that actors will be replaced by virtual counterparts, saying “I am very troubled by it…But it’s coming down, man. It’s going to happen” (Lyman, 2001, ¶ 8). The idea that actors will be out of work is probably absurd

(what would happen to the celebrity industry?), but seamless, virtual live-action may yet come to pass. It remains to be seen.

However, the worries about virtual kiddie porn, fake wars, and impostor actors are a departure from Manovich’s description above. The situation may be better described as “something which is intended to look exactly as if it did happen, although it really did not.” Audiences are aware that orcs and dragons do not exist but wars and naked children certainly do. Still, digital cinema which combines live-action and animation seems to have an unusual or weird quality and that is where this paper will focus. I only hope to have demonstrated that there are serious issues at stake. Even if

Beowulf producers are able call their film animation to dodge a restricted rating (which would significantly reduce its viewership and profits), it must be seen that animation, in

8 the digital era, is no longer kids’ stuff. Indeed, Manovich has pointed to “a new realism” that once more reopens the issue of representation and its relationship to the real.

Now, I am not so bold or giddy as to suggest the realism Manovich isolates is entirely new. His description of the impossible made to look “exactly as though it could have happened” can be applied to a whole host of pre-digital special effects and even magic tricks. But his use of the qualifier “exactly” is important and points to the precision with which it can be done. And digital technologies, no doubt, make it easier than ever. Those two points may be sufficient to justify calling the era of digital film new, or at least a significant advance and change in the way motion pictures are produced.

There is another important, if somewhat esoteric, historical precedent to the digital hybrids of live-action and animation. It is called rotoscoping. Patented by Max

Fleisher in 1917, the rotoscope projects live-action footage, frame-by-frame, onto frosted glass for an animator to trace over. The result is animation that has the motion characteristic of live actors. And it too has long been described in animation circles as weird, lifeless, or eerie (Thomas & Johnston, 1981). It is seen as the precursor to motion capture technology and, in fact, animators sometimes refer to mocap as “Satan’s rotoscope” (Langer, 2004). For all intents and purposes, it is not unreasonable to call the mocap technique just another generation of rotoscoping. This technology, as it is used in

The Polar Express, will play a central part in this thesis.

At its broadest, this thesis seeks to build upon Manovich’s work motivated by the direct but difficult question, “What is digital cinema?” It will ignore several parts that a complete answer would require. For instance, I am not concerned with how digital data

9 can be readily copied and distributed. Yes, movies can be sent to movie theatres via satellite, downloaded at home, and are searchable via mark-up language, but this will only receive a cursory treatment here. Nor am I writing about how digital cinema may be increasingly interactive (I’m not writing about gaming), or assembled on-the-fly from a database of coded, prepared scenes. Rather, I am interested in how digitization might change the aesthetics of film, and interrogating the “new realism” of digital, hybrid cinema. I want to know why, at least in some cases, digital film looks weird.

To this end, I hope to show that rotoscoping has a lot in common with digital film and may even be considered a type of digital film. I will turn to the limited literature outlining explanations of the strangeness of rotoscoping and try to build a complementary explanation fuelled by the insights of medium theory. Why medium theory? Because the base characteristic of digital film is its hybridity. If it is a hybrid of different media (i.e. live-action and animation), then it is prudent to examine the history and qualities of those media, each in turn. It is a way of looking at the parts that make up the whole, in the hopes of getting a better sense of how digital motion pictures operate more generally.

That is, I will look at the seemingly peculiar aesthetics of The Polar Express (as an exemplar of rotoscoping) to gain insight into digital film aesthetics as they are emerging today. Thus, I have articulated my research question: How can an examination of the uncanny phenomenon of rotoscoping in The Polar Express, through the lens of medium theory, inform an understanding of digital film?

I come to this problem as both scholar and practitioner. I have studied to be a 3D computer animator and admit my own interest in the rotoscopic technique was piqued in an animation class. An instructor offhandedly remarked, “For some reason, tracing over

10 film never really works. No one knows why” and then proceeded with a lesson in techniques that do “work.” Western, Disney-style animation has evolved to include a host of techniques to make characters move believably – so much that they have been adopted by schools and studios as dogmatic “principles of animation” (Thomas &

Johnston, 1981, chap. 3). So, there are times in the text when my understanding of animation is informed by personal experience, which I hope serves to complement and augment the somewhat limited literature on the subject.

CHAPTER 1: LITERATURE REVIEW

Medium Theory

Medium theory has enjoyed renewed interest since the digital era gained steam in the 1990s. Marshall McLuhan has been canonized as the patron saint of Wired magazine and its supporters at MIT. A Media Ecology Association has been founded and Donald

Theall, the leading authority on McLuhan, penned a biography with the telling title, The

Virtual Marshall McLuhan (2001). Medium theory, which for a period seemed a moribund or complete project, feels like a vital field again.

To make this even more apparent, W.J.T. Mitchell, addressed the first ever symposium for the long running Critical Inquiry journal by heralding medium theory’s enduring importance for its capacity to span disciplines. He is a supporter of “hybrid disciplinary formations” claiming,

We have needed, in my view, to think across the media and the arts, not in order to forsake their specificity but to locate and define that specificity in relation to a plethora of unique material practices and thus to trace their braidings (mixed media) and nestings (the appearance of one medium inside another). (2003, p. 334)

It is interesting to think retrospectively of how medium theory has evolved from Harold

Innis’s political economy and McLuhan’s enigmatic Gutenberg Galaxy (1962), to the more refined readings of Walter Ong and Joshua Meyrowitz, to the broad category

Mitchell describes. It is fascinating, too, that the merging of scholarly disciplines should mirror the mixing of media after the digital revolution. Mitchell even makes the

12 distinction between “media” theory (related to the media in popular parlance – television, film, advertising, etc.) and “medium” theory, which is essentially a return to

McLuhanesque roots where media is a ubiquitous and essential part of all human communication and interaction with the world. It is in the nest of theory he carves out that this thesis is positioned. I hope to have taken up the mandate Mitchell lays out in his symposium address, but also to revisit some of the crucial insights already established since McLuhan, particularly the dualism of orality and literal-mindedness and the affects of certain media on the sensorium. It is fitting that medium theory embraces interdisciplinarity, as this paper also belongs to the emergent field of animation studies

(or is it the changing field of film studies?), and visual studies (already encompassing many disciplines), with a debt owed to perceptual psychology.

As mentioned, the impetus for centering medium theory in this thesis stems from recognising digital film as a hybrid. If one is trying to understand a hybrid, it makes sense to trace the histories of each medium that comprise it. That is precisely what medium theory does well; it examines where media come from, how they operate and interact, and how they affect individuals and culture at large. Of course, a significant part of tracing a medium’s history could be to look at technological innovations and cultural practices. In the history of film for instance, developments of sound, colour, portable cameras, etc. have changed how we understand film and the look or aesthetics of film.

David Bordwell (1997) and others have traced these developments systematically and, indeed, this thesis continues the project to some extent by examining how digitization affects the object and practice of film. However, the project taken up by Bordwell is pursued at a micro level, and misses one of the great advantages of McLuhan’s medium

13 theory to extrapolate findings to a macro level. Joshua Meyrowitz (1994) writes, “On the macro level, medium questions address the ways in which the addition of a new medium to an existing matrix of media may alter social interactions and social structure in general” (51). Not only do technological developments expand filmic possibilities, but they also affect our relation to the medium and its ontology on a higher, cultural level.

Additionally, McLuhan’s observation that old media become the content of the new is a useful way to make sense of hybrid media. For instance, digitization allows the

“old” established media of live-action film and animation to be constituents (content) of the “new” medium of digital film. This phenomenon is very similar to the nesting

Mitchell observes, where media can both overlap and envelope each other.

A good jumping off point for a review of medium theory (even if it is cliché) is the most famous of communications aphorisms: McLuhan’s “the medium is the message”

(1964, chap. 1). This ambiguous phrase has been interpreted in many ways, as McLuhan no doubt intended. He quotes Francis Bacon, “Aphorisms, representing a knowledge broken, do invite men [sic] to inquire farther; whereas Methods, carrying the show of a total, do secure men, as if they were at farthest” (McLuhan, 1962, p. 102). Clearly,

McLuhan wanted his readers to inquire farther. What I take his aphorism to mean is that the technology used to communicate is just as, if not more, important than the content of the message. That is, users of technology are more affected by technology than the message it conveys. The most important example, of course, is Gutenberg’s printing press, which transformed (and, it is argued, gave us) Western civilization. The printing press itself was of more consequence than any book it produced, just as the Internet is more significant than any webpage. For McLuhan, print is the central axis that splits

14 three great epochs – the oral, the print/literate, and the electric. McLuhan defines media as “the extensions of man [sic]” suggesting they are extensions of the senses. The senses may be more significant to medium theory than is sometimes acknowledged, but a central contention is that different media favour or reconfigure the senses in ways that restructure consciousness. Walter Ong writes, “technologies are not mere exterior aids but also interior transformations of consciousness” (1982, p. 81). The idea of sensory configurations is the main tenet I wish to borrow from medium theory in this thesis and will explore it in depth in the analysis section. But first to gain some understanding of the orality/literacy split, and get some sense of how dominant media affect culture, it is necessary to look briefly at each of the three epochs in turn, including the transitional phases.

Primary orality is difficult for us to imagine, says Walter Ong (1982). Oral cultures can never look something up in a book and thus rely on memory to preserve their culture. They develop repertoires of stock phrases, myths, and narratives to keep alive what has been learned. Laws, for example, are not drafted out in long volumes, but implemented according to some basic dictums, like the ten commandments. Rhythmic poetry and frequent recitation are common aids to remembering instructive stories, genealogy, history, etc. Under this system,“[k]nowledge that is not repeated aloud soon vanishes,” observes Ong (p. 41). Another important factor is that spoken words, as all sounds, are temporal. Unlike written words that can be viewed at once, speech is evanescent – it comes and goes as it is spoken. Ong insists that this condition makes sound an “event” and places us at the centre of it. He says, “Sight isolates, sound incorporates. Whereas sight situates the observer outside what he views, at a distance,

15 sound pours into the hearer” (p. 71). This “centring” describes the acoustic space, or sphere, of oral people. Thus oral culture is one dominated, sensorially, by the ear, yet this results in a more unified or balanced ratio of the senses. Meyrowitz (1994) says,

“The oral world is one of rich involvement with and interplay of all the senses of hearing, sight, smell, taste, and touch” (p. 54). This is markedly different from literate print culture that favours the eye. Because sight “isolates” and puts the “observer outside,” it is suited to empirical science, which is rooted in detached observation and the logic of causality.

Before the emergence of print, most medium theorists identify a transitional, scribal phase. Meyrowitz is chiefly concerned with the ways media affect social roles, and writing is the first major step in redefining them. Oral cultures have relatively few class distinctions because everyone has access to the same oral repertoires, but writing creates a sharp division between the literate and illiterate. For instance, Neil Postman

(1982) argues that the division between childhood and adulthood used to be around age seven, when language is mastered; cultures with writing and education systems bump this partition up to a later age when reading is mastered. Because of their unique skill, medieval scribes had a monopoly of knowledge and power over the illiterate masses. In terms of sense ratios, the scribal (or manuscript) period is the beginning of a shift toward the visual. However, it is not until Gutenberg’s printing press and the mass dissemination of the printed word that sight really achieves dominance.

McLuhan (1962) stresses this point saying, “manuscript culture is intensely audile-tactile compared to print culture” (p. 28). The differences between scribal manuscripts and print are at least twofold. The first condition that makes print have such

16 a profound affect is the alphabet. Having a phonetic language that mimics the oral by breaking it into sounds is extremely important and makes Gutenberg’s invention feasible.

Certainly, not all scribal cultures use a phonetic written language. English, having a mere twenty-six characters makes print efficient compared to, say, Chinese writing systems that have thousands (where each symbol has to be learned individually). “Only the phonetic alphabet makes a break between eye and ear, between semantic meaning and visual code,” McLuhan declares (1962, p. 27). Secondly, and most obviously, print allows for mass production. The relative ease of learning a phonetic alphabet coupled with an explosion of available printed material transformed the Western world. The effect of reading print is to focus on the visual written page, which excludes the other senses and causes the reader to draw within. It gives a heightened sense of individualism opposed to the communal roles in an oral society. The idea of authorship exists for the first time as writers pen works that can be commodified and have a unique voice within the established oral milieu. Print has the effect of standardizing vernaculars and dialects in a way that unites large regions of people and brings about modern nationalism. It allows ideas to be drawn out and built upon, line by line, at the same time as it divides the world into discrete bits and words. This leads to linear “cause and effect” thinking and brings the flourishing of science, philosophy, and reason itself. Indeed, McLuhan contends, it brings about civilization as we know it. These are profound changes but the one of chief interest here is the way sight came to dominate the senses after Gutenberg’s invention. The way this domination invades even language itself has been recounted by

Meyrowitz (1994), Pallasma (2005), and other authors wishing to characterise occularcentrism. We have a “perspective” or “point of view,” we gain “insight,” and

17 even this section is a sort of “overview,” as though we have the vantage of a bird flying over the theory to see its full scope (of course, this is only a very cursory account).

Regardless, the emphasis on the visual in literate print cultures alters people’s very consciousness and continues to persist even into the electric age.

The second transitional phase, from print to electronic is one of the most difficult to get a hold of, perhaps because we are still in the middle of it. The full effects prophesized for the electric age may only be felt when virtual reality is realized, but we have begun to feel those effects nonetheless. Electronic media reconfigure social roles once again. For instance, Postman (1982) argues that television (for now the most influential of electronic media) is a “total disclosure medium” and makes knowledge readily available to wide audiences (chap. 6). That is, TV is easier to “learn” than reading. TV blurs the separation of child and adult worlds, once held firm by literacy, as children are exposed to sex, violence, death, and other adult themes. Different electronic media engage the senses in ways that loosen the stranglehold of sight. Meyrowitz (1994) summarizes, “Electronic media bring back a key aspect of oral societies: simultaneity of action, perception, and reaction. Sensory experience again becomes a prime form of communicating” (p. 57). McLuhan (1962) calls this a return to tribal culture (the “global village”), while Ong (1982) speaks of “secondary orality.” The word is once again made a temporal event rather than an object, through telephones, radio, TV, etc. Indeed, communication starts to resemble the pre-literate age. However, the new orality is markedly different; it does not have the time and space constraints of traditional face-to- face communication. The sheer scale changes things. Electronic media reach mass audiences, compete with other electronic signals, and make experience repeatable. All of

18 this occurs on a global scale now and messages from another part of the world seem just as likely as ones from a neighbour – McLuhan’s global village.

McLuhan prophesized a future where “communication will be tactile, post-verbal, fully participatory, and pan-sensory” says Theall (1995, p. 22). Although McLuhan did not know about VR per se, Theall uses this quote to speculate about the future of VR.

The notion of “post-verbal” communication may be suspect, but his description is informative as a glimpse into a possible future. The goal, it seems, is a more balanced sense ratio (“pan-sensory”). What McLuhan had to go on was the way early electronic media sampled reality. Samuel Morse’s telegraph allowed alphanumerals to be converted into bits of electricity. Daguerre’s photography sampled light into a portable, reproducible image. Television also sampled the visual and could send it across the country as an electric signal. Audio could be recorded and broadcasted via radio. It is thus the promise of the future that all sensory information may be sampled, stored, manipulated, and reconfigured. This is sometimes referred to as the “atoms to bits” theme, advocated by Nicholas Negroponte (1995), where all sensory data from the natural world is converted into the language of the computer, where it can be programmed and manipulated. Gordon Gow (1996) puts it poetically: “Through the bit, nature is sampled, translated into human symbols, and then animated. We turn the world into art” (p. 11). Digitization transposes the natural world. Digital cinema can thus be seen as a step toward VR and it juggles the senses in the process.

Thus, the governing concept usually taken from McLuhan’s writings is that the dominant medium of a given age restructures consciousness in a way unique to that medium and era. This has led some critics such as Raymond Williams (1967) to levy the

19 charge of “determinist” against him, but McLuhan dismissed such criticism. He found hope in understanding media; knowing how media function is a way to elide or mitigate their influence. He was fond of citing Poe’s Descent into the Maelstrom where the sailor escapes the peril of the whirlpool by observing its patterns and navigating through them

(McMahon, 2002).

