Computer Generated Animation as Product Design Engineered Culture,or Buzz Lightyear to the Sales Floor,to the Checkout and Beyond!

Automated perspective and the aesthetic of continuity

It has been suggested that the first 150 years of photographic history will be considered, in hindsight, a brief indexical irregularity in the millennial history of the image (Manovich, 2001: 200). With this in mind, it is worth noting that Comolli’s analogy between Renaissance constructions of the subject and 20th-century film viewers took place from within this indexical anomaly. From this position, Comolli elided an ontological distinction between the two forms of perspective in question: a seemingly subtle difference which gains greater significance reappraised in the light of contemporary CG imaging. While Renaissance perspective was the product of discovering mathematically incontestable laws of physics, its application was nevertheless carried out by human hand. By contrast, 20th-century film deployed perspective as a result of a mechanically reproduced recording process. Aware of this distinction, Ivins (1975) had already been careful to identify two strands through which forms of perspective should be understood to have developed. On the one hand, perspective was developed in the techniques of descriptive and prescriptive geometry articulated by the draftsperson. On the other hand, technologies of the photograph automated the process through which perspectival images were produced. Thus, the automated technologies of representational perspective and the manual techniques of descriptive perspective both coexisted in separate domains: the automated process of technical reproduction central to photographic perspective could not be utilized in the production of descriptive techniques deployed by the draftsperson when imagining non-indexical space. Until, of course, the development of the computer:

By automating perspectival imaging digital computers completed the process begun in the Renaissance. The automation became possible because perspectival drawing has always been a step by step procedure, an algorithm involving a series of steps required to project coordinates of points in 3-D space onto a plane. Before computers the steps of the algorithm were executed by human draftsmen and artists. The use of a computer allowed [sic] to execute them automatically and therefore, much more efficiently. (Manovich, 2003)

Thus, what had previously been two entirely separate strands of perspectival imaging finally became interrelatable with the automation that the digital computer ushered in. The rise of CG imaging marks a major development, not only in our capacity to fabricate a photorealistic image but also in our capacity to fabricate Renaissance perspective on an automated, industrial level. Thus the CG image breaks from the filmic image in a fundamental way. Like the radar technologies, it was initially derived from (Manovich, 1993), the CG image marks a shift to a scanned electronic image that abstracts the relationship between space and the image that represents such space. This differs from the indexical, ‘single referent imprint’ (Spielmann, 1999) technology of film which chemically records the actual space in front of the camera.

Paradoxically, this shift to an automated visual nominalism also results in a seemingly contradictory tendency towards an apparently more ‘film like’ animated image form. As Alice Crawford (2000: 113) has argued, the ‘introduction of the third dimension, or “z” axis, to animation makes possible, among other things, the introduction of highly filmic visual techniques that are too labour-intensive in analogue form.’ In this sense, the CGI image is both different from and yet seemingly more like the filmic image, depending upon the direction from which we are approaching the changes that the computer has ushered in. We can say that CGI features open up a nominalist space facilitated by the automated perspective of the computer algorithm. This is a space qualitatively different from the filmed image-making that came before it, a space that does not need to pay heed to the ‘camera’ in any physical sense, allowing the virtual camera to travel anywhere, at any time and on any scale. Equally, however, this is a space that is also qualitatively different from the traditional animated image-making that came before it. Not only is it three dimensional, it is filled with a level of detail common to film but impossible in animation constructed by hand.

Thus CGI ushers in a dimensional consistency foreign to traditional live-action film and cel animation. In CGI features, what is new is what I shall call an aesthetic of continuity. Manovich (1993) has described a ‘continuity of aesthetic’ in imaging where the aesthetics of differing audiovisual texts are increasingly blended by the digital composite. In addition to this ‘continuity of aesthetic’, we can add an aesthetic of continuity: the condition in digital imaging whereby the physics of space, time, gravity, force, movement and more are held in a new balance by the digital algorithms that govern them. Not only is there no need to edit together separate shots into sequences (should the animation team decide a continuous take would be preferable), but the physics of the space is continuous (unless decided otherwise for dramatic effect) with those experienced in real world film-making. Automated algorithms provide the spaces, objects and even ‘camera’ with a set of behaviours consistent with the physics of the real world at the same time as they allow for a plasticity ¹ in such rules only possible in animation. Consequently, the experience of the spectator is that of viewing an impossibly continuous, impossibly complex world that nevertheless appears to adhere to the laws of physics.

