156x234mm http://anm.sagepub.com Animation DOI: 10.1177/1746847709104643 2009; 4; 107 Animation Pat Power Animation Animated Expressions: Expressive Style in 3D Computer Graphic Narrative http://anm.sagepub.com/cgi/content/abstract/4/2/107 The online version of this article can be found at: Published by: http://www.sagepublications.com can be found at:Animation Additional services and information for http://anm.sagepub.com/cgi/alerts Email Alerts: http://anm.sagepub.com/subscriptions Subscriptions: http://www.sagepub.com/journalsReprints.navReprints: http://www.sagepub.co.uk/journalsPermissions.navPermissions: http://anm.sagepub.com/cgi/content/refs/4/2/107 Citations at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com/cgi/alerts http://anm.sagepub.com/subscriptions http://www.sagepub.com/journalsReprints.nav http://www.sagepub.co.uk/journalsPermissions.nav http://anm.sagepub.com/cgi/content/refs/4/2/107 http://anm.sagepub.com Animated Expressions: Expressive Style in 3D Computer Graphic Narrative Animation Pat Power Abstract The development of 3D animation systems has been driven primarily by a hyper-realist ethos, and 3D computer graphic (CG) features have broadly complied with this agenda. As a counterpoint to this trend, some researchers, technologists and animation artists have explored the possibility of creating more expressive narrative output from 3D animation environ- ments. This article explores 3D animation aesthetics, technology and culture in this context. Synthesizing research in CG, neuro - esthetics, art history, semiotics, psychology and embodied approaches to cognitive science, the nature of naturalistic vis-à- vis expressive visual styles is analysed, with particular regard to expressive communication and cues for emotional engagement. Two foundations of naturalistic 3D CG, single-point perspective and photorealistic rendering, are explored in terms of expressive potential, and the conclusion considers the future for an expres- sive aesthetics in 3D CG animation. Keywords 3D animation, creative, emotion, expressive aesthetic, naturalism, non-photorealistic rendering, perspective, realism animation: an interdisciplinary journal (http://anm.sagepub.com) Copyright © The Author(s), 2009. Reprints and permissions: http://www.sagepub.co.uk/journals/journalsPermission.nav Vol 4(2): 107–129 [1746-8477(200907)]10.1177/1746847709104643 article at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com Realism, naturalism and expression Realism and naturalism, ideas of art as an imitation of reality, are currently the primary ethos of 3D animation culture and technology. These issues are ‘far larger and more far-reaching than aesthetics or artistic convention’ (Kress and Van Leeuwen, 2006: 167), involving not only questions of aesthetics, but of ontology, epistemology and phenomenology. Their history is at least as old as Plato’s derogation of art as mimesis, and its subsequent defence by Aristotle. In modern times they again became prominent with the advent of photography, then the birth of cinema. Photography and cinema differ somewhat from painting and animation with regard to realism as, in lens-based arts, the indexical nature of the image is generally a given, whereas in both non-photographic imagery and animation the constructed nature of the imagery is salient.1 Other non-lens-based visual arts flourished subsequently by actively exploring denaturalization as both theme and technique. Since the late 1960s, when Roland Barthes’ analyses of the codes of reality effects and referential illusions undermined aspirations to realism and naturalism, contemporary cultural or semiotic theory has also aimed at denaturalization by revealing the socially coded basis of cultural phenomena which are taken-for-granted as natural. Ironi- cally, during the same period, naturalism has become the sine qua non of CG research, the achievement of photorealism being ‘the main goal of research’ in this field (Manovich, 2001: 199). ‘Different realisms exist side by side in our society’, but the standard by which we judge visual realism remains conventionally understood naturalism, that is photorealism ( Kress and Van Leeuwen, 2006: 158). In fine arts and animation, the term realism is often used interchange- ably with naturalism to define a style of visual or audio-visual mimetic representation that aspires to photorealistic or cinematic verisimili- tude. Andrew Hemingway (2007: 103) argues that the term realism is too confusing a term to apply to visual arts such as painting and suggests that, following E.H. Gombrich, the term naturalism (despite its own ambiguous associations) better reflects ‘the general idea of pictorial verisimilitude’. Though both terms are used where considered appropriate in this article, the term naturalism does seem somewhat less confusing, and better reflects the technological drive towards verisimilitude in 3D animation.2 Theories of art as expression also have a controversial history. Having been particularly out of favour in the second half of the 20th century, they have recently been revived due in part to advances in the study of emotion, like those by neuroscientists Antonio Damasio, Joseph LeDoux, Edmund Rolls and Jaak Panksepp, by psychologists including Jerome Kagan, Nico Frijda and Arnold Modell, and by theorists who deal specifically with emotion and expression in the arts, for example Jenefer Robinson, Noël Carroll, Greg M. Smith and Christopher Butler. 108 animation: an interdisciplinary journal 4(2) at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com In a narrative context, naturalist and expressive modes of represen- tation can be seen as dialectically related. Generally, illusionistic 3D attempts mimesis of an external (or cinematic) reality whereas expres- sive styles play more with the nature of mind and of perception, emotion, memory and imagination. However, in common with live action (as in German Expressionism or film noir, for example), a virtual visual reality can still be expressive in modalities such as lighting or sound, while a non-naturalistic animated narrative might not express anything successfully at all. But in animation as in painting, whereas some artists strive for visual verisimilitude, others prioritize expressive- ness, and these are aesthetically divergent styles, the former dealing primarily with denotation, and the latter, either consciously or intu- itively, with expressive connotation.3 There are resonances here with dialectics such as objective/subjective, logical/emotional and noumenal/phenomenal and with language, where prose can be contrasted with more poetic and expressive forms. Traditionally, animation has been one of the most expressive of the visual arts, but in 3D animation, quantitative has trumped qualitative, due in part to what Vivian Sobchack (2008) calls ‘the calculative and quantitative tendencies of the computer’ (p. 262). The issue in question here is whether or how an aesthetic culturally and technologically rooted at one end of this continuum can be taken for a creative stroll towards the other end. The exercise should prove worthwhile, because as Kostas Terzidis (2003: 58) suggests, the expressive has many advan- tages over the realistic and, whereas the computer-graphic quest for realism is essentially about completeness, ‘notions of incompleteness, imperfection, and subjectivity’ invite interactive participation and have an expressive value that can surpass this explicitness. As Michael Davis (1999), a specialist in Greek philosophy and translator of Aristotle sees it, mimesis should comprise ‘a stylizing of reality in which the ordinary features of our world are brought into focus by a certain exaggeration . . . like the relationship of dancing to walking’ (p. 3). He concludes that ‘the more “real” the imitation the more fraudulent it becomes.’ 