How Special Effects Work #1: The Sandman

It’s been over a week since I’ve blogged – towards the end of a semester, things seem to get much busier, and I start to feel a bit stupider, so it’s harder to sustain my more prolific bursts of writing. Perhaps my previous post was long enough to keep you busy (or fed up of me) for a while. Thanks to all those who’ve commented on my Cloverfield paper: I always enjoy receiving feedback, negative or positive, so feel free to add your thoughts to the discussion. Hopefully, things will ease up a little and I can devote a bit more time to this. To reboot things, I thought I’d play to my specialism and start a series of short posts about special effects. This is partly to expand upon and clarify points from Performing Illusions, and also to include some ideas that were too late to make the final cut of the book. I’ll be making these up as I go along from time to time, but I’m happy to take requests. It’ll be a good way for me to take notes as I go, and hopefully provide some interesting reading.

To kick things off, here’s a little analysis of a great sequence from a far less great film, Spider-Man 3. About halfway through describing the plot of a superhero film, I usually pause for breath and realise how ridiculous it all sounds. Teenage boy sprouts web-spinning glands and dresses up in natty spandex togs to fight crime. Meanwhile, some other dude gets exposed to some sciencey stuff that turns him into sand. It’s not exactly Death in Venice, but this kind of story has become so familiar that we barely bat an eyelid when some new fancy-dress vigilante takes to the screen. Stop and think for a moment. Peter Parker is at school. Then he gets bitten by a genetically modified spider and picks up some arachnid tendencies. Why are we not laughing this stuff out of town? Partly, I think, it’s because of the familiarity: we’ve seen a lot of superhuman heroic figures over the years, whether it’s Achilles, Aeneas, Hercules, Perseus, Jesus, Beowulf, Gawain, King Arthur, Superman, Batman, Iron Man, Barack Obama or token female Wonder-Woman. But also it’s because we understand the allegorical function of these characters. Whether it’s Superman as a refugee migrant who has to change his name and act like a local to gain acceptance in society (while secretly saving the world’s collective ass), or Spider-Man playing out his awkward years of bodily change and early-career anxiety, we know that these are not portrayals of how things really are, but re-imaginings of things that are easier to talk about and popularise if we dress them up in shiny clothes and pit them against a series of similarly allegorised embodiments of villainy/social evils.

That’s my starting point here, but I suppose it’s not strictly relevant, except to say that Spider-Man 3 operates (because it is a sequel) in a pre-existing alternative world where scientific exaggeration is an accepted form of expression, with certain agreed limits on what may occur: there’ll be no “magic” here, just scientific principles extruded to a degree that probably constitutes impossibility, all the while remaining anchored in a logical basis (however tenuous) that isn’t there to make incredible events believable, but comprehensible.

The scene in question is an origin sequence. We get to see how the Sandman came into being: as such, it offers a spectacle of incarnation, animating an apparently living body out of inanimate materials. It is structured between the bookends of these two states, beginning with an extreme, near-microscopic close-up of grains of sand, which gradually cohere into an image of the actor Thomas Hayden Church. This demarcation of the set-piece is a common trope in this kind of foregrounded spectacle – it has clear entry and exit points and stands alone as an autonomous performance, even as it offers some narrative information; It possesses a limited colour scheme of browns and greys (er… it’s sand-coloured), and the lack of dialogue or peripheral characters further enforces the self-containment.

Witnessing the birth of the Sandman, one of the pleasures comes from seeing a two-dimensional comic book character transplanted into a three-dimensional, digitally rendered figure. The Sandman is the perfect CGI character: the kind of particle-system modelling used to make swarms of particles take on shapes and patterns is something that computer-graphics are equipped to do – it would be extraordinarily difficult, if not impossible, to do this in stop-motion or another kind of pro-filmic object animation. So, while the scene references older media, it focuses on graphic qualities that exude novelty and technological specificity. The virtual camera (the scene is entirely computer-generated, so it’s not entirely accurate to think about the camera being situated within the scene) executes a slow track around the central focus of the emerging Sandman. The stressed dimensionality of the sequence thus puts further distance between this and two-dimensional animation, optical process shots and puppet animation where camera movements are much more difficult to pull off. In short, the scene’s novelty value is to be understood in terms of its differentiation from prior instances of animation and effects shots.

The long take is the core of this sequence. The sustained performance of a technical illusion would seem to imply its pro-filmic authenticity: the camera never needs to cut away to or fragment the trick to hide its mechanisms in montage. However, there is no longer a logical reason to attach such notions of presence and solidity to things we see onscreen, even when the camera’s unflinching eye seems to be hinting that there are no sleight-of-hand edits, nothing up its figurative sleeve; a virtual camera tracking through virtual space to “film” a virtual object never needs to cut, and those connotations of authenticity can just as easily be translated into indications of artifice, of a lack of presence or ostentatious virtuality. But digital effects still exploit our residual expectations of photographicness. You can see it in the use of artificial lens flare to suggest that the camera is physically present, its mechanism overloaded by the scene in front of it. Lens flare is a side-effect of a camera’s registration process, and in a virtual scene it is added in order to offset the true origins of the shot.:

These techniques subtly purchase your understanding of the sequence as a wholly situated moment, recoding it not as a flurry of algorithmic manouevres, but as a live recording of an event, where some of the unplanned markings of the photographic apparatus might come into play. To cut through my verbose description, the shot, which was actually constructed in a computer, is dressed up to look like it was shot on a set. So, computer-generated effects do not erase or evade the properties of photographic media. Instead, they extend those properties to supernatural lengths: the power of the illusion arises out of the distance between the acknowledged impossibility of the event, and impression of authenticity lent to it by the markers of a situated apparatus.