This piece was originally published exclusively to patrons on the 29th of November 2014 before being made available to all readers one week later. I intend for this to be the custom for one of my articles every month. To gain access to such patron rewards and to help make sure more articles like this exist in the future, please consider visiting my Patreon and becoming a patron.
A few weeks ago I gave a compositional walkthrough of a single room in Project Zero 2 – the projector room of the Tachibana House – illustrating how the game’s use of camerawork depicted an environment which is subversive of safe spaces.[i] Since Project Zero 2 is a horror title this meant highlighting the strengths of its use of perspective within the context of the game as a whole, which is to say how visualized or represented space translates to perceived dangers and securities.
Another game that uses a second-person perspective[ii] to wonderful compositional effect is Final Fantasy VII, with a ‘cinematic’ reputation that’s lasted the nearly twenty years since its release. This longevity is justified—even as children most of us wordlessly understood how its stellar use of cinematography contributed to the experienced narrative.
What I’d like to do today is revisit FFVII with this particular mechanism in mind, to glance composition off against storytelling for perhaps previously unnoticed nuances (at least, unnoticed from my own perspective). It’s worth stressing that it is a mechanism: the use of camera to frame and juxtapose characters, actions and scenery relates conscious decisions on behalf of its creators. Simply the very nature of modelling its pre-rendered backgrounds[iii] establishes how fused they are with intention—our visual relationship to the world is so fixed to narrative design that camera and scenery nearly become one thing.
Since it’s best not to assume everyone has played the game, a little bit of exposition is warranted. The story of Final Fantasy VII focuses on Cloud, an ex-military mercenary who at the game’s beginning is working for the eco terrorist group AVALANCHE in Midgar city. Nearly the entire world is under the rule of the Shinra Electric Power Company, owing to its monopoly on Mako energy, an oil analogue. The corporation/world-government’s highly industrial and capitalistic structuring has lead to widespread social oppression, together with its mining of Mako, the planet’s lifeblood, informing AVALANCHE’s resistance. Initially Cloud is only in it for the paycheck but before long the fight against Shinra connects back to an old personal adversary and war buddy, Sephiroth, whose own designs for humanity spell catastrophe.
So two fairly clear-cut examples of FFVII’s deliberate camerawork occur quite early on. The first is not ten minutes into the game: Cloud and co. are infiltrating the Sector 1 Reactor on a sabotage mission, and after Barrett tells the player-through-Cloud what’s what [fig. 1.1], the resistance group continues toward the reactor while Cloud remains immobile [fig. 1.2]. The camera descends towards him from its isometric position before swooping around his back and tilting upwards, showing us AVALANCHE’s looming destination [fig. 1.4]. Cloud then hurries on to catch up with his crew.
This simple camera movement does a couple of things for us quite effectively. The most obvious is that it shows us the reactor we intend to send sky high, the enormity of which conveys magnitude and severity onto AVALANCHE’s terrorist action, while conversely characterising the group as rather minuscule.
A second effect is to approximate the player to Cloud’s position in relation to both AVALANCHE and the reactor (and implicitly Shinra): the motion of the camera towards Cloud offers intimacy and insight into his perspective. Bearing in mind this occurs right at the start of the game when we know nothing about anything, this slight connection with our character, however briefly, has a lingering effect on our perspective into the narrative.
The second example of effective camerawork comes some hours later [fig. 2]. Shinra has just toppled the pillar holding up the upper plate of Sector 7[iv], and as we witness the plate’s imminent collapse from the underside and see the panic of civilians seconds from being crushed, the camera changes to this position high above [fig. 2.1].
From this distance we can see a full view of the devastation wrecked upon the Sector in terms of how altered the city’s landscape is now becoming, but from so far away we do not experience the human tragedy of the event. Magnificent, mile-wide clouds of fire are to us like fireworks. The camera pans gently away through a window, where the president of Shinra observes his handiwork as classical music filters into his quarters [fig. 2.2]. The final shot of the cutscene is an upward angle at Shinra headquarters, dominating, steadfast, untroubled by the havoc it’s unleashed on its people [fig. 2.3].