In Finding Nemo (2003), thousands of different shaped and coloured fish swim in the dappled and constantly changing light of a seemingly endlessly detailed coral reef. These fish do not simply swim like flying objects, however; their individual pace and trajectories are animated to function as closely as possible within the parameters of tropical fish movement while their collective location in the ‘underwater’ space is determined by the ‘surge’ and ‘swell’ tools developed and deployed by Pixar (CGSociety Features, 2011). A Shark’s Tale (2004) is marked by layers of complexity as thousands of fish and other sea creatures swim around a reef constructed from the built-up urban environment of central New York (complete with its own Times Square and branded advertising billboards for ‘Coral Cola’ and ‘Fish King’). Both Antz (1998) and A Bug’s Life (1998) take place in the multilayered, multidimensional spaces of insect nests teeming with life so detailed and plentiful it is difficult to take in.

In all these features, the animation supposedly deals with the natural world, but in all cases the teeming masses of animated objects betray a certain industrial and mass-produced nature. What is apparent in the mass-produced quality of these features, that betrays their synthetic nature, is the fact that nonhuman automation has clearly played a large part. The task of rendering and animating so many hundreds of thousands of objects, characters and environments would be so large for analogue animators as to be near impossible. Instead, what the viewer beholds is a composite of animated and simulated image forms only made possible by the synthetic means of computer automation.

In many of the CG features, this relationship between industrial automation and computerised perspectival imagery manifests itself through both the narrative and the formal construction. In most cases, the dénouement of the stories takes place in vast and complex environments. The complexity of these spaces, with the thousands of individual components, conveys the force of industrial automation that the computer has brought to the process. At the same time, these spaces are frequently already the location of industrial automation anyway. Thus Buzz and co. find themselves lost in the multidimensional space of an airport’s automated baggage handling system trying to rescue Woody in Toy Story 2 (see Figure 1e). In Monsters Inc (2001, see Figure 1b), Mike, Sully and Boo navigate their way through an almost identical environment hitching a lift on a door and riding a roller coaster amongst millions of doors in the ‘scare factory’. Like Buzz and Woody, Mike, Sully and Boo must negotiate their way around multiple automated production lines – all of which would not have been possible without the power of the automated algorithm. In Robots (2005, see Figure 1d), the viewer is repeatedly guided through such experiences. The first is when Rodney arrives in Robot City and first experiences the roller coaster/mouse-trap style transportation system. The second of these is the discovery of thousands of dominoes set up in Bigweld’s house. The third is to be found in the final scenes of the film when Rodney leads a robot rebellion on the automated production lines of the scrap yard furnace factory. In Cars (2006, see Figure 1a), we are treated repeatedly to views of a motorway system both complex and soullessly automated. Indeed, Lightning McQueen is lost on the motorway before stumbling across a small town that has been devastated and forgotten by the bypass motorway system. In an irony consistent across all of these features, automated industrial processes and the space they inhabit are presented as the enemy of our central characters. And yet these animated visual spaces are only made possible through exactly such automated industrial processes. In DreamWorks’ Bee Movie (2007, see Figure 1f) Barry spends his time (like his predecessors in Antz and A Bug’s Life) preoccupied by his distaste of the soul-destroying mechanised processes of the Beehive. Finally, The Polar Express (2004, see Figure 1c) repeats what is a familiar territory in the form of an industrial toy factory and its myriad of distribution slides (co-incidentally perfectly shaped and sized for the protagonists).

OPEN IN VIEWER

Figure 1.

The multidimensional navigation of space in CG features: (a) Cars (John Lassetter 2006); (b) Monsters Inc. (Pete Doctor, 2002); (c) The Polar Express (Robert Zemeckis, 2004); (d) Robots (Chris Wedge, 2005); (e) Toy Story 2 (John Lasseter, 1999); (f) Bee Movie (Steve Hickner, 2007).

We see in these images, then, an experience that, more than any other visual form, parallels the continuous perspective of a roller-coaster ride. The consequence of this is a constructed space that picks up and continues the work of the creators of perspective during the Renaissance. The development of perspective in the 15th century was accompanied by a slow but seismic philosophical shift towards humanism. Perhaps it should come as no surprise if the development of the computer-generated algorithm is accompanied by, and/or a product of, a similar philosophical shift towards industrial consumerism. In this case, like the beholder of a perspectival Renaissance painting, the Pixar viewer is placed at the centre of the visual world, but with CG sequences; however, the viewer also enters into a new relationship with the infinitely deconstructable and reconstructable objects and spaces on screen. Just as the Renaissance placed the humanist viewer at the centre of the physical/optical universe, so the CG sequence places the consumerist viewer at the centre of a constantly changing but nevertheless industrially fabricated roller-coaster ride. It is important to understand here that the use of perspective per se is not what causes a shift in the mode of spectatorship – that is something that has existed for 500 years. What causes the shift is the ability to link the fabrication of the perspectival image to the industrially reproducible process of the computer algorithm. As Manovich (1993) has argued:

The ability to generate three-dimensional stills does not represent a radical break in the history of visual representation of the multitude comparable to the achievements of Giotto. A Renaissance painting and a computer image employ the same technique (a set of consistent depth cues) to create an illusion of space – existent or imaginary. The real break is the introduction of a moving synthetic image – interactive 3D computer graphics and computer animation. With these technologies, a viewer has an experience of moving around the simulated 3D space – something one can’t do with a painting.

What is significant here is the way in which the computer-automated algorithm produces an audiovisual form that places the viewer/consumer at the centre of an industrially fabricated space. More than the technically reproduced indexical space of a film camera, this is a space that has been specifically fabricated on an industrial scale for consumption by the viewer. In this sense, Pixar animated features mainline the viewer into the processes of industrial fabrication and consumption simply through the act of spectatorship.

Perhaps the best analogy is that of the supermarket shopper. Like a child sitting in a supermarket trolley, the CG feature spectator moves through continuous aisles viewing the dizzying array of mass-produced product lines and their enticingly designed packaging (Figure 2). Here, as with the CG features, the industrially mass-produced and product-designed objects that fill the shelves are continuous with the space through which they are viewed – which is designed to be as conducive as possible to the act of purchase. It is a space designed specifically to aid our ease of progress through it. Similarly, the space of the CG feature, with its perspectivally continuous and digitally reproducible quality, has a certain ‘off the peg’ nature to it that reminds us of the industrial prefabrication used in out-of-town retail warehouse space and the consumer products that fill them. It is no coincidence that half of the action in Toy Story 2 takes place within a big box toy store. It is not simply that the big box toy store is the place which Pixar, Disney and DreamWorks executives most want their viewers to visit upon leaving the cinema; it is also that it is both the space and an object domain most suited to the dynamics of Pixar’s aesthetics. The same industrial design processes responsible for retail space and the objects that fill them are also responsible for CG features, spaces and objects. The Polar Express (Figure 2a), Toy Story 2 (Figure 2b), Robots (Figure 2c) and Bee Movie (Figure 2d) are just a few of the examples that recreate the Product Design Engineered Spaces and Objects of the Retail Store.

OPEN IN VIEWER

Figure 2.

The Product Design Engineered Spaces and Objects of the Retail Store: (a) The Polar Express (Robert Zemeckis, 2004); (b) Toy Story 2 (John Lasseter, 1999); (c) Robots (Chris Wedge, 2005); (d) Bee Movie (Steve Hickner, 2007).

This brings us onto the second aspect of the newly emerging automated CG visual nominalism which represents both a significant shift away from traditional film and animation processes at the same time as it reconnects to a process initiated during the Renaissance: the rendering of object forms. Concurrent with the developing perspective, the aesthetics of light source and object rendering also went through a radical shift after the 15th century. Rendering played an interrelated role in anchoring the object form described by the newly discovered rules of perspective. Because perspective allowed the object to be located in a mathematically definable space, it also allowed light sources to be located in space/s relative to the object. Thus, rendering used light sources to articulate the material substance of the object, located as it was in a physically identifiable spatial continuity. In the contemporary context, computer-automated rendering and the automated perspectival construction of CG spaces and light sources have seen a second renaissance in object representation. In the digital domain, objects are rendered according to algorithms that can articulate both complex multiple light sources and the effect of these light forms upon varying materials. This situates the CG image in a new relationship to contemporary practices of product fabrication. Beyond the often-noted prevalence of product placement and canny merchandising, what does not get discussed is the way in which the objects that populate CG features are themselves product-designed as a function of automated visual nominalism. This represents a step-change in the way we should think about ‘product placement’ in CG features since every object and space rendered within the frame is a product. It also represents a step-change in the kind of image the viewer must position themselves in relation to.