Admittedly, to recognize three epochs in human history is to speak in broad trends. McLuhan’s insights can get a bit more muddled at a localized, micro level.

Meyrowitz differentiates between medium theory and content theory, and astutely advises, “medium theory is most helpful when it is used not to supplant content concerns but to add another dimension to our understanding of the media environment. What is needed is a better integration of medium theory with other perspectives” (1994, p. 73).

The project of “Understanding Media” is a complicated one and McLuhan even observes,

“no medium has its meaning or existence alone, but only in constant interplay with other media” (1964, p. 26). His project is continued presently by the Media Ecology

Association, and the word “ecology” ought to raise a flag signalling the complexity and flux of interrelationships involved. But the task is not in vain and many fruitful insights can be gained in its pursuit. It will suffice to say that our senses are our primary (or only) way of experiencing the world and different media engage the senses in different ways.

Taking that premise as a springboard, this thesis asks: Does the medium of animation favour a certain sense, or configuration of senses, that differs from the sensory engagement of live-action film?

Embodied Vision and the Challenge to Visual Studies

The significance of multi-sensory experience is being recognized in the nascent field of visual studies. Inaugurated as an interdisciplinary umbrella for the study of art history, aesthetics, perception, film, videogames, etc., visual studies has tried to find cohesion by concerning itself with all things visual. Nicholas Mirzoeff dedicates the first six chapters of his edited book The Visual Culture Reader (2002) to a self-examination of the field. He prefers the designation visual “culture” because it foregrounds the political stake in the field. W.J.T. Mitchell makes the distinction that visual studies is the discipline and visual culture its object, “Visual studies is the study of visual culture”

(2002, p.87). Regardless, the young discipline is not yet comfortable in its own skin and is challenged to define itself, articulate its goals, and even justify its existence.

These are largely just the growing pains of an amalgamation of disciplines – certain academic territory is annexed, forfeited, and shared for the common good under the banner of interdisciplinarity. But another of those pains is the problem of visuality.

What seems like a self-evident label, the “visual” is recalcitrant under scrutiny.

W.J.T. Mitchell takes up this problem in his provocative “There Are No Visual

Media” (2007). He anticipates several exemplars of so-called visual media in turn and attempts to show they can never be pulled from the web of language and the other senses.

Silent film, for example, was always screened with accompanying live music or commentary, or was contaminated by subtitles or intertitles. “Sculpture is so clearly an art of the tactile it seems superfluous to argue about it” (p. 398). Photography “is typically so riddled with language…that it is hard to imagine what it would mean to call it a purely visual medium” (p. 398). Most significantly (to this paper, at least), he argues

21 that paintings are handmade and rhetorically asks “But what is the perception of the painting as handmade if not a recognition that a nonvisual sense is encoded, manifested, and indicated in every detail of its material existence?” (p. 397). The nonvisual sense he alludes to is touch – paintings bear a tactile trace of the artist’s handiwork. He credits

McLuhan’s notion of sense-ratios as the starting point for his argument. In the end, he claims that “all media are mixed media” (p. 399), yet he is quick to add that this does not preclude distinguishing one medium from another. His essay is to be taken as the questioning of one of visual culture’s central assumptions (the existence of visual media), but not an attempt to undermine the discipline. He only asks that the phrase “visual media” be put into quotation marks for a while in the name of scholarly rigour.

Embodied vision is foremost a theory of spectatorship. Even though the word choice “spectator” is misleading because it implies vision only, embodied vision is concerned with the audience and their reaction to film, more than the multisensory qualities of the film medium itself. Vivian Sobchack provides an excellent account of this phenomenon in her essay “What My Fingers Knew: The Cinesthetic Subject or

Vision in the Flesh” (2004). She is fascinated by the way popular film reviews often invoke multisensory descriptions like “Viscerally, it’s a breath-taking trip” or speak of

“white-knuckle” tension. She takes these as cues that the film-going experience is not the disinterested, textual reading that film scholars often ascribe and use in their own criticisms. Film theory has treated the derivation of meaning as a semantic and conscious process. But she is quick to remind us that we are embodied beings and our whole bodies

– our entire sensorium - are involved in all our experience.

Even at the movies our vision and hearing are informed and given meaning by our other modes of sensory access to the world: our capacity not only to see and to hear but also to touch, to smell, to taste, and always to proprioceptively feel our weight, dimension, gravity, and movement in the world. In sum, the film experience is meaningful not to the side of our bodies but because of our bodies. (2004, p. 60)

Phenomenologically speaking, she says there is a constant interplay between conscious registration and the orchestration of our bodily senses. What it takes Mitchell some time to argue (that all media are mixed), Sobchack takes to be obvious, asserting, “vision is not isolated from our other senses” (p. 64).

Sobchak, W.J.T. Mitchell and others recognize the importance of embodied vision, of expanding the long-dominant model of monocular looking and “reading” of visual texts. But they are a little unsure how to proceed. Indeed, it is a difficult project.

They glom onto obvious or exceptional cases to illustrate their point. Fine art installations that overtly try to engage bodily “reading” experience are favoured because they help crystallize the point. However, implicit in their approaches is the notion that some media or works are more visual than others. And it becomes difficult to parse out which ones are, and why. This thesis proposes to look at the case of animation and live- action to explore the nuanced differences between more purely visual media and media

(still often called “visual”) that are apt to engage the other bodily senses. The differences

– part of the embodied vision project – can then be extended using McLuhan’s concept that different sensorial configurations foster unique forms of consciousness. This will be explored more fully in the analysis section.

Digital Cinema and the Post-Film Argument

Digital film = live action material + painting + image processing + compositing + 2D computer animation + 3D computer animation

Lev Manovich, What is Digital Cinema?

The above is the condensed answer to Manovich’s pertinent question: What is digital cinema? It is an excellent starting point for thinking about digital film because it clearly identifies digital film as a hybrid and it may represent the best, or most cohesive, answer to the question yet. It also builds on the groundwork laid for the subject of digital imagery more broadly, especially the work of William J. Mitchell (1992) (not to be confused with W.J.T. Mitchell). In some ways, Manovich’s formula is a response to and extension of Bazin’s famous “What is cinema?” (1967). Bazin’s influence on film theory is not to be underestimated and Manovich borrows from his writings, at least in the way most contemporary film theory must, such as the importance of the indexical link between photos and the real world, outlined in Bazin’s chapter “the Ontology of the

Photographic Image.” But despite the nod to his predecessor, Manovich does not necessarily pick up where Bazin left off. In fact, he never even mentions Bazin in the essay. And Bazin, for one, never generated such an explicit formula as Manovich attempts.

To arrive at this formula, Manovich traces the archaeology of motion pictures.

He concentrates on the technological innovations, taking motion as the foundation for cinema. The names of early devices such as the kinetoscope and cinematograph highlight the privileged place of motion rather than, say, narrative. He actually begins by acknowledging that most contemporary discussion on digital film has focused on the

24 possibilities and implications of nonlinear narrative. Borrowing from Christian Metz he argues that narrative has become an integral part of film (p. 45). The vast majority of major motion pictures tell stories and he argues that live action, fictional, narrative films constitute a “super genre” that have dominated the twentieth century. Bazin too notes that film borrows from and extends the arts of theatre and literature (1967). Storytelling is supreme. But Manovich sets his trajectory away from narrative to focus on the technological developments of making pictures move.

Specifically he shows that the protocinematic devices were essentially animation devices. The images themselves were hand drawn to use with the early looping curios of the Praxinoscope, Phenakistiscope, and Zootrope. “Even Muybridge’s celebrated

Zoopraxiscope lectures of the 1880’s featured not actual photographs but colored drawings painted after the photographs” (p. 407). Not only were the images created by hand, but they were also animated manually. Magic lantern shows were performed, not automated, and the Zootrope’s barrel was twirled by hand or the Vivoscope’s handle cranked by the operator. The photograph was only combined with uniform, motorized motion at the end of the nineteenth century. Once the union of automated image production and automated motion was made, cinema – the live action, narrative super genre – was born and did not look back.

Cinema, Manovich says, “cut all references to its origins in artifice” (p. 408). An essential part of its identity is the implicit claim to capture reality on film, best exemplified in Jean-Luc Godard’s definition of cinema as “truth 24 frames per second”.

This quality, so often the discussion of film theory, is derived from the indexical nature of photography. Of course, the notion of live action as truth is problematic and has

25 divided theorists and filmmakers. James Monaco’s popular textbook How to Read a Film

(2000), for instance, outlines the division between “expressionism and realism” naming

Arnheim and Kracauer as champions of the two different schools. Arnheim’s Film as Art downplayed the realism of film recording, while Kracauer bolstered the potential of filmic verisimilitude. But the index as guarantor of truth persists and Manovich embraces the principle as at least closer to truth, saying poetically, “Cinema is the art of the index; it is an attempt to make art out of a footprint” (p. 406). Meanwhile, animation was relegated to the margins as cinema’s “bastard relative”. Without denying the success of

Disney and other animation studios, the twentieth century motion picture regime clearly belonged to live action cinema. But, Manovich proclaims, “In the 1990s, with the shift to computer media, these marginalized techniques moved to the center” (p. 409).

The process of digitization is a reduction. Not just from the infinitely smooth gradations of the analogue to discrete digital units, but also a reduction of different media to a common denominator. Once digitized, the computer does not differentiate between visual media. Computer animation, painting, live action film – all can be reduced to the common denominator of pixels, and software can be employed to meld and manipulate them. Live action material becomes only one of many raw materials that can be used to produce a motion picture. Thus, Manovich arrives at his formula:

Digital film = live action material + painting + image processing + compositing + 2D computer animation + 3D computer animation

He astutely gives the answer to his own question, “What is digital cinema?” “Digital cinema is a particular case of animation which uses live action footage as one of its many elements” (his emphasis, p. 410). After digitization, the privileged place of indexical

26 recording is lost. And he further summarizes the history of cinema to the present: “Born from animation, cinema pushed animation to its boundary, only to become one particular case of animation in the end” (also his emphasis, p. 410). In recognizing the central role of animation in film, he steers film studies in a new direction. Suddenly, animation studies – the academic “bastard” equivalent of the animated cartoon – gains newfound credibility and importance in film research.

Manovich’s essay has garnered considerable attention (it is absorbed as a chapter in his book The Language of New Media (2001), adopted now as a textbook in some post-secondary new media programs), but he never explicitly recognizes its academic roots. Essentially, he provides an updated version of what I have called the post- photography argument. The argument draws from a body of literature stabilized in the late 1980s and early 90s aimed at understanding the effects of digitization, as digital photography proliferated. The literature is perhaps best recognized in a collection of essays published as Photography After Photography (1996) and the thorough investigation undertaken by William J. Mitchell in The Reconfigured Eye (1992). While these varied works and authors may have unique takes on the changing nature of photographic images, they can be synthesized into a coherent argument they all, more or less, put forth.

Two major effects of digitization are highlighted: that images become 1) infinitely reproducible (copied) and distributable (shared) and 2), readily manipulated. The second effect is the most important to this paper. To be sure, this condition is not altogether new

– photographs have been manipulated since their conception. The point is one of degree; photos can now be easily manipulated and by nearly anyone with a home computer.

Given their mutability, digital photographs inevitably resurrect the old question, are photographs ‘Truthful’? Conventional wisdom recognizes the connection between mechanically produced photographs and their subjects. Wavelengths of light are allowed to enter the camera during the blink of the shutter, and react with the photosensitive film emulsion. The result is a ‘true’ optical representation of whatever stood before the camera – the causal or indexical link that seems inseparable from photography theory.

Barthes (1981) says the message of any photograph is: “This was.” Susan Sontag (1973) explains the power of photography, “a photograph is not only like its subject, a homage to the subject. It is part of, an extension of that subject” (p. 155). Indeed, “seeing is believing” is a guiding axiom to conventional thought – one purporting that that which is photographed did in fact exist or occur. One need only look to the courts, where (under certain circumstances) photographs have been used as incontrovertible evidence capable of deciding the fates of the accused.

But it is not just digitization that refutes photographic veracity. Traditional

(analogue) photographers are always making decisions about staging, lighting, mise-en- scène, composition, etc. and artistic license is exercised in the darkroom too. Landscape photographers, for example, encounter the difficulty of overexposing the sky to achieve good gradations in the ground. Frequently, skies are photographed separately, often at different locations, and the two negatives are composited together with gentle “dodge and burn” techniques. The resulting image confounds any claim to representational truth, but at least it has a sky. As Martha Rosler (1996) explains, “All these manipulations were in the service of a truer truth, one closer to conceptual adequacy, not to mention experience”

(p. 37). Those wishing to debunk the hypothesis of photography as truth are fond of

28 citing the now famous National Geographic cover. The February 1982 edition featured the pyramids of Giza, but moved closer together to achieve a more aesthetically pleasing composition that better fit the vertical “portrait” format of the magazine. A similar scandal erupted in 1994 when Time and Newsweek concurrently published a mug shot of

OJ Simpson. Time’s cover was modified to shroud the accused in shadow – “to make him look diabolical” (Coy, 1996, p. 74). Thus it would seem the debate is concluded.

Photos cannot – and never could – be trusted.

Yet the indexical nature of photography is obstinate and faith in photographic fidelity persists. A summary of the post-photographic argument may go something like this:

1. Photography has a privileged indexical relationship to its subjects. It is a

mechanical process that excludes the interpretation of the photographer and

therefore offers an objective ‘truth.’

2. Actually, traditional photographers make all kinds of choices – lighting,

composition, mise en scène, use of filters, cropping, compositing, etc.

3. These choices do not fully negate the ‘true’ indexical nature of photography.

Photographs still represent that which was.

4. Digital photographs are mutable in ways traditional ones are not. Content (pixels)

can be altered, swapped, deleted, composited, etc. More, all of this can be

done relatively easily with readily available software.

5. The easy manipulation of digital photographs will lead to the erosion of the belief

in photographs as ‘true.’

A typical conclusion to the post-photography argument is provided by Barbara Savedoff

(2000):

If we reach a point where photographs are as commonly digitized and altered as not, our faith in the credibility of photographs will inevitably, if slowly and painfully, weaken, and one of the major differences in our conceptions of paintings and photographs could all but disappear. (p. 202)

The jump between points two and three appears contradictory, but is made nevertheless.

To me, the conclusion is always qualified and has to offer up some ground, as if to say,

“Okay, we never could trust photos, but now (with digitization) we really can’t trust them.” Apparently, indexical truth is embraced, but qualified by the extent to which the photographer and printer interfere with the pure, mechanical photographic process.

The term “painting” as Savedoff, Mitchell, and others use it ought to be qualified.

It is used frequently in discussions of post-photography, and in the present text, to refer to a broad category of handmade images. It is not strictly oil or acrylic painting, for example, but could be any number of artistic media such as charcoal, pastel, conté, or, most importantly, pixels. Painting is the intentional act of making images as opposed to automated image production, chiefly with a camera. Paintings can have automated characteristics – what William J. Mitchell calls “algorithmic” properties (1992, p. 30) – when data about the subject is incorporated, such as perspectival measurements, or even holding up a pencil to one’s line of vision to measure relative proportions. But paintings are chiefly nonalgorithmic, “the product of many intentional acts” which “reveal a lot about what was in the artist’s mind” (p. 30). Thus, in the case of post-photography, the

30 digital photo can be taken as a starting point and edited or “painted” by an artist to better reflect their interpretation or desired message.

Regardless, the parallels between the post-photography argument and Manovich’s essay should be apparent. Digital photography blurs the lines between straight photography and painting, just as digital film blurs the line between film and animation.

To follow Manovich’s lead and reduce it to a formula, it might read:

Digital photography = photography + painting + image processing + compositing

Essentially, Manovich takes the argument and brings it to the realm of motion. And his writing has basically kept pace with the advance of the technology itself. William

Mitchell and others wrote just as digital photography became mainstream and photo- editing software became common, while Manovich wrote as the same technology advanced to manipulate moving images. This observation, I think, is useful in locating

Manovich in the present theory of digital images, both moving and static.