You’ll notice these two examples both use mobile camerawork, sweeping and zooming to elicit the desired effect, whereas the majority of camerawork in FFVII is largely fixed to a set distance and a set angle for each and every scene. When we think of camerawork being cinematic this is usually what springs to mind—dynamic, action, motion—such is the custom of a market-orientated critical language, but there’s no reason we can’t conceive of static camerawork in the same terms of the whats and hows of visual representation.[v]
Consider this scene after Reactor 1’s bombing [fig. 3]. The AVALANCHE crew have rendezvoused in a traincar and celebrate their victory, but Cloud’s apathetic attitude causes him to butt heads with leader Barrett. This conflict takes place in the dead centre of the screen in the midground of the shot, Cloud on one side, Barrett on the other, since it’s the central focus of the scene. When the tension abates, the resistance members exit the area one by one through the foreground into the next traincar. [fig. 3.1]
The following area is an already occupied passenger car; we’ve got a similar enough angle into the scene, favouring the right hand side a little this time. [fig. 3.2] As Barrett enters the car from the background and chases the car’s occupants away, he takes his position in the midground while the rest of AVALANCHE and Cloud come to reside in the extreme foreground. [fig. 3.5]
Barrett’s temper and reputation are here established in his relative colonization of the space. We know AVALANCHE fight for the oppressed people of MIdgar and for the health of the planet but this suggests his resistance force is not a universally welcome presence in the scheme of the city’s inhabitants—a fairly expected condition when it comes to fighting a deeply-rooted status quo. Given our angle on the scene, Barrett’s sitting arrangement then disadvantages him to Cloud, Jesse et al right up beside the camera’s position, appropriate to Cloud’s relationship to each character: we are at odds with Barrett but sympathetic to the rest of AVALANCHE. All this is achieved by moving the characters around the scene and keeping the camera static.
Now, room has to be made for a certain level of pragmatism for the player’s benefit, since many areas are designed to be traversed while also representing characters and environments in whatever way is semiotically appropriate. This is one of the reasons townplaces and ‘dungeons’ tend to be framed from an isometric angle, to give the player a slightly encompassing view of the area lending to its navigation [fig. 4], while smaller, tighter spaces allow for more intimate camera angles [fig. 4 inset]. Perhaps it’s not a coincidence that the shops of Sectors 5 and 7 take us closer to the slum’s inhabitants than, say, the more expensive (and larger) Wall Market, and even more so later on in Junon, given the game’s slant on dystopian capitalism.
The position of the camera in relation to the environment, the navigable space for the player, the visual space granted to characters onscreen, and their personal choreography all impact the player’s relationship to the gameworld through its visual representation. Poverty-stricken areas of Midgar are visually more ramshackle than the cleaner corporate Shinra environments, which translates back in to their navigability in a mechanical sense.[vi]
This overt contrast between Shinra and the people/spaces of Midgar juxtaposes the two forces of oppressor and resistance movement by relating each organization to a particular sense of space. AVALANCHE’s headquarters in the basement of 7th Heaven [fig. 5] is in a state of spatial disarray, it’s all cramped and cluttered and a tiny bit tricky for the player to navigate Cloud through given how the floorspace narrows in parts. Visually speaking it’s also rather messy: the angle of the centre table doesn’t match our view into the room, and the characters all sit around in a disorganized fashion.
The Shinra comparative, however, is extraordinarily neat and aligned symmetrically within the frame [fig. 6]. When first we see this room, the organisation’s president sits at the top of the table in the centre of the screen, clearly in a position of leadership over the assembled employees. It’s distinctly a corporate boardroom whereas the disarray of AVALANCHE’s base of operations signals a grassroots structure favouring no one character over any other.
