Screenwriting 101: Le Trou, and The True
I’m very angry that I’ve gone this long and nobody ever bothered to tell me about Le Trou, Jacques Becker’s exemplary 1960 prison-break movie. What am I paying you people for?
Longtime readers of this journal will know that I am strong supporter of a cinema that shows, simply and clearly, images of people doing stuff. And this past year has been a great, great year for movies that are primarily well-constructed dramas that show simple, clear images of people doing stuff. No Country For Old Men and There Will Be Blood being the latest examples. Control being another. Cloverfield another.
A movie that simply and clearly shows images of people doing stuff is almost always stronger than a movie that shows people sitting around yakking. Great dialogue is always a plus, but it should function as icing on a cake — the drama should be in the images.
And I’m going to go out on a potentially-heretical limb here and say that a movie that simply and clearly shows images of people doing stuff is better than a movie that elaborately and cleverly shows images of people doing stuff. That is to say, any movie where you walk out saying “what incredible camera moves!” has perhaps dropped the ball dramatically.
And another thing: death to backstory. I don’t care about what happened to characters before we meet them. I’m interested in what’s happening to them now. The past should only impact on the story to the extent that it informs the tensions between characters and their actions.
And one more, related thing: any time a character sits another character down and proceeds to tell all about himself, either that character is lying or you’re watching the work of a bad screenwriter. People don’t do that, people don’t sit around talking about who they are. People, in fact, do the opposite — they use language not to express their true feelings but to obscure them.
(This is, incidentally, why the Coen Bros write so compellingly on behalf of non-talkers — they’re the stronger, more compelling, more truthful characters. Language is a tool to obscure intent, but actions never lie. Think of how much less interesting Chigurh would be if there were a scene where he called up his old girlfriend and talked about how when he was young he had ideals, but then life destroyed his soul.)
Anyway, this is a roundabout way of telling you that Le Trou is pretty much a perfect movie. I had seen one other Jacques Becker movie before, Touchez pas au Grisbi, which I liked a lot, but it didn’t prepare me for Le Trou.
Here is the plot of Le Trou: five guys in a prison cell in Sante Prison in Paris attempt to tunnel their way out.
And that, pretty much, is it. There are five guys, and they want to get out of prison, and they proceed to try to do that. And we watch them.
They don’t talk a lot about stuff, there are no flashbacks, we barely know anything about them. This is not to say that they are undifferentiated. They are each very different men, they just don’t sit around yakking about their differences. They have vastly different faces. They move differently, they think differently, they react differently to stimulus, they bear up differently under stress. Maybe we like one more or less than others, maybe we don’t. The screenwriter has told us everything we need to know about the characters, as it pertains to the situation, that is, trying to tunnel out of this prison cell. There is no scene where they sit around and talk about their hopes and dreams and “what they’ll do” when they get out of the cell. They’re in a prison cell and they want out, if an audience doesn’t empathize with them about that, no amount of backstory will ever help.
At least half the movie is set in the cell. Only glimpses of the larger prison life are seen. The lighting is flat and unemphatic — it is no more atmospheric than an episode of The Rifleman. If the camera performs a move more complicated than a dolly shot I’ll eat my hat.
There is an absolutely brilliant, brief scene at the beginning where we meet an auto mechanic who talks to the camera about how this is a true story that happened to him. And later we’ll meet this guy in the prison and still later we’ll see why this simple, straightforward scene is crucial to the telling of the story.
There is an almost paralyzing level of suspense throughout the movie, but one early scene deserves special mention.
The men have been planning this tunnel. We don’t see them plan, there’s no scene where they all crouch around a diagram on the floor and talk about their plans, all that happened before the movie started. They planned it before we got there and now we’re going to watch them implement their plan.
This is the scene: one guy, the guy we met at the beginning of the movie, his name is Roland, looks a little like Anthony Quinn, goes to the wall of the cell, where a metal cot is folded up. He unscrews a bolt on the cot leg and we go in for a close-up. We see that he is missing two fingers. Why is he missing two fingers? Because he’s not an actor. He really is an ex-criminal. He’s the guy we met at the beginning of the movie, and apparently he is really missing two fingers. And now he’s showing us how he went about trying to get out of this cell. And we’re going to watch him do it. What could possibly be more fascinating?