Object rendering and visual nominalism: Where production line, product design and promotional aesthetic meet

John Lasseter, the director of the first two Toy Story films, has acknowledged the connection between the aesthetics of CGI and that of modern commodity forms. In an interview, he points out that: ‘Toys are a natural [choice for computer animation] the computer likes to make objects look plastic and perfect and manufactured’ (Kafner, 1997: 229). Lasseter’s observation goes beyond a basic acknowledgement of an aesthetic difference in rendering between one form and another. CGI, as the technically reproduced product of a process only made possible by computers, is both quantitatively and qualitatively different from its hand-drawn predecessor. For one, CGI is situated within a cultural and industrial context more closely integrated with, and reflective of, the operative processes of consumer capitalism. For example, the construction of CGI models in the animation process (using wire-frame models from object design software) is similar and sometimes the same as the CGI construction of consumer objects in industrial product design engineering. ² This collapse in the distance between image production and industrial product design is apparent when one considers the way in which CG animation is constructed. In the initial animation stage, objects and characters are literally built from scratch as wire frames. As Alice Crawford (2000: 114) points out:

The second step in CGI is to add or ‘map’ textures onto the objects one has modelled. This process works by fitting surface textures onto wire-framed models, similar to how one would stretch upholstery over a sofa, or tile a bathroom.

Crawford’s description makes explicit the connection between the way in which CG objects are quite literally fabricated and the processes of industrial production that bring other everyday consumer objects such as sofas and bathrooms into our lives. Though Crawford does not make the point, it is not an accident that the angle-poise lamp that introduces every Pixar film should have been taken up by the company as their corporate animated logo. The lamp is derived from the now famous first short produced by Pixar (Luxo Jr, 1986) which featured a ‘young’ lamp playfully irritating an ‘older’ angle poise. This short is indicative of the way in which Pixar connected into an already existing visual, industrial language of digital product design engineering from its earliest history (Figure 3).

OPEN IN VIEWER

Figure 3.

The industrially manufactured object personified in Pixar’s brand mascot Luxo Jr (John Lasseter, 1986).

Pixar is not alone in exemplifying this relationship. Surviving show reels from MAGI (the company that originally commercialised CG effects for public consumption in the form of US television advertising) demonstrate that industrial objects were a staple of early CG history (Gurevitch, 2010). While MAGI was eventually sold off, Pixar survived beyond its early history by successfully winning and completing advertising commissions (Figures 4a and 4b) for companies such as Tropicana, Listerine, Volkswagen and Kellogg’s (Pixar, 2011). In light of this, adverts for Listerine and Kellogg’s products functioned for Pixar as one more form of industrial computer-generated attractions: a form of transferable currency across an audiovisual economy more focused upon the digital attraction, interesting more for its illusory powers alone, than upon its status as a commercial visual form. The point here (visible in the similarities between Figure 3 and Figure 4) was not that Pixar animation functioned as advertisements, but that it functioned as attractions that could gain viewer attention based on the fact of their audiovisual novelty alone. Beyond studios themselves, a long-running tradition of new animators learning CG software by creating industrial objects such as bottles and cars continues to this day in the abundant Maya and Blender tutorials available on YouTube.

OPEN IN VIEWER

Figure 4.

Early Pixar advertisements: (a) Listerine advertisement; (b) All-Bran advertisement.

It is a short step to view the characters now populating CG features as a function of the consumer process: their very construction constitutes them as products from the outset. When engineers construct an object on their computer for mass industrial production, the visual object they work with is not simply an image of a product, it is a product. The software tools they work with, by virtue of their unique ontological foundation in an automated computing machine, cross the traditional boundaries between creative expression and industrial fabrication. The objects they fashion in this digital imaging domain already have their physical parameters articulated: size, structural integrity, aerodynamics and material construction are all worked out and dependent upon each other in a myriad of complexly interrelated computer algorithms. These are not simply images, they are industrial simulations.

Thus Pixar’s real-world merchandised toys are now created from the same processes of computer product design already required for, and completed in, the construction of the features themselves. With this in mind, John Lasseter’s (1995) description of the concept brief for Buzz’s character – ‘The goal was to make Buzz the toy that any boy would just die to have’ – is revealing, if unsurprising. In his description of the Pixar creative team’s methods John Lasseter states:

We focused on each toy, and looked at the toy, studied the toy, thought about how it was manufactured, with the quality of plastic, whether it was a cheap toy whether it was an expensive toy, all these things and combined these to draw the personality out of it.