3D computer graphics and photorealism There are strong historical, technical, commercial and cultural reasons for a dominant naturalist aesthetic in contemporary 3D CG. The homology of applied science and technology research and develop- ment ensured a legacy of ideologies of objectivity as opposed to subjectivity. As digital techniques have supplanted analogue tech- niques in many design and production contexts, including graphics and animation, 3D animation has co-evolved symbiotically and stylisti- cally with developments in 3D CG technology. There has been co- development and cross-over in technical advances for computer-aided design and manufacturing (CAD/CAM) and developments for use in Power Animated expressions 109 at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com 3D animation and entertainment, largely ‘determined by the needs of the early sponsors of this research – the Pentagon and Hollywood’ (Manovich, 2001:193). Whether they are for use in architecture, car design, military applications, medical imaging or feature animations, they all come under the rubric of 3D CG visualization, and can be traced back to Ivan Sutherland’s 1963 Sketchpad system that exempli- fied this ‘new paradigm of interacting with computers’ (Manovich, 2001: 102). Autodesk Inc., one high-profile contemporary example, develops systems for use in architecture, engineering, manufacturing, and media and entertainment. It develops CAD systems (such as Autocad) in tandem with 3D animation solutions (such as 3D Studio Max and Maya), and research and development in specialized graphics hardware and software are congruent across all these sectors (Figure 1). Though these markets are largely distinct, there are important historical, cultural and technical syntheses. For example, volumetric modelling and rendering using voxels (volumetric pixels) has been used for some time in areas such as medical imaging (visualizing MRI scans), but now, combined with physics simulation, this synthesis comprises a prominent research and development focus both for animation and effects for arts/entertainment, and in particular for water, ocean, cloud and other fluid or gaseous effects. Of the 10 tech- nical Oscars awarded in 2008, over half were for development of such dynamic fluid effects systems. Most of the commercial, educational, governmental/military organ- izations and individuals involved in 3D research and development are driven predominantly by an ethos of realistic or naturalistic visualiza- tion, and this is understandable in terms of goals for technical achieve- ment. SIGGRAPH is the major cross-industry professional organization for CG and its research proceedings point towards realism as a common goal (Manovich, 2001: 191). 3D CG animation software for arts/entertainment is currently focused on three main markets; 110 animation: an interdisciplinary journal 4(2) computer aided design (engineering, architecture, drafting vehicle and product design) computer aided manufacturing (CAD extended to control cutting, drilling and other output devices) medical imaging (visualization for fMRI scanning and other medical technologies) scientific visualization (mathematics, data visualization) simulation (training simulators for aircraft and other safety critical equipment) visual effects for movies (naturalistic effects for seamless integration with live action) interactive VR (remote control, virtual worlds for arts & entertainment) 3D animation for film & games (both naturalistic and expressive output for arts & entertainment) Figure 1 3D CG markets and applications (only those towards the bottom left are potential markets for expressive output). at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com animated movies (including features, adverts and shorts), 3D games and special effects generation (SFX). Although, to some extent, produc- ers of animation and games have a broad choice as to the aesthetic they choose, the special effects industry by its nature requires seamless integration with live action and thus depends on verisimilar naturalism. In this world of what Manovich (2006: 26) calls hybrid aesthetics, the goals of a naturalistic aesthetic for SFX have a strong influence on the world of animation. In spite of the fact that much in 3D animation has been technically determined, artistic innovation has also played a part, and many 3D animation and special effects companies have developed ad-hoc solu- tions to specific problems encountered by directors or designers that are often problems of aesthetics or style. Such advances have often disseminated into the wider CG community through forums such as SIGGRAPH, resulting in ‘the development of important algorithms that became widely used’ (Manovich, 2001: 194) and, despite commercial competition, a range of top-end systems has evolved with broadly similar functionality (for example, variations on sub-division surfaces, inverse kinematics, fluid effects, particles and dynamics). There have always been technically gifted iconoclasts with an eye for aesthetics (for example, John Whitney, Ed Catmull, Chris Landreth), who were as much concerned with artistic as with technical advances. Landreth (2004), an engineer turned artist who worked on the devel- opment of Maya and whose animated short Ryan won an Oscar in 2005, calls this a renaissance field, bringing together artists, program- mers, musicians, engineers and other eclectic talents to develop new kinds of storytelling. As these systems evolve they are gradually becoming more accessible to artists and, as 3D CG output becomes more pervasive, artists are becoming more attuned to their creative potential. Landreth sees this as a process of democratization and fore- casts that ‘individuals, not just large studios, will soon be able to develop huge works of art, such as CG feature films, on their own’. More affordable digital systems and tools with more intuitive interfaces and better educational resources play their part in this increased accessibility, and Norman Klein (2000) cites animation students who want their work to look ‘haunted . . . as an antidote to the hygienic digital screen’ (p. 35). Despite the naturalistic orthodoxy, all of this signifies the ongoing evolution of a more eclectic and expressive aesthetic in 3D CG animation environments. Aesthetic expression and emotion The concept of expression or the expressive is ubiquitous in the arts, but ‘few terms are as poorly understood’ (Robinson, 2005: 231).4 The quotidian use of expressive, as in gesture/facial expression, points to emotion as underpinning expression, and despite the fact that ‘the Power Animated expressions 111 at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com Expression Theory of Art came in for widespread and formidable criticism’ from the 1950s onwards (p. 231), contemporary theories of aesthetic expression have emerged based on recent psychological and neuroscientific research in emotion, such as that advocated by philos - opher of aesthetics and psychology, Jenefer Robinson, in Deeper than Reason (2005). Suggesting a ‘New Romantic Theory of Expression’, Robinson argues that although all works are expressive in some respect, some are what she refers to as central cases of expression (p. 266) while other secondary cases are more peripheral. Emotions are processes that involve an initial fast, unconscious but coarse affective appraisal of the immediate environment involving low- level neural circuits, particularly the amygdala, that result in physio - logical responses affecting attention, motivation and action tendencies. This rapid response is accompanied by a slower cognitive appraisal that assesses the appropriateness of the quick-and-dirty affec- tive appraisal and that monitors, labels and feeds back into the overall emotional process (p. 231). This view is consistent with LeDoux’s (1998) fast low-road and slow high-road theory of emotion, and with Rolls’ (2005: 452) explanation of that perennial philosophical conun- drum, the affective paradox of fiction.5 Through aesthetic engagement, the arts can educate us emotionally by initially evoking instinctive emotional responses, followed by cognitive monitoring and reflection upon them, with aesthetic reflection comprising a later part of this process. Understanding in the arts is dependent on affective embodied experience, and expressive qualities of artworks are ultimately ‘quali- ties that can be grasped through the emotions they arouse’ (Robinson, 2005: 291–2). Oxford Professor of English, Christopher Butler (2004), writing of emotions and the arts, observes that: ultimately it is these emotional responses which count for our pleasure or pain; it is our emotions and moods, apart from physical pain, that contribute most to our sense of the happiness, and the sadness of our lives. (p. 36) Both Robinson (2005: 292) and Butler (2004) suggest that it is wrong to equate expressive qualities in an artwork with named emotion labels such as happy or angry, as artistic expression of emotion evokes complex emotional reactions in audiences that cannot easily be labelled, that are often the very raison d’être of the creation. Edvard Munch’s archetypal Expressionist painting The Scream might be char- acterized as expressing anguish, for example, but evokes much more complex states (including aesthetic pleasure) which may be ineffable outside of the work itself. Robinson argues (2005: 292) that successful artistic expression arouses appropriate emotions in audiences, and quoting from Coleridge’s Dejection: An Ode, she suggests the purpose of expressive art is, From outward forms to win The passion and the life, whose fountains are within. 