And he takes the cot leg over to his cellmates and they look at it and say that “it will do.” Do what? What are they going to do with it?
And they go to the corner of the cell and they pry up the floorboards. (Apparently, prison cells in Paris have floorboards. Live and learn.) And we watch them do it.
And from here on out, the camera doesn’t move. It’s a high shot, looking down at the corner of the cell, and we don’t see the prisoners faces and there is no dialogue to speak of (so to speak), and this is what we see:
Roland’s hands with the missing fingers pry up the floorboards, exposing the cement floor beneath. And then he takes the metal cot leg and starts pounding at the cement. And it makes a terrible racket, and we’re sure that everyone in the prison can hear it, and we’re sure that the guards are going to come rushing. But they don’t. And what’s more, we don’t cut away to show the potential of any of that happening. And the pounding of the metal cot leg barely leaves a scratch on the cement. And so Roland decides he’s going to have to pound harder. And so he does. And tiny little chips of cement come chipping off.
And we realize, we are watching a real pair of hands gripping a real metal bar, pounding away at a real cement floor. And if those hands gripping that bar don’t succeed in creating a hole in that small patch of concrete (Le Trou means, literally, “the hole”), everyone in that cell is going to die. So we sit there in ever-escalating suspense as tiny chips of concrete give way to bigger chips of concrete, and there are no cuts (although I think there are a couple of dissolves), and Roland gets tired and hands the bar over to cellmate Manu, and we don’t see either of their faces when this happens, and Manu pounds away at the floor for a while, and after a while he hands the bar back to Roland, who pounds away some more.
This shot goes on for four minutes. Four minutes of hands gripping a bar, pounding at a cement floor. Until finally, somehow, a hole begins to form, and we realize that we’ve just watched, spellbound, real hands making a real hole in a real cement floor.
This, to me, is cinema. What could illustrate the lives of the men in that cell better than that shot? What could better dramatize their want? Not long scenes of them talking about their former lives, that’s for sure. Whoever these men were, who they are now are men in a concrete cell, trying to get out, and that’s all there is to it. The floor has to be real cement, because the floor is, in a sense, a genuine antagonist — the men are struggling against their immediate physical reality. If there were cuts, or if the floor were not convincingly real cement, we wouldn’t feel the same level of suspense, nor the same level of triumph when they finally break through. The missing fingers are a compelling detail, but they are ultimately icing on the cake, production design — the scene would have worked whether Roland was missing his fingers or not.
The rest of the movie goes on to fulfill the promise of that shot, but I’ll leave that for you to discover and enjoy.
(There is a similar scene in Grisbi — Jean Gabin, a gangster, comes home after a hard day of gangstering and walks into his kitchen and prepares his dinner. And if memory serves, in a single, sustained shot we watch him sit down and cut some bread and get out the butter and spread it on the bread and eat it.)
The brilliance of the scene at the beginning with Roland, where we see him as an auto mechanic, is that because of it, we grant that the men get out of the cell, and that knowledge, paradoxically, raises the level of suspense. Every chance the men take after creating that hole in the floor of their cell, every step they take, is somehow made more urgent by our knowledge that Roland, somehow, is today a free man. We don’t see the other men in the prologue, and we gather that some are actors and some are not (and some were not actors when they shot the movie, but went on to become actors because of their work in Le Trou), and we wonder if some lived and some died and how this is all going to turn out. And the reality of the situation turns out to be nothing we expect it to be.
Although most of what we learn about the men is in their faces and physiques, there is one character who talks. And we learn about his life outside of the prison. And so naturally we don’t trust him, and maybe we shouldn’t. And that’s all I’ll say about that.
Hmm. I must’ve not gotten the check yet.
I think you’re absolutely right. I’ve heard of Grisbi but have never seen any of Becker’s work and until today I never even heard of Le Trou. It sounds like it would stand alongside the works of Dassin and Melville. I’m a sucker for heist films (and their cinematic cousins, escape films) for the very reasons you outlined. I like it when you get just enough info to feel like you’re in on the job but not so much that you can’t be surprised by the more clever parts of the plan. I’ll definitely keep my eyes open for this one.
