Saturday, May 14, 2011

Fight scenes and drama

We saw Thor today. It wasn't the best superhero movie that's come out in the last few years, but I'll be honest: when it comes to superhero movies, I am pretty easy to please. I liked the Ang Lee Hulk and I liked the Ed Norton reboot. I liked Iron Man a lot and Iron Man 2 about as much. I am not sure I will like Captain America but I will probably give it my best. I like these movies because they are fun and silly and dramatic, and often funny, and because they are generally more good-hearted than other sorts of blockbuster.

One thing I think a lot about while I watch superhero movies is how to portray action in film. I try not to be crazy in this particular way on most subjects, but I honestly believe I would be a spectacular action director. It's not so much because I love action sequences -- although when you do them right, they can be one of the best parts of any film -- but because the standard practices in action movie-making are so awful, it wouldn't take a particularly grand talent to do better. There are a lot of different and related issues that bother me when it comes to modern action movies, but they come back to one principle: drama. Simply put, most people (most writers, most directors) don't understand how to create drama, and without drama, your fight scene will suck, no matter how big your robots are and no matter how fast your robot ninja assassins may be.

It's especially tragic because while it takes most people (especially me) a damn long time to figure out, drama is very simple and it's easy to produce. All you need is a character who wants something. Then you need an obstacle -- another person, a force of nature, a personal limitation, etc. This obstacle is going to get between the character and his or her desire. The character will overcome the obstacle or he won't. This will lead to some sort of resolution. That's drama. You can have complicated iterations on this (maybe five different characters want the same thing, maybe they want five different things, maybe they think they want five different things but really it's all different facets of the same thing) or you can have very simple ones (the character is, say, a guy with a hammer, and his goal is to break all of the windows in his office building, and his obstacle is that there are a lot of windows) but this basic structure is going to be in the vast majority of stories, and it's especially going to be in your action movie.

Of course most people who have thought a little bit about stories will know all this, and movie people especially know about this because they tend not to have the luxury of deluding themselves about what makes for entertaining art and what doesn't -- they need to make their investors' money back. But movie people can be as lazy as anybody else, and drama requires hard work and a lot of honest self-assessment, so they tend not to bother. Here, for instance, is the climactic sequence from The Dark Knight. Now, I like The Dark Knight pretty well, and most people I know like it even more than I do. But can anybody tell me what the hell happens in this scene? It's dark, the stakes are ill-defined, I have no idea how many guys are in there with Batman, and even if I did know I doubt I would believe it. I'm not saying the scene would have been easy to shoot -- it would be, doubtless, very difficult, and I'm sure it was quite elaborate and carefully planned and the people involved understood it completely. But that's all just busy work designed to make them feel as if they did their jobs, when they failed to do the one thing that counts: to tell a story, a dramatic one, with stakes. The movie as a whole accomplishes that, but its action scenes never do. There's nothing to define Batman as a character here except for the fact that he's winning the fight. There's nothing to define the bad guys at all. And, apart from the fact that he's probably going to save some dudes from the Joker, we're not even sure why he's fighting in the first place. Seeing the whole movie helps a little with these problems, but even in context the sequence is such a dark, indistinct mush that it's nearly impossible to remember what we're there for (I forgot both times I saw the film).

Here's another example from The Bourne Supremacy (and this one, YouTube will let me embed!):

The Bourne films are sort of notorious for this: the cameras are so shaky and the editor is cutting so frequently that it's legitimately difficult, at times impossible, to discern what's going on. When Jason Bourne escapes the guys at the beginning of this clip, how does he do it? Well, he hurts them somehow. I don't know what he did to those guys, but they didn't seem to like it. Then he gets in the car and it's only because you saw him get in and then you see it move that you know for a fact he's started the car and is now driving it. Some cops seem to be driving around in an unrelated part of town but then they're right behind him. Dudes are shouting at each other, dudes are just showing up left and right, there's an SUV on sort of its own adventure, I just don't know what's going on. It's awful. You end up watching the actors' faces for cues -- does our guy think this is good news? Does he look happy or scared? But they all have this grim expression of determination. It takes magnificent levels of concentration to know anything about the story here, which is especially troubling in a film where -- as is the case in most action movies -- we know the hero is going to win. If we know the hero is going to win, then the only question worth asking is how. How the hero wins becomes literally the only thing we're there to find out, and the Bourne movies are an especially acute case of this because the whole point is seeing how this one awesome dude will do a lot of seemingly impossible things against all odds. But you can never actually see what he's doing, so the movies are terribly dull and I hate them.