I will conclude this section with a brief critical look at Manovich’s formulation of digital cinema. He says digital film is a combination of a number of things: live action material, painting, image processing, compositing, 2D computer animation, and 3D computer animation. So, it can be inferred that digital cinema is some combination of the ingredients he lists. The recipe is incomplete. It does not prescribe the amount of each ingredient and, in fact, it is hardly necessary that all of them be included. One has to assume that even one of those ingredients is enough, say just live-action material, so long as it is digitized. Shooting on a newer, consumer grade camcorder, for example, creates a digital film. It does not take an astute reader to note that Manovich’s list is not exhaustive either. He makes no mention of sound, for instance. And he could expand it

31 further to include more obscure technical images like satellite weather forecasting simulations. But this is not a major flaw in the formula, just a matter of prudence and concision; including all kinds of digital visualizations and processes would make the list too long and unwieldy.

Another possible criticism is that his ingredients are qualitatively different in that some are better described as processes rather than raw visual materials. That is, image processing and compositing are things a digital artist does to the more material components of video footage or animations. Manovich tries to be inclusive but makes a categorical error. So, is his formulation bankrupt and useless? A hundred digital chefs could take his recipe and produce a hundred different visual dishes. I would suggest his formula is still useful and even integral to this thesis, if only for two important insights it affords. One, his key observation is that digital images are hybrids. In particular, animation scholars like Steve Reinke (2005) (not to mention myself) have latched onto his formula because it highlights the growing significance of animation as a key ingredient of digital moving images. Of course, digital images are by no means the first hybrids. There is a long tradition of mixed media in art and design. But (and this is the second point), it is implied that digitization makes such mixing easier than ever. The common binary computer language allows the various components (mainly live-action footage and animation) to be combined readily and seamlessly. My hypothesis (and departure from Manovich’s formula) is that the parts that make up the hybrid each have their own traditions. If the medium is the message, then so-called multimedia ought to have multiple messages. Or, at least, overlapping, nested, and unique messages.

It is worth pausing a moment to consider what sort of formula Manovich might have come up with for traditional (non-digital) cinema. How might he have answered

Bazin’s “What is [traditional] cinema?” (1967). Above, I conjectured about a formula for still, digital photography; a similar one for traditional cinema may read something like this: cinema = theatre + photography + music + practical special effects + montage editing. I write this not to suggest it is complete or even cohesive. Theatre is not rightly a material that goes into film, and editing is more of a process than material. If anything my hypothetical formula acts as a mirror to reveal how problematic and almost haphazard

Manovich’s formula can be. My point is to show that traditional cinema too is a hybrid.

Bazin rightly pointed to the debt cinema owes to theatre and literature (1967). We look to the old to understand the new. This notion is embodied more deeply and astutely in

McLuhan’s suggestion that old media becomes the content of the new (McLuhan, 1994).

Still, I think Manovich is correct to observe that digital media are somewhat unique hybrids owing to the ease with which “old” media can be combined and altered.

Manovich’s formula may be best reduced to an even simpler formula, by keeping its two most essential elements, live action film and animation. This is essentially an updated version of digital (still) photography as a merging of photos and painting, with the temporal or kinetic component now accounted for. The two formulas read as such:

(static) Digital Photography = Photography + Painting

(kinetic) Digital Cinema = Live Action Footage + Animation

Manovich is not so concerned with the erosion of truth associated with digitization, as the post-photography writers invariably are. However he still recognizes the indexical nature of live action filming and says that it is lost. He simply refrains from the value

33 judgement, or “sky is falling” mentality, of writers like Savedoff. Indeed, as film goes digital it may be less trustworthy (as objective documentation) than ever, but traditional film should never have been trusted so wholeheartedly in the first place.

Towards a Definition of Animation

Animation studies is a fledgling discipline. Jayne Pilling, editor of A Reader in

Animation Studies (1997), notes it is only in the 1990’s that “animation seems finally to have emerged from its previously very marginalised status” (iv). And it was only in 1991 that Alan Cholodenko introduced “the world’s first book of scholarly essays theorizing animation” (p. 9) with The Illusion of Life: Essays on Animation following shortly after the 1988 founding of the Society for Animation Studies (SAS). It may be because animation studies is such a young discipline that the nature, and even the definition, of animation is not particularly well understood. However, given the central role animation is to play in this discussion (and if one is following Manovich, to all discussions of film to come), it is prudent to provide a good working definition. It needs to be understood what differentiates animation from live-action film.

Animation has long been considered a sub-genre of film. But it has hardly been taken seriously by the academy. James Monaco’s (2000) popular text How to Read a

Film scarcely mentions animation and only discusses Disney as an economic and production powerhouse. Animation bears the stigma of kid’s stuff – no doubt owing to the child-oriented Disney feature length film and a tradition of Saturday morning cartoons. Thus, what scholarship exists, with a few exceptions (such as Eisenstein’s essays on Disney compiled by Jay Leyda), treats animation as a ground only fertile for

34 sociological and ideological investigation, such as the representations of race or women.

This is in part due to animation’s seeming predisposition to embrace character types and caricatures which then become useful objects of study for the way they lay bare collective cultural stereotypes, attitudes, and assumptions. Such investigations are important but are not exclusive to animation per se and thus serve to highlight how unexplored the field of animation studies remains in the academy. More recently, animated television shows directed at adults have become increasingly common, sparked by the unprecedented success of The Simpsons (1989). Popular adult-oriented cartoons such as Southpark,

Family Guy, and the Cartoon Network’s Adult Swim programs have come to the fore as successful animated shows. Some of their success is even owed to the preconceived notion that cartoons are for kids, which they exploit for comedic effect. In Southpark especially, watching grade school kids swear and confront adult situations is part of the joke. So film studies has acquiesced to a renewed interest in animation partly because cartoons seem to be taking themselves more seriously, having passed some rite of passage into the adult (and scholarship-worthy) world. However this is only one impetus, and the lesser next to Manovich’s proposition that animation has become an inextricable part of film.

There exists a sort of hierarchy of definitions within the connoisseurship of animation. The idea of a moving cartoon is seen as the most conventional and pedestrian.

This was the assumed definition when shows such as The Simpsons were mentioned above. Another common definition focuses strictly on movement. For instance, this is the assumed definition as adopted into the nomenclature of current web development software. Adobe’s (formerly Macromedia’s) Flash software speaks of animation

35 whenever an object moves between keyframes. A flashing or scrolling bit of text on a webpage, for example, is said to be animated. Working animators on the other hand, as

Stefan Kanfer (1997) notes, are fond of quoting the dictionary definition – “to give life, to bring to life.” That is, the task of the animator is not simply to make images move (a machine can do that!), but requires the investment of a creator – a mortal artist – to breath life into the image and make it become real. Following in this vein, nearly every western animator has a copy of The Illusion of Life: Disney Animation (1981) on her library shelf.

This lush, full-colour tome traces the development of the Disney style and the unique techniques necessary to make the characters come to life (adopted widely as the

“principles of animation”). For instance, it describes Norman “Fergy” Ferguson’s famous flypaper sequence from Playful Pluto that marked a turning point in animation.

According to the book and animation lore, it was the first time a drawn character seemed to be “thinking” and reacting in a believable way that made audiences emotionally invested in his plight. The idea of giving life is illustrated plainly in the Max Fleischer classic Out of the Inkwell series where the draftsman puts pen to paper and the creation literally jumps off the page. Clearly, this definition is the one most favoured by animators because it lends a certain credence to the God-like creativity to their profession, and is one of the most difficult things to do well.

Of course others see “character animation” (Disney’s modus operandi) as only one particular brand in a wider category of hand-moved images. If commercial animators favour the illusion of life definition, artistic-leaning connoisseurs are fond of Scottish-

Canadian animator Norman McLaren’s cryptic account of the essence of animation:

Animation is not the art of drawings that move, but the art of movements that are drawn. What happens between each frame is much more important than what exists on each frame. Animation is therefore the art of manipulating the invisible interstices that live between frames. (cited in Ward, 1998, p. 10)

For McLaren, the art of animation is not simply drawing a series of images (or moving objects), but knowing how much and in what ways to move them. No doubt this process can be learned and Disney’s team made great strides to develop conventions that seem to work. These conventions are taught at practical animation schools around the world and will be explored later in this chapter. But McLaren also goes so far to say, in an interview in the Canadian National Film Board documentary The Eye Hears the Ear Sees

(1970) that the particular skill of knowing what to do between frames is what separates the wheat from chaff of animators. He argues it is an intuitive process – even a kind of performance – that cannot be taught. The mystery surrounding creativity remains a steadfast bedfellow to all art.

So, it is apparent that McLaren’s definition bears one other assumption: animation is an art. This position was argued early on by animation pioneer Windsor McCay. He had high hopes for the medium as an art and lamented its loss to commercialization even before Disney applied assembly line logic to production. Addressing his colleagues at a dinner after he had stopped producing animation, he reportedly protested, “Animation should be an art. That is how I conceived it. But as I see what you fellows have done with it, making it into a trade…not an art, but a trade, bad luck!” (cited in Kanfer, 1997).

It is not particularly informative to raise art history’s long-standing query “is it art?”

37 especially in this postmodern age that has narrowed the gap between high and low culture. Suffice to say art and commerce are not mutually exclusive.

Just as Howard Becker (1982) is able to speak of “Art Worlds” as social constructions, so too can animation be seen as a particular artistic or production mode validated and upheld by social institutions. For most people in North America, mention of the word animation is apt to conjure up images from Disney movies. The frequent conflation of animation with the Disney style is testament to its institutional hegemony.

More broadly, it could be said that animation is not a particular technique (for instance drawing cartoons on cels) but is whatever viewers recognize as animation. This approach is useful for conquering the problem of how understandings of animation change with the advancement of different techniques and forms (just as this paper proposes to trace the latest developments). It effectively denies the notion of an “essence” of animation and reminds critics that animation as we know it is a particular form that has been shaped by social forces. David Bordwell (1997) is adamant in his critique of film theory that seeks to discover the essence of film, which he terms the “Standard Version” account of the development of film style.

[T]he idea of cinema’s history as an unfolding potential treats the medium as holding at the outset the seeds of future growth. Yet the later developments to which the historian points are always a mere sampling of the uses that have been made of the medium; if the historian picked different instances, she might be forced to posit a different essence ab initio.” (p. 32)

While Bordwell may be right to controvert the idea of a medium’s essence, his position amounts to saying that film (or animation, or whatever) is whatever historians say it is. In opposition, one could argue, it still is, because the historians need something

38 to write about and the institutions need something to build upon. Yet assuredly, the meaning of animation, like all meaning, is largely a social construction.

Bordwell and Kristin Thompson, in fact, offer a useful, if conventional, technical definition of animation. According to them animation is “any process whereby artificial movement is created by photographing a series of drawings, objects, or computer images one by one. Small changes in position, recorded frame by frame, create the illusion of movement” (cited in Bouldin, 2004, p. 6). Such a definition is fairly inclusive and allows for more marginal animation techniques such as stop-motion, claymation, drawing or scratching directly on film acetate, pinscreen, sand on glass (like the beautifully executed works of Caroline Leaf), etc. However, as Bouldin notes, and Bordwell alludes to in the earlier quotation, such a definition focuses on technique and may need to be revised as those techniques develop or change. Not to mention it champions the goal of

“movement” as opposed to “the illusion of life” that character animators are sure to contest.

A useful strategy may be to define animation in opposition to live-action.

McLaren’s definition focuses on the work done between frames, highlighting the conscious effort made to create the movement. It is a steady, deliberate process, a kind of choreography that is sometimes likened to dance. Oppositionally, live-action is characterized by continuous filming of a continuous action, typically of a real-world subject. This logic has led Edward Small and Eugene Levinson (cited in Bouldin, p. 24) to think of animation as single frame cinematography. Bouldin, however, thinks it is better to set up animation and live-action on a continuum, instead of in opposition to each other. Following Maureen Furniss (1998), she posits that the two poles of the continuum

39 are marked by a tendency towards mimesis (live action) or abstraction (animation), rather than exclusively different practices. However, there is no reason to think that they cannot be oppositional and on a continuum. The frame-by-frame technique better lends itself to abstract representation, while continuous filming is more likely to yield mimesis. Furniss is right to suggest that “there is no one film that represents the ideal example of ‘mimesis’ or ‘abstraction’ – everything is relative” (cited in Bouldin, p. 23). Animation and film belong together in what can best be called motion pictures.

In his own contribution introducing The Illusion of Life: Essays on Animation

(1991) (likely titled to find itself on the shelf next to the popular Disney history book of the same name) Cholodenko explains the close kinship of animation and film. He blazes the trail for Manovich’s essay (though unacknowledged by Manovich) by also tracing the history of protocinematic devices and reveals that they were essentially animation devices. The zoetrope, praxinoscope, and others first featured sequentially drawn images, rather than photographs. It is animation that begets film and “In this sense, animation would no longer be a form of film or cinema. Film and cinema would be forms of animation” (p. 10). This point is echoed by Manovich, when he writes “Born from animation, cinema pushed animation to its boundary, only to become one particular case of animation in the end” (Mirzoeff, 1998). Manovich admits the word “animation” may not sit comfortably with many people in this context, and indeed “motion picture” may be more apt. Perhaps the colloquial “movie” has never seemed so concise a designation.

So, there are many definitions to choose from. One guideline may be that if it is a motion picture and it is not live-action, it is probably animation. Manovich’s formula

40 that introduced this chapter is central to this thesis, and although he seems to say that digital cinema effectively is (a kind of) animation, such a definition will only be more confusing. However, an important distinction throughout this paper will be the opposition of that which is handmade and that which is mechanically produced. This will be the central tenet to differentiate animation from live-action, just as it is used implicitly to distinguish painting from photography. Paintings are handmade, animation is hand- moved! It may appear problematic to call CGI, or computer animation “handmade” but it need not be complicated. Computer animation can be entirely automated after parameters are set, as in running certain simulations, but it usually involves user input of varying levels. Thus CG can be seen as a hybrid between automation and hand manipulation, and can be called handmade in a general sense. Live action film is moved by machine – by the regular (twenty-four frames per second) sampling of moving subjects. So, in the context of motion pictures, animation describes images and objects that are made to move by hand, where each sample, or the interstices between samples, is chosen by a human animator.

The Rotoscope Phenomenon

It is not only recent film critics that are discontent with motion capture, the successor to rotoscoping. Even though rotoscoping was employed frequently after its invention in 1915, animators always had mixed feelings about it. Frank Thomas and

Ollie Johnston (1981), pioneering animators at Disney, relate their dissatisfaction with the technique,

[W]henever we stayed too close to the Photostats [live- action reference], or directly copied even a tiny piece of human action, the results looked very strange. The moves appeared real enough, but the figure lost the illusion of life. There was a certain authority in the movement and a presence that came out of the whole action, but it was impossible to become emotionally involved with this eerie, shadowy creature who was never a real inhabitant of our fantasy world. (p. 323)

They make some effort at an explanation which highlights the necessity of personal investment and expression in art, but offer the curious remark that recalls Jean

Baudrillard’s simulacrum or Walter Benjamin’s mechanical copy without aura, “No one knows for sure why a pencil tracing of a live action figure should look so stiff and unnatural on the screen, unless there simply is no reality in a copy” (p. 323). Their attitude seems typical of animators in general for whom rotoscoping is just another technique, perhaps useful in certain circumstances and for particular effects, but ultimately one of the many experiments (a failed one) towards desirable animation techniques.

The “uncanny” is a useful way to describe the rotoscoping phenomenon, however

I adopt the term with some hesitation given its multiplicity of meanings. Popularized by

Freud, the term remains ambiguous even among authors today. Michael Arnzen, introducing a special issue of Paradoxa addressing the uncanny, says the term’s meaning

“is multifaceted, but generally refers to an experience (either subjective or textual) of the

‘strangely familiar’” (1997, p. 319). The analysis section will demonstrate that descriptions such as eerie, strange, and weird are commonly applied to rotoscoping, and the term uncanny may belong with those adjectives, but the affect of rotoscoping may not necessarily be characterised as “strangely familiar” by all viewers.

Freud approaches the problem by compiling several examples of the uncanny

“evoked by persons, things, sense-impressions, experiences, and situations” (¶ 4). He builds on Ernst Jentsch’s work, which says the phenomenon rests on “doubts whether an apparently animate being is really alive; or conversely, whether a lifeless object might not e in fact animate” (cited in Freud, part 1, chap. 2). This is the premise that Mashiro Mori

(1970) builds upon in his concept of the “uncanny valley” and may be the most relevant to the task at hand. However, Freud is ready to dismiss Jentsch’s account, preferring to explain the uncanny as a fear of blindness or the dread of castration. If this account has merit, it will not be pursued presently. Freud’s other explanation – the uncanny as double

– may be more relevant. But he uses doubling broadly to refer to recurrences of themes or situations, as well as doppelgangers or twins. He describes a situation where one no identifies strongly with another to the point “doubt as to which his self is” (part 2, chap.