Can’t believe that I never told you about Le Trou
I think I was keeping it a secret…shhhhh- nobody tell Todd…
In the same vein- but way more Christian…hate to keep harping on Bresson, but you should check out
Un condamné à mort s’est échappé ou Le vent souffle où il veut (A Man Escaped or: The Wind Bloweth Where It Listeth)
It’s worth its weight in title length.
I’ve got nothing against Bresson — I just watched Diary of Johnny Cash, what more do you want out of me?
One of the masterworks of the genre, of course, is the 25-minute silent heist sequence in Rififi.
Your description of the prisoners digging the hole reminded me of the heist scene in Rififi, a movie that is in the DNA of every heist movie ever made. I love that Dassin shows the characters talking about going through the ceiling; how they plan to muffle the alarm and break the safe, etc. and just when you think you how it’s going to go down, then he reveals why Cesar would bring an umbrella to a jewelry heist which proves integral to the plan. That Cesar is played by Dassin himself makes this little touch all the more special. You may have already seen it but Ad ogni costo AKA Grand Slam is another worthy entry in that genre.
while reading your thoughts on dialogue vs. action, i suddenly had the urge to ask your thoughts on the movie version of Who’s Afraid of Virginia Woolf?
Granted being based on a play means it plays by a different set of rules, but considering the film features characters talking a whole lot and doing relatively little (aside from great scenes like George pulling out his fake gun and pretending to shoot Martha), through your own guidelines the movie probably shouldn’t work. The film lives and dies by dialogue. But i feel like the movie works, and it’s a real treat to watch.
So basically my question is…can the icing ever actually be the cake?
It’s been a long time since I’ve seen Woolf. The play, as I recall, is very dense and full of incident. The fact that there are so many discrete “beats” of relatively short duration may help it cinematically.
Now then: the characters of Woolf are academics, and to them, an evening of sitting around drinking and talking is action, which is an important thing to note. This is what they do for a living, is talk.
The other thing to note about Woolf is that most of that talk is utter bullshit. The action of the narrative, so to speak, is peeling back all these layers of bullshit to see what the “final truth” is. And talk is necessary to that end.
There are writers, Tarantino comes to mind, who write dialogue that seems to be the point of the whole scene, but it’s always coupled, in my mind, to the protagonist’s goal. In True Romance, Clarence may gas on about kung-fu movies, but his goal is not to impress us with his knowledge of Hong Kong cinema, his goal is to get the girl into bed. And when his father tells the mafioso the long story about Italian bloodlines, his goal is not to show off but to earn himself a painless death. There is, and should be, always a subtext to the fancy dialogue.
In the only negative review I read of There Will Be Blood, the reviewer cited the scene where Day-Lewis sits there with his pseudo-brother and talks about what he’ll do after he has a lot of money — the fact that he plainly states “I don’t like people” and it appears to be true.
I don’t agree with that reviewer, but it’s interesting that that was one of the main gripes — a scene of backstory and dreams — no complaints about any of the clear images of mining or drilling or surveying.
So do you think this has something to do with character’s doing things connecting them to a physical reality — they touch stuff around them which helps viewers want to believe in the reality of the stuff on the screen?
The scene in Blood where Plainview talks about the magic house from his childhood I also felt was the weakest in the movie, and the most pedestrian. Because the movie is highly sophisticated, I also suspect it is a red herring — the house is only what Plainview thinks he wants. It’s a classic “death of my kitten” speech, which is always bad news, and I’m sure PT Anderson knows that.
The reality on the screen may have something to do with it, but I think it’s more basic than that — film is a visual medium and excels in showing physical activity. The more specific that activity is, and the more connected it is to the protagonist’s goal, the more interesting it will be.
Was going to post this in the protagonist article, but this’ll work too…
So I’m seeing now that, contrary to what I used to think, it’s possible for a protagonist to start the story with a want already in place. (I haven’t seen Le Trou, but from your description, it sounds like the goal is “escape the prison” and it was there before the story even started.) I used to think, for whatever reason, that the protagonist didn’t have a goal — at least one that was relevant to the story — until at least a few minutes in. That is, the protagonist goes about his everyday life, maintaining the status quo, until something happens that upsets it, and the goal comes out of that. This probably goes back to my favorite movie of all time, Jaws — Brody doesn’t appear to have a goal (other than “do his job”) until he types out S-H-A-R-K-A-T-T-A-C-K. But it looks like it doesn’t always have to be like that. (Probably isn’t like that as often as I thought.)