When it comes to action, then, we can establish a few principles by the negative examples offered above: above all else, we need to be able to see our character clearly. The more of him or her we can see, and the better we can see the character, the more we'll understand the drama of the scene. Secondly, we need to know the character's goal. It's best if during our climactic fight scene the goal is often in the frame with the character, just out of reach. Thirdly, we need to know exactly what the character has to do to get what he wants, or at least we need to know what success would look like if it were possible. One of the maddening things about the Bourne movies is that the character is constantly moving with tremendous purpose, and yet we've got no idea where he's going. How will we know when he gets there?

This all sounds perhaps a bit reductive, but think of the power of sports. Nothing could be simpler. In basketball, there are some guys in a relatively small rectangle who need to get a ball into one of two circles that are only slightly larger than the ball. The ball has to go through these circles from above, each team only gets to use one of the circles, and there are some basic rules designed to keep the game from becoming a brawl. Beyond that, all of the complicated strategies and maneuvers you see, and all the years of physical training and practice and so on that lead them to this point, are designed around the basic problem of how to get that ball to go through that circle.

I personally like Football better because the situation is even more stark. There's this really long rectangle of grass. One team wants to take the ball to one end of the rectangle. The other team wants to get that ball to the opposite end. There are a few basic rules designed to keep them from killing each other, and to give them time to think and strategize, but all you really need to know is that each team wants to go in the opposite direction with that ball. That's drama. It's simple, elegant, beautiful, and many people will spend their entire adult lives thinking, writing, reading, and talking about moving that ball. They will watch a series of almost indistinguishable games wherein that ball moves up and down the field, slowly, over the course of hours. They will love every second.

Action sequences need to be more like football.

To show you what I mean, let's look at the climax in The Deer Hunter, a scene wherein protagonist and antagonist barely move, wherein the camera is terrifically steady, wherein there is in fact very little that could really be called "action." This is one of the most dramatic scenes in film history, and it's more exciting than every summer blockbuster from the past ten years combined. (It should go without saying that the remainder of this post will spoil the living fuck out of The Deer Hunter, which is an amazing movie you should really see sometime.)

Here's what you need to know to understand this movie: Michael, played by Robert De Niro, wants to bring his friend Nick back from Vietnam. Nick, played by Christopher Walken, wants to stay in Vietnam playing Russian Roulette until he dies. The rules of Russian Roulette are as follows: You put one bullet into one chamber of a revolver. The revolver has six chambers. You spin the chambers so you don't know where the bullet is. So does your friend. Then you take turns aiming at your own head and pulling the trigger. The first time you do this, the odds are one in six that you'll die. The second time, it's one in five. The third time, it's one in four. The fourth time, it's one in three. The fifth time, it's one in two. If you pull the trigger the sixth time, you'll die. You have to hope your friend friend finds his bullet before you do. This is all you need to know about this scene, although the rules they play by are even more simple. (The most relevant stuff begins at 2:33.)

Of course what makes this scene great is that everything that comes before it invests each second with tremendous emotional power and truly terrifying stakes. But you could put this scene, as is, without any context, at the end of any movie, and it would still be really powerful for any audience member who understands Russian Roulette. Michael doesn't want to play the game because if he plays it long enough, either he or his friend Nick will die. This is a certainty: he's seen the bullet. He has to get Nick to come home before that happens. Nick, on the other hand, is determined to keep playing the game. And we can see from the deadness of Nick's eyes that he'll probably get what he wants, that the odds are against Michael, that Nick is going to play until one of them dies -- and that if Nick isn't the one that dies, he'll go on playing until he is.