Asymmetry of composition reflects disorder, in this case organizational disorder, conferring an absence of control from those whose space it is designated. AVALANCHE’s headquarters ties them compositionally to the slums and the oppression that disorganizes and interrupts their existence.
(I want to briefly go on a tangent regarding the Shinra conference room. The first time you see it in-game is when Cloud, Tifa and Barrett are looking down through a ventilation grate—the camera angle adjusts to the one shown above once our perspective divorces itself from that of our heroes. Anyway, during the scene the scientist Hojo enters through the foreground and stands just to the right of the centre. He then turns around, approaches the dead centre of the table and faces the camera head-on [fig. 7], before exposing that he intends to breed Aeris to maximize her potential as research subject. Hojo’s placement on-screen showcases his disregard for President Shinra’s incumbent authority, and foreshadows his role as the true villain of the corporation. Hojo is the core agent responsible for Mako energy, Jenova, and Sephiroth.)
So we can see straight away that there are differences in how each space is visually represented in terms of order and chaos and how this might relate back onto their narrative context. The angle with which 7th Heaven is viewed closely matches that of the slum exterior, as does the layout of objects of interest across the town. Notably it is a rather sparse area: one entrance and exit, a bar (doubling as AVALANCHE’s HQ), a shop, and a tiered weapons shop/training ground. That can be accounted by its function within the game’s progression—it’s the first proper ‘town’ area you come to, maybe an hour into your playthrough, so the designers want to give the impression of this place offering some downtime and freedom but don’t want to throw too much your way too soon.
The architecture of this place is all over the gaff, and if we map navigable terrain to the screen the positive space that results is equally disorganized [fig. 8 above]. There’s not a neat line between the entrance and 7Th Heaven or the bar and any of the other shops and there’s a lot of junk on your path to break the space up visually and mechanically. For nearly every location within the slums the navigable paths available to you as a player are consistently irregular to the angle of the shot, while more cramped spaces offer little room to walk around but invite the player to a closer level to Cloud and the slum’s inhabitants.
We can then compare this to how positive, navigable space is represented within Shinra-designated areas such as the president’s chamber [fig. 9].
Here a far away aerial view pans to encompass the whole of the room—a significant space when compared to, say, 7Th Heaven’s ground floor, no clutter or debris of life to serve as an obstacle, even if it is ostensibly empty of value to the RPG player. Walking from one end of the room to the other offers us a smooth ride with plenty of mechanical freedom (you can move up! and down!) but large scales that dwarf Cloud’s footsteps principally frustrate our desire to reach our destination and progress the game [fig. 10]. Thus the president’s chamber sharply contrasts with the ragged paths we’re met with elsewhere, but the more abundant and accessible space is framed disinterestedly.
Until we approach the president’s desk, the point of action for the scene above [fig. 9 and 10, inset]. Now the camera switches to a closer view, placing the president centrally and at a distance, as usual, with the focus of the shot resting instead on his subordinates. Again there is an enormous amount of space for characters to position themselves in the foreground, although the now decreased scale affects an interest in the choreography of the president’s underlings.
Incidentally, when I first played FFVII I continually overlooked President Shinra as a character, such as forgetting when I had seen him before and glazing over his importance in context of the gameworld. It could be down to inattentiveness on my part but I think the cinematography played a certain amount into it. As I said, when the president appears in a scene he is generally framed centrally but located a fair distance from the camera, which denotes him as a central figure relative to his surrounding executives and events but grants him little screen presence personally, despite basically being the ruler of the world. This is likely an intentional move to downplay him as an individual, aided by his blatant unnaming, and convey him instead as an icon of Shinra-ism. So when President Shinra is soon killed and replaced by the more charasmatic and likeable Rufus, who remains as a secondary antagonist for the bulk of the story, President Shinra swiftly becomes a distant memory. There is something to be said here of the treatment of President Shinra and Rufus as antagonistic characters in their own rights in relation to the story’s themes of existentialism, but it falls outside the scope of this article.
