Seeing Movement: On Motion Capture Animation and James Cameron’s Avatar

Seeing and Being in Avatar

Much has been written on the themes of Avatar’s ecological concerns and politics (Monbiot, 2010; Žižek, 2010), particularly on the film’s portrayals of the destruction of indigenous populations and environments and, correspondingly, its ‘great white saviour’ narrative. While these ideas form the film’s dominant themes, the narrative arc of Avatar is also very much concerned with the dialectic of seeing and being. Its central plot device of the avatar – the figuration of Jake Sully (Sam Worthington) which drives the narrative – plays precisely on this opposition. Like major and minor musical keys, seeing and being in Avatar transpose each other in harmony and counter-point. I argue that this is done in three ways: (i) by the act of seeing; (ii) by being woven into the fiction of Na’vi culture; and (iii) by being a metaphor resonating through the film’s plot.

Firstly, the act of seeing is important because it prescribes experience. Looking at his avatar for the first time, floating inert in a tank, Jake remarks, ‘looks like him’, meaning Tommy, his deceased twin brother whom he has replaced on Pandora. Fellow avatar user Norm (Joel David Moore) replies: ‘no, looks like you’. This exchange indicates the significance of the visual: the reality of the avatar is understood by how it is seen. In fact, the avatar looks like both Jake and Tommy as they are twins, yet Jake relates the avatar to his brother because he is mourning his death. Norm, on the other hand, relates the avatar to Jake because the point of the mission is for Jake to take it: ‘this is your avatar now, Jake’. Experience, particularly in Pandora, is often assigned to sight, as signified by numerous close-up shots of eyes. Every significant experience in Pandora undergone by Jake concludes with a close-up of his eyes, a visual motif that also bookends the film, with the second shot showing Jake awakening from a dream of flying, and the last showing the opening of his eyes after the transfer of his consciousness to his avatar form. These close-ups of the eyes mark the first transition from being/experience to seeing. Many others are scattered throughout the film: after Jake falls asleep for the first time in Pandora (and his first experience in his Na’vi avatar); after learning about the Toruk Makto, or Na’vi who control the dragon-like leonopteryx, which Jake later becomes; after the Home Tree has been destroyed; and just before Jake dies in his human form in Pandora. When Grace (Sigourney Weaver) admonishes Jake that he should better understand Pandora with Neytiri’s (Zoe Saldana) guidance, her instruction is for him to see: ‘This isn’t just about eye–hand coordination out there … Try to see the forest through her eyes.’

Yet what appears to be sight frequently turns out to be blindness, undermining the validity of the experience it prescribes. Jake’s (and by extension the audience’s) sight is often erroneous. Stranded in the Pandoran jungle, Jake, like any well-trained marine, ignites a torch so that he can see. Nevertheless, he still fails to see his immediate danger, is attacked by viperwolves and has to be rescued by Neytiri. Significantly, once the threat is averted, she extinguishes the torch with obvious irritation, ignoring Jake’s reactions of chagrin. As the fire sizzles in the river, the scene is plunged into darkness, and for a moment we share Jake’s discomfort at not having the firelight. However, even as he retrieves his torch, Jake, and we along with him, realize the fallacy of that sight by light – the forest is not dark, but glows with bioluminescence, which can paradoxically only be seen by not seeing (with light). Our own visual senses are sharpened as well – we see more clearly the spots on Jake’s and Neytiri’s faces, features previously obscured by day and firelight. This episode, the first in the film to take place in the darkness of Pandora, thus clues us in on the precariousness of vision in this world: it is not all that we can see. Similarly, in the fictionally important Hallelujah Mountains, where the final battle takes place and in the middle of which the Tree of Souls is located, sight is paramount as flight instruments fail due to the area’s Flux Vortex, an effect from Pandora’s magnetic field forces. The pilot Trudy (Michelle Rodriguez) calls flying the Hallelujah Mountains VFR – ‘visual flight rules’: ‘it means you gotta see where you’re going.’ The tactical strategy for the final battle between the human SecOps and the Na’vi rests on precisely this opposition between seeing and not seeing: due to ‘the flux’, the pilots and gunners have to see with their eyes rather than with their instruments; as Jake says, ‘they’ll have to fire on line of sight.’ Yet the humans’ sense of sight is completely undermined, for the Na’vi count on precisely the fact that the SecOps pilots cannot see them, hidden in ambush above. The importance of the trading between seeing and blindness in the film is thus shown clearly in the set-up of this climactic battle.

