top of page
  • Writer's pictureAlex Harper

Constructing Space in Ghost in the Shell 2: Innocence.

Still from Ghost in the Shell 2: Innocence (Oshii, 2004). Via Listal.

The deliberate use of space in cinema is a long standing tradition that has metamorphosed across the years with our understanding and interpretation of space as a concept. Space is interdependent on the characters and action within it, as well as the narrative surrounding it, and as globalisation and inter-connectivity have redefined and revolutionised what one can consider as “space”, so has our usage of it in cinema.

One film that deals intimately in these concerns is Ghost in the Shell 2: Innocence (Oshii, 2004). The belated sequel to its oft more discussed predecessor, Innocence tells of a character’s and a society’s search for individual identity and for a feeling of solid, physical existence within a deeply interconnected world—a world in which space is at once collapsing and expanding with the new and overwhelming prevalence of an online “Net.” As an animation, the space of Innocence is carefully constructed from the ground up, rather than simply captured. Each image must be considered to be deliberate, and therefore the space it depicts more meaningful. The use of space feeds into the films theme of transposing the living world from private and personal to something more corporate, bringing along a sense of something omnipotent—a global conspiracy.

The visuals of Innocence are a constant reminder of this loss of a private world: observational high-angle shots continually place unsuspecting characters as distant, off-guard and under surveillance, while visualisations of the Net permeate the characters reality like a societal connective tissue. This spatial juxtaposition between a solid “real” world and an intangible interconnected net contributes to this vital sense of paranoia and instability—the essence of this new world working against its inhabitants by its very nature. The protagonists, Batou and Togusa, sit in a car, still, as fluoric holographs of computer networks and information rupture onto the screen. No reaction from the duo, just interaction—this is normal. Similarly, the same characters are placed within a stylised non-physical space during a conference: blocks of information dominate the central space of the screen as characters discuss from the perimeters, before information and people alike dissolve to reveal an almost empty room, with two people standing insignificantly within. Humans no longer dominate the space around them. When we consider the thematic backbone of Innocence—the deeper the connection to the Net, the greater the risk of manipulation to our physical lives is—this merging of the characters physical spaces and the non-physical reality of the Net raises the question of whether the space inhabited by the characters can be considered “safe.” There is no clear demarcation between whether the physical or the intangible spaces could be considered safe or unsafe for the characters. The point here is that when both merge together, the dual space becomes unstable, and able to be utilised for whatever means—good or bad. In the opening sequence of the film, Batou stalks a suspect through a claustrophobic alleyway. From Batou’s point-of-view, channels of information from the Net permeate his view, turning his vision into an HUD—his eyes bounce from potential threat to potential threat, zooming in and scanning everything. Each threat dissipates under scrutiny, and the intensity of the scene is neutralised—the Net’s influence turning an unsafe physical space into a safe one.

As for the negative, the non-physical space causing the physical to become unsafe, we are best to look at the market sequence. When Batou enters the shop, the “camera” pans around him to lock him into position, then, a warning: “You have entered the kill zone”, as Batou proceeds with caution through the space that visually appears no different from any other “realistic” space seen before. Each shopper faces away from him, faces—and intentions—hidden; the image depicted from above with a sense of objectivity. The 2D and 3D elements of the animation aim to achieve a sense of physical realism to the scene, from the depth of the objects lining the shelves, to reactionary movements of bodies and objects as Batou crumples into his surroundings when under attack. The controlled realism of the scene fractures—phantom gunshots ring as the image continually distorts into colour and video-noise. The space is narratively unsafe, and how we are perceiving the world it is deeply wrong. The information channels we saw through Batou’s eyes are now disintegrating and unable to decipher where the danger is, before the true nature of the scene becomes apparent—Batou, his mind and eyes hacked, but now restored, being restrained by a colleague. There are no aesthetic differences between the manipulated and the true space of the market, and we are left to rely on the narrative to determine which one is which. This scene is the crystallisation of Oshii’s idea, that the space experienced is not always trustworthy, and not always real. It also feeds into the greater conspiracy—the individuals perception of the world is not always theirs: there has been a breakdown between the private and public spaces of the world.

