Carlito’s Way (1993) Review
9/10
A Puerto Rican gangster gets out of prison on a technicality, announces he’s finished with the life, and spends the rest of the film being pulled back in by people who need him to be who he was. The 9 is for a crime film built entirely on dread instead of surprise. The opening shot is Carlito bleeding on a station floor while a mural of a beach advertises the retirement he won’t reach. Everything after that’s a countdown that the audience can see and the character cannot. The withheld point is Penelope Ann Miller’s Gail, who’s written thinner than the men around her and is asked to represent the film’s entire idea of a future. Miller is good. The part isn’t built to the same standard as the rest.
Few crime films get under my skin the way this one does. I know from the first frame that Carlito dies on that platform, and I spend two hours hoping he’ll make the train anyway. Pacino plays him so decently that I forget the ending I’ve already been shown, right up until the film reminds me.
The Ending Comes First
The film opens on its own last scene, in black and white, with Carlito shot and dying while a poster of a Caribbean beach hangs above the platform. Then it goes back and tells you how he got there.
That structure removes suspense and replaces it with something better. You’re not watching to find out whether he escapes. You’re watching each decision arrive and knowing, with more information than the man making it has, exactly how much it costs him. So every small courtesy Carlito extends becomes unbearable. He gives a young cousin a ride to a deal. He takes a call from his lawyer. He does the loyal thing, the reasonable thing, the thing a man trying to be decent would do, and the film has already shown you the platform.
The ride for the cousin is the moment I dread most. Carlito is doing his young cousin a favor, the decent thing, and I want to reach into the screen and drag him out of the car.
De Palma had used this shape before and never this cleanly. The point of a foretold ending is that it converts the audience’s knowledge into pressure, and the pressure only works if the character’s choices look reasonable in the moment.
Telling the reader the ending is a trade. You give up surprise and you buy dramatic irony. That’s a stronger and longer lasting effect if the story is about character instead of plot. The requirement is that every doomed choice must be defensible. If the audience can see the mistake and the character looks stupid for making it, the structure collapses into impatience. Carlito’s decisions are all the decisions of a man behaving well. Notice how specific the opening image is. A beach poster, a station platform, black and white. The reader has to be able to recognize the ending when it arrives, so plant something visually unmistakable and never show it again until the last reel.
Sean Penn Drives the Film
Penn plays David Kleinfeld, Carlito’s lawyer, under a wig of tight curls and a pair of glasses that make him look permanently sweaty. It’s a physical transformation and it’s not the point of the performance. Kleinfeld is the film’s real antagonist and he never presents himself as one. He’s Carlito’s friend, the man who got him out, the one person with a legitimate claim on his loyalty. Every catastrophe in the film arrives through him, and every one of them arrives because Carlito is honoring a debt.
What Penn understands is that Kleinfeld is a drowning man. He’s been taking his clients’ money and their habits, and the further he sinks the more he needs Carlito to pull him out. The character has no plan. He improvises his way into murder. Penn received a Golden Globe nomination and lost, and the performance has been steadily reappraised since. It’s the more difficult of the two leads. Pacino is playing a man with a code. Penn is playing a man with none, and making him sympathetic anyway.
The scene that turns the film is on a boat. Kleinfeld asks for help with something indefensible, and Carlito goes, because the man once opened a cell door for him. Nothing in that exchange is raised above conversational volume, and it’s the moment the ending on the station platform becomes inevitable.
Penn is the best thing in the picture for me. His Kleinfeld is so needy and so sweaty that I can’t decide whether to pity him or throw him overboard, and that boat scene makes my stomach drop, because Carlito’s loyalty is the weapon being used against him.
De Palma Builds Suspense Out of Geography
The film’s reputation rests on two sequences, the pool hall shootout in the first act and the chase through Grand Central at the end.
Both work the same way. De Palma establishes the physical space with total clarity before anything happens in it. Where the exits are, who’s standing where, what’s behind which door. Then he lets the audience watch the trap close with full knowledge of the geometry.