Furthermore, seeing is deemed an important part of Na’vi culture, beginning with the important greeting, Oel ngati kame, meaning ‘I see you’, or, as Norm explains, ‘I see into you.’ Conversely, one of their most offending insults refers to the lack of sight, such as rival Na’vi Tsu’tey’s mocking comment on Jake’s initial inability to ride a direhorse: ‘this alien will learn nothing, a rock sees more.’ Before transferring Jake’s mind into his avatar at the Tree of Souls, Neytiri kisses his eyes, presumably because of the link to the importance of sight in Na’vi culture. Yet, at the same time, the ultimate experience for the Na’vi lies in being and feeling, invoked particularly by tsaheylu, the intensely intimate neural bond between Na’vi and all other creatures of Pandora that lies not in seeing, but being with it, reaching a closeness of experience so as almost to be at one with it. Neytiri instructs Jake in his first experience of tsaheylu with the direhorse: ‘Feel her. Feel her heartbeat. Her breath. Feel her strong legs.’ The bond, so linked to experience, is also linked to blindness: when Jake picks up the end of his neural queue – the conduit for tsaheylu – for the first time, staring at it in wonder, Grace says to him, ‘don’t play with that, you’ll go blind.’ Again (beyond, at least, the gendered metaphor of the phallus in the myth of masturbatory blindness in Western culture), seeing and feeling trade off each other, continuing the dialectic in the film.

Finally, sight becomes a metaphor. Lost and shunned after he reveals his betrayal, Jake narrates: ‘Outcast. Betrayer. Alien. I was in the place the eye does not see.’ To be expelled is not to be seen. Blindness applies even to mechanical vision: when Sully smashes the camera lens of the bulldozers razing down the Omaticaya clan’s Home Tree, its operator, manipulating the machine remotely from a control room, cries out, ‘dude, I’m blind!’

Yet, Jake’s most significant experiences in Pandora lie not in seeing, but in movement. When he first occupies the avatar’s body, his appreciation of his new world is clear: he breathes in the air, he digs his toes into the mud, he crunches into the fruit tossed to him by Grace. Literally springing out of paralysis, Jake is overjoyed simply at his new sense of being. This is subsequently ramped up when he discovers flying on his ikran – ‘I was born to do this’ – dovetailing into his sense of purpose and unwittingly underscoring his destiny to be a Toruk Makto. The Na’vi’s sense of being and being in their environment is particularly contrasted against the humans enclosed inside the steel exoskeletons of the Amplified Mobility Platform suits donned by the SecOps forces of the Resources Development Administration (RDA). Covered by a protective shield, they are screened from every experience of being in Pandora – they do not feel any part of it; they are unable even to breathe in the air. To Jake, Neytiri accuses ‘Sky people’ (humans) of not being able to see: ‘Sky people do not learn, they do not see.’ Jake replies: ‘Well, then teach me how to see.’ ‘No one can teach you to see’ is the response. This exchange is not about seeing in the literal sense, but as a metaphor for being and living, and no one can teach you how to live.