Lasseter’s account reveals the extent to which the conception of manufacturing methods and consumer practices are integrated into the creative production processes of the Toy Story features. In her work on the relationship between Hollywood and Madison Avenue, Janet Wasko (1994) quotes an anonymous television industry executive offering advice to programme makers, which she correctly argues has also now become applicable to film. Programmes, the executive asserts, ‘must have a sense of adventure and create an environment which the child or adult can relate to on many plateaus’ (p. 207). Chief among these plateaus is the requirement that ‘the situation must develop into toyetic applications – characters which have a personality that can be easily transferred to dolls and playset environments.’ Tellingly, the contemporary creative process Lasseter describes takes place in reverse order to that of the television executives. Pixar derive their characters’ personalities from the already constituted consumer products (be they physical or simulated), whereas the TV executive suggests that characters’ personalities and forms be developed with a view to turning them into consumer products later. The difference here is the place that the consumer process occupies in the hierarchy of the production process of the visual text. In the executive’s description of the programme production process, the consumer product is secondary to the creative process; characters are constituted in such a way that they can then be translated into consumer objects. In Lasseter’s description, the consumer product is primary; it is the source of his characters, their personalities and their environments.

Thus the aesthetics of all CG features betray the way in which the objects and characters that populate them are deeply embedded within contemporary processes of industrial product fabrication. As William Schaffer (2004: 86) has observed of A Bug’s Life:

Despite the fact that Pixar’s modellers spent hours studying real ants before constructing their virtual counterparts, these bugs evoke the author’s memory of plastic pull-apart insects that were advertised on TV during his childhood and had to be pieced back together …

Here we are reminded of Crawford’s (2000) description of the way in which wire-framed models are wrapped in textures reminiscent of furniture upholstery or home improvements. Similarly, Schaffer (2004: 87) describes the construction of Sully’s coat in Monsters Inc ‘like a sort of beautiful nylon toilet brush’ and other characters such as Nemo in Finding Nemo, ‘still strongly recall squeezable toys’.

In this way, the nature of industrial fabrication is firmly embedded within the simulative processes of CG animation. Indeed, the logic of industrial fabrication permeates the animation process in multiple ways. Like production lines which outlive the life of a mass-produced object, the computer programs utilized and the objects and spaces created to populate CG animated features are not junked at the end of a production but recycled, amended and updated for each new feature. Like the development of other industrialised consumer product lines, the CG animation process is set up to be flexibly redirected when the fabrication of ever new objects and environments are called for. Animators interviewed in the DVD extras talk often of the ways in which computer algorithms developed for an animation process in one film have been adapted to suit another. ³ For this reason, the representation of machined and manufactured objects comes closest to visual verisimilitude in CG aesthetics – because CG aesthetics are so deeply embedded within the processes of industrial fabrication responsible for synthetic objects in the first place. ⁴ Written in code at the heart of CG animated features is a whole system of industrial relations borne out of contemporary consumer capitalism (Figure 5). In this sense, the tension between Buzz and Woody in the first Toy Story feature is not simply a narrative construct but symptomatic of a fundamental underlying aesthetic reality in CG animation. Buzz does not just usurp Woody’s place in the diegetic hierarchy of the movie – his aesthetic existence represents the culmination of a 20-year and more process in which Pixar developed the rendering of industrial objects to a fine art. Buzz and his seductively perfect, industrially manufactured form, functioned for Pixar not as the beginning of a brave new era in CG animation (as many commentators claimed on the movie’s release), but as the product and showcase of years of research development already undertaken in the domain of industrial design, rendering and advertising.

OPEN IN VIEWER

Figure 5.

The CG Character as an Industrially Manufactured Object in Toy Story (John Lasseter, 1995).

Further analysis reveals another relationship in which newly industrialised animation processes have led to a reinforced link with the world of product consumption that did not exist in traditional film and animation. In these highly rendered images of industrially manufactured objects, we start to see an intriguing convergence between the aesthetics of the animated feature and the aesthetics of the advertising industry. CG features overflow with scenes in which perfect, industrially manufactured forms are fetishistically lingered over. In the opening scenes of Cars, for instance, the viewer is introduced to Lightning McQueen in what looks more like a 1980s Ferrari advert in which perfectly polished body work glistens as it is caressed by the light (Figure 6). Likewise in Robots, the prevailing aesthetic, from the outset, is one which renders the gleaming, industrially produced surfaces of the robot world in the kind of glistening detail usually reserved for advertising. In Ratatouille (2007), the rat’s fur and the farmhouse world the characters inhabit is finely detailed but it is not until Remy arrives in Gusteau’s Parisian restaurant kitchen with its cornucopia of gleaming, machined utensils and shining pots and pans that the aesthetic comes to life. It is as if the full reality of big city life is reserved until the moment at which the high-definition, reflective aesthetics common to advertising are unleashed.

OPEN IN VIEWER

Figure 6.

The Industrial Aesthetic of High Concept film, Advertising and CG Features: Lightning McQueen’s bodywork is fetishistically rendered and represented in Cars (John Lassetter, 2006).

OPEN IN VIEWER

Figure 7.