The Grand Central sequence runs from a station platform in Harlem, through moving subway cars, onto the terminal escalators. Trains were re-routed for the shoot so Pacino could move between them. The escalator section gave the editor a problem, which was that the escalator is very long and the audience mustn’t start noticing. His solution was to cut so the tension keeps accumulating across the ride. Watch it and try to count the time. You cannot, because the sequence keeps introducing a new threat at the exact interval where you’d otherwise start doing arithmetic.
I think the Grand Central chase is De Palma at his absolute best, and the escalator is the stretch I’d show anyone who doubts him. Somewhere in the middle of that ride I stop breathing.
Suspense is a function of the reader knowing the layout. Confusion isn’t tension. If they can’t picture where the danger is coming from, they can’t dread it.
Establish the space before you need it, in a calm moment, using the character’s ordinary movement through it. By the time the shooting starts the reader should already know how many doors there are.
For a long sequence, the risk is the reader stepping outside it to notice the length. Introduce a new element on a regular beat, and make each one change what the character has to do next.
A Gangster Film About Wanting Out
The genre normally runs the other way. A man rises, acquires, overreaches and falls, and the pleasure is in the acquisition. Carlito starts at the end of that arc and wants only to stop.
His plan is small and specific: a stake in a car rental business in the Bahamas, enough money to get there, a woman to go with. He states the amount he needs and works toward it. The modesty of the ambition makes the film sad instead of operatic.
Pacino plays him with the volume down, which given the same year’s Scent of a Woman and the earlier Scarface is a deliberate choice. He’d already made the loud version of a Latino gangster for the same producer a decade before. This is the correction.
The film’s argument is that the world doesn’t permit the exit. Everybody Carlito owes something to needs him in his old role, and his own code of behavior, the thing that makes him worth watching, is the mechanism that keeps dragging him back.
This is also where the film’s one weakness shows. Gail is supposed to be the reason the Bahamas matter, and the script gives Miller so little to play that I believe in the car rental business more than I believe in the romance. I’d have traded a few minutes of almost anything else in the film for two more real scenes with her.
The narration reinforces it. Carlito talks to us throughout in the past tense, calm and a little rueful, a man explaining himself after the fact. He’s not confessing and he’s not pleading. He’s walking through his own reasoning, and the reasoning is always sound. That’s the most damning thing about it.
Why It Was Underrated Then
The film cost thirty million dollars, opened in November 1993, and took around sixty-three million worldwide. Reviews were lukewarm. It was treated as a lesser retread of Scarface by a director and star repeating themselves for the same producer. That reading was wrong on the merits and understandable in context. Scarface had become a cultural object by 1993, and a Pacino crime film from the Bregman office looked like a sequel to a phenomenon. What changed is that the two films stopped being compared. Scarface is about appetite and it’s enormous and vulgar on purpose. Carlito’s Way is about restraint and it’s quiet, mournful and formally exact. They share a producer and almost nothing else. The reappraisal is now near total. De Palma has called it among his best work, Penn’s performance is routinely cited, and the Grand Central sequence turns up on every list of staged chases. Thirty years is a long time to wait for a correct verdict, and it arrived.
Work gets misread when it resembles something famous. If your book sits near a phenomenon in setting or subject, expect the comparison and expect to lose it in the short term. The defense is formal difference. Carlito’s Way survives the comparison because its structure, tempo and moral argument are opposite to the film it was measured against, and those differences became visible once the noise faded.
Don’t sand off the differences to look less like the famous thing. Sharpen them. The features that make a work hard to place on release are usually the features that make it durable.
Should you watch Carlito’s Way?
Yes, and watch Penn. The film is remembered for Pacino and for the Grand Central chase, and I think its most interesting work is a lawyer in a bad wig destroying a man by asking him for favors he can’t refuse. It loses one point from me, for a heroine the script never finished writing, and everything else in it justifies the 9 with room to spare.
