Inception

What holds it together is architecture, in the literal and structural sense. The film teaches you its rules through images before it explains them in words, most memorably when a Paris street folds up and over itself like a closing book, and when an endless staircase loops back on its own beginning. These moments are built to be legible: Nolan wants you to understand the geometry of the trick so the later payoffs land. That clarity is a deliberate choice, and it comes with a cost the film knowingly pays. There is a great deal of talking here, characters narrating the machinery so the audience can keep up. When the dialogue is working, it is a briefing; when it isn't, it is a manual.

Leonardo DiCaprio carries the emotional freight as Cobb, the thief who leads the crew, and he keeps a private grief alive under all the planning. It is a performance that has to hold rational calculation and raw feeling in the same body, and DiCaprio never lets either go slack — the strategist's composure and the widower's exhaustion read in the same face. Around him the ensemble is sharply cast. Joseph Gordon-Levitt brings a dry, buttoned composure that turns into physical comedy once gravity abandons him, fighting through a corridor with his suit still immaculate. Tom Hardy, as the crew's shape-shifting forger, sparring his way through the exposition, times his needling so the briefing scenes play as verbal fencing rather than instruction. Ken Watanabe, Elliot Page, Cillian Murphy, and Marion Cotillard each anchor a distinct register, and Michael Caine and Pete Postlethwaite lend the edges some gravity.

The craft is the argument, and the argument is practical. That rotating hallway fight, with Gordon-Levitt tumbling through a corridor as the floor becomes the ceiling, was staged on a genuine hundred-foot spinning set, and editor Lee Smith trusted the footage enough to let a long stretch of it run as a single unbroken take. The café explosion was air cannons, not pixels. Shooting on 35mm anamorphic with passages in 65mm and VistaVision, Nolan blends grand aerial sweep with handheld urgency, and rides speed-ramping and extreme slow motion to make the different velocities of the dream levels physically readable.

The climax is the showpiece: roughly twenty minutes cutting between a van falling off a bridge, that spinning corridor, and an assault on a snow fortress, all of it timed to collapse together. It is a feat of parallel construction that few blockbusters have matched for sheer control of simultaneity.

Is it as strange as a dream actually is? No, and that is a fair objection. Nolan's dreams feel real rather than surreal; the film reasons where it might have hallucinated. If you come wanting the unhinged imagery of a Buñuel or the visual fever of stranger films, you may find this too orderly, too explained. But judged as what it is, a puzzle-box spectacle built to be understood and re-entered, it operates at a level of ambition and execution the form rarely reaches. It is worth your time, and it rewards a second run.

I want to take the dissent seriously, because it is real and it is smart. The sharpest charge against Inception is not that it's dumb but that it's the opposite: too legible, too diagrammed, a dream world that behaves like an engineering problem. Wael Khairy makes the case cleanly, arguing the dreams feel disappointingly real when they should feel surreal, and pointing at The Cell as the film that got the haunted imagery Nolan left on the table. David Denby's Buñuel comparison lands the same punch from a different angle: where Buñuel slid you into reverie without a word, Nolan needs pages of dialogue to hold the trance together. I think both are correct about the mechanism and wrong about the verdict. The explaining isn't a failure of nerve; it's the price of the structure. You cannot cross-cut four levels of collapsing time and expect an audience to feel the stakes unless they know the rules cold. Nolan chose comprehension over mystery. That is a defensible trade, not a botch.

Where I'd push back on the film's defenders is the originality claim, which is the most overheated part of its reputation. Keith Garlington calls it one of the most original pictures ever made, and I don't buy it. The parallels to Satoshi Kon's Paprika are specific enough — plot beats, scenes, characters — that documented scholars have laid them side by side, and you can add Solaris to the ancestry. The heist skeleton is ancient. What's original is the assembly, not the parts, and that's a genuine achievement without needing the inflated label. Nolan himself has admitted the film's deeper resonances weren't fully premeditated, which I find more honest than the fans do.

On DiCaprio, I side hard with Sheila O'Malley, who called his work here one of the year's most overlooked, buried under plot. The scene of Mal's death — DiCaprio collapsing, screaming, and somehow not tipping into melodrama — is nearly impossible to play, and he plays it. Jon Negroni's read is right too: what makes Cobb compelling is the war between his logic and his impulse, and that war is legible on DiCaprio's face even when the script is busy narrating something else. The performance is the film's beating heart, and it's telling how easy it is to talk about the rotating hallway for an hour before mentioning it.

What would I cut? Some of the mid-film briefing. There's a stretch where the movie stops to re-explain its own physics, and I think it underestimates its audience there. Khairy's most useful complaint is that the film is too busy to sit inside its own best images — it launches the Paris café fold and then hurries past it. I'd have held on that a beat longer. There's a version of this that's ten minutes leaner and loses nothing but redundancy.

And yes, the ending. The spinning top, the cut to black before it falls or doesn't. I've made my peace with it as the correct choice: the point was never whether Cobb is dreaming but whether he's stopped caring, and he walks away before it settles. The film hands the anxiety to you and refuses to resolve it, which is either the one place it trusts the audience completely or a cheap tease, depending on your mood. I lean toward the former.

The legacy is its own footnote. This became the shorthand — the film every layered-reality show gets compared to — and the trailer's BWAAAM reshaped how a decade of blockbusters announced themselves. Villeneuve keeps it among his favorites; Stephen King ranked it high the year it landed. There's a minority view that this is where Nolan's work began to dip even as his fame climbed. I understand the argument and I don't share it. The control on display here is total, whatever the coldness costs.