Advertising aesthetics in Cars (John Lassetter, 2006). The pristine aesthetics of a country drive: more than coincidentally reminiscent of a car advert.

This is not to say that the high-definition aesthetics of advertising are common only to CG features. As theorists from John Caldwell (2000) to Terry Flaxton (2011) have pointed out, the development of high-definition imaging across audiovisual industries is a symptom of contemporary production culture more widely. Nevertheless CG animation has emerged as the holder of a unique relationship with high-definition aesthetics precisely because of its location in the heart of computer generated culture. Key to understanding this is a nuanced and long running relationship between industrial fabrication and exhibition and audio-visual aesthetics that stretches back to before the commercial emergence of CG.

In the 1990s, Justin Wyatt (2003: 30) attributed the polished, reflective, high definition look of high concept films ⁵ to a general advertising aesthetic of the time. These films, he argued, ‘tend toward sleek, modern environments mirroring the post-industrial age through austere and reflective surfaces’. This went hand in hand with the need to premise the marketing and distribution strategies of high concept film upon strong images that could translate from film to promotional poster to television trailer (p. 112). The convergence in the high definition look of film and advertising image forms was also noted by Michael Schudson (1993) in his discussion of advertising aesthetics. Quoting Barbra Rosenblum, Schudson argued that:

If anything, advertising looks more real than it should. As Barbra Rosenblum writes in her study of professional photography, advertising photography uses ‘crisp focus’ to create ‘a dense and busily detailed surface. Light is used in conjunction with focus to create a hypertactile effect. Things look real; in fact, almost too real.’ (p. 217)

While Wyatt argued the sharply reflective, austere surfaced, high definition aesthetic form in film to have originated in advertising, Rosenblum and Schudson attributed the same thing in magazine and television adverts to ‘cinematic’ imaging. Wherever the origin, Wyatt, Schudson, Rosenblum and, more recently Caldwell, are all united in identifying the presence of highly detailed, richly reflective aesthetics in contemporary imaging.

Approaching this issue from a different direction, Manovich (2001: 138–142) has suggested more recently that a similar overlap of high definition image forms in different audiovisual industries is taking place in the digital domain. The digital composite, he argues, facilitates an ‘aesthetic of continuity’ across multiple forms: television advertising, music video, gaming and film. Strikingly, Manovich’s description of digital technology’s impact upon audiovisual imaging echoes Schudson and Rosenblum’s observations that in advertising things ‘look real; in fact, almost too real’:

Common opinion still holds that synthetic 3-D images generated by computer graphics are not yet (or perhaps will never be) as ‘realistic’ in rendering visual reality as images obtained through a photographic lens … this common opinion is mistaken. Such synthetic photographs are already more ‘realistic’ than traditional photographs … Typical images produced with 3-D computer graphics still appear unnaturally clean, sharp, and geometric looking … So although we normally think that synthetic photographs produced with computer graphics are inferior to real photographs, in fact, they are too perfect. But beyond that we can say also that, paradoxically, they are also too real. (Manovich, 2001; 201–202, emphases added)

What Manovich, like Schudson and Rosenblumn, attributes to ‘the real’ is based upon the capacity to reproduce the image in as high a level of definition as possible. Thus, the strand that runs through Wyatt, Schudson, Rosenblum, Caldwell and Manovich’s analyses of their respective image forms is the identification of an ever increasing level of definition in the image and a recognition that such an aesthetic operates equally across differing forms of cultural production and different forms of technological distribution. In all cases, advertising and the synthetic industrial objects that constitute its subject matter are cited as a driving force of this aesthetic shift. Significantly, the relationship between the image as a promotional form and the image as an ever higher definition attraction that appears ‘too real’ has a history dating back to the inception of cinema.

Steven Botomore (1999) has detailed the way in which early audiences attended attractions to experience the fabled ‘train effect’ in which audiences were reputed to be panicked by the realism of the trains they witnessed approaching. Bottomore asserts that these tales of panic were myths invented for promotional effect in newspaper features and the accompanying merchandise that surrounded even early cinema. Bottomore’s work suggests two points crucial to our own understanding of the relationship between CG imaging and advertising aesthetics. Firstly, he argues that it was not just the verisimilitude of the new moving images that attracted audiences and made early cinema both an attraction and a promotion: it was the experience of beholding something the viewer knew to be impossible, no matter how real it appeared, that attracted audiences. Secondly, Bottomore’s work suggests that the process by which the audiovisual image is continually developed to contain ever-higher levels of definition – and therefore ever higher levels of verisimilitude – is both unceasing and inherently self-promotional (pp. 177–216).

