8MM (1999) Review
5/10
A widow finds a reel of film in her late husband’s safe. On it, a girl is killed. She hires a surveillance specialist to establish whether it’s real and what happened to the girl, and he goes looking.
The 5 is for a film with a strong first hour, a good supporting performance, and a complete collapse of nerve in the last forty minutes. The question it opens with is serious and the answer it delivers is a man in a mask being stabbed.
Walker wrote Se7en and the connection is instructive, because this script has the same shape and doesn’t have the same discipline about what it withholds.
For an hour I’m hooked. It’s a patient, grim procedural about a man putting a name to a dead girl from records, and it feels like it could sit on the shelf beside Se7en. Then it turns into a revenge picture with a masked villain, and every bit of nerve it built goes out the window. That leaves me annoyed in a way a plain bad film never could, and the 5 is me splitting the difference between the film I admired and the one I got.
The First Hour Is a Procedural
Tom Welles works cases through documents. Missing persons records, motel tones, a bus company, a dentist, and the film spends real time on how somebody identifies a person from almost nothing. That material is the best in the picture. He establishes who the girl was by cross-referencing a missing persons file with a physical detail, travels to meet her mother, and the scene in that kitchen is the film’s high point.
The bus company sequence is the best of the procedural material. He works out which route she would have taken, finds a depot record, and the film shows him being patient with a clerk who has no reason to help him and eventually does.
The detail is a set of teeth. That’s the sort of thing that makes an investigation feel real. A dental record, a photograph and six years, and the film spends the time to show him arriving at a name instead of being handed one.
The mother sequence works because it refuses to be about the investigation. She talks about her daughter, produces a diary, explains what she thought had happened, and the film lets the scene be about a woman whose child left and didn’t come back.
Amy Morton plays her, and I think her eight minutes are better than anything else in the film. She doesn’t grieve for the camera. She’s a woman who has been managing an absence for six years and has a system for it, and seeing her walk through that system moves me more than any of the violence later on.
She also asks him one question the film is honest enough not to answer. Whether he’ll tell her what he finds, and he says he will, and both of them understand in the moment that this isn’t a promise he should be making. The diary is the object at the center of the scene. She produces it, he reads it, and what’s in it is unremarkable teenage writing. That’s the film telling the audience that the girl was a girl and not a subject.
Spend time on the mechanics of finding somebody. Records, cross-references and a physical detail is more absorbing than any amount of intuition, and it establishes competence without a single line of dialogue about it. The failure mode is the investigator who simply knows things. Let the bereaved scene not be about your plot. A mother who talks about her daughter instead of about the case is the scene that makes the case matter. A character with a system for managing loss is truer than one visibly breaking. Six years of practice produces routine, and routine is more affecting than distress.
Phoenix
Max California works in an adult bookshop, knows the industry, has read a great deal, and agrees to guide Welles through it for money.
The performance is the only alive thing in the film. Max is funny, contemptuous, well read, and entirely clear about what he does for a living and what he thinks of the people who come in.
The bookshop is where the film establishes him properly. He’s reading something serious behind the counter, a customer interrupts him, he handles the transaction without comment and returns to the book, and the film delivers the whole character in ninety seconds. Phoenix is the reason I’d sit through the first half again. He’s funny, sour and smarter than everyone around him, and the film perks up every time he walks in. I wish the script had trusted him with more than the job of tour guide. What the script gives him is the film’s one good argument. He states plainly that the material exists because there’s demand for it, that the demand is ordinary, and that Welles’s horror is a tourist’s reaction to something that’s a customer base. He also gets the line that should have been the film’s thesis. That’s about dancing with the devil and the devil not changing. It’s delivered well and the film doesn’t build toward it.
The Restraint Is Correct
The film never shows the reel properly. There are fragments, a few frames, a reaction, and the audience is given enough to know and not enough to see.
That’s the right decision and the film holds it for an hour. Welles watches the film repeatedly and the camera stays on his face. That puts the audience in the position of reading somebody else’s reaction.
The projector sequences are the best staged material. A man in a room, a reel running, the sound of the mechanism, and a face doing something the audience has to interpret.