I’m working on a screenplay that has vague structural similarities to Die Hard, and I used to think that McClane’s goal is to stop the bad guys. But thinking about it — it isn’t, is it? It’s actually “get back together with his wife” — the bad guys are just in the way of that, kidnapping her and threatening to kill her, which would crimp the reconciliation, to say the least. And that goal pre-exists the beginning of the film. (Unless I’m totally forgetting something, which, despite the number of times I’ve seen Die Hard, is completely possible.)
It’s a small thing to be sure, and probably blindingly obvious to some, but fascinating to me nonetheless.
— Kent M. Beeson
The thing about Le Trou is that it gives us an apparent protagonist, “the new guy,” whose want seems to be “to fit in with his new cellmates.” Then the movie shifts its focus half-way through Act I and becomes about Roland and Manu and their adventures in tunneling. By the time Act III rolls around, the new guy has become the antagonist, another deft, elegant sleight-of-hand by the screenwriter.
You are correct about John McClane — his goal is to reunite with his wife. He must battle terrorists in order to prove to his wife that he is a good man. Die Hard is a brilliant script, a classic and one of my all-time favorites, one that really blew my mind when I saw it. It was a real bolt from the blue, and I’ll probably do a post about it some time in the future.
I don’t think that the protagonist necessarily needs to be “at rest” at the beginning of the narrative. Quite the opposite, usually the later you can conceivably enter the story, the better. Mamet suggests throwing out the first 20 pages of whatever script you’re working on and see if the story still works.
The “20 pages rule” is not only good for screenplays, I think, but finished films as well. An anecdote: when my wife and I put in the DVD for Collateral (the Michael Mann movie with Tom Cruise as the villain, for anyone who might not recall), somehow — sitting on the remote? — we skipped past the opening chapter, so the movie started with Cruise entering a building, coming out, and getting into the cab. *We never noticed anything wrong*. Everything was completely coherent. It wasn’t until the end, when the love interest-in-jeopardy angle came into play, that I thought maybe something was up, but my wife was clueless until the end, when I went back and checked. And when I checked, I was shocked to find that the 13 minutes that we skipped were, for the most part, completely unnecessary — basically a lot of backstory for Jamie Foxx. (I felt sorry for people who saw the whole thing and had to listen to Foxx describe his lifelong dream twice in a row.)
— Kent M. Beeson
a request
I’ve seen Grisbi but not Trou, though it’s on my Netflix queue. Looking forward to it.
I love your analyses and have a request. Actually two requests. I’m fascinated by two movies, both which have a different narrative structure than the standard (perhaps because both have multiple protagonists). They’re both brilliant movies, and I’m dissecting them in my own head, but I’d like to ask you take these on (when you feel so inclined.)
First is Michael Mann’s THE INSIDER, a movie I love unreservedly though I know folks who don’t. For me, the movie works like gangbusters, some sort of hyperintellectual JAWS-about-ethics kind of movie. Dare I say– I like it better than ALL THE PRESIDENT’S MEN, though I’m sure I’m in the minority on that. The first half I feel is about Crowe’s character and the second half is largely about Pacino’s character– and it’s interesting the way the narrative reins are handed off at the midpoint by The Big Interview. It works despite the fact that not a whole lot happens, it’s almost all talking, and what’s at “stake” (at least up until the Freedom of the Press thing in the last third) is really not a whole lot– people die from smoking, and this dude wants to say how a company boosted nicotine. Why do I care? And yet I do. Would love your take on this movie.
The second is Fincher’s ZODIAC, which I just saw for a third time (the HD DVD of the director’s cut is totally amazing), but I have no earthly idea how it works scriptwise, all I know is it does. Basically the first half or so is a police procedural without any clear protagonist (maybe Ruffalo’s cop?) but the second half is clearly about Graysmith obsessively compiling the facts. It almost strikes the constructivist in me that the real “movie” here is in the second half and perhaps could have done without the entire first half– but the movie works so well I can’t really agree with that statement. I think it has a five-ish act structure, but would love to know your thoughts on this.