1). There is a “doubling, dividing and interchanging of the self,” which may possibly apply to rotoscoping imagery as a double of the live-action source material. Yet this account begs why images of people – whether photographs, painted portraits, or filmed actions, are not particularly uncanny. For this reason, the notion of the double is taken as only modestly applicable to the phenomenon of rotoscoping.

It can be seen the uncanny has many meanings. Indeed, it is difficult to say if the

“uncanny” is the best term to use in the first place. Yet, the ambiguity of the term may be a strength, and other authors have not hesitated to use it. So the term is used presently, in a sometimes ambiguous way, to describe an equally ambiguous phenomenon.

This section will provide some of the explanations of the uncanny phenomenon pertaining to rotoscoping in hopes of better characterising animation’s relationship to the

43 real, and digital cinema as a hybrid of live-action and animation. That is, the following are the reigning attempts to explain why rotoscoping and mocap look strange or uncanny.

It will draw from the rather limited scholarship on the subject of rotoscoping

“uncanniness,” particularly a special edition of Animation Journal (Langer, 2004) devoted to the rotoscope and its affects. The edition is guest-edited by Mark Langer of

Carleton University who is something of an expert on the subject and, in his own writing, tries to explain the phenomenon in terms of Freud’s concept of the uncanny. As suggested, I am wary of the relevance of this psychological approach and it will only be pursued as it is taken up by Mori’s concept of the uncanny valley. Note that the uncanny valley, as it applies to animation, has not been well theorized in a body of literature, and I will fill in these gaps here in the literature review. So, I am ignoring much of the uncanny psychology, and instead prefer to focus on issues of representation. I will begin with what may be the best explanation, or at least the most common.

The Conventional Approach

The standard explanation for the rotoscoping phenomenon is as a clash of stylistic conventions. Introducing the journal mentioned above, Langer describes rotoscoping in which there is,

a simultaneous presence of the drawn and the photo- indexical, in which the rotoscoped or Rotoshopped [Rotoshop is a proprietary software program developed by MIT alumnus Bob Sabiston] body is not so much fused with the human body as it is mapped over it. Indeed, it is this fluctuating nature of co-presence, or, as Jose Capino puts it ‘the appearance and disappearance of the real body’s image beneath and between its traces’ (p. 6, his emphasis)

It is that simultaneous presence of two stylistic histories that is difficult for audiences to make sense of. In a newspaper article about Beowulf, Eric Lichtenfeld, a film professor at Loyola Marymount University, puts it succinctly, "Movies like this often look just enough like animation and just enough like live action that you don't feel like it's either one, which makes it that much harder to invest in the reality of the story." (cited in

Breznican, 2007). All media have their own established cultural codes learned by audiences and creators. Beholding something that looks like a cartoon evokes a whole host of expectations for “cartoonness” based on past experience. E.H. Gombrich (1959) argues that the simplicity of caricature (or in this case animated cartoons) assures its success, “because its lack of elaboration guarantees the absence of contradictory clues”

(p. 284). But when those clues, or conventional cues, are co-present with filmic ones that oscillate and overlap, they are difficult to hold in the mind at once. The experience of watching rotoscoped pictures may even be a little like what psychologists call cognitive dissonance where two contrary beliefs held at the same time. Not that the two styles are necessarily contrary, but they are distinct from each other and evoke different agreements about relations with the real. From this perspective, it makes sense that a viewer would call the experience eerie or strange.

This explanation holds great currency when championing cultural construction as the pre-eminent moderator of experience and panacea for all meanings. But when looking for something “in the medium,” the explanation falls as flat as the rotoscoped characters. Explaining the eerie phenomenon in terms of a clash of conventions fails to explain why it existed before animation was conventionalized. Of course, this is not to say that animation is without stylistic precedent before the twentieth century. Indeed, the

45 animated cartoon has a long lineage borrowing from comics and caricaturists such as pioneer Rodolphe Töpffer, and even to the history of drawing back to Lascaux. But I have defined animation as the hand moved, and I mean to focus on the conventions of movement in motion pictures, which has a much shorter history. The illusion of life celebrated by animators is achieved largely through conventions established after the invention of the rotoscope. Thomas and Johnston (1981) even suggest those conventions were developed to transcend the limitations and ghostly effect of the rotoscope, “once a movement was understood it easily could be incorporated into cartoon terms. We had made the big break with rotoscope” (p. 323). The “break” they speak of at once acknowledges how rotoscoping helped to understand motion, but also the break away from dependency on the technique.

So, although the mixing of established cultural cues may be a significant part of explaining the uncanniness of rotoscoping, it is not entirely satisfactory.

Newness and Novelty

If the mixing of conventions is one explanation, another is that rotoscoping itself is a convention not yet assimilated by audiences. Kim Williamson, film critic for

Boxoffice Magazine (2004), thinks newness is at least part of the phenomenon. She writes of The Polar Express’s creepiness, “Whatever it is, older members of the audience might find the characters’ look downright creepy. (Children, being more adaptable and unschooled by decades of filmviewing, are less likely to be put off by this.)” (¶ 3). She implies that the next generation, exposed to more frequent use of motion-capture, will be

46 immunized to its affects and the phenomenon will be relegated to a history of other film techniques and curio, once shocking, now commonplace.

A possible counterargument is that the rotoscoping technique, approaching its centennial, is hardly new. But its date of origin is less significant than its frequency of use. Relatively, it has not been used much having been largely supplanted by the hegemony of Disney’s style in the West. It has made occasional appearances in the mainstream, such as in Ralph Bakshi’s animated The Lord of the Rings (1978) and Cool

World (1992), but not with much commercial success. Indeed, most audiences have never heard of the Fleischers’ invention.

In an excellent article, Lisa Bode (2006) likens the rhetoric surrounding rotoscoping’s eeriness to the grumblings of early film critics. Particularly in Russia, she says, there was a tendency to regard characters on the screen as lifeless shadows.

It seems that what the early Russian response to the cinematograph and the recent English-speaking writing on the digital actor have in common is their use of an image of a human figure (its uncanny effects and aesthetics) as an allegory of ongoing human transformation in modern industrial societies. (p. 185)

She argues that “the history of image technologies is punctuated by moments of strangeness that emerge and then recede again,” and concludes, “If the past tells us anything, these perceptions of the digital actor as an uncanny figure will not be permanent” (p. 185). She offers an argument more nuanced than the parenthetic explanation of Williamson; not that rotoscoping is a short-lived fad like rollerskates or foam resin “Crocs” sandals, but a particular representation of the mechanically human

47 that will be accepted when the culture is good and ready. She may very well be proven right, but her decision to focus on the human deserves more attention.

The Human, the Machine, and the Uncanny Valley

New York Times film critic Manohla Dargis attributes the “unlifelikeness” of The

Polar Express characters to motion capture technology, but observes the same technology was used to good effect with the Gollum character in Peter Jackson’s The Lord of the

Rings trilogy (2001-2003). “With the actor Andy Serkis providing its outlines, Gollum came across as more or less persuasively real because the character is a non-human creature a-prowl in a fantasy world” (2004, ¶ 8, my emphasis). It seems to be a recurring theme for critics to identify mocapped human characters as particularly problematic, whereas animals or fantastic creatures are deemed acceptable. Many of the critics enjoyed the scene in Polar where a golden boarding ticket is blown out a passenger car window, wafts underfoot of a pack of scurrying wolves and is seized in the beak of a swooping eagle before miraculously returning to the train. Empire Magazine even calls it

“the single most lyrical passage of CGI yet painted on a screen” (¶ 4). The scenes with wolves and birds, apparently, do not to raise the same concerns as the human characters, but nor were they motion-captured, so it is difficult to locate the source of uncanny discontent – is it because of mocap per se or only when used for human representation? I do not know of any instances of live animals being motion captured, but it will likely happen and the results may be interesting.

Regardless, there may very well be a fault with Zemekis and his team for attempting to make the human characters too human. In the seemingly disparate field of

48 robotics, scientists have long been wary of making their creations resemble the human.

In his influential paper titled “The Uncanny Valley” (1970), Masahiro Mori theorized the strange relationship between viewer empathy and the human-likeness of robots. He observed that, unlike many phenomenon that follow a y=f(x) relationship (such as a common acceleration curve mapping velocity over time), human empathy is not so simply plotted. He explains, “I have noticed that, as robots appear more humanlike, our sense of their familiarity increases until we come to a valley. I call this relation the

‘uncanny valley’” (¶ 2). Mori says there is a positive relationship between familiarity and resemblance up to a point where onlookers deem the robot either too real, or not real enough. At that point, the relationship plummets and viewers are left with a sense of unease or creepiness.

Figure 2.1

Mori indeed focuses on humans, but gives some attention to animals (the stuffed animal on the graph), suggesting the sense of familiarity between human and animal is never as

49 heightened as between two people. The two axes on his graph are translated as familiarity and human likeness, but the poles in the relationship are the living (human) and the inanimate, or machine. That is, the actual curve of the graph moves from

“industrial robot” up to “healthy human.” What is fascinating is that animators and roboticists, who usually occupy very different social spaces and seemingly have disparate goals, just might meet each other at the edge of the uncanny valley.

I have already suggested that rotoscoping is a simultaneous representation of the handmade and the machine-made. Interestingly, Langer identifies the same two human versus machine poles as co-present in rotoscoping, but labels the live-action as organic

(human) and animation as mechanical. Specifically, he says, “By incorporating live- action (the organic) with the mechanical (cel method drawings), the rotoscoped image became a hybridized product…” (2004, p. 5). Most theorists, myself included, would contend the opposite designation of poles. Langer may have in mind the notion that most commercial animation is produced according to an assembly line logic, with thousands of drawings being produced by many animators working busily at their drafting tables; the assembly line could be called mechanical, but that is a stretch. More likely, he may be thinking of the “jerky” movement characteristic of early animation, compared to the smooth results yielded by rotoscoping. While such jerky movement may be interpreted as mechanical (sometimes that adjective is used in criticism), it is better called “crude” or

“undeveloped.” The term mechanical is highly misleading because such drawings, though mechanically photographed in sequence after the fact, are not mechanically produced. Drawing, by most standards (as I have tried to demonstrate) is a highly personal and organic procedure. And so is animation, where an animator decides the

50 movement between frames. By contrast, live action film is very mechanical. Manovich concurs, speaking of the transition from still photography to film, where mechanical recording of images in time became possible with the motorized camera and projector.

“Irregularity, non-uniformity, the accident and other traces of the human body, which previously inevitably accompanied moving image exhibitions, were replaced by the uniformity of machine vision” (1998, p. 407). It is live-action that should be designated as mechanical.

I would concede that regardless of how the poles are differentiated, whichever side is mechanical or handmade, rotoscoping stands as a hybrid of the two. Mori’s uncanny valley is of equal relevance to the animator as to the creator of robots because rotoscoped and mocapped imagery risks falling into that valley. I wish to extend Mori’s argument to its logical conclusion. It may be that the poles identified by Mori and

Langer do not go far enough; perhaps it is not a matter of human versus machine, but a matter of life and death!

The labels on the graph that read “corpse” and “zombie” are telling; as is the word

“animation,” considering the dictionary definition – to give life. Plus, the eventual goal of robot research is artificial intelligence, sliding into the even loftier goal of artificial life. In this context, all of the comparisons of rotoscoped films with horror films, peppering the film critic’s reviews of The Polar Express, seem less like unfair hyperbole.

Stella Papamichael of BBC Films, for instance, writes “Forget Christmas cheer – kids will run screaming from The Polar Express after seeing Tom Hanks looking like he’s been killed, embalmed and resurrected by lightening” (¶ 1). Several others describe the

Polar characters as zombie-like, or like the monsters of Invasion of the Body Snatchers.

Indeed, zombies, undead vampires, dolls-come-to-life, and wax museums are common subjects in the horror genre. Much like the creepy mocap characters, all of these mythological re-animated monsters are, by most standards, creepy. They occupy an unnatural earthbound purgatory between life and death. The production notes for Polar state the goal of the filmmakers as exactly what Mori cautions not to do.

‘The problem with traditional animation for a project like this,’ says Zemeckis, who isn’t averse to employing the technique in its rightful place, ‘is that it falls short in depicting authentic human characters. With exaggerated images, fantasies like Snow White and the Seven Dwarves, or cartoons, it’s great. But I was looking for something more realistically alive.’ (¶18 )

Despite his best intentions to make “realistically alive” characters, Zemeckis seems to have steered his team into the uncanny valley. Witness the animator turned murderer. Or at least, creator turned Dr. Frankenstein.

In her insightful review, Manohla Dargis describes what she perceives as a division between animated filmmakers’ goals:

On one side of the divide are Pixar visionaries like Mr. Bird and the Finding Nemo co-director Andre Stanton, who either know they can’t recreate real life or are uninterested in such mimicry, and so just do what animators have always done: they imaginatively interpret the world. On the other side of the divide are filmmakers like George Lucas who seem intent on dispensing with messy annoyances like human actors even while they meticulously create a vacuum-sealed simulacrum of the world. (¶ 10)

She clearly puts Zemekis and Doug Chiang (Polar production designer and former Lucas compatriot) in the second category. She does not seem aware of Mori’s theory, but is able to draw a similar division nonetheless. In fact, she is astute to speculate that Brad

Bird and the Pixar team “know they can’t recreate real life.” One of the best resources I have found on this subject is from a section of Brad Bird’s The Incredibles (2004) DVD special features, entitled Building Humans. Tony Fucile, Bird’s friend and animation supervisor explains, “…we’ve introduced more, maybe a little more caricature in the humans, whereas I think in a lot of other films, the humans tend to get…natural looking and the closer you get to the real human it looks off, it looks weird” (2004, 18:50).

Again, it is difficult to say whether Fucile is aware of Mori’s theory, but as a veteran animator and character designer he appears acutely aware of what is acceptable to audiences. Bill Wise, character supervisor, echoes, “The characters were stylized cartoons that had to be believable humans. With our kind of faux realism, the closer you get to real, the creepier” (Robertson, p. 20). In The Making of The Incredibles (2004)

Pixar director John Lasseter explains people’s sensitivity to evaluating human representations, “Our audience looks at a human being in the mirror every single day and if you don’t do it just right, it won’t be believable to the audience” (18:30). The precipice above the uncanny valley is acknowledged bluntly by Rick Sayre, Supervising Technical

Director, who says of the character design, “It needs to not be disturbing but yet still feel like it’s alive, and that’s a pretty tough thing to do” (in Building Humans, 19:33). The

Pixar team has been hugely successful, perhaps not by jumping the uncanny valley to the next “real life” peak, but for recognizing the danger below and holding back to settle comfortably on the top of the Mori’s first peak. Indeed, that is precisely what Mori prescribes, “We hope to design robots…that will not fall into the uncanny valley. So I recommend designers take the first peak as the goal in building robots rather than the

53 second” (¶ 10). Zemekis, on the other hand, tries to make the leap to the second peak and tumbles down into the pit of monsters.

This is not to say that Mori’s theory is fact. David Hanson, one roboticist who has brazenly challenged Mori’s cautionary tale calling the idea, “really pseudoscientific, but people treat it like it’s science” (in Ferber, 2003). Hanson, who ignored Mori and built a robotic head as humanlike as he could, seeks to traverse the valley. Indeed that is the challenge to artificial life researchers and consequently to those in Hollywood who wish to create a convincing synthetic actor. Already, there have been instances of secondary characters or CG “stunt doubles” onscreen that pass for the “real” thing in the eyes of most spectators. Mark Langer, for one, “believe(s) that true photo-real animation is just around the corner” (2002, ¶ 12). Many others may concur, but whether a compelling, photo-real, synthetic “star” actor will emerge and satisfy audiences remains to be seen.