112 animation: an interdisciplinary journal 4(2) at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com This resonates with the everyday use of expressive (as in facial expres- sion) and contrasts with the naturalistic focus on an objective reality that is without. Expressive arts need to be experienced emotionally if they are to be properly understood. Butler (2004) sees understanding and emotion as being aesthetically interdependent and sees expressive form as ‘a provocative rhetoric’ (p. 20) that aesthetically guides our attention in experiencing works of art. Like the experience of hearing a funny joke compared to an explanation of it, experiencing a work of art and knowing about it are qualitatively distinct phenomena. ‘Wagner’s music is better than it sounds’ was Mark Twain’s (1924) twist on this phenomenon. Creative expressive signification The genre comprising the rapidly growing body of 3D CG animated features for children or family audiences, led by Pixar’s Toy Story (John Lasseter, 1995), and including titles such as Dreamworks’ Shrek (Andrew Adamson and Vicky Jenson, 2001), Sony Pictures’ Monster House (Gil Kenan, 2005) and Warner Brothers’ Happy Feet (George Miller, 2005), is the main focus of attention here, together with some contrasting work that may point the way towards a more expressive 3D aesthetic. This dominant genre shares not only a common techno- logical genesis but exhibits many consistencies in content, form and style. One way these can be summarized is in terms of semiotic modality markers or cues, as proposed by Kress and Van Leeuwen (2006). Building on ideas from Habermas, Bourdieu and Bernstein, Kress and Van Leeuwen outline four reality principles or coding orientations that modulate the motivated signs comprising modality markers or cues within specific social contexts (p. 165). In 3D animation, modality cues are generally interpreted through the dominant, common sense, natu- ralistic coding orientation, with high modality aspiring to naturalism. Stylistically, most 3D CG features favour high modality cueing for movement (e.g. motion capture data), relatively high modality cueing for form (detailed but stylized 3D character models), high modality dialogue soundtracks (high-profile actors) and low modality character- ization (e.g. talking tortoises or dancing penguins). Synthetic reality effects are uneven, and some ‘privileged signs of realism’ (Manovich, 2001: 196), for example fluid effects, are high modality cues that might compensate for others, such as human form. Due to our cognitive sensitivity to the latter, lower modality stylized cues can be more aesthetically effective or expressive, and are less likely to cue dis - sonance as in, for example, the uncanny valley effect (Power, 2008). Besides the dominant naturalistic coding orientation, Kress and Van Leeuwen (2006: 165) also posit technological, abstract and Power Animated expressions 113 at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com sensory coding orientations that modulate modality cues differently within specific social contexts (Figure 2). Whereas high modality would be ascribed to audio-visual verisimilitude in a naturalistic orien- tation, to accuracy in a technological orientation (e.g. in technical diagrams), or to generalization in an abstract orientation (e.g. in pie- charts or abstract works of art), the sensory coding orientation is affectively based, and congruent with an expressive aesthetic. It is an orientation or context in which high modality or value might be ascribed to non-naturalistic qualities that are tacit, suggestive, exagger- ated, affective, connotative, evocative, or in some way expressive. In its sensory coding, traditional animation is often closer to theatre than to cinema, and ‘animated narration recalls the fluency of mise-en- scène in contemporary theatre’ (Hernandez, 2007). The constraints of space and live production often require theatre to be more expres- sively inventive than film, and from moment to moment or scene to scene, whole sets or scene props might transmogrify magically, and a trunk might become a bed, or a coffin or a car. Julie Taymor (1998), a writer/designer/director of theatre, film, musicals and opera, who has successfully adapted animation for stage,6 sees art as essentially about transformation, and argues that an artist must transform and distort reality in order for an audience to be transformed. Echoing Coleridge’s lines (cited earlier), she sees an expressive approach as having the potential for powerful impact, enabling more active and creative inter- action by audiences in making their own aesthetic and imaginative connections. Taymor also sees the attempt to recreate external reali- ties as a fundamental mistake, and believes instead in internal realities as the only reality we can really know. This too is precisely the premise 114 animation: an interdisciplinary journal 4(2) naturalistic orientation photorealism perspective recorded sound motion capture documentary full colour technological orientation accuracy legibility denotation appropriate detail restrained use of colour abstract orientation abstraction organization formal design selective use of colour sensory orientation expressiveness connotation sensory design caricature symbolism defamiliarization neuroesthetic cues examples of high modality markers (cues for high value) Figure 2 Coding orientations & modality cues (following Kress and Van Leeuwen, 2006). at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com of playwright and writer Michael Frayn’s (2007) book The Human Touch: Our Part in the Creation of a Universe. ‘Ultimately, a profound evaluation of artistic expression must involve both the world at large, which is its inspiration, and the human brain, which is capable of being inspired’ (Harth, 2004: 115), and contemporary neuroscientific research, in particular recent findings in neuroesthetics, sheds new light on many of the classical and gestalt principles of expressive art. Ramachandran’s (2004) neuroesthetic concepts such as peak shift, isolation, metaphor and problem solving point to how an expressive aesthetic can facilitate cognitive, creative and emotional engagement. Ramachandran’s concept of isolation, for example, is equivalent to Scott McCloud’s (1994) idea of amplification through simplification and to the Minimalist design aphorism less is more. The neural basis of this is a bottleneck of visual attention; ‘there cannot be two overlap- ping patterns of brain activity simultaneously’ (Ramachandran, 2004: 52),7 and realistic imagery has a poor signal-to-noise ratio that can distract attention. The brain, as a complex dynamic system, responds to stimuli through associative Hebbian resonance in its neural networks that dynamically activates multimodal attractors affecting (and affected by) the complex reflexive interplay of phenomena such as emotion, attention and memory. Emotion drives attention, and stylized or expressive imagery can isolate and accentuate rhetorically, guiding and focusing attention by amplifying the signal, and through metonymic and synesthesic connotation and resonance, can act as a multimodal neural hyperstimulus, capable