Motion capture and Avatar

In this section, I argue how motion capture technology used in the making of Avatar, particularly via the ‘swing camera’, echoes the seeing/being opposition. In making this argument, I am not positing a self-reflexiveness in the film about how its technology might specifically mirror the film’s thematic concerns of seeing. Rather, I connect the two as an implicit relationship, one which marks a shift to an alternative visual regime for cinema from seeing to being. In that sense, connecting motion capture and seeing in Avatar qua technology and theme are synapses to broader ideas I want to explore in the viewing of cinema – a more intuitive experience of movement, experience, being and seeing.

Mocap in Avatar was particularly plugged for creating the realism of the Na’vi people and particularly the many close-ups of the lead characters. With the usual breathlessness of marketing hype, much of it was branded revolutionary and groundbreaking, acclaimed to be ‘a milestone’ (Chmielewski, 2010) and ‘a new era of motion capture technology’ (Lindsey, nd). However, The Polar Express (Robert Zemeckis, 2004), Monster House (Gil Kenan, 2006) and Beowulf (Robert Zemeckis, 2007) already all predate Avatar as entirely mocapped full-length films. Furthermore, A Christmas Carol (Robert Zemeckis, 2009), released two months before Avatar, had already used a head-rig system to record the minute facial movements of the film’s actors similar to that trumpeted as groundbreaking in relation to Avatar (see McConnon, 2007).

Nonetheless, mocap was used in interesting and noteworthy ways in Avatar. One major aspect of such use is James Cameron’s employment of the ‘swing camera’, a ‘virtual camera that could show him a low-resolution view of Pandora as he shot the performances’. The virtual camera

has no lens at all, only an LCD screen and markers that record its position and orientation within the volume relative to the actors. That position information is then run through an effects switcher, which feeds back low-resolution CG versions of both the actors and the environment of Pandora to the swing cam’s screen in real time. (Thompson, 2010)

The primary purpose of this camera was to allow Cameron ‘to shoot a scene simply by moving through the volume [the large mocap stage]. Cameron could pick up the camera and shoot his actors photographically, as the performance occurred’ (Thompson, 2010), and ‘adjust and direct scenes just as if shooting live action’ (Auza, 2010). The upshot of this swing camera is that what Cameron was viewing through its lens was entirely different from what was in front of it. If the profilmic were the human actors on a mocap stage, to all intents and purposes Cameron was seeing (and directing) through the virtual camera blue Na’vi aliens in Pandora. This upends the camera’s vaunted purpose and role as a visual device – recall, for instance, Vertov’s triumph in his proclamation of the perfection of the cinecamera to ‘penetrate deeper into the visible world’ (Vertov, 1926: 125). Instead, the virtual camera here is about the real time production of its own reality, not unlike the making of machinima in a video game, where the profilmic of the virtual camera are objects generated in accordance with the physics and aesthetics of the game world; indeed, Cameron likens the virtual camera to ‘a big, powerful game engine’ (Waxman, 2007). This powerful rendition of space afforded by the virtual camera has been previously observed, such as Dave Kehr’s (2004) remarks in 2004 on the ‘literally infinite choice of camera angles’ of virtual cinematography akin to games: ‘[The filmmaker] can place his virtual camera at any point in the 3-D space, much as players of video games … can do.’ Likewise, Tobey Crockett (2009) characterizes every point of digital space in the virtual camera as a posthuman subjectivity possessing of agency and authorship. However, the virtual camera in this case, by directing mocap recording in real time, does more than give Cameron liberation of space and physics; more significantly for this argument, the swing camera also gives Cameron the immediacy of the actors’ performance by placing him in the midst – in the sense of his directing – of the virtual camera’s software-generated reality. In other words, Cameron was no longer seeing through the camera; he was partaking – feeling, being – in its own reality. In this reality, the visible – the human actors in front of the camera – is rendered invisible; the human actors are not seen through the camera, only sensed.

