Before I talk about screenwriting, I want to talk about fractals for a moment.
Fractals are objects where the overall shape resembles the shape of the component parts. (Yes, yes, I know that’s a vast oversimplification of the proper mathematical definition, but I don’t think I need to get into a discussion of the Hausdorff-Besicovitch dimension to make my point about the three-act structure.)
A classic example of a fractal shape in nature is a tree. View a tree from a distance, and you see a thick central trunk with branching parts. Get closer and look at one of those branches, and you’ll see that it, too, has a thick central trunk with branching parts. And those branching parts probably contain similar branching structures of their own.
To me, the best film plots are usually fractal. That is, the whole film, seen from a distance, has a three act structure. And each of those three acts can be divvied up into three sub-acts. And each of those sub-acts can be… Well, you get the idea.
Let me use, as an example, one of the best-plotted films ever written: Raiders of the Lost Ark. I’m going to get into the plot in some detail, so SPOILERS follow.
Here’s the overall act structure:
ACT 1: Indiana Jones is a cynical archeologist who alternates between finding expensive artifacts he can resell and giving boring lectures at a university.
ACT 2: He is suddenly thrust into a much bigger game when he is given the job of getting the Ark of the Covenant before the Nazis can get their hands on it.
ACT 3: Having failed at his concrete goal of keeping the Ark out of the hands of the Nazis, Jones must succeed in his larger goal of stopping them from using it to win the war.
Now, let’s take a look at Act I. You’ll notice that it, too, breaks down pretty neatly into three parts:
SUB-ACT 1.I: Indiana Jones steals a little golden idol, despite interference from his rival Belloc. As far as we know, this is the status quo of Indie’s life. Act out: Indiana and his pilot fly away to civilization.
SUB-ACT 1.II: Indiana Jones returns to his boring day job at the university, establishing a new status quo. Act out: Indie’s old friend Marcus Brody tells him that two US agents want to meet him.
SUB-ACT 1.III: Indiana Jones meets with government officials who want him to combine the academic knowledge he displayed in II with the hands-on adventurism of I to recover the Ark. Act out: Brody tells Jones that the US agents are hiring him to go after the Ark.
And as it happens, each of those sub-acts can be broken down even further…
1.I.A: Indiana Jones is slowly and steadily getting closer to stealing a gold idol. Act out: Indie has his hands on the idol… and then the bag of sand he has put in its place begins to sink into the altar.
1.I.B: As a series of death traps whir around him, Indiana Jones madly tries to get out of the temple alive. Act out: Indie discovers his rival Belloc waiting for him outside the cave. As so often happens a the end of an act, Indiana has succeeded in his concrete goal (getting out of the cave alive) but now finds that he still has work to do to achieve his larger goal (getting the idol back to Brody).
1.II.C. Indie tries to escape Belloc and the tribe of Hovitos. Act out: Indie and his pilot fly away to civilization.
And you know what? each of those sub-sub acts can be broken down further still–although I’m going to leave that as an exercise to the reader.
I don’t want to oversell this. For one thing, although the sub-acts and sub-sub acts are very similar in structure to the overall acts, they aren’t the same, any more than a branch is an exact duplicate of a tree. And I have no idea if screenwriters Lawrence Kasdan, George Lucas, and Philip Kaufman plotted things out in such a formal schematic way; it’s entirely possible that they were just going by their instincts of when the story needed to take a turn. However they arrived at it, the fractal nature of the structure means that, just when the audience gets used to the status quo, the rug gets pulled out from under their feet. But because these constantly shifting micro-acts fit together into larger and larger macro-acts, we never feel like the script is episodic; there is a clear sense that we are building to something.
And, at long last, this brings me to The Incredibles, and Alex Epstein’s argument that it has nothing to do with the three-act structure. (Warning: SPOILERS follow for The Incredibles.)
Alex claims that the film has a 7-act structure, but he’s confusing sub-acts with acts. What he describes as Acts 1 (“The Incredibles fight crime”) and 2 (“Mr. Incredible has a dead end job”) are actually the first two sub-acts of Act I. (The third sub-act, which he omits, is “Mr. Incredible tries to alleviate the tedium of his life by doing good on the sly with his old buddy Fro-Zone.”)
What Alex labels as Act 3, 4, and 5 are actually the three sub-acts of Act II, and his Acts 6 & 7 are actually the first two sub-acts of Act III. (The last sub-act of Act III, in case you’re wondering, starts off at the track meet.)
Hey Jacob,
I don’t take the following as your claim, but just to be sure: do you assert fractal-like structures are the *most* satisfying form of story? If so, how do you account for highly satisfying, non-fractal story structures? If not, do you believe there is another, more abstract structure which may facilitate an understanding of story?
Lee
Screenwriting isn’t like tic-tac-toe, where there’s some mathematical algorhythm that guarantees an optimal result, and I’m therefore reluctant to issue any universal, arbitrary dictums about what makes the most satisfying script.
For one thing, there are plenty of satisfying films that are character- or mood-driven, rather than plot-driven.
But I am willing to go so far as to say that, in the best plot-driven movies, the plots do generally fit this fractal pattern. I certainly wouldn’t claim that the writers of these films have consciously set out to fit this pattern; rather, I think they have an instinctive sense of when to change things on various levels to keep the plot twisting in interesting ways.
Notice that I’m hedging my bets by saying the best plot-driven movies generally fit this pattern. I can’t think of any that don’t–but I’d be interested in any counter-examples you can come up with.
It seems to me as if you are using the word “fractal” incorrectly. Good Script Bad Scripts deals with the love of the 3-part sequence in early hollywood; at its most basic, any dramatic situation can be broken down to conflict introduced/conflict engaged/ conflict resolved (I do think that this is a very strong storytelling method, but not the only one). What would make these sequences “fractals” is if their structure and content were simpler or metaphorical renditions of the larger screenplay- the essence of fractals being repetition more than miniaturization.
Seeing as Vertigo takes a wild turn, I’d hazard a guess that it’s a plot-driven film that doesn’t fit the fractal theory.
How about it fits the midway turning point idea: succesful films often change story/ type halfway through. Put more simply, they amp up in the middle.
Alien is an example. For the first half it’s a science fiction mystery that develops into a monster-in-the-house horror.
I’d also add that the midway idea works better as a tool rather than a theory, not to be followed religously but employed when useful and relative. Many films don’t have a clear change in story gear. Indiana Jones and the Last Crusade starts off as an action adventure and stays that way.
Jake
What you’re referring to as “sub-acts,” when I was studying screenwriting at USC, we were taught as “sequences.”
The way they taught 3-act structure there was that the 1st act was made of 2 sequences, the 2nd act is made out of 4 (2 before the midpoint, and 2 after), and the third act is 2 sequences – for a total of 8. (The 2nd act is generally twice the length of acts one and three.)
And then those sequences are broken down into scenes (a normal Hollywood movie has approximately 30 of these.)
There’s no need to use new terms like sub-act or sub-sub-act — “sequence” and “scene” seem to work well enough for me 🙂
David
David, those are all fair points. I find it a useful mental tool to break things down into sub-sub acts and so forth, but I shouldn’t present it as anything more than new phrasing for an old idea.
Additionally, it’s been pointed out to me that I’m really conflating two ideas. It’s possible to have a fractal script that doesn’t fit in the three act structure, and it’s possible to have a three-act script that isn’t at all fractal.
In any case, Jake’s point about tools vs. theories is well taken. Anybody who claims to have a foolproof method of writing a script is lying; the most anyone can say is, “Here’s what works for me.”