of encapsulating an entity’s essence in a blended aesthetic gestalt. Active audience engagement can intensify these effects, and Semir Zeki (1999) argues that artistic hyper- stimulation of areas of the visual cortex through expressive cues such as use of creative ambiguity, expands the viewer’s imagination and invites participation in constructing meaning. Recent brain imaging research has also compared responses to natu- ralistic video imagery, and then its rotoscoped, expressively animated equivalent. Rotoscoped from video, Linklater’s Waking Life (2001) embraced a deliberate visual stylization for expressive effect, using imagery as metaphor, reflecting characters’ altered states of mind. Evidence from the research suggests that, whereas naturalistic live- action evokes brain responses that characterize recognition and mind- reading, expressive animated footage is more likely to activate areas associated with emotional reward (Power, 2008). In other experiments, brain imaging of subjects shows that the amygdala, a centre of emotion, responds more strongly to impressionistic than to naturalis- tic faces, and that expressive works distract conscious vision while engaging more directly with emotions (Cavanagh, 2005). The qualitative aspects of imagery are those that are expressive (Green, 2007), and a central case of expressive work is imbued with aesthetic cues (for example, exaggeration, isolation or defamiliariza- Power Animated expressions 115 at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com tion) that, in contrast to a literal or verisimilar depiction, conveys a poetic or metaphorical psycho-verisimilitude that can evoke appropri- ate emotional responses in an audience. An animator might aspire to an expressive aesthetic in any of several different modalities; through use of form, music, dialogue, lighting, colour, movement, setting, narra- tive dynamics, or through the complex isomorphic or metaphoric interplay between these. Expressive aesthetic cues can apply in any modality, for example peak shift (i.e. caricature or exaggeration), and Gooch (2002) notes that in imagery ‘the human responses to color, motion, form, highlight, outline, and depth are all susceptible to peak shift effects’ (p. 194). Expressive effect can be amplified by creating a resonating harmony or counterpoint between different modalities,8 and this involves metaphor. Metaphor is a creative fusion of similarity and difference that may have its evolutionary origins in symbiogenesis. Biosemiotics provides insights into how the capacity for creative joining together of different, even competing, phenomena in nature (evidenced in most cells in our bodies) may emerge as semiosymbiogenesis in culture through a capacity for metaphor (Wheeler, 2006: 137). Metaphor is also one of Ramachandran’s (2004) neuroesthetics principles: ‘in many ways . . . the most important’ (p. 56), and he sees it in the brain’s capacity for cross-modal connectivity, an exaggerated form of which, synesthesia, is relatively common amongst artists (p. 74). Such cross- modal metaphoric and metonymic associations might enable a staccato sound to resonate with sharp edges in imagery or to evoke edginess as a feeling. Similarly, for both artist and audience, the gestalt of an expressive aesthetic emerges through the poïesis of sensory interplay, through the sensorium’s synergy of metaphorically harmo- nizing modalities. Discussing the semiotics of feeling, Modell, a Harvard Professor of Psychiatry, agrees that the connection between sensation, emotion, feeling and meaning is based primarily on the cross-modal associations of metaphor and metonymy. ‘Metaphor mediates, cate- gorises and thus organizes the perception of bodily sensations’ (Modell, 2006[2003]: 145), including emotion, and not only transfers meaning but transforms it. In cross-modal metaphoric poïesis, the energy inherent in a hand-drawn character can echo inner emotion, the sorrow experienced in a tragedy can be transformed into aesthetic pleasure, or representations of others can imaginatively and empathet- ically become ourselves. Artists and theorists, for example McCloud (1994) and Sobchack (2008), have highlighted the expressive qualities inherent in drawn lines or brushstrokes. Another insight into this source of expressive energy is provided by recent research on the brain’s mirror-neurons’ capacity for active simulation, that helps explain the nature of the expressive human warmth evoked by hand movement and vibration through a drawn line, or the embodied energy indexically evident through a thumb imprint on a clay model. Expressiveness and empathy 116 animation: an interdisciplinary journal 4(2) at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com are closely linked (Green, 2007), and simulation theory suggests that the quality of the artist’s gestures embedded in the work can induce empathetic engagement through active simulation, and that ‘visible traces of goal-directed movements’ will activate the mirror-neuron system (Freedberg and Gallese, 2007: 202). This is described as ‘feeling the movement behind the mark’, and helps explain some of the aesthetic appeal of expressive work that foregrounds expressive strokes, fingerprints, gouges, or any indexical artefacts of embodied gesture involved in its construction (Power, 2008: 43). 3D CG is usually indexically dehumanized through absence of such qualities. Marjane Satrapi, writer and co-director of Persepolis (2007), who uses pen-and-ink says that there’s a perfection about computer generated animation that doesn’t look ‘natural’ and a ‘coldness’ she doesn’t like, compared to the expressive ‘vibrations of the hand’ that give life to hand-drawn animation (Satrapi, 2008). Klein (2000: 24) observes a similar phenomenon with ani-morphs in which ‘the audience is supposed to sense the hand intruding’, and proposes an aesthetic that foregrounds production methods. Aardman’s supervising director Richard Goleszowski insists that audiences can tell the differ- ence between CGI, drawn and stop frame, and that in contrast with the automated, synthetic, even plastic-looking nature of much 3D animation, if ‘you know it’s a hunk of plasticine and occasionally you can still see the fingerprints – some of the process is revealed and that actually helps you tune in to the character’ (Strike, 2007). Such index- ical expressiveness ‘captures the ontological spirit of form and its shaping forces’, as Terzidis puts it (2003: 1). ‘It manifests form’s meaning, significance and quintessence.’ Creative expression in 3D CG animation Expressiveness is about ‘personality, individuality and idiosyncrasy’ (Terzidis, 2003: 1), and expressive works in animation are more likely to be independent or auteur-type works, whereas many of the more formulaic animated features belong at the mainstream end of the spectrum. Though many of the latter will have an identified auteur such as Lasseter or Byrd, they are less likely to be driven by personal expe- riences or by strong empathy with others’ experience, as is for example, Ari Folman’s Waltz with Bashir (2008), about personal experiences connected with the 1982 Sabra and Shatila massacres in Lebanon, or Landreth’s Ryan (2004), about Canadian animator Ryan Larkin, an Oscar nominee in 1969, who had fallen on hard times. It is intriguing that such expressive animation should come as documen- tary, traditionally the archetypal realist form. Independent animation will not necessarily be more successfully expressive than mainstream output, but in common with the film and music industries, cultural and economic factors, such as the scale, automation and economy of Power Animated expressions 117 at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com production and distribution, have a huge impact on how creatively or idiosyncratically expressive 3D animated output is likely to be. Culture and economics influence narrative style too, and no matter how clever the graphics, formulaic storytelling will tend to be not only less expres- sive but less creatively satisfying. Creative scriptwriting in Hollywood is undervalued9 and use of formulaic narrative structures such as Joseph Campbell’s a hero’s journey in a fairytale setting lead to lack of originality