Mori’s concept of the uncanny valley is very useful to explain the eerie phenomenon of rotoscoping and mocap. Even if some of the more stylized examples of rotoscoping, such as the Fleischer’s Betty Boop or Koko the Clown shorts, can hardly be considered photo-real in static appearance, their movements are quite “real” in the sense that they are derived from live-action footage. However, if the explanations provided thus far are satisfactory for explaining the phenomenon, I wish to add one more to build on this groundwork, drawing from the insights of medium theory.

CHAPTER 2: METHODS

My methodology can be largely deduced by recapitulating my research question:

How can an examination of rotoscoping in The Polar Express, through the lens of medium theory, inform an understanding of digital film? It will involve two distinct parts. The first will establish the object of analysis, the rotoscoping phenomenon. Up to this point the supposed “eeriness” or “uncanniness” of rotoscoping has been implied, but not properly demonstrated. Again, by “uncanny” I do not mean to invoke a specialized, operational definition because I think Freud and others use it in too many varied ways that may or may not be relevant in this instance. Instead I only mean it as a broad category used vaguely by critics, but also used interchangeably with terms like eerie, weird, strange, etc. That is, I am not trying to prove that rotoscoping is uncanny (by whatever definition). I simply want to show that film critics have reached a consensus in describing it as uncanny, or at least weird, strange, eerie, etc.

The second part is an analysis of the rotoscoping phenomenon using medium theory. This analysis is not intended to supplant those already put forth in the literature review, but rather to provide a complementary explanation. Further, I hope these analyses will provide some groundwork to theorize about digital film more generally.

Proving the Object / The Object as Method

Borrowing from Lev Manovich, I have established that digital film is a hybrid form, made easy by the codification of images into pixels. Related to this premise is my suspicion that digitization ought to yield some unprecedented results. The infinite

55 variations of digital hybrids may beget some strange and wonderful offspring. The digital screen can be freak show or beauty pageant. Combining rather disparate traditions of particular media (in this case animation and live-action cinema) may yield some surprising forms, of which rotoscoping may be the most readily identifiable. There are many movies to choose from that exhibit what I consider such forms. Final Fantasy: the

Spirits Within (2001) was the first attempt at a feature length computer animated picture with photo-realistic characters and environments. Richard Linklater’s Waking Life

(2001) and A Scanner Darkly (2006) are two films that use proprietary software

(Rotoshop) to trace over live-action footage making it look like a sort of cartoon. The

Polar Express uses motion capture to animate computer models. These films are not chosen at whim. I have scoured film reviews, from popular sources such as Roger

Ebert’s Chicago Sun-Times column, and identified a common thread. Invariably, with these particular films, the reviews mention the way they were produced and the resulting unusual look of the films, even if they do not always agree on their relative success as narratives. The Polar Express has been chosen as exemplary of digital cinema because of its clear and simultaneous merging of animation and live action film techniques, bridged through digital computer technologies.

So, my hypothesis is that digital films may have some unusual looks or aesthetics and it will be part of my task to prove it. Such a seemingly subjective claim (that digital film looks “odd” of “eerie”) may appear difficult to prove, however this may be the most straightforward part of this paper. It will be undertaken through a cursory content analysis of film reviews of The Polar Express. I say “cursory” not to skirt a rigorous inquiry, but because this investigation does not demand an overly close reading of the

56 reviews. It is not essential, for instance, to understand how the film critics construct their arguments in depth, or even to perform keyword counts – only to establish that the critics reach consensus that the film has some unique qualities worth exploring. This is what I mean by “proving the object” – to demonstrate that I have identified a phenomenon worthy of explanation. To show that the rotoscoping phenomenon is not just something I dreamed up. Explaining why the film looks odd is ultimately more challenging than establishing it as somehow odd in the first place.

To prove my object, I will use the Movie Review Query Engine website

(www.mrqe.com), an online database (it claims to be the largest) of movie reviews. It provides the option of sorting the reviews by “popularity,” which is taken to mean the most widely read film reviews as opposed to, say, by the “star” rating or success rating awarded to the film. I will take a sample of the ten most “popular” search results for The

Polar Express and use those reviews as a way to gauge a general response to the film. I will look for trends that might suggest a readily identifiable aesthetic, unique to the film.

Note that the film was also released in 3D IMAX format, and I have elected to ignore the reviews specific to the IMAX version, believing it would unnecessarily complicate the analysis.

Studying the film The Polar Express will do double duty; it is both the object of inquiry and a method at the same time. Ultimately, this paper seeks to offer a better understanding of digital film. To this end, the film Polar is used as an exemplar, or sort of case study, and it is hoped that an analysis of it will allow me to speculate more widely about digital film in general. In this sense, all case studies, when situated in a broader context, are both object and method. Particular cases (such as a film) are microcosms

57 that can be extrapolated to better understand macrocosms (i.e. digital film at large). That is, the film and reviews of the film will be examined to demonstrate that it is an “eerie” film, and this point will be used to speculate about the peculiar nature of digital film more broadly.

The Medium is the Method: How to “do” Medium Theory

In my introduction, I made a congenial proposition. If digital films (and especially all digital or “new media” for that matter) are hybrids, it makes some sense to trace the traditions of each media stream as it meets the ocean of hybridity. This, I suggest, will result in a better conceptualization of the particular hybrid in question. But what do I mean by tradition? Another paper might find it fruitful to define tradition as a cultural practice and systematically parse out the different practices of the animator set against those of the live-action filmmaker. This has already been briefly covered in the literature review, referring to the uncanny phenomenon as a clash of styles or cultural practices. Again, this explanation carries great currency, but will not be pursued further here. Rather, the impetus of this thesis is to provide an alternative, complementary explanation. I hope to round out the account by tracing the traditions of each hybrid component from a medium theory perspective.

In what way can medium theory be considered a method? It is an approach, framework, or even science for understanding phenomenon. Arjen Mulder (2004) writes:

Marshall McLuhan called media theory a science because, like all sciences, it can explain all of reality – but only from a single angle or on a single level. Physics can interpret essentially every phenomenon in the universe as obeying natural laws. Biology considers everything from a microbe

to an ecosphere as a living and evolving system. Chemistry sees bonds between atoms and molecules forming and breaking everywhere. In the same way, everything can be interpreted culturally, or sociologically, or historically. Media theory understands every phenomenon and process as determined by the media with which it is brought into being, interprets itself, is perceived and stored, and finally made into something else. (p. 15-16)

Of course, a common criticism levelled against McLuhan was that he did not develop or utilize a cohesive, systematic approach – something the hard sciences and social sciences properly attempt to do. In some ways that was a strength of his approach. He was fond of the analogy that a fish does not know it is in water, recalling the way we are usually oblivious to the air we “swim” in, or more pointedly, to the media environments we are constantly engaged with. Following the poetic strategy of the Imagists, he preferred to utilize what he called “probes” – the cultural equivalent of dropping dynamite into a pond to see what would float to the top. He wanted to challenge assumptions to gain a fresh view and did not hesitate to board sometimes fanciful trains of thought to see where they might take him. Playboy (1969) suggested his methods were “erratic and inconsistent,” and that he engaged in “free-for-all theorizing.” I am NOT proposing to follow such a reckless, scattered, shotgun approach – however admirable it is in its own way. After

McLuhan, scholars have had the benefit of his more plausible insights to build medium theory into a more accepted and respectable “science.” Walter Ong and Joshua

Meyrowitz, for example, made great strides to solidify the best of what McLuhan had to offer, and this paper attempts to continue that tradition.

How, then, does one do medium theory? Emulating McLuhan could become unnecessarily complex, if it is even possible. I wish to keep this is as simple as possible.

For my purposes, doing medium theory involves three successive stages that build on one another. Simply put, they are 1) examining how different media engage different senses,

2) how media “restructure consciousness” (to borrow Ong’s phrase, 1982) in particular ways, and 3) how media affect cultural trends more broadly. I will not start from scratch.

Instead, I will largely draw from work already done by medium theorists to flesh out the three stages.

Potentially, the three areas I mention could be huge undertakings. For instance, in the first stage, a logical approach might be to compile a list of all the parts (inputs) of hybrid digital film, to trace their sensory engagements, and develop a taxonomy of all parts. Then, the taxonomy could be used to better understand individual digital films, by referring to how much of each ingredient is put into its particular recipe. W.J.T. Mitchell

(2007) has made such a proposition after challenging the notion of purely visual media:

The claim that there are no visual media, then, is really just the opening gambit that would lead toward a new concept of media taxonomy, one that would leave behind the reified stereotypes of “visual” or “verbal” media, and produce a much more nuanced, highly differentiated survey of types of media. (p. 400)

By Mitchell’s own description, such a taxonomy would build on what Hegel called the

“theoretical senses” – sight, hearing, and touch. These would be combined with the

Peircean sign triad – icon, index, and symbol – to create a matrix of (at least) six variables.

If such a framework is even possible, it will not be attempted here. As mentioned, one of the charges frequently laid against McLuhan was his lack of systematic inquiry. It is difficult to take up his ideas and be thorough and resolute. The idea of sensory ratios

60 especially, which I believe carries great currency, may be impossible to quantify. To say a movie or a particular scene is three parts visual, two parts tactile, and one part olfactory, is likely impossible. But the advantage of his insights is that they provide understanding not available otherwise. I will be content to speak in trends, and for now employ a more limited taxonomy. I propose this rather unabashedly because I perceive a danger in being too specific in this sort of analysis. The argument for differing sense ratios, I suspect, is more compelling at the level of trends and starts to break down in contradictions when one tries to quantify the ratios exactly at a micro-level. That is, it can be useful and meaningful to speak of a film evoking a tactile sensation in the viewer, but the claim becomes an enormous problem to try and measure just how tactile the experience.

To avoid such a pitfall, I hope to employ a more pragmatic model by dichotomizing approaches to image making. Roughly, the two camps correspond to either hand-produced or machine-produced images and, in particular the differences between animation and live-action film. It is hoped that such a bifurcation (which is not necessarily dialectical) will be a more convincing and accessible framework than the nuanced and potentially messy taxonomy Mitchell has in mind. That is, I am dividing handmade and machine-made media to establish two camps that can be followed more succinctly through the three stages I have described.

Methods Summary

In sum, digital cinema is a hybrid with two significant parts: live-action footage and animation. Hybrids, I hypothesize, ought to have some strange or unique qualities from their more “essentialist” parts. The Polar Express appears to have such qualities and this

61 will be confirmed through a content analysis of popular film reviews. Further, The Polar

Express uses motion-capture as a descendent of the technique of rotoscoping, which may have a lot in common with digital film. As such, it will serve as a precedent and useful object of analysis to better conceptualize digital film. So, the function of this film analysis is twofold. Digital film, I suggest, can be better understood by tracing the traditions of its parts, which are mainly animation and live-action film. Polar is an example of such a digital film. It is also useful to look for precedents to phenomena, and rotoscoping is a historical precedent to digital hybridity worthy of exploration. Polar, in addition to being a digital film, is a good example of rotoscoping/mocap, giving it a second use in my analysis. In this paper, these traditions will be analyzed using medium theory, particularly the concept of sense-ratios. Rather than attempting a complex taxonomy of media and their sense-ratios, I will dichotomize the traditions into handmade (animation) and machine-made (live-action) media.

CHAPTER 3: ANALYSIS

Part 1: The Polar Express Reviews: Technology Before Story

The Polar Express is one of those films that divide people. Not in the same way that, say, The Last Temptation of Christ (1988) did; but there are certainly mixed reviews.

The division, however, appears to lie along two axes, technical production and story. Of the ten reviews I selected, which are the ten most popular according to the Movie Review

Query Engine, all mention both aspects. Except, that is, for Peter Travers’ from Rolling

Stone who readily dismisses the entire film (awarding one star out of four) based on its technical failure. The reviews cited were all published in 2004, the year of the film’s release. The reviewers and their respective employers are as follows:

1 ReelViews (James Berardinelli) [3.5/4]

2 Chicago Sun-Times (Roger Ebert) [4/4]

3 The New York Times (Manohla Dargis)

4 Rolling Stone (Peter Travers) [1/4]

5 Boxoffice Magazine (Kim Williamson) [3/4]

6 Time Out, London (DA) [no rating]

7 Empire Magazine, UK (Ian Nathan) [3/5]

8 Salon (Stephanie Zacharek) [no rating]

9 Reel.com (Tim Knight) [3/4]

10 BBC Films (Stella Papamichael) [2/5]

I will not go through each review individually, as that is unnecessary to the task at hand. But it will be prudent to give a little background to the film. I will take my cue

63 from the rather formulaic background provided in each review. Given their consistency, one has to wonder just how much influence the press release packages distributed to reviewers shape their articles. Regardless, all the critics are privy to and make mention of the mode of production and are familiar with the source material, Chris Van Allsburg’s award-winning 1985 children’s book of the same name. The film is directed by Robert

Zemeckis, which is significant given his penchant for effects-laden movies and his interest in incorporating animation techniques. He tends to work with pretty good, or at least likeable, populist stories. He directed Back to the Future (1985), Who Framed

Roger Rabbit (1988), and Forrest Gump (1994), all with great domestic success, commercially and critically. Forrest Gump is an oft-cited landmark film for visual effects. Famously, Tom Hanks’ character was made to interact with archival footage of

John F. Kennedy by compositing his body into the scene. The president’s mouth was then “painted in” to match the over-dubbed speech. Equally famous is the floating feather, featured in the bookend scenes that open and close the film. Lev Manovich describes the feather as, “something which is intended to look exactly as if it could have happened, although it really could not” (In Mirzoeff, 1998, p. 410), which begins to describe the peculiar aesthetic of digital film this paper attempts to grapple with. Roger

Rabbit would also prove an interesting object of study because it explicitly combines animation and live-action, though using more traditional, analogue techniques. Roger

Ebert rightly identifies the connection, noting how in Roger Rabbit Zemeckis “juxtaposed live action with animation, this time [in Polar] merges them, using a process called

‘performance capture’” (Nov. 10, 2004 paragraph 3). I prefer to distinguish these films as heterogeneous and homogeneous blending, forming opposite ends of a spectrum.

Roger Rabbit leans towards heterogeneity, while digital technology in Polar increasingly makes homogenous blending possible.

As Ebert mentions, the film uses the technique of “performance capture” in the filmmaking process. Of course, this is the other bit of background information deemed essential by reviewers (and the film’s marketers), next to Van Allsburg’s book. But there seems to be some confusion about the technique, which points to a larger debate. First, here is how it works. Motion capture is a technique where an actor dons a sensor-laden suit and performs actions in a specially equipped studio. Motion trackers – usually optical cameras, but also magnetic or acoustic systems – sample and record the sensor coordinates in three vector space, in real time, into a computer. This data is then translated onto a doppelganger 3D character model (for those unfamiliar with computer character animation techniques, it helps to envision a virtual puppet with virtual hinges placed at all its joints). However the actor moves, the character model moves. In the film, Tom Hanks provides the performance to be captured for no fewer than five characters.

In my view the term “performance capture” is a publicist’s manoeuvre to flog the technique as a new and improved version of the (implicitly) obsolete process of motion capture. The film’s production notes say, “Beyond mere motion, this highly developed system was designed to capture every discernable movement and the subtlety of human expression from an actor’s performance, down to the slightest nuance or flutter of an eyelid” (¶ 22). They even call it a “giant step beyond current standards” (¶ 22). At least one reviewer, James Berardinelli, must have been enthusiastic to receive his press release package, parroting “The techniques used in The Polar Express will allow filmmakers to

65 make a giant leap forward” (¶ 1). The other critics are more, well, critical. Regardless, the principle is the same as mocap, only so-called performance capture (percap?) uses more sensors, including pins attached to the actor’s faces to capture facial expressions.

The confusion I alluded to, not surprisingly, concerns whether this process is animation or live-action. Again, the production notes acknowledge this debate, saying.

Those who have seen the final footage attest that it defies easy categorization. Familiar comparisons fall short of the mark. Often it’s described in terms of what it is not – as in not traditional animation, not merely motion capture and not strictly live action. An art form in its own right, Performance Capture effectively breaks new ground to offer images like nothing seen before. (¶ 30)

What is interesting is to observe how this subtly arises in the film reviews. Reelviews and the New York Times seem ready to embrace the technology and say that the characters are played by Tom Hanks. Reel.com’s Tim Knight felt compelled to put the word “played” in quotes, acknowledging the ambiguity of the process. Boxoffice gives a traditional breakdown of the film preceding the actual review, including running time, the Motion

Picture Association of America (MPAA) rating, and opts for the traditional animation crediting, i.e. “voiced by Tom Hanks…”. Salon also says, “featuring Tom Hanks as the voice of…”. Roger Ebert provides a similarly matter-of-fact breakdown, but gives credit under the designation “Body movement performers.” If nothing else, these subtle discrepancies should point to the newness of this phenomenon and the reviewer’s current inability to decisively address the ambiguity. Or, maybe it just highlights the dogged compulsion to put things in established categories even when they do not fit neatly.