Re: a request
I haven’t seen The Insider in a long time, but Zodiac, it seems to me, is a narrative with multiple protagonists (3 to be precise), but the first two’s arcs end earlier than the third. I haven’t sat down to analyze the act structure but it sounds like a worthwhile way to spend an evening.
For some reason, I’m reminded of an early episode of The Wire, in which McNulty and Bunk are investigating a murder scene. They do all the usual CSI stuff – finding shell casings and paint fragments, matching wound patterns to bullet trajectories – but the dialogue throuought the entire scene consists exclusively of variations on the word “fuck”. We see the detectives pointing out clues to one another, making inferences, rejecting hypothesis and piecing together disparate clues, but their verbal comments are limited to “fuck”, “fuck it”, “fuckity fuck” and “motherfucker”.
I’m sure David Simon wrote it that way to make a point. The expositional dialogue we’re used to hearing in police procedurals makes no sense for the characters, who are presumably experienced professionals who have worked together for years and hardly need to spell everything out to each other. Further, the audience is not dim, and can perfectly well follow the characters’ trains of thought without having the dialogue provide a Dummy’s Guide. Other crime writers could learn a lot from this scene.
That sounds like a very good scene.
One of the things that impressed me about House was how the dialogue wasn’t “dumbed down” to stop and explain everything for the non-medical-specialist audience. Then I found out that, to anyone in the medical field, the dialogue in House is painfully dumbed down, because real doctors in those situations would never need to constantly remind their colleagues why their argument makes sense and explain how this disease ties in with that infection and so forth.
Apollo 13 did a good job with handling jargon and making it dramatic. It, and the “fuck” scene you cite, show that a lot of language in dramatic works is meaningless — what counts is not what the words mean to the audience but what they mean to the characters — it’s not important that Chase is right about the disease and Foreman is wrong, it’s important that Chase “show Foreman up” or “impress Cameron” or “get points from House” — the accuracy of his observation couldn’t be less important.
you know, what’s funny about this to me is that, “taken to the extreme” this viewpoint yields dialogue that a lot of viewers seem to think of as self-indulgent. it’s as if an author is self-indulgent to let characters talk only to each other rather than to us, and then to further expect us to empathize with them even though they’re not handing us exposition.
i have a few (anime, /blush) works in mind when i say this, and i’m actually not meaning the comment so much as “see, i agree it only matters what X means to the characters!” as wondering if there is a line between off-putting self-indulgence and respecting the viewer enough to let the story play out without (coming right down to it) breaking the 4th wall (because that’s all that unnecessary exposition and dumbing down dialogue really is, isn’t it?). i mean, if a script is strong enough, the phsyical events of the story are almost always intelligible to the audience, but is it really 100% true that characters should forget about us and think only of themselves? i’m not sure.
for most works, if the subject matter and action of the story is compelling enough, the viewer will be drawn in enough to imbue the dialogue with character-based meaning, even if they (viewers) don’t really understand the dialogue… but is “i’m letting John Protagonist talk and the viewer is CHILDISH to expect me to supply the meaning for them!” screenwriter laziness, or on point? in real life, it’s difficult to say with certainty if objective truth and personal bias overlap when it comes to understanding someone’s words, but should that ambiguity be left in art? or more to the point, if you as a writer see that ambiguity, is it lazy or heroic/appropriate for you to leave it in?
also, i’m drunk, so if this is ridiculous, my apologies!
There are a lot of issues you bring up here, which I should probably get to in time, but let me start by telling this story:
In 1999 or so I was in Paris for the first time, and drunk on it. What a gorgeous freakin’ city. Having loved French movies for a long time, I decided I was going to go see a Real French Movie, In French, Without Subtitles, just to have a pure French moviegoing experience.
I picked Patrice Leconte’s La Fille sur le Pont. I’d seen a couple of Leconte’s movies before and I’m a big Daniel Auteuil fan. And the movie started and the plot got underway and I got totally sucked in, found the movie engrossing and mysterious and lyrical and very powerful, in spite of the fact that I understood perhaps five or six words in the entire thing.
Later, back in New York, the same movie opened near my house and I went to see it again, this time with subtitles. And I found that all the things I found weird and mysterious and lyrical were about 100% more pedestrian and ordinary than I had originally thought, because the subtitles explained why everything was happening.