in narrative (McClean, 2007: 219), a scenario all too common in CG features. Many artists or directors of moving image projects choose anima- tion over live action explicitly for its expressive potential, and Linklater’s Waking Life (2001) and Folman’s Waltz with Bashir (2008) are just two examples (extensive live-action was shot for both). The latter was shot first in studio on video and subsequently animated using a combination of cel animation, Flash, 3D CG and live-action roto- scoping (Figure 3). The producers pitched to 40 international TV station managers at the Toronto Documentary Film Festival, but only two expressed any interest, eventually supporting the film. ‘The other 38 attendants at the pitch did not comprehend why a film should be animated’ (Folman, 2008). In contrast, reaction to the finished docu- mentary seemed to justify Folman’s animated approach. In Variety, Leslie Felperin (2008) observed that a subject that might have been just another war documentary, was ‘transmuted via novel use of anima- tion into something special, strange and peculiarly potent’. Jonathan Romney in The Independent (2008) wrote that choosing to ‘depict his quest in impressionistic, often dream-like animation initially seems like an outrageous poetic liberty – but it makes his film all the more personal and gives it the urgency of a true cri-de-coeur’. Emotional rather than visual realism is the stylistic choice for Folman and also for Chris Landreth in his Oscar-winning documentary short film Ryan (2005). Landreth is interested in what he calls psycho- realism, ‘in co-opting elements of photorealism to serve a different purpose; to expose the realism of the incredibly complex, messy, 118 animation: an interdisciplinary journal 4(2) Figure 3 Still image from Waltz with Bashir (Ari Folman, 2008). at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com chaotic, sometimes mundane, and always conflicted quality we call human nature’ (Landreth, 2004). To achieve this he uses 3D graphics to reflect the characters’ states of mind, and the 3D scenes include expressive non-photorealstic rendering and multiple, warped, non- linear, simultaneous perspectives (Figure 4). There is some indication of expressive advances in CG features. Disney-Pixar’s Wall-E (2008) is unusual, in that in contrast to most movies of this genre that rely on high modality dialogue voiced by Hollywood actors, for much of the film, the blips and whirs of robots are the only dialogue. Ben Burtt’s sound design is minimalist and highly expressive, and Wall-E was nominated in all sound and music cate- gories for the 2009 Oscars, and won the Oscar for best animated feature. Dr Seuss’s Horton Hears a Who! (Chuck Jones, 2008) by Blue Sky Studios, the computer animation subsidiary of 20th Century Fox, is adapted from Dr. Seuss’s book of the same name. An animated version was first produced and written by Theodore Geisel (Dr. Seuss) in 1970, and directed by Chuck Jones. Dave Torres, the lead animator, says the film enabled them to ‘push the boundaries of expressiveness for an animated character’ (Bekins, 2008). Co-director Steve Martino tried ‘to bring Chuck Jones-style animation into 3D’ using the original pen and ink drawings as stylistic inspiration, and describes ‘a zone of Seussian exaggeration’, where everything has a ‘very hand-drawn, free-flowing style’, with Who-ville being stylistically ‘the very opposite of the recti- linear and symmetrical graphics that computers are fond of making’ (Bekins, 2008). While the finished film fits comfortably in the main- stream genre, aspirations to more expressive output are clearly evident amongst some of those working in CG features. Of course, 3D CG is modular and object-oriented (both under the hood and in the animation process), making 3D metamorphosis, figure- ground reversals and some distortion effects difficult, even counter- intuitive. Despite being digital, it lacks some of the flexible simplicity of pen and ink, and the ‘dream of plasmatic freedom’, signified for Power Animated expressions 119 Figure 4 Still image from Landreth’s psychorealistic Ryan (2005). at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com Eisenstein in the fluidity of the animated line (Sobchack, 2008: 262), can prove somewhat less lucid in 3D space. Alternative perspectives Two of the primary cornerstones of 3D animation are single-point perspective, the sine qua non of 3D CG technology, and its co-signifier in naturalistic illusion: photorealistic rendering. In exploring how 3D animation is expanding its stylistic horizons, it is worth examining briefly how both these conventions have recently been explored in terms of their expressive potential. ‘The system of perspective is fundamentally naturalistic’ (Kress and Van Leeuwen, 2006: 131), and 3D animation is defined primarily by consistent linear single-point perspective. Nevertheless, this conven- tion is paradoxical, as it offers a point-of-view, often socially deter- mined, that is encoded as though it is ‘subjective, individual and unique’, while it simultaneously rests on ‘an impersonal, geometric foundation, a construction which is a quasi-mechanical way of “record- ing” images of reality’ (p. 129). Thus, socially constructed viewpoints are naturalized, and single-point perspective remains one of our most pervasive abstractions. Nevertheless, experimental evidence shows that ‘no culture-dependent learning is required for adequate percep- tion of perspective images’ (Zorin, 2002: 119), and linear perspective is a good approximation of the human visual system and will always be the norm when depicting 3D scenes in 2D. Although the earliest documented observation of perspective has been dated to approximately 4000 BC (Coleman and Singh, 2004: 129), it failed to take off until the early renaissance in Europe and remained dominant in Western art until the 20th century when, unshackled from service to verisimilitude by photography and perhaps inspired by non- Euclidean perspectives or by Einstein’s theory of relativity, modernist artists painted convolutions of space-time rather than just space. Cubists and later Surrealists (in particular René Magritte) played thematically with perspective and challenged conventions of natural- ism and realism. Alternative projections don’t just belong to ancient and modernist art, and to non-Western traditions,10 but to mainstream animation too; in cartoons, it became a convention for foreground char- acters and backgrounds to be treated differently, not just stylistically but in terms of their projection. Many experimental alternatives to single-point perspective systems have been tried in CG with varying degrees of success, including systems for computing Escher-like impossible scenes, 3D warps and combined multiple projections (Agrawala et al., 2002: 158). Andrew Glassner, a CGI technical researcher turned writer-director, has worked on what he calls non-linear or free-form optics for computer graphics (Figure 5), as detailed in a paper titled ‘Cubism and Cameras’ (2000). 