Innocence has a simultaneous divergence and convergence of these two spaces: the private personal and public impersonal. They are represented very differently, yet they are both limited by their representative lack of safe boundaries. A trademark of Oshii’s work, characters will inhabit a shared personal space as they converse, yet their physical actions towards each other are limited. Confined in a car, Togusa and Batou talk at length about philosophy and identity, yet they simply stare forward as they do, rarely looking over at each other. Any emotional drive in the scene comes from the rotating shot selection and the editing rather than physical action. The idea of the personal space and the private conversation is subverted, as the lack of interaction within it raises the idea of discourse happening in the ether. Emotions do not affect what they say: the topic of their discussion, something greater than themselves, is only fit to inhabit an impersonal space, and therefore the characters become detached from the physical world as they talk. This idea is expressed most distinctly in the spatially rich morgue sequence. In a sterile, white morgue (or, this Sci-Fi worlds equivalent) Batou, Togusa and a mortician named Harroway discuss first criminal then philosophical issues. The spectator never observes the morgue as an objective whole, instead each character is perceived in close-up, occupying their own distinct space within the room, while the others are placed in the wide negative space behind. Only the robotic details are presented objectively within the room. As characters remain relatively rooted to their personal space and ideologies, Togusa accounts for the greatest spatial shift in the scene. Togusa, the closest to being truly human—that is, the least technologically altered—forms the link between the two arguing characters: considerably, a person navigating a room of “objects.” Furthermore, only Togusa’s breath is visible in the cold environment; it may be comparatively inhospitable for him, yet he is the one to compromise on the philosophical debates and push through the impersonal space of the morgue. This is one aspect that substantiates the difference between personal and impersonal space. Personal space in Innocence is realistic, logical and claustrophobic—seen in action set-pieces and the criminal investigation; impersonal space dwells in grandiose, unrealistic spaces, and evokes philosophical debate in its characters.

As the film progresses to its late stages, and the ideologies shift from looking at the individual to the collective, the characters inhabit far more of these impersonal spaces. Travelling towards the northern city, spaces become immediately and dramatically more fantastical: the open, baseless and vast cathedral a far-cry from the tight personal quarters of the city. As the characters become separated from other people, the world gets larger and less sure around them—the film is now becoming an exploration of the space in which the character inhabits. In the dollhouse-styled castle, they become mere figurines depicted in omnipresent viewpoints in an imposing space; imposing in size colour, visual symbolism. As with the market scene, time and space here begins to fractures and reconstructs with no warning—this time looping to the beginning of the scene with a replacement and rearrangement of the objects within the space. The characters grasp on their own place in reality is scrambled. There are notable differences to the intense, jolt back to reality of the market scene: in the castle, the characters are oblivious, drifting through the space as the scenes idly slip over each other, again and again, before a slow realisation of what is really happening. Their perceptions have been warped, but their reaction is not the same as the fear of the market. When manipulated, personal space in Innocence becomes suspicious and terrifying, as the characters mind cannot comprehend the loss of their perception, while in impersonal space, the characters willingly lose themselves within a contextual stream of conscious thinking. This implies that the physical disturbance of real-world space is more damaging than the manipulation of impersonal, less tangible spaces. Like the philosophy that partners it, impersonal space is less concrete, more malleable, and more ungraspable than its realistic personal counterpart.

As mentioned at the beginning, the fact Innocence is an animation is of major significance—as spaces are not simply captured, but created. If each shot created is deliberate, then each images physical space is purposefully designed. Oshii takes advantage of this with an animating methodology designed to imitate live action cinematography in terms of composition, movement, and even technical limitation. Reality in animation is mostly unattainable, yet the cinematographic techniques employed in Innocence develop a more filmic way of observing the space within the film. Simulated long-takes, depth of field changes, animated dolly shots and whip pans—a particular favourite for the action sequences—all make an appearance. Each is used in a natural way—to stall time and space, dictate the plane of observation among other things—and the fact they are natural is important. It reads like any other film; we have a nascent understanding of the spaces we are seeing because we have seen these filmic constructs before. The morgue sequence remains one of the most interesting scenes to analyse in regards to this metric due to its strikingly cinematic nature. Each image in this scene is designed to look as if it were shot across different focal lengths—at the most extreme end, appearing as if shot on a very wide angle, almost fish-eye lens. At its widest, the faces in the image narrow, the lines of the room bend, and distortions appear—a cinematographic realism scene throughout the film, such as a zoom in, dolly out shot of Batou in the northern city. This unusual practice allows the spectator to witness the animated space in a more representative, and thus, realistic way with regards to “filmic reality.” As each character occupies their designated—and distant—spaces in the morgue, they are animated through the wide angle method, close up, consequently stretching the background around their faces so that we may see the other characters standing in the room. This also exacerbates the amount of negative space between the characters: pushing them further from each other as they debate. The editing, rotating around the room and refusing to settle on a master or establishing image for the scene, only adds to the sense of each individual separated in this impersonal space.

Alongside the cinematographic techniques, Innocence merges 2D and 3D images to evoke some form of quasi-realism to the image quality, as well as to make the spaces in the film feel more alive, and the dangers more present. There is a notable difference in when the different animation styles are used: 2D for characters and some environments; 3D predominantly for impersonal spaces such as the Net, and the castle. In these impersonal spaces, the 3D is less useful for conveying realism than it is in displaying the opulent and philosophical elements of the film in some manner of aesthetic glory—highlighting yet again the many different methods used in Innocence to depict space in a narratively and thematically meaningful way. In Innocence, space is utilised by both image and narrative for a mutually beneficial impact. The thematic reasoning between the claustrophobic personal and the limitless impersonal space, always astute, always effective.

198 views0 comments

Recent Posts

See All


bottom of page