Anyway, as I mentioned before, Shinra-designated environments tend to reveal a pattern of visual symmetry often coinciding with wide and typically empty navigable spaces [fig. 11], in contrast to the confines of the lower plate’s slums. This is perhaps never more present than the front entrance to the Shinra main building [fig. 11.1]—after the player has exited from a cacophony of navigable space confounded by deliberately obfuscated visual pathways, the ‘climbing the plate’ section of the game as a metaphor for institutional barriers of oppression[vii], the first area they’re met with on the upper crust is the entrance to the Shinra base of operations. It’s a lovely broad shot with plenty of room for Cloud to stretch his legs after the climb.
What I want to note here is the repetition of visual symmetry throughout Shinra spaces. Navigable space is a clear luxury permitted of the decadent higher class and corporate elite of Midgar, substantiated by its absence among the lower class citizens. But to remark as I have on symmetry as signifying order and asymmetry relating chaos makes for something of a bland observation, especially seeing as that doesn’t tie quite into the text as-is. Shinra are clearly the villains if you follow the story, so what is the game saying by framing these shots to characterize their spaces as orderly?
Well, although the slums by and large prefer jaunty camera angles and diagonal visual throughlines, shots favouring environmental symmetry do occur in these areas, albeit as rarities. One especially notable example is the church in Sector 5, which we first visit in this wonderfully choreographed scene [fig. 12].
Separated from AVALANCHE after their attack on the Sector 5 reactor, Cloud regains consciousness to find himself in the company of a familiar flower girl, Aeris. As Aeris interchangeably fawns and tuts over the damage done, she speaks of how this church is unique in being the one place in Midgar’s slums where flowers can grow. The two initially begin in the foreground of this mid-range shot [fig. 12.1], while her tending to the flowerbed takes Aeris closer to the camera and then farther into the midground of the shot. Cloud follows suit, and they make their formal introductions. Meanwhile, Reno of the Turks arrives in the background from the church’s entrance [fig. 12.2].
As soon as the pair acknowledges the intruder, the camera switches to a high angle on the opposite side of the room [fig. 12.3]. Now Reno is positioned in the foreground, quickly joined by a squad of Shinra soldiers. Cloud and Aeris relinquish their central position and move into the background to make their retreat through a door on the far wall. The camera then switches back to the initial position as Reno moves from the midground to the centre foreground, trouncing through the flowerbed in his path. Here he stalls to make an observation—“They were… Mako eyes”—which we are given to assume is a note of some importance considering its placement in the scene. Reno then exits through the foreground before briefly re-entering to tell his men not to step on the flowers.[viii]
This scene is rather like the one of AVALANCHE commandeering the traincar in that it depicts a ceding of ownership of space, establishing the intruder with power through their gradual spatial and visual dominance of the screen. Unlike the traincar we have a curving flow of the action here to accommodate the tonal shift midway through: a private exchange between Cloud and Aeris is interrupted, whereupon the visual power dynamics flip to privilege the Shinra operatives over their targets. Once Cloud and Aeris depart, the camera resumes its intimating position but now Reno moves into central focus, thereby claiming and subverting their former positions. He wins the scene.
Notice, however, the aerial shot of the church bisecting it into four neat quarters. Despite the debris and scattered furniture about the place, this angle lends to the room a sense of visual symmetry unusual to a location in the slums. This area, more than any other within the game, is an Aeris-designated space. It is spiritually significant to her as a haven away from the corruption and infertility of Midgar. To Cloud and to us it becomes significant as the location of our proper introduction to her character, but this is secondary to its symbolism to Aeris’ narrative arc.