If there are problems coping with just how mocap technology works, there is even bigger uncertainty surrounding the resultant look. Almost without fail, the film is said to

66 have a “creepy” or at least unusual look, largely attributed to the use of performance capture. However before invoking those descriptions, I will be guarded and provide other possible explanations for the creepiness; mainly, the rather creepy mood or tone of the film’s story. Ebert provides a good introduction, “’The Polar Express’ has the quality of a lot of lasting children’s entertainment: It’s a little creepy. Not creepy in an unpleasant way, but in that sneaky, teasing way that lets you know eerie things could happen” (¶ 1). This does not reveal why, but later he adds, “It has a haunting, magical quality because it has imagined its world freshly and played true to it, sidestepping all the tiresome Christmas clichés that children have inflicted on them this time of year” (¶ 11).

The story begins with the unnamed boy (credited as Hero Boy) waking from his sleep to a loud steam train halting on his quiet suburban street, which suggests the events are a dream and the story is imbued with dreamlike qualities. It has the quietude of a moonlit night and moments of imminent peril such as when the train careens off its tracks and across an open icefield. There is the apparition of a hobo, whose outstretched hand lingers a prolonged moment each time he departs by evaporating into the night air. The story has moments of warmth, sure, but is by no means holly jolly.

The restrained glacial palette of blues and greys, broken by hearthen glows of light, help lend an eerie mood. The colours mimic those used in the book and Manohla

Dargis of the New York Times echoes, “The illustrations in Mr. Van Allsburg’s book have a patina of nostalgia, but there’s something distinctly melancholic about them as well.

The oil pastels the artist used for the pictures soften even the hardest edge,” adding, “the images seem at once moody and mysterious” (¶ 4).

Next, I suspect there is a certain phenomenon unrelated to rotoscoping or mocap, but fairly common in animation, where sequences or characters are reused onscreen

(called “instancing” in CG parlance). The hot chocolate sequence, a forgettable song and dance number midway through the film, features eerily similar waiters doling out hot cocoa to the kids on board the train. They move with exacting, algorithmic symmetry that would be impossible in real life. Similarly, the hordes of elves at the North Pole look like they too came off the assembly line where they make the toys. It is in these scenes that Freud’s account of the uncanny as “doubling” may have some currency. A similar sensation comes, I would say, in watching uniformly built ballerinas perform

Tchaikovsky, or witnessing the stringent conformity of a military parade. The military invocation is apt, I think, because the scene of congregating elves at the North Pole’s town square recalls a “political rally,” according to Ebert (¶ 8). More pointedly, Dargis exclaims, “Tots surely won't recognize that Santa's big entrance in front of the throngs of frenzied elves and awe-struck children directly evokes, however unconsciously, one of

Hitler's Nuremberg rally entrances in Leni Riefenstahl's "Triumph of the Will" (¶ 2). The

North Pole town (or commune?) is likened to a “munitions factory… conceived along the same oppressive lines as Coketown” from Dickens’ Hard Times (Dargis, ¶ 2).

There is even a visitation to a sinister, total surveillance room where children around the world are monitored by elves on closed-circuit televisions. Apparently, there are elves busily at work compiling databases to determine whether the children should be judged naughty or nice. All the while, vintage Christmas music plays hollowly in the background of the vacuous room, much like in George Romero’s Dawn of the Dead where shopping music plays dully as zombies assail the door of an abandoned mall.

So, yes, there is reason to say the film is creepy from the onset. However, the most consistent and compelling reason for the creepiness still stems from the filmmakers’ use of motion capture. Admittedly one review, from Berardinelli at ReelViews, does not quite say as much. Rather, he writes,

The result [of mocap] allows the characters to appear much like real human beings while still retaining a slightly "animated" look. (The humans are intentionally not as real as those in Final Fantasy - a movie that gave some viewers the creeps). (¶ 6).

Indeed, a selection of Final Fantasy: The Spirits Within (2001) film reviews would make a good complementary object of analysis for this thesis. It was universally regarded as creepy because of its photorealistic human character portrayals. It is worth noting that, according to Beradinelli, Polar barely skirts the risk of creepiness that exists when a too real aesthetic is sought, as in Final Fantasy. He implicitly acknowledges a line between creepy and acceptable that seems to exist matter-of-factly. Ebert, as mentioned, thinks the film is creepy partly because of the motion capture. He says, “The characters in ‘The Polar Express’ don't look real, but they don't look unreal, either; they have a kind of simplified and underlined reality that makes them visually magnetic.” (¶

3). The magnetism he speaks of is at least recognition of the peculiar aesthetic of the mocap process.

The other critics reach a consensus about mocap creepiness. According to Peter

Travers:

Sadly, nothing in The Polar Express seems touched by human hands. The eyes of the characters, from the boy to Santa himself (also Hands), have a glazed look that is almost spooky in an Invasion of the Body Snatchers kind of

way. The result is a failed and lifeless experiment in which everything goes wrong.” (¶ 1)

Several of the critics identify the eyes as an area of particular concern. Tim Knight says,

“It’s uncanny how realistic the characters in The Polar Express appear, save for one area: the eyes” (¶ 4). Even though the production notes say there were sensors on the performers eyelids, they do not seem to translate to the characters convincingly. This is one way that the film critics, however cursorily, attempt to explain the phenomenon.

More to the point, Knight does not simply say that the characters are realistic except for the eyes – even the rest looks uncanny.

Adding to Travers’ Invasion of the Body Snatchers (1956) reference, others seem ready to put the film in the horror genre. Stella Papmichael writes, “Forget Christmas cheer – kids will run screaming from The Polar Express after seeing Tom Hanks looking like he’s been killed, embalmed and resurrected by lightening [sic]” (¶ 1). Stephanie

Zacharek of Salon provides the most cynically over-the-top, but still astute, review. She colourfully describes the motion capture process:

A special tool, kind of like a blackhead extractor, is used to remove the souls of real-life actors so their faces and bodies can be cloned and freeze-dried in a handy digital format. Then, a special reanimation process brings them to "life." Voilà -- the stuff nightmares are made of…(¶ 2)

She likens the child characters to the tots from Village of the Damned (1960) and the main character (Hero Boy) to the telekinetic twins in Brian De Palma’s The Fury (1978),

“the ones with such potent supernatural powers they can make people bleed just by looking at them – you expect his brain to explode any minute” (¶ 4). Continuing the nightmarish theme, the Time Out review echoes, “the ‘motion captured’ cast…look half

70 dead. Perhaps that’s what scared the wits out of several kiddies during the media screening” (¶ 1).

In a more moderate description, Ian Nathan say that mocap “grant[s] the characters genuine human movement and expression, albeit more doll-like than photorealistic” (¶ 3). The overwhelming point and consensus of the reviews is that there is something odd or peculiar, if not creepy, about the mocap process. It is perhaps best summed up by Dargis, who exclaims, “The largest intractable problem with ‘The Polar

Express’ is that the motion-capture technology used to create the human figures has resulted in a film filled with creepily unlifelike beings” (¶7).

I will be satisfied with this evidence to demonstrate that it is not my personal observation alone that there is something odd about the mocap process. Whether the oddness is a quality common to all digital film remains to be seen. I have posited, following Manovich, that digital films are best identified as those that combine live- action and animation and mocap (as well as its predecessor, rotoscoping) may be just a particular, extreme example of such a hybrid. But I suggest it is a good starting point and worthy of exploration.

Part 2: Medium Theory into Practice

Once again, a central tenet to this thesis is borrowed from Lev Manovich – that digital cinema is a hybrid of animation and live-action film. This definition has in turn been borrowed, and updated, from the post-photography argument claiming digital photos are a hybrid of painting and photography. My argument, simply put, is that the uncanniness of digital media is due, in part, to a clash of different thinking fostered by

71 these streams of established media. But this begs the question, what is the difference in the first place? How does animation really differ from film and, before that, how are paintings different from photographs?

In the literature review I already put forth the argument that they have different conventions – culturally determined styles and preferences that have developed over time as these unique forms of representation were established. This final chapter seeks to go a level deeper and understand these forms from a media theory perspective. By no means is this an effort to downplay the highly significant role of social stylization, but media theory, from the start, always looks to media essentialism. It posits that paintings and photographs are fundamentally different, not because of how they are understood culturally, but because they have different essential, medium-specific qualities.

The differences may appear obvious. Indeed, in each medium’s most common incarnations, the glossy surfaces of photographs are readily distinguished from the pigment-smeared canvases of gallery paintings. A child can tell them apart. Yet there are photos that look quite “painterly” and, of course, there are photorealistic paintings.

Gerhard Richter goes so far as to say, “I’m not trying to imitate a photograph; I’m trying to make one [using paint]” (cited in Costello 2007, p. 80). It turns out the differences may not be so obvious. Certainly, many of the differences are cultural, as Allan Sekula

(1982) has argued. However, I am proposing to temporarily sideline all of the cultural and stylistic differences to focus on the media themselves.

When McLuhan calls media the “extensions of man” (1964), he highlights the fact that media extend – or engage – the senses. This is why I have proposed to begin a

72 medium theory analysis by looking at how photos and paintings engage audience’s senses. McLuhan writes,

If a technology is introduced whether from within or from without a culture, and if it gives new stress or ascendancy to one or another of our senses, the ratio among all of our senses is altered. We no longer feel the same, nor do our eyes and ears and other senses remain the same. (1962, p. 25)

With only five primary senses, and a multiplicity of media, it would seem there must be considerable overlap or even redundancy in sensual engagement. The field now called

“visual studies”, for instance must posit that there are a good number of visual media, all engaged in similar ways. And, indeed, I would venture to claim that conventional wisdom regards photos and paintings very similarly; they are both, by most accounts, visual media.

It seems obvious that these media are visual because “viewers,” foremost, look at them. This notion is reified by W.J.T. Mitchell, who initially explains “I understand the use of this phrase [visual media] as a shorthand figure to pick out the difference between

(say) photographs and phonograph records, or paintings and novels” and then counters

“but I do object to the confident assertion that ‘the’ visual media are really a distinct class of things, or that there is such a thing as an exclusively, a purely visual medium” (p. 95).

He demonstrates that the designation of visual media appears obvious and pragmatic, but is ultimately misleading. Instead, he prefers to speak of mixed media that simultaneously engage different senses, just as McLuhan recognised sense ratios. His argument cautions against judging paintings and photographs on the same plane, as visual media.

I want to argue that photos are inherently more visual, whereas paintings are more tactile. The difference comes mainly from how each is made, which is to say (as I have hinted) that photographs are mostly automatically produced, whereas paintings are handmade. This occurs to the extent that the audience (or participant or viewer) recreates the process of creation when they engage the resultant artefact. Joyce Brodsky makes this same point, “that some of the complex actions entailed in an artist’s practices are somehow embodied in the work; that the work is the trace of those activities” (2002, p.

102).

In essence, a photograph is produced by a camera, which simulates a single perspective eye. When the usual rules of photography are followed in taking a picture, the resultant photo looks like what the photographer (or more precisely, the camera) saw.

And the viewer is able to simulate that process readily and see what the camera saw when looking at the print. To follow Brodsky’s point, “If that is the case, it is equally reasonable to think that in perceiving them some of those operations will be understood in an embodied manner, and some may even be performed during the interactivity (2002 p. 102). Thus, when a camera snaps a record of a landscape, the audience for that printed snapshot is similarly able to envisage the same landscape as a distanced observer. The camera simulates the eye and produces an (mostly) automatic record of what is seen.

By the same token, when beholding a painting (the word “behold” is apt, as behealden originally means “to hold”) the viewer does some of the “work” of the artist.

She can feel the slight ridges of the textured surface without actually touching it and even feel the gesture that made each stroke. Art historians often recount how photography emancipated painters from the task of visual mimesis to concentrate on the painting itself

74 as an object, or the artist’s experience of the thing or event painted. Cezanne, for instance, claimed to paint as though he held the objects of his still lifes (Gombrich, 2002).

When a painter paints, the audience is able, to some extent, to empathize with the subjective experience of the artist as she produced the work. Certainly, the experience of producer and audience are not identical and may hardly be isomorphic, but even if the audience draws explicitly on their own experiences in formulating their reaction, it is felt nonetheless. The painting process “records” the work of the artist and produces a handmade painting the viewer is able to feel and experience. Or, as W.J.T. Mitchell succinctly puts it, “seeing painting is seeing touching” (2007, p. 397).

These examples draw from (an admittedly oversimplified) context of “looking at art” but can be considered more broadly as engaging media. McLuhan, of course, would speak of engaging the printed word foremost, or ads, games, the telegraph, or other media. I am not attempting a complex taxonomy of sensory engagement, but rather a general division between the visual and the audile-tactile. The automatically produced photograph is a more visual media (largely because it is made by a simulated eye) and the handmade painting is more tactile because it is handmade. Of course, this division is not definite. As Mitchell (1992) says, the division is helpful “particularly if we do not insist on a clearcut dividing line between paintings and photographs but think rather of a spectrum running from nonalgorithmic [handmade] to algorithmic [automatic] conditions

– with ideal paintings at one end and ideal photographs at the other” (p. 30). The boundary is blurred, as a viewer is able to feel emotionally what the photographer felt in selecting and composing a certain scene and is able to look at a painting as an object. But

75 the distinction is there nonetheless. This incongruity is better dealt with in the next question.

How do media restructure consciousness? The phrasing of this question is borrowed from Walter Ong who, writing on the difference between orality and literacy, claims, “More than any other single invention, writing has transformed human consciousness” (1982, p. 76). It should be noted that I am not pursuing a psychological explanation, as the question may suggest. Neither Ong nor I mean to examine so strictly how different synapses may fire in different patterns in the brain, for example. Rather, I think the dichotomy Ong draws between orality and literacy can also divide painting and photography. Meyrowitz (1994) provides a concise summary of McLuhan and Ong’s most important division. “The oral world,” he says “is one of rich involvement with and interplay of all the senses of hearing, sight, smell, taste, and touch” (p. 54). Foremost, it is the world of the audile-tactile, experienced in a lived “hands on” way. Later, “print, even more than writing, undoes the tribal balance of the senses. The importance of the simultaneous aural surround yields to the dominance of the sequential sense of sight” (p.

56). After print, the literate culture yields to occularcentrism.

It is the occularcentric world that begets photography. When vision dominates, the left side of the brain takes centre stage. Privileging the observational and rational modes of thinking fuels the mimetic imperative and the trend towards naturalistic representation. The literate mind is able to take a step back, embrace Euclidean geometry, and perfect the techniques of perspective and foreshortening before integrating developments in chemical science that eventually culminate in the invention of the daguerreotype, photography, and film. Of course, the history is not nearly so simple as

76 that, but photography shares a lineage with print and science. Vilém Flusser makes the same argument, though he considers photos within a larger category of technical images.

The technical image is an image produced by apparatuses. As apparatuses themselves are the products of applied scientific texts, in the case of technical images one is dealing with the indirect products of scientific texts. This gives them, historically and ontologically, a position that is different from that of traditional images. (2000, p. 14)

For Flusser, technical images are fundamentally set apart from traditional images (such as paintings) because they are inextricably bound to science and the literate mind.

And photography even echoes back to its literate origins because it too emphasizes sight. It is like a direct line or shortcut to representing what the eye sees, neatly arranged on photo paper in the way the literate mind has come to see the Euclidean world. This is not to equate photos with the mythic innocent eye, but to suggest they belong to the same way of thinking as rational inquiry, nurtured by the invention of print, and played out in the Enlightenment.

