The Lord of the Rings: The Two Towers · essays & theory

A reading · through the lens of theory

There is a moment, two-thirds in, when the camera holds on Gollum beside a dying fire and lets him argue with himself. The cut goes back and forth — Sméagol, then Gollum, then Sméagol again — as though two actors were sitting where there is only one creature. Nobody else is in the frame. The whole apparatus of the film, all that machinery and money, narrows to a single ruined face changing its mind. Start there, because that face is where this enormous engine of a movie does the one thing Gilles Deleuze would never have expected a blockbuster to do.

But first the engine, because The Two Towers is the movement-image at full throttle, and worth seeing as such. Deleuze used that phrase for the basic logic of classical cinema: a character perceives a situation, acts to change it, and the action resolves the situation — perception flows into movement flows into a new state of the world, and the cut exists to serve that chain. Helm's Deep is the textbook page. Watch how cleanly it runs. A fortress, a coming horde, defenders too few; the wall holds, the wall is breached, the line falls back, and at first light a single decisive act turns everything over. That is the action-image in what Deleuze called its large form — Situation, Action, new Situation, the circuit at full power.

The siege also gives us his synsign, the encompasser: a milieu so vast it seems to demand the deed before any one man chooses it. The storm of ten thousand Uruk-hai massing under the wall is not a backdrop. It is a pressure. It calls the act out of Aragorn and Théoden the way weather calls out an instinct. And the fighting itself is pure binomial — Deleuze's word for force against force, the duel that organizes the whole. Wall against ladder, breach against shield-wall, the line that holds and fails and reforms.

None of this is invented from nothing, and the film is honest about its debts. The breach-and-fallback dramaturgy — a tiny garrison holding a walled compound against successive waves — is Rorke's Drift from Zulu, scene logic lifted almost whole. The rain, the mud, the telephoto coverage of a fortified position swamped by numbers, is Kurosawa's Seven Samurai. The way thousands of bodies resolve into color-coded graphic masses sweeping a hillside is the battle grammar of Ran, here scaled up by Weta's MASSIVE software, which gave each digital soldier a crude agent-brain so the crowd could behave rather than merely tile. And the climax — Gandalf cresting the eastern ridge at dawn with Éomer's riders, the cavalry arriving in the last possible second — is montage-organic-american, the convergent last-minute rescue that D.W. Griffith standardized a century earlier: two lines of action cut in parallel, tightening until they collide and save. Even the rhythm of the thing is Deleuzian. He noticed the action-image breathes, alternating the vast space of the situation with the taut single thread of the act — respiration-space. Helm's Deep keeps cutting from the immensity of the field to one man on one stretch of wall, and the cutting is the lungs.

So the engine is magnificent, and entirely classical. Now back to the face by the fire.

When the film stops to watch Gollum, it stops acting. The close-up of his arguing face is what Deleuze called the affection-image: a feeling registered on a face but not yet discharged into any deed. For the length of that scene nothing happens in the plot sense; we simply read intensity gathering. More precisely it is a potisign, an icon of power — the threshold face, where micro-movements of the brow and lip build pressure without release. This is the real achievement of Andy Serkis's performance capture, and it is worth being plain about why it matters. For seventy years synthetic creatures could share a frame with actors but not a feeling. King Kong gave O'Brien's stop-motion ape eyelines and pathos; Harryhausen's skeletons in Jason and the Argonauts could trade blows convincingly with men in the same shot. What they could not hold was the flicker — the half-second of expression crossing a face. Weta digitized exactly that. Gollum is the first wholly synthetic figure who can sustain the affection-image, and the film knows it, which is why it gives him the camera and the silence.

And in that scene something genuinely strange leaks into this otherwise classical film. Gollum is two identities that do not cancel — Sméagol who wants to be good, Gollum who wants the Ring — and the film makes it undecidable which is the real one. Deleuze had a name for this figure, borrowed from the modern cinema of time rather than action: the-forger, a being of serial, shifting identity, where truth and falsehood stop being a settled question and become something the character is still producing. Everywhere else The Two Towers obeys the contract of the action film, where acts resolve and good and evil hold still. In this one creature that contract dissolves. The face by the fire is the place where a time-image flickers inside a movement-image epic.

That is the film's quiet significance. Loudly, it proved that computation could rebuild the action-image at a scale the medium had never sustained — that Ran-sized battles and convincing massed combat could be staged from agent-driven crowds and composited creatures, and that the Harryhausen dream of a synthetic body holding eyelines with a human one had finally come true. Quietly, in Gollum, it did something its genre did not require: it let a digital face carry a self that would not resolve.

So next time, watch the battle for the machine — it is one of the great ones, and you can trace every move back through Zulu and Seven Samurai and Griffith. But watch Gollum for the leak. Everywhere else the film tells you what a thing is. By the fire, it admits it doesn't know — and hands the not-knowing to you.