120 animation: an interdisciplinary journal 4(2) at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com He insists these are much more than curiosities, and that just as the cubist painters found new angles to communicate, his ‘animated, fluid form of cubism’ using free-form optical models can do likewise with synthetic images and animated sequences’ (p. 1). Agrawala et al. (2002, amongst others) have also developed systems for simultaneous 3D multiprojection, as artists often use multiple projections for ‘express- ing a mood, feeling or idea’ or for improving comprehensibility of the scene (p. 155). Karan Singh, who had worked on the development of Maya, was R&D Director on the production of Ryan (2004). He was motivated by the surreal storyboard and inspired by the artwork of people like Picasso, Dali, Albright and Bacon to develop a system (subsequently implemented in Maya), that could render out multiple simultaneous projections and camera angles to a single frame. These warped non- linear projections helped express the characters’ psychological perspectives and could also be used to create cinematic mood or create a sense of uneasiness in the audience (Coleman and Singh, 2004). Capacity for simultaneous multi-projections needs to be built into the compositional space and many problems remain to be solved with such systems.11 However, the seamless integration of multiple perspec- tives with appropriate lighting, shadows and effective artistic control may provide a projection palate freed from the hegemony of the quasi- objectivity of perspective, enabling the animator to choose dialogically what aspects of the scene to accentuate, from what angle and to what aesthetic effect. Expressive (or non-photorealistic) rendering Alternative rendering algorithms can be used to output a naturalistic scene in eclectic styles from cartoon-style to Canaletto. Some similar off-the-shelf effects are common in 2D paint applications such as Adobe Photoshop, and simpler image-based post-processing tech- niques can also be used in 3D. However, effective non-photorealistic rendering in 3D applications is a more complex affair. Non- photorealistic rendering (NPR) is the collective name for a range of Power Animated expressions 121 Figure 5 Image from a storyboard for a thriller using expressive projections. Artwork by Tom McClure from Glassner (2000). at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com techniques that have been developed to render 3D models and animations in alternative modes to the photorealistic norm for the genre, and the term expressive rendering is synonymous and seems preferable to the negatively couched default term NPR. Durand (2002a) argues that, although computer graphics has long been defined as a quest to achieve photorealism, ‘as it gets closer to this grail, the field realizes that there is more to images than realism alone. Non-photorealistic pictures can be more effective at conveying information, more expressive or more beautiful’ (p. 11). Typically, NPR flattens output from 3D animations and offers a plethora of output effects. Amongst these have been simulated artistic media (e.g. watercolours, oils, charcoal or stained glass), painterly styles (e.g. impressionist, pointillist or Van Gogh), and even cartoon or toon- style rendering that has been integrated into mainstream 3D applica- tions such as Maya (Figure 6). In Ratatouille (Brad Bird, 2007), a brief 2D effect where a book illustration of the famous chef Gusteau comes to life and talks to the protagonist Remy was created using a regular 3D scene output with a toon-style render. If well handled, expressive rendering has the potential to mean- ingfully expand the palate of artistic effects and can open up expres- sive opportunities for animation artists. On the down side, many of the effects implemented using NPR have been trivial or gimmicky. There is a difference between images of expressive or artistic merit and ones that look vaguely artistic, and switching on a Van Gogh style filter will most likely generate kitsch.A paper titled ‘A Real-Time “Boiling Style” Nonphotorealistic Rendering System for Low Fidelity Animation’ describes an NPR system that ‘produces jittery style drawings’ that mimic ‘low-fidelity animation conventions to produce rendered models that look hand drawn rather than machine inter- preted’ (Hesselgren and Naftel, 2003).Without denigrating the system in question, the description does beg the question as to whether algorithms that automate ‘jittery style drawings’ that ‘look hand drawn’ can help output genuinely expressive animation. Animation 122 animation: an interdisciplinary journal 4(2) Figure 6 Expressive rendering – contrasting naturalistic and expressively rendered duplicate objects from a single 3D scene (using a Maya toon-style render). at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com that mimics expressiveness, perhaps, but much depends on imple- mentation and on how much control the animator has over the output. Although some NPR work has focused on derivative artistic effects or effects derived from other media, expressive rendering effects do offer creative options for animators if well implemented. Much depends on whether the effects are relatively pre-determined or auto- mated, or whether they offer interactive, intelligent and intuitive tools for expressive artistic control. Further creative developments are feasible, with recent research focusing on findings from cognitive psychology and empirical research on perception, including neurosci- entific work on visual perception, which is one of the most advanced fields of neuroscience. Conclusion: evolving an expressive 3D aesthetic Mainstream 3D animation culture continues to be driven by a natura- listic agenda, and the convergence of live-action SFX and animation is one of the driving forces. For example, James Cameron’s Avatar (mostly CG with some live-action and due for a 2009 release), uses seamless performance capture, so that the actors can be directed in real-time with the director simultaneously viewing the actions of the CG characters set in 3D CG environments. The aim is a seamless synthesis of the real and the virtual, even during direction and produc- tion. Avatar is made in stereoscopic 3D for 3D projection. Assessed as a gimmick during the early 1950s and again in the early 1980s, the 3D projection phoenix has arisen again, perhaps to persist this time, particularly for mainstream feature animation, as the technology has improved immeasurably and 3D CG production techniques are highly suited to flexible and economically viable production in this mode.12 Heightened realism as immersive spectacle is the primary aesthetic goal of 3D projection and it is another indication that Hollywood is still hot on the trail of the virtual reality grail. Physics simulation, seamless performance capture and 3D model generation from video, for example, are all particularly active research and development areas, and all are linked to the naturalistic agenda. As the technology gets faster and more sophisticated, the limitations that have so far precluded attempts at fully photorealistic 3D anima- tion will eventually be surmounted. But material constraints can have creative advantages, and often lend a medium its charm, requiring artists/animators to adopt expressive solutions. Durand (2002b) outlines three ways to handle them: by elimination (finding technolog- ical solutions); by compensation (alternative hacks or workarounds); or by creative accentuation, which he says ‘can bring important richness to pictures’ (p. 27). Accentuation foregrounds constraints stylistically for expressive aesthetic effect. Less is more can apply to Power Animated expressions 123 at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com creative freedom too, and the capacity to do work with few material constraints in a digital environment may prove to be a mixed blessing. These developments are fine as long as more expressive approaches are not culturally and technologically sidelined. Presently there are many modes in which 3D animation can be used as an expressive storytelling medium. Expressive 3D models can be designed by choice using distorted geometry for non-naturalistic modelling effects. Anima- tion data (including motion capture data) can be augmented stylisti- cally either by keyframing, or by using algorithms enabling, for example, hyper-energized, fluid, languid, robot-like, rubbery, sticky or other effects. As outlined earlier, expressive effects can be achieved through implementing warped or multiple cubist-like camera projec- tions, or through expressive rendering techniques or post-rendering effects. 