In this respect, mocap in Avatar achieves more than merely enhancing the film’s realism; it also reflects on wider questions about how we engage with cinema and the nature of our experience of seeing it. For instance, in view of seeing and revelation in cinema (Bazin, 1967; Keathley, 2006; Turvey, 2008), what, then, can we make of a film in which we are consciously reminded at every turn about not seeing (the actors), but feeling (their movements)? To take this further, what more can be made of Avatar’s other much-hyped technological breakthrough in 3D stereoscopy, whereby seeing is arguably diminished by the supplementary medium between image and audience in the form of 3D glasses and its dimming effect, a shadowing of projected light not unlike the extinguishment of a torch fire but only in order so that we can really see, be it glowing flowers or a third dimension of depth (see also Brown, 2012, this issue)? In one of the most thrilling sequences of the film, Jake swoops up, down and around on his ikran in tandem with Neytiri. The sense of movement – of being – was so strong in that scene I lurched with them in a visceral response, feeling in the movement with them, light-headed in the vicarious flight like the giddiness of a soaring Icarus but with no sun in the dark cinema hall to melt the wax of my feathers. My eyes were wide open and glued to the screen as I took in the aeronautics of Jake and Neytiri in my stomach, on my skin, in my ears, and, last but not least, behind the dimming filter of the 3D glasses. I felt like a child again on my first terrifying roller-coaster ride, eyes squeezed shut in my seat as my car was flung sideways before hurtling downwards – not seeing, only feeling.

Motion capture: movement versus moving object

How, then, does this mocap achieve this synapse from seeing to being? The crux lies in the nature of what is recorded by the camera. Before mocap, the process of filming a referent, whether using film or a digital medium, centred on the recording of light and sound. The modus operandi of the recording process is/was the following equation: light-gathering camera lens + microphone picking up sound = imprinted image combining profilmic visual and aural elements. The image operates in ocular and auricular terms. We both see and hear the image.

Mocap technology significantly changes this, both in terms of what it records and in terms of how it does so. For while mocap (obviously) involves the recording of motion, some questions remain. What exactly is motion? And is recording movement different from recording any other moving object as it happens with a conventional camera?

The Oxford English Dictionary defines motion as ‘the action of moving; movement’. Movement, in turn, is defined as ‘the action or process of moving; change of place, position, or posture; passage from one place or situation to another; activity’. The key lies in the notion of ‘change’ and its inherent idea of derivation, for change is always only arrived at through the passage of one state to another. In other words, it cannot be determined in itself but only via an underlying element progressing from point A to point B.

Movement, being in essence change (of place, position or posture, to take on the OED definition), is therefore also derivative. It likewise does not exist in itself, but only in the progress of an underlying entity from one place/position/posture to another. There can only be movement when, for example, my finger depresses a key on the keyboard. In these terms, it is important to distinguish between (i) the underlying originator (my finger); and (ii) the derivative (movement) that results from my finger progressing from one state to another when it presses down on a key. The two entities, although sharing a relationship (the derivative deriving its existence from the originator), are nonetheless distinct. The former is the data of the originator (my finger), including what it looks like, smells like and so on. The latter, on the other hand, is the data of the derivative (the movement) derived from the action of the originator finger. This data is essentially presented as positional and orientational coordinates, and, when recorded by mocap cameras, is transmitted as algorithms into the computer. The conceptual key here is the separation of derivative from originator, i.e. movement from body, a separation which music critic Kodwo Eshun (1998: 176) similarly seizes on to describe the detachment of beat from music in Breakbeat science:

Breakbeat science, as I see it, is when Grandmaster Flash and DJ Kool Herc and all those guys isolate the Breakbeat, when they literally go to the moment of a record where the melody and the harmony drops away and where the beats and the drum and the bass moves forward. By isolating this, they switched on a kind of electricity, by making the beat portable, by extracting the beat. I call it Motion Capturing … They have a guy that’s dancing slowly, and each of his joints are [sic] fixed to lights and they map that onto an interface, and then you’ve got it. You’ve literally captured the motion of a human. (emphasis added)

As Grandmaster Flash, in Eshun’s description, isolates the beat (‘by making the beat portable, by extracting the beat’), mocap likewise distils movement from the moving body. The accomplishment of mocap technology in cinema, then, is to efface the visual (and sometimes also aural) qualities of the object and apply such isolated movement to the image, so that the moving image is now not of the audiovisual but of the kinaesthetic, with somewhat different accompanying sensorial constituents, as I will now turn to examine.