Crucially, we very likely know how this will end. It wouldn't be satisfying for Michael to die because it wouldn't change anything (Nick would just go on playing anyway) and it wouldn't be realistic for Nick to come home (we want to believe it, but we can't: he's too far gone). We know that Nick will probably lose. So we watch Michael to see what he'll do to save Nick, to see how far he'll go for what he wants. We watch to see the face he'll make while he pulls the trigger, and to see how he'll carry the burden once Nick is gone.

The situation is so simple. Nick is not only the antagonist, he is also the object of Michael's desire; Michael hates Nick for putting him in this position, for thwarting him, but he also wants Nick. He wants to save him. The question is not whether someone will die so much as who, and the answer is something most people will know instinctively. It becomes about how he carries that burden.

To return to Thor for a moment, this is part of what makes the film effective. We don't entirely understand the powers each character brings into each scene, which means that the action sequences are not quite as simple as Russian Roulette. But we always know at the outset how the fight will be resolved: we know that an entire planet of frost giants can eventually best Thor, that Loki can defeat the trusting Heimdal (temporarily), that mighty Thor can best the monster that his friends cannot, and that Thor will defeat Loki. None of these scenes has a surprising outcome. The simplicity of Thor's hammer is also the strength of the film: the question becomes how these things will happen, and, partly because of the bright and colorful superhero movie's mandate to show beautiful bodies in motion, we always see more or less exactly how. This is effective storytelling, and it's exciting action. (It also helps that none of these scenes are very long.) It's not as good as The Deer Hunter, but it's not bad either.


  1. I think my problem with the batman fight scene is its narration. It uses narration to explain wtf is going on and helps me care. I know that if Morgan Freeman, the narrator God will hire for his documentary of human history, wasn't telling me what was going on, I wouldn't give a shit.

  2. Yes, they tried to fix it that way and it just makes things worse.

  3. I liked Thor. I think the Batman and Iron Man movies have been the best of the superhero bunch, though.

    I've also grown tired of shaky cameras during fight scenes. Unfortunately, I don't think their prevalence will decrease.

  4. I would be interested in seeing a short film of the man trying to smash all windows in his office building.

  5. Yeah Tim, I wrote that and then I thought, "Wait, that sounds pretty good."

  6. I was high as a kite when I saw Thor. It was 11am (because that's the discount showing) and it was 3D (because that's the craze) and I'd swallowed some X (because that's what I had). I did not buy popcorn. It is possible that these (the X and lack of popcorn) are the reasons that I feel Thor sucked. But I think Thor sucked just because it was a shitty movie. I got a half stock when I heard Kenneth Branagh would be directing the film. But then I saw the film and it was like I had never had a boner in my life.

    My problem is the exact thing that you say makes the action scenes work. And I say, yea, the action scenes work but the movie is shit nonetheless. The plot is entirely obsessed with identifying the stakes for the main characters and structuring the hierarchy of ass kicking ability (as you so neatly outline) so that when the action does happen it is entirely comprehensible (and, as you say, predictable). I vote for lyricism in film. Action scenes included. Yea there are a lot action scenes that rely on the gimmicks of quick cuts and handheld seizures, but don't write off that entire aesthetic because the scripts that most often use it aren't taking the time to worry about characterization anyway.

    I guess maybe you have never been a fight. I've been punched many times and I cannot always identify where the punch came from or even where it landed. Often I am bloodied and running and crouching behind a wall and I have no idea how this came to pass.

    The thing with Thor is that the script is so meticulous about getting us to understand these parallel universes and their inhabitants and their relative asskicking abilities (the big picture) that we never get to sink into any of the specific emotional moments that make tragedies work.

    So what am I saying
    1) Thor was not a good film
    2) Don't be so harsh on the quick cut in action scenes

    My favorite action scene of recent years

    And yea, this is more like what you're talking about. So maybe I don't disagree with you as much as want to temper your harsh critique.

    So, really what I'm saying
    1) Thor was not a good film