To take the first point, early cinema audiences sought the pleasure of viewing ‘realistic’ images of waves lapping on a shore, or clouds passing over the sky, projected before them. Part of the pleasure (and almost without exception part of the promotional rhetoric surrounding this experience) was derived from the cognisance that the waves lapping upon the screen could not possibly be doing so. In contemporary audiovisual imaging, this process has not changed, it has merely been passed on into the developing field of computer imaging. As Lasseter (2003) attests:

At Pixar, we like to think we use our tools to make things look photorealistic, without trying to reproduce reality. We like to take those tools and make something that the audience knows does not exist. Every frame they know this is a cartoon. So you get that wonderful visual entertainment of, ‘I know this isn’t real, but boy it sure looks real’. I think that’s part of the fun of what we do.

Lasseter’s analysis here leads us on to the second point: not only has CG imaging become the inheritor of the process by which the experiences of impossibility and verisimilitude are finely balanced, CG imaging has also inherited the promotional role that accompanies this process. For viewers of the ‘train effect’ at the turn of the 20th century, the shock registered in the encounter with an image apparently so real that it could cause panic became a publicity generating event story. While the veracity of this story is questionable, for Bottomore what is important is the way in which new levels of verisimilitude went hand in hand with new levels of media coverage and publicity. For the historian of early film, Bottomore’s argument suggests that from the inception of early cinema the production of arresting images for entertainment and the utilization of these images as publicity generating events went hand in hand.

Like the film image, the history of CG imaging is closely bound up with the history of the promotional process. When the US military set their computer labs to the task of developing commercial civilian applications for their newly developed raytraced imaging techniques, their first use was found in television advertising and corporate promotional videos (Lisberger; 1982). ⁶ While companies such as Levis and AT&T used CGI to capture the US public’s attention for their products, aerospace and construction industries of all kinds used CGI as a promotional tool that could envision the proposed final product. It was from their experiences in these promotional fields, including television advertising that the first CG animated feature, Tron was commissioned from companies like MAGI, Digital Effects and Abel to be made by Disney. Pixar itself operated for many years in the advertising industry.

With this in mind, one could see Pixar’s choice of cars as the subject matter for their fifth film as a foregone conclusion: a perfect marriage of form and content that was inevitable. Cars, more than any other mass-produced industrial object, have from the days of Henry Ford been intimately intertwined with what Andrew Wernick (1991) has dubbed ‘promotional culture’: a culture in which industrially produced objects are both cause and consequence of an endlessly circular promotional process. In this process, object forms and their advertised images are collapsed:

The production of cars as signs is a special case of the way in which, since the industrial revolution of the late eighteenth century, all mass-produced consumer goods have come to intersect with the world of meaning. That is: their visual appearance is designed to be continuous with the advertising through which they are mass marketed … Overwhelmingly, from billboards to showrooms, cars have been advertised by being shown, giving pseudo-auratic texture to that endless parade of vehicles on the actual highway, which itself has served as one vast ad. (p. 70)

In Pixar’s Cars, it is not just that the industrial object of the car intersects with the world of meaning. In Pixar’s CG animation, the industrial object and the industrial image are one and the same thing. They do not intersect with meaning: they are the meaning. In this audiovisual space, the distinction between the object, the image and the advertising image through which imaged and actual objects gain meaning is considerably reduced if not removed completely. More specifically, the CG aesthetic and the advertising aesthetic increasingly reflect one another: something that can be seen in the loving rendering and representation of Lighting McQueen’s bodywork.

Tellingly, this relationship between the advertising aesthetic and Pixar’s CG aesthetic has another facet in the choice of Lightning McQueen as subject matter. As Wernick argues, the visual appearance of the car has indeed come to be continuous with the advertising through which it is mass marketed. Crucially, however, this advertising has turned increasingly to the use of CGI in recent years for a number of reasons. As Wernick continues:

For imaged cars … looking modern has come to mean not just looking fast and airborne, but being linked to computers and all that they connote … The introduction of computer-related imagery into the promotion of cars has amounted to more … than just an updating of the way technical progress is approvingly portrayed. There has been a profound change in psychological emphasis: from conquest to experience, from outer to inner space. (p. 81)