Experiencing movement: being in the world

If the data recorded by mocap is different from that recorded by conventional cameras, a second question arises: how do we experience this data – movement – as recorded by mocap?

As children, we are conventionally taught our five senses (sight, hearing, smell, taste and touch); we can also turn to Aristotle for an account of this Western five-sense model: ‘In the Psychology we have given a general account of the objects corresponding to the particular sense-organs, to wit colour, sound, smell, flavour, and touch’ (Aristotle, 2007[c. 335 to 323 BC]). These are called the exteroceptive senses, i.e. the senses by which we perceive the external world – the ‘perception of the layout and changes in layout of adjacent material surfaces, that is, the perception of objects and events, respectively’ (Turvey and Carello, 1995: 402). In cinema we apprehend the audiovisual data of an object immediately through our eyes and ears, while others have argued for a hapticity in viewing cinema – a visual feast so intimate and tactile that sight becomes, synaesthetically or otherwise, touch (see Barker, 2009; Beugnet, 2007; Bruno, 2005; Lant, 1995; Marks, 1999; Sobchack, 1992).

In contrast, movement – as derived from a moving object – is something we cannot sense wholly with exteroceptive sensors. We can see, hear and touch a moving object progressing from point A to point B, but we cannot correspondingly experience its movement – its change as derived from a positional differential – in only those terms. An illustration: we can hear a falling tree from the whistle of its canopy through the air to the thud as it hits the ground, and we can see it topple from an upright to a horizontal position, but we cannot see, hear, or touch its change from whistle to thud. Rather, that sense of movement, or kinaesthesia, is apprehended by a cooperation of several proprioceptive and exteroceptive senses comprising receptors in the muscular and vestibular systems, respectively. The brain ultimately reconstructs movement in the body and in the environment from these sensory data. Briefly, the sense of movement is a combination of various sensory receptors consisting of:

All these receptors and sensory modalities in different parts of the body work in concert to apprehend movement, rather than a result of any discrete contribution from each. As Henri Poincaré (1952: 58) puts it, a change of an object’s state or position ‘is always translated for us in the same manner, by a modification in an aggregate of impressions’ (emphasis added). Henri Bergson (2004: 246), describing the movement of his hand from point A to point B, similarly points to this awareness in the body as he notes ‘two things in this movement’:

An image which I see, and an act of which my muscular sense makes my consciousness aware. My consciousness gives me the inward feeling of a single fact, for in A was rest, in B there is again rest, and between A and B is placed an indivisible or at least an undivided act, the passage from rest to rest, which is movement itself. (emphasis added)