It’s like someone once said about the Mona Lisa: if DaVinci had put, at the bottom of the painting, “She’s smiling because she’s hiding a secret from her lover,” nobody would care anything about the painting.
David Mamet puts it a slightly different way: if you’re out on the street and suddenly a car pulls up to the curb and two guys get out and grab another guy on the street and start beating the crap out of him, that’s probably going to be the most exciting thing you see all day. If those same two guys get out of that same car and go up to the same guy and say “You there! You are in deep debt to our employer and he is losing patience with your disrespect!” the dramatic impact is greatly lessened.
It occurs to me that this rule also applies if the two guys are looking for you instead of the other guy.
It’s not just dialog, either. I have a friend of mine that’s a historical military buff and gets really bothered by a lot of movies where the German or the Turks or whatever are using the wrong guns for the era. Likewise, I’m a tournament-level Go player, and I cringe at the scenes in Pi or A Beautiful Mind where they play Go, even though I know they had help from strong players — they just seem to have ignored it during filming.
Of course, this probably has nothing to do with the writing. But it does call to mind another phenomenon in which the character says something wrong (i.e. gets a scientific fact wrong) and I start this spiral of conjecture — does the character know this is a lie? Does the writer know? Does the writer know we know? Are we supposed to know the character is wrong? …etc.
I know someone who refused to see Gladiator because there was a sandal in the preview that was not historically accurate.
This gets into another problem of filmmaking separate from screenwriting, the fact that technical information is inherently dramatically useless. It’s nice when the science of a movie matches up with the drama, but in the development process, when push comes to shove, the science goes out the window if the drama demands it.
The screenwriter can do all the research he or she wants, and work it into the script as diligently as possible, but then on the shooting day there’s going to be a director with a million things on his mind who’s not going to care about any of that — the shot will read better from a certain angle and so the Go game will be restaged at a moment’s notice and that will mean that one of the players has to make a bonehead move and the director will say “Well, nobody will notice.”
I disagree that technical information is dramatically useless. (Although I grant it’s usually not a screenwriting issue.) I think is just a question of playing the numbers — incorrect details will usually ruin the dramatic effect for someone. You wouldn’t dress someone up in a Roman Centurian outfit for a Vietnam epic, because everyone would immediately know that was wrong. The finer details are just quibbling over price, as it were.
One tries to get enough details right that it works for the majority of the audience, which is usually enough. I do like to think that every one of us has at least one area of expertise that, when dealt with inaccurately, can drive one up the wall.
Incidentally, here is a fine example of the issue at hand.
I’ve always said, before you and I are dead there will be a movie about how George Washington and Abraham Lincoln team up to fight Hitler.
But movies like Spartacus and Braveheart and Gladiator do it all the time, compress centuries of history into a year or so and pretend like they’re doing us a favor. And those movies are all very well-regarded.
Not to mention Citizen Kane and the pteradactyls.
I hope, given your hypothesis, that you live a very long time.
Then think of this: less than sixty years separate the events of The Untouchables and its movie version, and yet the only facts that are preserved from one to the other is that there was once a treasury agent named Eliot Ness and a gangster named Al Capone and a city called Chicago. Every other aspect of the story has been turned into a cartoon.
Not to mention Jesse James, who was being turned into a cartoon character while he was still alive.
I don’t think we need to look nearly that far in the past for examples of cartoonification. We are in the middle of a presidential race, after all. Every person running has at least two cartoon images of themselves out there that are better known than the real person (who we can’t really claim to know even just a little.)
I imagine there’s some image-creation going on in Hollywood, too.
One of the most striking realizations I’ve had in my adult life is how little of what we take for granted that we know about the world is actually true.
Good commentary, as usual. I wasn’t too fond of the B story this time out, but I suspect that’s because we haven’t spent enough time with the Moppets (who, it must be said, aren’t the most pleasant characters to be around). I did notice that they’re another of Jackson & Doc’s bantering pairs (like 21 & 24, Watch & Ward, and Pete & Billy), so maybe we’ll get to see more of them bickering with each other (as opposed to violently striking out at The Monarch’s henchmen, which hardly endears them to us).