The projector is also the film’s only recurring image and it justifies the repetition. Each time he watches it the film gives the audience slightly less of the screen and slightly more of his face, until by the fourth viewing the reel is entirely offscreen.
What’s good about this restraint is that it makes the subject the demand instead of the act. The film can’t show you what somebody paid for and can show you the man who paid. Christian, played by Peter Stormare, is where the film almost gets there. He’s a director of this material, treats it as a career, complains about budgets and talent, and the ordinariness of his professional grievances is the most frightening thing in the picture.
Stormare’s scenes are where I think the film might pull it off. A man grumbling about budgets and talent while making this material disturbs me far more than any mask, and I’d have built the whole last act around him.
Show the reaction and withhold the act. The reader constructs something worse and you haven’t put it on the page. This is the opposite of the Ms. 45 calculation and both are defensible. What’s not defensible is showing the thing and also claiming to be appalled by it.
A face reading something is a scene. Two minutes of somebody watching is more demanding of a reader than any description.
The subject of a film about demand is the customer. That’s available and almost nobody takes it, because a customer is ordinary and an atrocity is dramatic.
Then It Becomes a Revenge Film
In the last forty minutes Welles locates the men involved and kills them, and the film’s entire tone changes.
The violence isn’t the problem. The problem is that a film about complicity, demand and the ordinariness of a market resolves itself by identifying four individuals and removing them. That converts a systemic subject into a hunt. The most damaging thing about the shift is that the film stops asking about the customer. Ninety minutes establish that this material exists because people pay for it, and the last forty are about the people who made it. That’s a smaller and easier question. Machine, the masked figure, is the clearest symptom. An anonymous man in a leather mask is a horror object instead of a person, and when he’s unmasked the film discovers nothing behind it. That’s intended as a point and lands as an absence.
The intended point is a real one: there’s no explanation and no monstrous face underneath. The execution undermines it, because a film that’s spent forty minutes building a masked figure has already made him one, whatever is under the leather.
Here the film drops the thread I cared about. A dark ending is fine with me. A film that spends an hour on the customer and then decides the problem is a guy in a leather mask lets the audience down. It feels like somebody got scared of their own movie. The phone call to the mother is the one part of the ending that works. Welles asks permission before doing the last thing, she gives it, and the scene is the only moment in the third act with any weight.
James Gandolfini is also wasted. That matters because this arrived the same year as the first season of The Sopranos. Eddie Poole is a mid-level operator with his own brand of cowardice, and the part gives him one good scene in a car and nothing else.
Compare Se7en
Walker wrote both and the difference between them is entirely a matter of what is withheld at the end. Se7en doesn’t let its detective win, doesn’t show the atrocities, and closes on an outcome the protagonist can’t act against. This film shows the atrocities partially, lets the protagonist win, and returns him to his family. The reported history is that this script was considerably revised and that Walker was unhappy with the result. Whether or not that’s the full account, the finished film has two halves with different intentions and the seam is visible at almost exactly the sixty minute mark.
Schumacher is the other variable and his record supports the reading. He’s a director of considerable facility with no consistent position on anything. That serves some material well and is fatal to a script that needed somebody to hold a line. Both halves are competent, and that makes the film more frustrating to me. These are two films spliced together, and Welles’s investigation asks a question the revenge plot couldn’t care less about. I’d trade the entire revenge half for twenty more minutes of Welles working through records.
Whether Walker’s original draft resolved it differently isn’t something the public record settles, and the useful observation is available without the biography. Two intentions are detectable in the finished object regardless of how they got there.
A work with two intentions will read as a failure even when both halves are competently made. Readers detect the change of purpose and stop trusting the direction. Identify what your book is asking. Anything that answers a different question is in the wrong book.
Compare your ending to the strongest version of the same material. Here that version exists, by the same writer, and the comparison isn’t favorable.
Returning a protagonist to safety after this subject is a statement that the subject was survivable. Decide whether you believe that before you write the last scene.
Should you watch 8MM?
Half of it, and that’s an irritating answer to have to give. The first hour is a good procedural about identifying a missing girl from records, with an eight-minute scene with her mother that’s the best thing in the film and Joaquin Phoenix livening up every scene he walks into. The last forty minutes answer a question the first hour didn’t ask, and that bait and switch is exactly why I can’t go above a 5.