In Cars, this relationship could be said to reach its apotheosis. Not only does the feature portray a series of computer software engineered cars, but they are rendered and animated onscreen in an aesthetic continuous with the visual prerogatives of advertising: the form that is also responsible for the next stage of a car’s creation as promotional object. In this sense, the absolute and complete centrality with which the subject of promotion is placed in Cars is a consequence of the multiply interrelated audiovisual conditions under which the feature, the advertisement, the object and the CG aesthetic come to co-exist. It is in this sense that the distinction between the object, the image and the advertising form is reduced if not removed completely. A concrete example of this can be found in an early introduction to McQueen. McQueen’s early direct address to ‘camera’ is mediated through his ‘advert’ for ‘Rusteze’. While this ‘advert’ is clearly a fabrication of the narrative, the nature of the computer rendering means that there is little to distinguish the aesthetics of this mock advert from the rest of the film. In a 10-minute section towards the end of the film in which McQueen and Sally go for a drive, it is difficult to tell the difference between the sequence as a feature film or a car advert. It is not only that McQueen and Sally drive playfully along the pristine mountain roads of countless car advertisements, nor that the constantly canting ‘camera’ angles are set up to mirror such adverts that recreate the apparent sense of freedom and fun attendant with winding mountain drives: it is also that the ultra clean aesthetics of perfect blue skies and polished reflective bodywork take the viewer straight back to the CG imagery that has been entering car advertising for the last two and a half decades. With this in mind, the etymological origins of the word ‘automation’, central as they are to the renovation of visual nominalism, are extremely revealing:

The term ‘automation’ was finally coined in 1947, and in 1949 Ford began the construction of the first automated factories. Automation was made possible by the development of digital computers during World War II and thus became synonymous with computerization. A decade later, the automation of the process of constructing perspectival images of both existent and non-existent objects and scenes was well underway. (Manovich, 1993)

Here, then, we have an explicit connection between the processes of computer automation, contemporary industrial production (of automobiles, no less) and automated perspectival image forms. Corresponding with Wernick’s assertion that the car has been symbolically integrated with the computer in recent years, we have a final revealing example of the way in which even the processes of production have come to closely interrelate through visual nominalism.

Beyond the checkout

Cross referencing Comolli’s work with Manovich’s more recent observations, we could revise the history of perspective during the Renaissance and say that its invention performed another, so far unidentified, function. The Renaissance viewer of a perspectival painting was not only placed at a newly created humanist centre of the visual universe, they were simultaneously informed of the primacy of engineering and manufacture to that visual universe. In Comolli’s terms, the social machine, premised as it is upon representation, was now constructing, and constructed from, a scientifically rationalised, engineered and manufactured world. Today’s CG features continue this process, not just in representing a space manufactured for the viewer (as was the case in Renaissance spectatorship) but in extending the axes of the perspectival image to encompass time. Contemporary viewers do not simply view a still perspectival image manufactured for them, they view a form manufactured for them in every dimension. These spatial and temporal constructions offer much to our understanding of the impact CG imaging is having on both live action and animated film and the viewing positions still developing.

From the construction of space to the rendering of complex objects located in that space, CG features segue with the advertising and promotional forms that have long functioned at a visual crossroads with the manufactured object. Witness, for instance, the dancing cars that are equally at home in a Citroën advert or a Transformers film, the Gladiatorial Coliseum that could equally function as a promotional bid for a world cup stadium. From science fiction film sequences that view like Nasa’s last funding bid to Volkswagen adverts that incorporate Gene Kelly singing in the rain, the list of engineered and reengineered visual forms grows (Gurevitch, 2010). In turn, movies of all kinds (not just animation) increasingly become manufactured objects in their own right. This requires a revised understanding of ‘product placement’ and its function within the movie feature, not least because it poses the question: what does ‘product placement’ mean when the feature itself is a product design engineered object? CG animated features constitute a relationship between processes of manufactured imaging and processes of imaging manufacture that go beyond product placement. Where product placement took the industrial object and ‘inserted’ it into the filmic narrative, the CG feature is the industrial object around which the narrative is constructed. In this sense, Toy Story and its successors have ushered in a paradigm shift on the scale of Gunning’s (1990) articulated development of narrative film between 1906 and 1911. In this case, the cinema of attractions returns with a vengeance, no longer functioning ‘underground’ only to emerge between the cracks of narrative but as an equal, feature-length partner of the narrative. In traditional understandings of product placement, the industrial object is treated as an invasive, infectious and distorting form that threatens narrative purity. With Pixar film, the narrative is the industrial object. The significance of a newly renovated visual nominalism is not just that it sparks a new round of integration and innovation in science, engineering and manufacture; it is that it draws our cinematic culture into this relationship too. With 3D television on the horizon, expanding the visual nominalist image literally into the space of the living-room, the industrially manufactured object will increasingly move its sales floor further into the domestic space, skipping the checkout and beyond.