To this extent, sensing movement might still not be very different from seeing or hearing; all still require, to some degree, a profound physiological engagement, albeit in different senses (exteroceptive and proprioceptive). Yet, even the above is only a simplification of our sensation of movement, as anecdotal evidence reveals. For instance, neurophysiologist Alain Berthoz writes of his astonishment upon visiting a high school for visually impaired children in Montgeron, France, where he discovered how successful the school’s blind children were at sports – playing ball, fencing, etc. They had been taught to ‘see’ movement not with the sense of sight, but with what he calls ‘blind vision’, or an as yet unclear combination of perceptual and muscular senses. As Berthoz (2000: 51) recounts: ‘this fascinating experience convinced me that ophthalmological tests examine only a tiny fraction of visual function and, in any case, totally ignore perception of movement’ (emphasis added). Furthermore, French Resistance leader Jacques Lusseyran, blind from an accident since the age of eight, writes in his memoirs of how he could ‘see’ by sensing objects: ‘trees and rocks came to me and printed their shape upon me like fingers leaving their impression in wax’, producing ‘sensations as definite as sight or hearing’ (Lusseyran, 1963: 31–33). In their disabilities, Lusseyran and the blind children experience movement – the prescribed motions of playing sport, the ‘approaching’ of objects – in perhaps less conventional ways, but they nevertheless sense it in terms of complex physiological responses (via bodily sensors and receptors) as well as in intricately multimodal engagements of the world and – more importantly – of being in the world. They perceive movement by placing their bodies, in defiance of their disabilities, in the very centre of activity – of play, sport, and connecting with others. Thus, not only is movement ordinarily apprehended by a particularly acute engagement of physiological sensors and receptors, but I further suggest that it is also an experience which one may perhaps only access by being in and immersed in the world. For – referring to my argument in the previous section – movement is a process, a passage of change; by its very nature it exists purely and only in fluctuation. Might not the only way to perceive change be to be part of it, to similarly exist in that fluctuation? How to catch the wave upon the shore? Only if one is a part of that wave. Tim Ingold (2005: 103) invokes this very idea in his analysis on how to experience wind, a particularly apt example as the experience of wind is also an experience of movement (wind being the movement of air over differently pressurised areas of the planet):

We do not touch the wind, nevertheless things feel different when it is windy compared with when it is calm. For we touch in the wind. Wind is an experience of feeling … In our movements of action and perception we respond to the wind, as other creatures do. Soaring in the sky, the seagulls were feeling and responding to the wind. (original emphasis)

The experience of movement is thus an aggregated physiological response as well as – or perhaps leading to – the experience of being in it and being a part of it. As Ingold puts it, experiencing movement is not of touching something, but touching in something, of feeling and responding to it. If mocap records pure movement – as in movement isolated from the object – then the experience of that movement is the same experience of feeling in the wind. In mocap, then, a character may soar to the greatest heights, plunge to the darkest depths, be flung to all corners of the planet, this or others. In this cinema, we are not what Roland Barthes (1989) would term ‘hypnotized’. On the contrary, we are in a pure rush of movement, responding to it, being in it.

Now-ness and the immaterial ghost

In mocap, the experience of movement in being – in the immersion of the action of movement – also manifests itself in terms of time. This is because, as I will argue, mocap occupies a limbo position between performance and cinema, both of which have distinct temporalities. Cinema always points to an anteriority, to something that was. The imprint of light through the lens that marks the light-sensitive surface leaves a trace that points to past people and events or, as Roland Barthes (1977: 44) calls it, the chemical photograph’s ‘having-been-there’-ness: ‘[the photograph] establishes not a consciousness of the being-there of the thing (which any copy could provoke) but an awareness of its having-been-there.’

Of course, film is also invariably viewed in the present, so that its pastness becomes, in Philip Rosen’s (2001) words, ‘past presence’: the image refers to both its imprint of the profilmic from a past time as well as to the presentness in which we view it. Yet, we almost always find cinema’s greatest significances in its pastness. For Barthes (1977: 45), the inherent pastness of the photograph overpowers its present: ‘This kind of temporal equilibrium (having-been-there) probably diminishes the projective power of the image …: the this was so easily defeats the it’s me.’ Paul Willemen (1994: 241) similarly uses the same connection between trace and pastness in the photographic image to underpin his idea of cinephilia as revelation, as ‘something [that] shone through into the film’. For Willemen, the connection between object and image – the fundamental nature of the trace of light on a surface, making its mark in a specific time and space – forms the premise of that cinephiliac revelation.