Standing in opposition, painting is connected to orality and the current electric age (Ong’s secondary orality). As Meyrowitz summarizes, “electronic media bring back a key aspect of oral societies: simultaneity of action, perception, and reaction. Sensory experience again becomes a prime form of communicating” (1994, p. 57). The differences between the electric age and traditional oral cultures (mainly to do with time and space constraints, according to Meyrowitz) are not particularly relevant. What is relevant is that paintings are connected to “sensory experience” as Meyrowitz uses the term – to the interplay of the senses. Being handmade, painting is more closely related to experience and empathy. The experience of the artist becomes, to some extent, encoded

77 in the work and the audience, to some extent, is able to empathize with it. While the cool-minded scientist stresses objective distance and cautions against the pitfalls of involvement and empathy, the painter embraces these qualities and must work hard if she wishes to suppress the personal in her handmade work.

Discussing the differences between painting and photography, William J. Mitchell frames this dichotomy in terms of “intention” and “objectivity.” He uses freehand sketching as an example of intentionality, “every freely made mark that the artist chooses to execute is the realization of an intention, and the result is usually something that has a strongly personal character” (1992, p. 30). An individual’s signature is personal enough to be legally binding – as a guarantee that the author “was” there. The personal character of mark making stands opposite to the impersonal objectivity of mechanically-produced photographs.

Instead of looking at precisely how each medium – photography and painting – restructures consciousness, I have categorized them according to the broad dualism already posed by medium theory. Roughly they correspond to the left and right brain, or to a literate, rational culture opposite an oral, experiential one. The dichotomy is permeable, but useful nonetheless. In his introduction to McLuhan’s Understanding

Media, Lewis Lapham (1994, p. xii) tries to summarize the great dualism by charting some of the primary dichotomies. A selection of them follow:

Print Electronic Media

Visual Tactile

Mechanical Organic

Sequence Simultaneity

Eye Ear

Classification Pattern Recognition

Typographic man Graphic man

Photography / live-action film Painting / animation

Lapham identifies qualities related to the unique forms of consciousness McLuhan postulated rather than listing specific media that may belong on either side (other than print and electronic media). However, it should be possible to include specific media in his schema and so I have added the lower italicized categories to illustrate the differences

I have drawn. So far, I have focused on photography and painting, but I hope it is obvious enough how film and animation correspond. Live-action film is the mechanical or automatic generation of sequential photographs and animation is a sequence of handmade images. More, as I have defined animation in the literature review, the interstices between those images is also determined by an artist or animator; the images are hand-moved. Just as sketching has a deeply personal character, so too does animation. There is a discernible “mark” of the artist deciding how one image flows into another. It is this immaterial work that McLaren perceives between frames and regards as the essence of animation and mark of the animator as artist. Finally, it can be shown that animation and film belong to the two categories of orality and literacy, respectively, as articulated in medium theory.

It remains to ask how each medium affects culture at large. If following the territory already mapped by medium theorists, a possible answer could relate to social organization, or what Meyrowitz calls the “role triad” (1994). For instance, he speaks of how information access patterns, enabled by different media, constructs differences in group identity, socialization, and hierarchy. However, this would do little to shed light on the uncanny in digital film. Instead, it will be instructive to follow the trajectory set by William J. Mitchell and proponents of the post-photography argument, who invariably tie their discussion to the veracity of images. That is, I wish to extend my argument into the realm of truth claims. Mitchell and others claim that digitization “dramatically changes the rules of the game” (p. 31) and enables a shift from objectively true photographs to unverifiable personal statements that “no longer have the power to convince us [as truth]” (p. 31). The shift is framed as a loss of truth – one that could once be held on a glossy 8x10, supposedly. After digitization, truth concedes to artifice.

Such a shift in truth-value undoubtedly signals a sea change in culture at large.

And this could be an adequate answer to the third medium theory question I posed in the methodology: how do media affect cultural trends more broadly?

However, I want to argue that each stream of the dualism detailed above actually produces its own “truth” and it is a bias that Mitchell and others should privilege one over another. The first is an objective truth and the second an experiential truth. For some time now, arguments for truth in photography have invariably relied on the notion of indexicality. Photographs are true, the argument goes, because they are traces of the objects photographed. The same is said of film. For example, Manovich says that digital cinema makes film a kind of animation, and with that, “It is no longer an indexical media

80 technology, but, rather, a subgenre of painting” (p. 406). In this and similar arguments, of course, it is implied that indexicality is a guarantor of truth and that handmade media

(i.e. painting) are not indexical. Indexicality is the golden link or linchpin, which is broken after digitization and marks a change from veracious images to suspect ones.

This begs the question, what is an index?

This may appear as a departure from medium theory, but if the index is what separates paintings from photos, then it should also be related to the boundary cleaving

Lapham’s two columns above. I do not wish to get too far into what Peirce may have meant by “index” in his taxonomy of signs, as that could demand another thesis entirely.

Indeed, James Elkins’ edited book Photography Theory (2007) transcribes a lengthy roundtable discussion among several scholars debating this very thing. I do, however, think I can put forth a straightforward definition and clear up some confusion surrounding the debate of what exactly constitutes an index, and how it might guarantee truth in photography. W.J.T. Mitchell makes a significant observation to this end. He writes, “Every icon or image takes on… an indexical component the moment we ask how it was made” (2007, p. 400). At their broadest then, indexical signs reveal a causal process; they point to how they came to be. Making sense of indices inevitably involves some detective work. If the body is still warm, the victim died recently; if there is a photo of a bridge, at some point, the camera lens was focused on that bridge; if there is a painting, someone painted it. It seems that in the context of photography theory, the index is often invoked as a magical explanation or way of separating photos from traditional images. This, I believe, can be very misleading as paintings are no less indexical for the way the visible brushstrokes reveal the patterns of the working artist and

81 index her intention. These marks can even be an important way of detecting forgeries.

Paintings are indices too. But, of course, the photography theorists have something more in mind when they label photos “indexical.”

To my understanding, they refer to two particular things that are often confused, conflated, or at least are not regarded consistently. And both are often put forward in the argument for photographic truth. The first is the most important and undoubtedly separates photos from paintings. Simply put, in order to photograph an object, the object must exist out there in the world in the first place. William J. Mitchell paraphrases philosopher Roger Scruton’s comments on the inability to photograph an angel, “the non- existence of angels need not prevent you from painting a picture of one, but it certainly prevents you from taking a photograph of one” (1994, p. 29). Roland Barthes is of the same opinion, “the thing was” (1981). Rightly, calling a photo indexical is really acknowledging it as process – but a specific kind of process that requires a real-world object and produces a likeness of the object.

But often proponents of photographic truth invoke another aspect of the photographic process: its objectivity. As Edgar Allen Poe has said of photography

“…the closest scrutiny of the photogenic drawing discloses only a more absolute truth, a more perfect identity of aspect with the thing represented. The variations of shade, and the gradations of both linear and aerial perspective, are those of truth itself in the supremeness of its perfection” (in Mitchell, 1992, p. 4.). But such detail is only achieved by a process where, according to Bazin, “between the originating object and its reproduction there intervenes only the instrumentality of a nonliving agent. For the first time an image of the world is formed automatically, without the creative intervention of

82 man” (1967, p. 13). The fine, exacting detail unmarred by human interpretation, not the fact the “this was,” is praised as virtue and guarantor of truth. Note that in this account it is not verisimilitude which makes photos true. Yes, photos often look very much like their subjects – even to the point where image and referent are conflated. However, it is the objectivity essential to producing such a likeness that is important, and what makes photos “true” by this account. The machine-eye is privileged over the selective, fallible hand of the artist.

Both accounts are often at work. In Antonioni’s Blowup (1966), for example, a murder is unwittingly captured on film, which becomes important evidence for two reasons. First, it guarantees the gunman was there, and secondly, the details of the murder can be ascertained after the fact by enlarging the photo even though the photographer was oblivious to the incident at the time. Both accounts are very good reasons to regard the photo as “true” evidence. The first, that “this was” is incontrovertible, I think. The second, that photos are true because they are objective and indiscriminately record all details, is more contentious. Why? Because it does not rightly reveal the experience of the photographer or the audience.

The objectivity of photographs is highly valuable in a lot of situations and contexts, especially scientific or legal ones. But it can be a limitation or liability in communicating artistic accounts. E.H. Gombrich (1960) identifies the limitations of photography in capturing experience, saying if they did it well,

[E]very [portrait] snapshot would have a greater chance of impressing us as a satisfactory representation of a person we know. In fact only a few snapshots will so satisfy us. We dismiss the majority as odd, uncharacteristic, strange, not because the camera distorts, but because it caught a

constellation of features from the melody of expression which, when arrested and frozen, fails to strike us in the same way the sitter does. (p. 292)

Interestingly, he feels that part of what is missing is the motion of the sitter that a portrait artist, such as Rembrandt, is able to compensate for. But he need not go so far. Often a quick gesture drawing is able to capture the mood or quality of a model’s position better than a snapshot. Or at least, differently, than a snapshot. Ultimately, paintings and handmade works are better positioned to communicate an experiential truth; better at capturing an experience had by an artist and conveying an experience that an audience is able to identify with, or project onto. At base, my argument is that photographs are more objective because they are less expressionistic, and paintings are more expressionistic because they cannot be objective. Each privileges and communicates a different kind of truth.

To return to medium theory, I suggest that the truth attributed to photographs is due in large part to their perceived objectivity, which in turn connects them to the same lineage as literate, scientific thought. As Flusser says, photos are engendered by scientific texts. It is little wonder that the introduction of the human mark, made easy by digital technologies, should be perceived as a compromise to objective integrity.

However to regard this shift as a loss of truth highlights the bias to rational thought pervasive since Gutenberg. The intentionality of paintings makes them categorically different but, as thinkers like Gombrich show, they are capable of relating an experiential truth. It is that experience which we can readily relate to because we have “been there.”

Or, as Meyrowitz (1994) explains, “While written and printed words emphasize ideas, most electronic media emphasize feeling, appearance, mood.” Adding, “The major

84 questions [in the electronic age] are no longer ‘Is it true?’ ‘Is it false?’ Instead we more often ask, ‘How does it look?’ ‘How does it feel?’” (p. 58). It is handmade media that are best at expressing just how something “feels” and communicating an experience.

I trust it is not too much of a leap to extend the arguments made from the static media of photos and paintings to the motion pictures of live-action and animation. The chief division I have drawn is between automatism and the handmade, and it holds up whether the pictures are moving or not.

To summarize, different media engage the senses differently. This engagement assumes a necessary, but not straightforward, link between the producer and audience.

Photos and film are foremost visual media because they are made by a mostly automatic process (the camera, which simulates the eye). Paintings and animation are mostly made by hand and are chiefly audile-tactile. Photos can be said to restructure consciousness by protracting occularcentrism and continuing the trend of linear, objective, rational thinking. Paintings, being audile-tactile, invite empathy and privilege experience. Each way of thinking fosters a different kind of truth. Photos offer an “objective truth”, revealing the world “as it is” by minimizing human bias. Paintings are defined by the human mark and are better positioned to give an “experiential truth” capable, at their best, of relating and understanding the human experience. These two trajectories have been traced back starting from the bare essentials offered by Manovich’s formula for digital cinema, as a hybrid of live-action footage and animation. Plainly illustrated, the trajectories may look something like the following:

Figure 3.1

Eye  Literacy  Objectivity  Photography  Live-action Digital Cinema Ear  Orality  Experience  Painting  Animation

I want to argue that the collision of the two modes of thinking may result in the uncanny experience of digital film. In The Polar Express, the objective capture of an actor’s movement is transcribed overtop of characters that are predominantly handmade, albeit in a computer. It may be that the audience is attune to the objectively true recording and has a difficult time reconciling it with handmade elements. The simultaneity of the objective and subjective makes it difficult to either watch as a disengaged voyeur, or as an empathetic participant. The mixed signals are not just the socially constructed, stylistic conventions between live-action film and cartoons, but the qualitatively different media of the mechanical and handmade. This difficulty in reconciling contrary media, not just contrary messages, helps to explain the uncanny experience.

Of course, I am not suggesting that the paths cross for the first time with the development of digital cinema. They are both always active and cross and intertwine in everyday life. But they may be operating simultaneously and in deeply enmeshed ways in digital film. At the highest level, where the different media of live-action and animation, and their profoundly divergent histories, give rise to different “truths,” it is understandable for one to call the experience uncanny. Holding two truths in the mind, or experiencing the world in two distinct ways – as passive observer and active participant – seem like the perfect conditions for witnessing what Manovich described as

86 that “which is intended to look exactly as if it could have happened, although it really could not.”

CONCLUSION: REFORMULATING DIGITAL FILM

To revisit my original research question, some conclusions can now be drawn. I have asked: How can an examination of rotoscoping in The Polar Express, through the lens of medium theory, inform an understanding of digital film? Most of my analysis has focused on rotoscoping and its successor motion-capture with the hope that it may shed light on digital cinema at large. By reinterpreting Manovich’s formula from the shaky makeup of several disparate elements,

Digital cinema = live action material + painting + image processing +

compositing + 2D computer animation + 3D computer animation

To the more concise and manageable:

Digital cinema = animation + live-action

I have made the connection between rotoscoping and digital film more apparent. Both are foremost hybrids of animation and live-action film. In this respect, there is good reason to think they may have some common characteristics, not the least of which is that they may appear uncanny or eerie in some way. This is admittedly a reductive approach and it must be acknowledged that rotoscoping is a particular way of integrating animated and live-action techniques. Just because rotoscoped results are often deemed uncanny does not mean all digital film will be. However, I suspect that some digital effects laden movies, like the Star Wars prequels, may appear strange or uncanny at times to audiences. D.N. Rodowick (2007), for example, detects the uncanny in digital cinema, generally. He writes,

[W]e are confronted with something new in the image, something that disturbs the perceptual defaults of the chemically based analogical image…A subtle shifting of gears is taking place in our current ontology, in our relation to the world and to others, as mediated through technologically produced images. What we find to be uncanny and unsettling, I would suggest, is the spatial similarity of digital images to the now antecedent practices of photography and film (my emphasis, p. 98).

For Rodowick, it seems all digital film is uncanny. His thesis is that the ontology of film has changed and the uncanny is produced because we still try to approach it using the same, established film theory of analogue media. We do this because digital films often look an awful lot like analogue films. That may be a helpful explanation, and I am not trying to supplant it or any of the other theories of the rotoscoping phenomenon, but rather seek to augment them with my media theory account. The overriding point is that rotoscoping and digital film are related due to their most basic makeup, and resulting uncanniness. So, I will hazard to say that, at least to the extent that the strangeness of rotoscoping is explained as co-present objective and subjective experience, the same might be said of digital film at large. There are two traditions relating to the oral and literal at work in digital images.

If I have been reductive in reformulating Manovich’s definition, I hope I have made enough headway to rebuild it, using the insights outlined in the analysis. My simplified version defines digital cinema as a hybrid of animation and live-action, and this is essentially an update of the definition of digital photography – as a hybrid of photography and painting. The most crucial dividing line, I have tried to show, is between automatism and handmade media. But there is another quality of film, perhaps so obvious, that it is often overlooked or goes unsaid. Film is a series of consecutive

89 images shown in rapid succession. Traditionally, it is thought or assumed that those images are photographs in the case of (live-action) film, and drawings in the case of animation. But that need not necessarily be the case. Animation as “drawings that move” is widely accepted and with good reason. Animators have traditionally spent countless hours at a drafting or light table producing drawings. Recall even Norman

McLaren’s claim that “Animation is not the art of drawings that move, but the art of movements that are drawn”, which still assumes the process involves drawing. But I have taken pains to denounce that definition; animation is not about drawing, but about the hand-moved. Once drawing is excised as an essential part of animation, a useful sketch of digital cinema can be made. I propose the following, revised formula:

Digital cinema = painting + photography + live-action + animation

Techniques of combining those components, such as compositing and rotoscoping, I deem to be categorically different from the parts themselves and are not included. Of course, they are still integral to digital filmmaking, but only need be implicit in the formula in the interests of proper taxonomy and brevity. Again, if one wants to include things like image-processing, they may as well include camera operation or other processes that go into filmmaking as a broader process. I hope that the above formula is more straightforward than Manovich’s but it can be improved upon further. It can be seen that the four elements might be divided along two axes. If the axes are not easily guessed, the first divides the static (still) images, paintings and photos, from the kinetic

(moving) ones, live-action and animation. The second divides automated from handmade production. Taking those categories into account, I propose the following matrix as a way to approach digital cinema:

Figure 3.2 Static Kinetic

Handmade Painting Animation

Automated Photography Live-Action

This is my condensed answer to the broad question, what is digital cinema? My intention in presenting the schema this way is to highlight the important distinction between automated and handmade production and imply the histories I have traced above in my analysis. That is, it may be helpful to remember that the handmade remains connected to

Ong’s conception of orality, while the automated is tied to literacy. Paintings and animation are associated with the audile-tactile, intention, the organic, experiential truth, etc. Photography and live-action, on the other hand, are associated with the visual, causality, the mechanical, objective truth, etc.