3D is often digitally integrated with live action and other animation techniques, and this hybrid aesthetic is one obvious way forward for expressive animation. Mixing live action and animation is hardly new (the Fleischer brothers composited live action and animation in the 1920s in the Out of the Inkwell series starring Koko the Clown). Orig- inating assets in digital format simplifies manipulation and synthesis, however, as evidenced by many modern hybrids: for example, Luc Besson’s Arthur and the Invisibles (2007) and the Stuart Little series. Both Horton Hears a Who! (Chuck Jones, 2008) and Kung Fu Panda (John Stevenson and Mark Osborne, 2008) are 3D animated movies, but both use brief 2D animated sequences to code dream/fantasy altered states of consciousness. In the former there are two, one based stylistically on Dr. Seuss’s pen-and-ink illustrations from the book and another based on anime, and there is one in the latter, styled on Chinese shadow puppetry. The flexible nature of a digital environment enables this fruitful fusion of styles and a simultaneous dialogical synthesis of alternative animation traditions for expressive aesthetic effect. Looking to the future, there are a number of developments that would help enable the evolution of an expressive aesthetic in 3D animation. Interface design is an area in which ‘the amount of user vs. computer control is an exciting issue’ (Durand, 2002a: 17), and many animation systems are not intuitive or easy to use, as focus has been on high-profile developments in naturalistic functionality and effects, with comparatively little development in intuitive interfaces. There is the real danger that computerization produces what Terzidis (2003) calls Whorfian effects,13 where ‘through the use of commercial applications and the dependency on their design possibil- ities, the designers’ work is at risk of being dictated by the language- tools they use’ (p. 69). This conundrum can be observed regularly when novel or ‘cutting-edge’ effects (for example, morphing) become fleet- ingly fashionable and are consequently over-used in adverts or films to the extent that they soon become clichéd or passé. Terzidis suggests 124 animation: an interdisciplinary journal 4(2) at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com that, on the other hand, what he calls algorithmic design enables creativity and discovery of new forms and novel phenomena (p. 68). This involves more open-ended, complex or programmable options being available to the animator or designer as opposed to default presets or a relatively finite range of ready-made effects. Such algorith- mic design capabilities are already enacted on some systems (through scripting, for example), but their current effectiveness depends on an animator’s capacity (or at least that of one of the team) to program them effectively. It is a lot to ask of animators (or anyone else) that they should excel in visual and motion design, character design and storytelling, and then be capable programmers as well, so the implementation of intuitive interfaces for algorithmic design processes is critical. Many modern interfaces are still encumbered with numerical or other non-intuitive forms of input, often into a bewildering host of variables with indeter- minate functionality. Interfaces with iconic or indexical relationships to real-world equivalents enabling embodied interaction can be more intuitive and easy to use for expressive effect, providing they do not bind the artist to the constraints of their real-world equivalents. What might be called a magical-realist approach to interface design would be optimal in this context, with functionality based on real-world analogues where helpful, but with ‘magical’ digital capability besides. Although limited gesture-based input already exists, for example on paint systems, complex gesture-based interfaces (and those with haptic feedback) are likely to become more common, and evolving versions of these can be seen with recent products such Nintendo’s Wii, and those using Apple’s MultiTouch and Microsoft’s Surface technologies. Gesture-based interfaces for 3D animation systems would be particu- larly apposite and would lend a reflexive element to animating – the animator animates character gestures in software through using gesture. Use of gesture on input could facilitate more expressive and intuitive computer-based animation, including virtual gesture-based aesthetic phenomena such as feeling the movement behind the mark. In terms of theory, emotion is an important basis for understanding expressiveness. Despite their hypothesizing the concept of a sensory coding orientation, Kress and Van Leeuwen (2006) acknowledge that affect/emotion has been ‘too thin a thread in the tapestry’ that comprises their influential semiotic theory of imagery and visual design (p. 267). Emotion is both psychological and visceral, embodied minds and their physical and cultural environments are complex systems, and can best be explained, as Antonio Damasio argues (Liston, 2001), by blending theories at several levels of organization, from mole- cules and cells, to large-scale systems, and physical, social and cultural environments. Symbiotic paradigms such as complexity theory, phenomenology and biosemiotics can embrace the connectedness of biology and culture and provide fertile environments in which embodied approaches to expressive aesthetics can flourish. Power Animated expressions 125 at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com Although we are ‘inextricably shaped by embodied action’ (Gibbs, 2005: 276), ‘our linguistic, imaginative and symbolic capacities provide a degree of freedom from the demands of our environment’ (Modell, 2006[2003]: 138). While the semiotic freedom of creativity is the fullest expression of human communication (Wheeler, 2006: 133), naturalis- tic or other conventions impose ‘limitations of conformity on sign- making’, constraining its ‘semiotic scope’ (Kress and Van Leeuwen, 2006: 12). By qualitatively transcending its calculative and quantitative genesis, the contemporary expressive turn in 3D CG animation has a creatively liberating potential through which we may be moved, even transformed by the reflexive resonance of the human touch. Acknowledgements The author wishes to thank the editor, Suzanne Buchan, and the anonymous reviewers, for the detailed feedback that was invaluable in finalizing this paper. Notes 1 Photorealism as a style in painting and Hyperrealism, a more recent and expressive variant, add somewhat to the potential for confusion in terminology. 2 Although 3D computer graphics has its own compound term virtual reality that is relevant in this context, it suffered from over-hyped usage in the 1990s and still suggests headsets and real-time simulation in virtual space. 3 It is worth pointing out that denotation can be seen as just a dominant naturalizing connotation, an illusion perpetuated by a naturalist ideology. 4 Labels for art movements such as Expressionism, Abstract Expressionism and cinema’s German Expressionism muddy the terminological waters considerably. 5 This concerns how we are able to have strong emotional responses to fictional situations we know are not true. Rolls suggests, like Robinson, that though the emotions may be real enough in an arts context, there is slower top-down cognitive mediation of the emotion that contextualizes response. 6 Taymor’s multiple award-winning adaptation of Disney’s animated film The Lion King (1994) for the stage was produced by Disney Theatrical Productions, who have similarly adapted other animated works for Broadway including The Hunchback of Notre Dame (1996) and Beauty and the Beast (1991). 7 This phenomenon is the basis of visual illusions such as The Rubin vase/profile figure/ground illusion and the Necker Cube ambiguous line drawing illusion. 8 Conversely, dissonance between modalities can evoke effects such as tension or irony. 9 The 2007/8 100-day US writers’ strike had a minimal effect on animation production (with some exceptions such as The Simpsons), as for historical reasons many animation writers belong to the Animation Guild whereas 126 animation: an interdisciplinary journal 4(2) at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com the strike was called by the more live-action oriented Writers Guild of America. 10 Many other cultures used alternatives to linear perspective, e.g. Japanese and Chinese artists typically used oblique parallel projection in their drawings. 11 Multiple projections, metamorphosis (morphing), image distortion and NPR effects can all be added relatively easily as post-render or 2D image effects. However, these can be qualitatively different from more challenging 3D implementations that are integral to the scene and rendered as such. 12 On the commercial front, it also helps differentiate a lucrative theatre release market from a domestic one, at a time when high definition technologies such as Blu-ray are leading to market convergence. 