On the other hand, performance is always in the present. It exists in itself. Being unrepeatable, performance can only be in its own present. As Peggy Phelan (1993: 146) writes:

Performance’s only life is in the present. Performance cannot be saved, recorded, documented, or otherwise participate in the circulation of representations of representations: once it does so, it becomes something other than performance … Performance occurs over a time which will not be repeated. (original emphasis)

Performance is unrepeatable because it occurs in a trinomial conflation of action, space and time – to perform is to show doing in both space and time. Action is also un-replayable because it is in continuity – an action must continue to another action. Richard Schechner (2002: 23), taking into account practice and rehearsal, writes of performances as ‘twice-behaved behaviors’, or ‘restored behaviors’. Yet repeated behaviours are different from each other; every re-play would simply be another performance: ‘Even though every “thing” is exactly the same, each event in which the “thing” participates is different. In other words, the uniqueness of an event is not in its materiality but in its interactivity.’

Performance and cinema are thus antithetical to each other in that action and object are opposed: one is continuity, the other is discreteness. The former endures by contradictorily consuming itself into non-existence – as Phelan (1993: 148) notes, in performance ‘there are no left-overs… [performance] saves nothing; it only spends.’ The latter leaves a trace by virtue of its imprint of light on film, which, by its mechanical nature, can be reproduced endlessly, so that its pastness not only revisits the viewer in his or her presentness but theoretically continues in infinite surplus.

Mocap, then, in its recording of movement/derivative rather than object/originator, challenges both cinema and performance. On the one hand, mocap, being a record of action, problematizes the ontology of the image as trace. Movement cannot be imprinted on a surface; more than the fact that it is not visible, it is also by definition and nature in transition. Rather than being ascribed to a specific object in a specific space and time, as is the case with traditional cinema, movement conveys the change of a specific object through a continuum of space and time. Yet one cannot imprint a continuum; one can only map points within that continuum.

On the other hand, mocap also problematizes the non-repeatable nature of performance, for mocap is precisely the saving, recording and application of a particular action (in a particular space and time) to be repeated in other spaces and times. One can argue that this may no longer be a performance as it is no longer in the particular space and time it originally took place. Hence, watching a record of a performance does not count as watching that performance because what is recorded and re-watched are the objects themselves, visible in a different space and time. However, mocap differs by recording a specific action, as opposed to a specific object (as mentioned above, to perform is to ‘show doing’). As action occurs through a temporal and spatial continuum, mocap records the entire action–-space–time conflation, to be made repeatable in other spaces and times. This had been previously unachievable because action, being a continuum, could not be recorded as imprint or trace; action must necessarily be in the present, because transition – the derivative, the progression from one point to another – falls into the between-ness of the past reaching into the future. As Deleuze (2005b: 79) writes:

The past is constituted not after the present that it was but at the same time, time … has to split the present in two heterogeneous directions, one of which is launched towards the future while the other falls into the past. (emphasis added)

It is in that splitting that action occurs, springing from the past, launched into the future. Mocap is thus about the record of action in its presentness, yet made reproducible without being past. The experience of mocap is not only of being in action but also lies in this constancy of presentness that is always repeatable in any space and time. If Barthes (1977: 44) advocates ‘a new space–time category’ for the photograph in its ‘spatial immediacy and temporal anteriority’ (the ‘illogical conjunction between the here-now and the there-then’), then I suggest mocap might be the follow-up category: spatial and temporal immediacy, for in mocap lies the presentness of action to be repeated in any spatial and temporal framework.

In this seeming contradiction of repeatable performance, I argue that the experience of movement in mocap becomes supercharged with the feeling of being, not only to experience movement but also to grasp the now. Seeing is thus consigned to even greater irrelevance here: being is all, for appreciating presentness surely requires an openness and aliveness to experience, to grasp the split of the present before half of it falls into the past and the other half hurtles into the future. In this process, the materiality of the body is also effaced as being now trumps visibility: what matters is the experience of being; the visible object can be anything or anybody. A Vanity Fair article, for example, surmises that ‘in time, it should be possible to build a scan of Marilyn Monroe from her performances in existing films [with mocap], integrate that scan into a fresh story and hire, say, Halle Berry to give the performance’ (Biskind, 2004: 189). If cinema is resurrection – ‘just as the cinema animates its still frames, so it brings back to life, in perfect fossil form, anyone it has ever recorded, from great star to fleeting extra’ (Mulvey, 2005: 18) – I suggest that mocap is possession: not only is the dead Monroe re-animated, but she (in terms of her physical movements) also manages to take over a living and breathing body in the form of Berry and, in this way, assert presentness in hijacked mortality. In thinking through imaging technologies in relation to time, death, resurrection and ghostliness thus come into play.