This gives us a clue as to what visual symmetry might represent within the context of FFVII’s compositional styling, connecting back to the core themes of identity and existentialism. It could signify order along the lines of one’s spiritual cohesion, expressed narratively as ideological dedication. Which is not to say it’s a firm rule of how visuals are depicted in FFVII, just that it might be understood as a loose motif to help us better access the text. As a guideline, the more resolute a character or organization and the closer their means and actions coincide to the values they purport, the less fragmented their personal identity and the more orderly their representation on-screen.[ix]
And there lies the difficulty in combating a capitalist dystopia with such a tight grasp on the status quo, in the case of AVALANCHE. We exist as subjects of its oppression in spite of our political resistance, leading to spiritual fragmentation through self-denial of our contaminated parts.
So if personal resolve to ideology is what is signified by ‘order’ through visual symmetry, it follows that Shinra would claim a great degree of symmetrically-framed environments given its authenticity to its capitalist (and corporate) values. So then, the church of Sector 5 symbolizes a reservoir of floral life, and thereby planetary life, making it an appropriate environmental avatar for the character of Aeris.
And what of the lead villain, Sephiroth? What scene might best relate his existential fulfilment through its composition?
But to consider that properly, we’ll need to examine the scenes leading up to the mental collapse and rebirth of Sephiroth as Final Fantasy VII’s primary antagonist. That we’ll do in part 2.
[ii] I’m using “second-person perspective” in perhaps an idiosyncratic way to describe a style of camerawork to which I believe it is appropriate. We are all familiar with first-person and third-person perspectives and how they are attributed through categories of camerawork, although the implied narrative effects of their respective cameras might be dubious. Rather than throwing out the common way we discuss this aspect, I suspect there’s a secret wisdom in how we talk about personal/narrative identity through visual perspective, so I’m running with it. If you’re interested, I wrote on this topic here: http://normallyrascal.com/2014/11/11/framing-identity-or-how-can-i-be-both-lee-and-clementine/
[iii] Since pre-rendered backgrounds have fallen out of fashion, a quick explanation: pre-rendering was a technique used to digitally showcase gorgeous-looking scenery without needing the computer to put in a load of effort at runtime. It’s prepared long in advance, so designers would put care into choosing a visual angle into the scene that the rest of the development team could take into account. We can contrast this with more modern methods of cinematography dedicated to orchestrating scenery action and points of interest along a expectations of a player’s manual camerawork, as in many first-person perspective games. The latter often comes across as overtly player-centric worldbuilding, unfortunately, lending to feelings of solipsism rather than a convincing gameworld. It’s a trade-off of the form.
[iv] Midgar is literally a two-tiered city: it is composed of an upper layer suspended high above the ground level of the city by these tremendous pillars, presumably to accommodate overpopulation. Poor folk live on the ground level, lacking sunlight and fresh air and all that you might expect would come from such living conditions. Richer folk and the Shinra elite live on the nicer upper layer.
[v] As good a time as any to reference Mark Filipowich’s articles on representation of space and abstraction through camerawork: http://big-tall-words.com/2014/11/17/thoughts-on-representations-of-space/ and http://big-tall-words.com/2014/07/25/the-narration-and-abstraction-of-the-camera-in-games/ For a much more concise analytical application of these principles, see Zolani Stewart’s piece on expressionism and Sonic Adventure 2: http://fengxibox.blogspot.ca/2014/11/expressionism-and-sonic-adventure-2.html
[vi] FFVII was built for the PlayStation, with its fairly traditional directional button layout. There’s a button each for ‘up’, ‘down’, ‘left’ and ‘right’, but no button for ‘diagonally down and right’, for example. In a very simple way, it means you only have to press one button to move horizontally or vertically across the screen but you need two buttons to move diagonally. Environments framed at an isometric angle therefore require this tiny extra bit of mechanical effort on the player’s behalf to be traversed.
[vii] See Cameron Kunkelman’s article, In Praise of the Worst Design Moment in Final Fantasy 7.
[viii] Comedy gold.
[ix] I will mention this again in Part 2 but consider Cloud’s positioning on-screen as he recounts the events in Nibelheim 5 years ago: his visual obfuscation behind the terrain nicely hints towards his stance as an unreliable narrator—and of course, his self-delusion.