With digitization, the four categories are blurred more readily than ever. They can be combined in several ways. Some examples of possible combinations:

• Painting + Animation = most cartoons, Disney films, The Simpsons, etc.

• Photography + Live-Action = most traditional film, Citizen Kane, etc.

• Photography + Animation = Monty Python cut-out animation, Norman McLaren’s

Neighbours, etc.

• Painting + Live-Action = “traditional” rotoscoping, several Betty Boop cartoons,

Waking Life, etc.

Paintings combined with photography leads to the definition given to digital photography. Note that this scheme does not allow for an exclusive combination of animation and live-action, as that hypothetical combination would have no material, image base. But, of course, there can be three or four elements combined and hybrids of hybrids are possible. A film like The Polar Express is likely a combination of all four categories. In that movie, the base for many of the textures used on the models is photographic and then digitally tweaked for the desired aesthetic. Tom Hanks’ face is photographed and then retouched before the texture is mapped around the computer model. And even the model of the conductor character is likely designed and extrapolated from photos of the flesh-and-blood Hanks’ physiognomy. In this respect, the textures are best characterised as digital photographs. Likewise, the actors’ performances are motion-captured (moved via live-action), but these movements are inevitably tweaked by animators to meet the desired aesthetic of the filmmakers. So, live-action and animation are utilized as well. The Polar Express is a hybrid digital film par excellence, because it is comprised of paintings, photographs, live-action, and animation.

Medium theory has proved useful in making sense of digital hybrids. It ultimately demonstrates that there are two ways of experiencing the world – as observers and participants. Likewise, there are corollary modes of representing the world, as when making images. Images can be objectively informed by limiting the bias of human interpretation; or, they can embrace subjectivity and reflect the experience of the artist.

Our primary way of accessing the world is through our senses and McLuhan reminds us that different media favour different senses, which ultimately privilege one of the two modes of engaging the world. It is hoped that discussions of digital media will benefit from bringing this account to bear. When thinkers like W.J. Mitchell and Lev Manovich identify digital media as combinations of automated and handmade processes, it should help to keep those deeper histories in mind.

Finally, I can attempt an answer to my original question, How can an examination of rotoscoping in The Polar Express, through the lens of medium theory, inform an understanding of digital film? The uncanny phenomenon of rotoscoping has provided an access point to digital film at large, by offering a precedent to motion pictures that hybridize automated and handmade components. Medium theory offers a useful reminder of some of the crucial differences between automated and handmade media.

These media are frequently combined, but in parsing them out into two streams of orality and literacy, it is easier to make sense of their hybrids in the electric, and now digital, eras. It also serves to question the truth claims of so-called indexical media, where automatically produced images are considered true for eliminating human bias (being objective), but do not consider the experiential truth of the human experience. Building from these media theory insights, Manovich’s original formula can be revised and reconstructed. A brief re-examination of cinema as a succession of static images leads to a useful schema with which to categorize and consider digital films. This schema can now be used as a starting point to answer the elusive question, what is digital cinema? It is a combination of automated and handmade media, both static and kinetic, put together in new and interesting ways that will continue to fascinate audiences and scholars alike.

REFERENCES

Aida, J. (Producer) & Sakaguchi, H. (Director). (2001). Final fantasy: The spirits within [Motion Picture]. USA / Japan: Square Pictures.

Allsburg C. V. (1985). The polar express. Boston: Houghton Mifflin.

Amelunxen H. V., Igluhaut S., & Rötzer F.(Eds). (1996). Photography after photography: Memory and representation in the digital age. Amsterdam: OPA.

Arnzen, M. (1997). The return of the uncanny. Paradoxa, 3(3-4), 315-320.

Avery, R., Gaiman, N. (Producers) & Zemeckis, R. (Director). (2007). Beowulf [Motion Picture]. USA: ImageMovers.

Barthes, R. (1981). Camera lucida: Reflections on photography (R. Howard, Trans.). New York: Hill and Wang.

Baurdrillard, J. (1994). Simulacra and simulation (S. F. Glaser, Trans.). Ann Arbor: University of Michigan Press.

Bazin, A. (1967). What is cinema?. Berkeley: University of California Press.

Becker, H. (1982). Art worlds. Berkeley: University of California Press.

Benjamin, W. (1999). Illuminations (H. Zorn, Trans.). London: Pimlico.

Berardinelli, J. (2004). The polar express [Review of the film The polar express]. Retrieved September 13, 2007, from http://www.reelviews.net/movies/p/polar_express.html

Blenkin, J., Cunningham, C. (Producers) & Jackson, P. (Director). (2005). King Kong [Motion Picture]. USA: Big Primate Pictures.

Bode, L. (2006). From shadow citizens to Teflon stars: Reception of the transfiguring effects of new moving image technologies. Animation: An Interdisciplinary Journal, 1(2), 173-189.

Bouldin, J. R. (2004). The animated and the actual: Toward a theory of animation, live-action, and everyday life. Retrieved from Dissertations and Theses database. (AAT 3132381)

Bouldin, J. R. (2004). Cadaver of the real: Animation, rotoscoping and the politics of the body. Animation Journal, 12, 7-31.

Beowulf raises question: what is animation?. (2007). The Associated Press, retrieved January 16, 2008 from http://www.msnbc.msn.com/id/21825491/

Brodsky, J. (2002). How to “see” with the whole body. Visual Studies, 17, 99-112.

Bordwell, D. (1997). On the history of film style. Cambridge: Harvard University Press.

Breznican, A. (2007, November 12). Old Beowulf new tricks: Cutting-edge film could herald new digital era or end up history. Calgary Herald, p. C1.

Cholodenko, A. (Ed.). (1991). The illusion of life: Essays on animation. Sydney: Power Publications.

Clooney, G. (Producer) & Linklater, R. (Director). (2006). A scanner darkly [Motion Picture]. USA: Warner Independent Pictures.

Costello, D. (2007). After medium specificity Chez Fried: Jeff Wall as a painter; Gerhard Richter as a photographer. In J. Elkins (Ed.), Photography theory (pp. 75-89). New York: Routledge.

Coy, W. (1996). In photographic memory. In H.V. Amelunxen, S., Igluhaut, & F. Rötzer (Eds), Photography after photography: memory and representation in the digital age. (pp. 66-72). Amsterdam: OPA.

Disney, W., Sutherland, J. (Producers) & Gillet, B. (Director). (1934). Playful pluto [Motion Picture]. USA: Walt Disney Productions.

D.A. (2004). The polar express [Review of the film The polar express]. Retrieved September 13, 2007, from http://www.timeout.com/film/reviews/81393/the_polar_express.html

Dargis, M. (2004). Do you hear sleigh bells? Nah, just Tom Hanks and some train [Review of the film The polar express]. Retrieved September 13, 2007, from http://movies.nytimes.com/2004/11/10/movies/10pola.html

De Luca, M., Polstein, C. R., (Producers) & Levinson, B. (Director). (1997). Wag the dog [Motion Picture] USA: New Line Cinema.

Denise, D., Hanks, T. (Producers) & Zemeckis, R. (Director). (2004). The polar express [Motion Picture]. USA: Castle Rock Entertainment.

Ebert, R. (2004). The polar express [Review of the film The polar express]. Retrieved September 13, 2007, from http://rogerebert.suntimes.com/apps/pbcs.dll/article?AID=/20041109/REVIEWS/41006005/1 023

Elkins, J. (Ed.). (2007). Photography theory. New York: Routledge.

Ferber, D. (2003, September). The man who mistook his girlfriend for a robot. Popular Science, 263(3), 60.

Finerman, W., Starky, S. (Producers) & Zemeckis, R. (Director). (1994). Forest gump [Motion Picture]. USA: Paramount Pictures.

Flusser, V. (2000). Towards a philosophy of photography (A. Mathews, Trans.). London: Reaktion Books Ltd. (Original work published 1983)

Freud, S. (1919) The uncanny. Retrieved September 12, 2007 from http://www- rohan.sdsu.edu/~amtower/uncanny.html

Gibeaut, J. (2000). Image is everything. American Bar Association Journal, May.

Gombrich, E. H. (2002). Art & illusion: A study in the psychology of pictorial representation (6th ed.). New York: Phaidon.

Gow, G. A. (1996) Digitizing McLuhan: Transformation theory and communication in convergence. Retrieved March 8, 2006 from http://www.sfu.ca/~gagow/capcom/mcluhan.html

Gustafson, R. (2007). Beowulf [Review of the film Beowulf]. Retrieved December 15, 2007, from http://www.parentpreviews.com/movie-reviews/beowulf.shtml

Kanfer, S. (1997). Serious business: The art and commerce of animation in America from betty boop to “toy story”. New York: Da Capo Press.

Kaplan, C. (Producer) & Linklater, R. (Director). (2001). Waking life [Motion Picture]. USA: Fox Searchlight Pictures.

Kennedy, K., Marshall, F. (Producers) & Zemeckis, R. (Director). (1985). Back to the future [Motion Picture]. USA: Universal Pictures.

Knight, T. (2004). The polar express [Review of the film The polar express]. Retrieved September 13, 2007, from http://www.reel.com/movie.asp?MID=131806&Tab=reviews&buy=open&CID=13#tabs

Langer, M. (2002). The end of animation history. Retrieved February 7, 2006 from http://asifa.net/SAS/articles/langer1.htm

Langer, M. (2004). From the guest editor. Animation Journal, 12, 3-6.

Levinson, B. (Director). (1997). Wag the dog [Film].

Lyman, R. (2001, July 8). Movie stars fear inroads by upstart digital actors. The New York Times. Retrieved August 17, 2007, from http://query.nytimes.com/gst/fullpage.html?res=9F07EFDE1338F93BA35754C0A9679C8B6 3&sec=&spon=&pagewanted=3

Mancuso, F. (Producer) & Bakshi, R. (Director). (1992). Cool world [Motion Picture]. USA: Paramount Pictures.

Manovich, L. (2002). What is digital cinema? In N. Mirzoeff (Ed.), The visual culture reader (2nd ed.) (pp. 405-416 ). New York: Routledge.

Manovich, L. (2001). The language of new media. Cambridge: MIT Press.

McLuhan, H. M. (1969, March). Marshall McLuhan: A candid conversation with the high priest of popcult and metaphysician of the media [Interviewed by Eric Norden]. Playboy, 26-27, 45, 55-6, 61, 63.

McLuhan, H. M. (1962). The gutenberg galaxy. Toronto: University of Toronto Press.

McLuhan, H. M. (1994). Understanding media: The extensions of man. Cambridge: MIT Press.

McMahon, K. (Director). (2002). McLuhan’s wake [Motion Picture]. Canada: National Film Board.

Meyrowitz, J. (1994). Medium theory. In D. Crowely, & D. Mitchell (Eds.). Communication theory today. (pp. 50-77). Cambridge: Polity Press.

Miller, G. (Director). (1970). The eye hears, the ear sees [Motion Picture]. Canada: National Film Board.

Mirzoeff, N. (Ed.). (2002). The visual culture reader (2nd ed.). New York: Routledge.

Mitchell, W. J. (1992). The reconfigured eye: Visual truth in the post-photographic era. Cambridge: MIT Press.

Mitchell, W. J. T. (2004). Medium theory: preface to the 2003 critical inquiry symposium. Critical Inquiry. 30 (2).

Mitchell, W. J. T. (2007) There are no visual media. In O. Grau (Ed.), MediaArtHistories (pp.395-406). Cambridge, MA: MIT Press.

Monaco, J. (2000). How to read a film: The world of movies, media, and multimedia (3rd ed.). New York: Oxford University Press.

Mori, M. (1970). The uncanny valley (K.F. MacDorman & T. Minato, Trans.). Energy, 7(4) 33- 35.

Mulder, A. (2004). Understanding media theory: Language, image, sound, behavior. (L. Martz, Trans.) Rotterdam: V2_NAi.

Nathan, I. (2004). The polar express [Review of the film The polar express]. Retrieved September 13, 2007, from http://www.empireonline.com/reviews/ReviewComplete.asp?FID=10037

Negroponte, N. (1995). Being digital. New York: Knopf.

Ong, W. J. (1988). Orality and literacy: The technologizing of the word. New York: Routledge.

Pallasma, J. (2005). The eyes of the skin: Architecture and the senses. Hoboken, NJ: John Wiley & Sons.

Papamichael, S. (2004). The polar express [Review of the film The polar express]. Retrieved September 12, 2007, from http://www.bbc.co.uk/films/2004/11/30/the_polar_express_2004_review.shtml

Pilling, J. (Ed.). (1997). A reader in animation studies. Sydney: John Libbey & Co.

Portman, J. (2007). Angelina Jolie feels exposed in beowulf [Review of the film Beowulf]. Retrieved September 13, 2007, from http://canada.com.dose.ca/topics/entertainment/story.html?id=07e3be86-f975-4c95-a587- af85c68195a0

Postman, N. (1982). The disappearance of childhood. Toronto: Random House.

Reinke, S. & Gehman, C. (Eds.). (2005). The sharpest point: Animation at the end of cinema. Toronto: YYZ Books.

Rodowick, D.N. (2007). The virtual life of film. Cambridge, MA: Harvard University Press.

Rosler, M. (1996). Image simulations, computer manipulations. In H.V. Amelunxen, S. Igluhaut, & F. Rötzer (Eds.), Photography after photography: memory and representation in the digital age. (pp. 36-56). Amsterdam: OPA.

Rouve, P. (Producer) & Antonioni, M. (Director). (1966). Blowup [Motion Picture]. UK: Bridge Films.

Rubinstein, R. (Producer) & Romero, G. A. (Director). (1978). Dawn of the dead [Motion Picture]. USA: Laurel Group.

Sekula, A. (1982). On the invention of photographic meaning. In V. Burgin (Ed.). Thinking photography. Basingstoke, England: Macmillan.

Savedoff, B. E. (2000). Transforming media: painting, photography, and digital imagery. In Transforming images: how photography complicates the picture (chap. 5). Ithaca, NY: Cornell University Press.

Sontag, S. (1973). On Photography. Toronto: McGraw-Hill Ryerson Ltd.

Sobchak, V. (2004). What my fingers knew: The cinesthetic subject, or vision in the flesh. In Carnal thoughts: Embodiment and moving image culture (chap. 3). Berkeley: University of California Press.

Theall, D. (1995). The micro as the medium and the message: Synaesthesia, the harmonization of the senses, and the mechanics of art, Beyond the word (pp. 21-108). Toronto: University of Toronto Press.

Thomas, F. & Johnston, O. (1981). The illusion of life: Disney animation. New York: Abbeville Press.

Travers, P. (2004). Polar express [Review of the film The polar express]. Retrieved September 13, 2007, from http://www.rollingstone.com/reviews/movie/6165819/review/6636947/polar_express

Ward, P. (1998). Understanding animation. New York: Routledge.

Warner Bros. Pictures. (2004) The polar express: Production notes. Retrieved July 7, 2007, from http://polarexpressmovie.warnerbros.com/movie_prodnotes.html

Williamson, K. (2004) [Review of the film The polar express]. Retrieved September 13, 2007, from http://boxoffice.com/moviedetail.aspx?movieid=3808&flag=New

Williams, R. (1967). Paradoxically, if the book works it to some extent annihilates itself. In Gerald E. Stearn (Ed.), McLuhan hot and cool (pp. 188-191). New York: Dial Press.

Ufland, H. J. (Producer) & Scorsese, M. (Director). (1988). The last temptation of Christ [Motion picture]. USA: Universal Pictures.

Zacharek, S. (2004). The polar express [Review of the film The polar express]. Retrieved September 13, 2007, from http://dir.salon.com/story/ent/movies/review/2004/11/10/polar_express/index.html

Zaentz, S. (Producer) & Bakshi, R. (Director). (1978). The lord of the rings [Motion Picture]. USA: Fantasy Films.