To that end, Gilberto Perez (2000: 28) describes cinema as ‘the material ghost’, derived from the retention ‘of the world itself, something material’ sourced precisely from a tangible world of referents which can be seen and heard. However, as mocap shifts cinema from visible light – of objects in the world – to non-visible movement, the material for the image correspondingly becomes immaterial, similarly transferring from the brutality of the corporeal to the ethereality of the kinaesthetic – that which we need to feel. This immaterial ghost might, then, be the second order of phantoms for cinema, not from what we can see, hear and touch of the living, but from what we can only sense from aggregated receptor sensors and being in the world. While mocap imagery, charged with the non-visible yet palpable, is sourced from actors in the world, it plunders them of every physically visible aspect, recording literally thin air (through which action moves). Stacey Abbott (2006: 97) writes of how science fiction film today ‘serves to make the invisible visible’: they use ‘digital technology not to rupture the boundaries of the body but rather to stretch and extend the body beyond its usual limits’. I suggest that mocap does the opposite: it uses digital technology to eliminate the body, to expose the ghost in the shell. Mocap in science fiction films such as Avatar signals the shift not just away from the body, but its abandonment altogether. Motion capture renders the visible invisible.

Conclusion

In this article, I have shown how the dialectic of seeing versus being is a resonant theme in Avatar, before connecting it to a mirroring opposition in the use of motion capture technology. Broadening that idea out to cinema generally, I have argued for the sense of being and its significance in experiencing movement, particularly in relation to the increasingly prevalent technology of mocap in its unique recording of non-visible movement as opposed to the audiovisual reality of the object. This notion of being is brought about not only by the physiological sensibilities of feeling movement, but also by mocap’s prevalent temporal sense of the present. In the process, I suggest a shift in the viewing framework of cinema, not in terms of the exteroceptive senses – seeing, hearing and feeling – but of other sensory receptors and of muscular awareness, of being in the movement. I want to end with a thought on the significance of this idea of being. Why being? What is the importance of this transfer of experience? With this shift from the visible to the non-visible, might this be the first step towards a Neuromancer-esque vision (Gibson, 1995) where bodies are inter-changeable, if movement – if our sense of being – can be mapped onto other bodies, so that identities become truly fluid (from Andy Serkis to Gollum; from Zoe Saldana to Neytiri – and we have only just begun!), emergent and in constant flux through changing time and space?

Perhaps motion capture, in its emphasis on being and now-ness, presents a unique unconsumable entity, one that is not pinned down with a specific pastness, instead remaining free with the indeterminacy of movement and the sensory experiences of presentness, a shifting visual code which nevertheless retains the internal knowledge, the sensory experiences, the bodily-ness, the physicalities, the lived-in-ness of the body. Ghostliness, then, in the case of the immaterial ghost, is not the void of limbo, a shadow’s image, rootless and wandering, where the Derridean spectre looms large – one with no history, no time and no linearity of succession, consigned only to between-ness (see Derrida, 1994: 10). The immaterial ghost, rather, is a power for the future, proffering agency through its transmission of being, resisting objectification in its non-visibility and challenging commodification of capital in its lability of object. The last shot of Avatar, then, in which the mocapped eyes of Jake Sully’s avatar pop open, bright with purpose and potential, is not just about the transformation from human to Na’vi, but also about a new signification of self – of simultaneous resistance and agency, of being and being invisible, of an effortless moving between worlds, realities and forms.