13 The controversial Whorf or Sapir-Whorf hypothesis in linguistics suggests that different language patterns give rise to different patterns of thought and that thought is determined to a great extent by language. References Agrawala, Maneesh, Zorin, Denis and Munzner, Tamara (2002) ‘Artistic Multi - projection Rendering’, in Perceptual and Artistic Principles for Effective Computer Depiction (Siggraph Course Notes), pp. 155–92. San Antonio, Texas. URL (consulted 19 May 2008): http://people.csail.mit.edu/fredo/SIG02_ ArtScience/DepictionNotes2.pdf Bekins, Russell (2008) ‘Giving a Whoot About Horton’, Animation World Magazine. URL (consulted 16 May 2008): http://mag.awn.com/index.php? ltype=pageone&article_no=3580&page=1# Butler, Christopher (2004) Pleasure and the Arts: Enjoying Literature, Painting, and Music. Oxford: Oxford University Press. Cavanagh, P. (2005) ‘The Artist as Neuroscientist’, Nature 434: 301–7. Coleman, Patrick and Singh, Karan (2004) ‘Ryan: Rendering Your Animation Nonlin- early Projected’, in Proceedings of the 3rd International Symposium on Non- photorealistic Animation and Rendering, pp. 129–56. Annécy, France: ACM. URL (consulted 8 July 2008): http://portal.acm.org/citation.cfm?id=987678& jmp=indexterms&coll=portal&dl=ACM Davis, Michael (1999) The Poetry of Philosophy: On Aristotle’s Poetics. South Bend, IN: St Augustine’s Press. Durand, Frédo (2002a) ‘An Invitation to Discuss Computer Depiction’, in Percep- tual and Artistic Principles for Effective Computer Depiction (Siggraph Course Notes), pp. 11–25. San Antonio, Texas. URL (consulted 19 May 2008): http:// people.csail.mit.edu/fredo/SIG02_ArtScience/DepictionNotes2.pdf Durand, Frédo (2002b) ‘Limitations of the Medium and Pictorial Techniques’, in Perceptual and Artistic Principles for Effective Computer Depiction (Siggraph Course Notes), pp. 27–45. San Antonio, Texas. URL (consulted 19 May 2008): http://people.csail.mit.edu/fredo/SIG02_ArtScience/DepictionNotes2.pdf Felperin, Leslie (2008) ‘Waltz with Bashir Movie Review from the Cannes Film Festival’, variety.com. URL (consulted 24 May 2008): http://www.variety.com/ index.asp?layout=festivals&jump=review&id=2531&reviewid=VE1117937141 Folman, Ari (2008) ‘Waltz with Bashir: Production Notes’, Waltz with Bashir. URL (consulted 23 May 2008): http://waltzwithbashir.com/film.html Power Animated expressions 127 at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com Frayn, Michael (2007) The Human Touch: Our Part in the Creation of a Universe. London: Faber & Faber. Freedberg, David and Gallese, Vittorio (2007) ‘Motion, Emotion and Empathy in Esthetic Experience’, Trends in Cognitive Sciences 11: 197–203. Gibbs, Jr, Raymond W. (2005) Embodiment and Cognitive Science. Cambridge: Cambridge University Press. Glassner, Andrew (2000) ‘Cubism and Cameras: Free-Form Optics for Computer Graphics’. URL (consulted 13 March 2009): http://research.microsoft.com/ research/pubs/view.aspx?msr_tr_id=MSR-TR-2000–05 Gooch, Bruce (2002) ‘Ramachandran and Hirstein’s Neurological Theories of Aesthetic for Computer Graphics’, in Perceptual and Artistic Principles for Effective Computer Depiction (Siggraph Course Notes), pp. 193–204. San Antonio, Texas. URL (consulted 19 May 2008): http://people.csail.mit.edu/ fredo/SIG02_ArtScience/DepictionNotes2.pdf Green, Mitchell S. (2007) Self-Expression. Oxford: Oxford University Press. Harth, Eric (2004) ‘Art and Reductionism’, Journal of Consciousness Studies 11: 111–16. Hemingway, Andrew (2007) ‘The Realist Aesthetic in Painting’, in Matthew Beaumont (ed.) Adventures in Realism, pp. 103–24. Oxford: Blackwell. Hernandez, Maria Lorenzo (2007) ‘The Double Sense of Animated Images’, Anima- tion Studies 2. URL (consulted 28 April 2008): http://journal.animationstudies. org/category/volume-2/maria-lorenzo-hernandez-the-double-sense-of-animated- images/ Hesselgren, A. and Naftel, A. (2003) ‘A Real-Time “Boiling Style” Nonphotorealistic Rendering System for Low Fidelity Animation’, in Proceeding (396) Visualiza- tion, Imaging, and Image Processing, Benalmádena, Spain. Calgary, AB: Acta Press. http://www.actapress.com/PaperInfo.aspx?PaperID=14295&reason=500. Klein, Norman M. (2000) ‘Animation and Animorphs’, in Vivian Sobchack (ed.) Meta-Morphing: Visual Transformation in the Culture of Quick Change, pp. 21–39. Minneapolis: University of Minnesota Press. Kress, Gunther and Van Leeuwen, Theo (2006) Reading Images: The Grammar of Visual Design. London: Routledge. Landreth, Chris (2004) ‘Maya Masters – 2004’, Autodesk. URL (accessed 8 July 2008): http://www.autodesk.com/eng/etc/mayamasters/masters/chris_landreth. shtml LeDoux, J. (1998) The Emotional Brain. London : Weidenfield & Nicolson. Liston, Conor (2001) ‘An Interview with Antonio R. Damasio’, The Harvard Brain, Vol. 8. URL (consulted 30 Oct. 2008): http://evans-experientialism.free webspace.com/damasio.htm Manovich, Lev (2001) The Language of New Media. Cambridge, MA: MIT Press. Manovich, Lev (2006) ‘Image Future’, animation 1(1): 25–44. McClean, Shilo T. (2007) Digital Storytelling: The Narrative Power of Visual Effects in Storytelling. Cambridge, MA: MIT Press. McCloud, Scott (1994) Understanding Comics: The Invisible Art. New York: HarperCollins. Modell, Arnold H. (2006[2003]) Imagination and the Meaningful Brain. Cambridge, MA: Bradford Books. Power, Patrick (2008) ‘Character Animation and the Embodied Mind–Brain’, Animation 3(1): 25–48. Ramachandran, V.S. (2004) A Brief Tour of Human Consciousness: From Impostor Poodles to Purple Numbers. New York: Prentice Hall. 128 animation: an interdisciplinary journal 4(2) at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com Robinson, Jenefer (2005) Deeper Than Reason: Emotion and Its Role in Litera- ture, Music, and Art. Oxford: Oxford University Press. Rolls, Edmund (2005) Emotion Explained. Oxford: Oxford University Press. Romney, Jonathan (2008) ‘Cannes Film Festival in Review’, The Independent. URL (consulted 24 May 2008): http://www.independent.co.uk/arts-entertainment/ film-and-tv/news/cannes-film-festival-in-review-830174.html Satrapi, Marjane (2008) ‘Persepolis: Production Process’. URL (consulted 26 April 2008): http://www.filmeducation.org/persepolis/production-process.htm Sobchack, Vivian (2008) ‘The Line and the Animorph or “ Travel Is More than Just A to B”’, animation 3(3): 251–65. Strike, Joe (2007) ‘Creature Comforts, American Style’, Animation World Magazine. URL (consulted 16 May 2008): http://mag.awn.com/index.php? ltype=search&sval=style&article_no=3308&page=3 Taymor, Julie (1998) ‘Theater and the Imagination’. URL (consulted 23 June 2008): http://www.ted.com/talks/view/id/277 Terzidis, Kostas (2003) Expressive Form. New York: Spon Press. Twain, Mark (1924) ‘Mark Twain’s Autobiography’, Vol. 1, Project Gutenberg. URL (consulted 17 July 2008): http://gutenberg.net.au/ebooks02/0200551h.html Wheeler, Wendy (2006) The Whole Creature: Complexity, Biosemiotics and the Evolution of Culture. London: Lawrence & Wishart. Zeki, S. (1999) Inner Vision: An Exploration of Art and the Brain. Oxford: Oxford University Press. Zorin, Denis (2002) ‘Perception of Pictures’, in Perceptual and Artistic Principles for Effective Computer Depiction (Siggraph Course Notes), pp. 115–45, San Antonio, Texas. URL (consulted 19 May 2008): http://people.csail.mit.edu/ fredo/SIG02_ArtScience/DepictionNotes2.pdf Patrick Power is a Senior Lecturer in Digital Media and Design at London Metropolitan University. He worked for over a decade in the creative industries, specializing in graphics, video and 3D animation, before completing an MA in Interactive Multimedia at the Royal College of Art. He has since worked for 12 years as an academic in Higher Education, and his research interests include animation, emotion, play, narrative and the synthesis of science and the arts. He currently teaches theory and practice of digital media design at London Metropolitan University, specializing in creative 3D animation. Address: Digital Media & Design, London Metropolitan University, Room 315, Ladbroke House, 62–6, London N5 2AD, UK.[email: p.power@londonmet.ac.uk] Power Animated expressions 129 at AJOU UNIV on December 4, 2009 http://anm.sagepub.comDownloaded from http://anm.sagepub.com