In my experience, this is the point where screenplays are won or lost. Almost anyone can have, and has had, a cool idea for a movie. A dog accidentally gets issued a credit card — there, I just had one myself!

And a similar number of people could sketch out a basic three act structure for that movie: Act I, the dog gets the credit card and goes hog-wild, buying all kinds of things, Act II, the dog’s bills come due, and he finds he must get a job in order to pay for all the things he bought, Act III, the dog, through his experience, learns that he was happier not having a credit card after all.

Putting together an outline starts to get a little more difficult, but the treatment is where the rubber hits the road.

Basically, a treatment is a prose version of your screenplay, your screenplay, in a way, told as a short story.

You may ask, hey wait, if I was any damn good at prose, what the hell would I be doing writing a screenplay? If I could freakin’ write prose, I’d be freakin’ Raymond Carver, I wouldn’t be scraping around trying to write a screenplay! And I understand your pain, for I have felt it myself.

Okay then, let’s not think of it as prose, let’s think of it this way: you’ve just seen a really cool movie, and you can’t wait to tell me about it.

The movie’s just let out and you’re totally buzzed about it and you meet up with me at a good restaurant afterward and you have to tell me about the movie and get it all out before the food comes. Go!

“There’s this guy, Rob, and he lives in New York, and he’s got a job working for some big-deal corporate thing, and he’s all psyched because he just got transferred to Japan, and he can’t wait to go, but guess what? There’s this girl, Beth, and she’s got this totally awesome apartment on Columbus Circle that belongs to her dad, and Rob has been friends with her off and on for a long time but now he’s, like, totally fallen in love with her, but he decides he can’t really tell her that, because, right, he’s about to leave for Japan forever. So he’s conflicted about that, and then the night before he leaves, all his cool friends get together and throw him a surprise party, and Beth is there, but she’s with some other guy, some douche we don’t know, and all his other friends are there, and his best friend is, like, shooting the whole party with Rob’s video camera, so Rob can’t, like, say anything to Beth, even though he really wants to, because he’s leaving the next day and everyone wants to say goodbye and his dorky best friend Hud is shooting everything with this video camera. And get this — the whole movie? is shown from the point-of-view of Hud, through Rob’s video camera. So it’s this really cool cinema-verite kind of thing, we pick up random pieces of behavior, and we see Hud is trying to put the make on this girl Marlene, and he’s really not doing his job very well, and it’s totally funny and awesome and everything, because Hud’s, like, taping over this tape that Rob made about his one date with Beth, so like he unknowingly is, like, being a total dick. And anyway, Beth leaves the party with the douche and Rob gets really upset and everyone’s concerned and Rob’s brother Jason or somebody tells him he absolutely has to go after Beth, and then you know what happens? A GIANT MONSTER SHOWS UP AND STARTS BLOWING STUFF UP!!”

There. The above paragraph is, in essence, a treatment for the first 20 minutes of Cloverfield. I, personally, would not hand this in to a producer for consideration, but BELIEVE ME, I’VE READ WORSE.

Now, look at that paragraph again. It’s not great prose, it’s barely prose at all, but it describes the plot with a kind of propulsive energy and sense of movement, and what’s more, it does it in a language that gets across the 21st-century, internet-generation sensibility (or at least my 46-year-old’s interpretation of same). And that’s all a treatment really has to do. It has to tell the story, the whole story, and get across the general feeling of the movie. If you were writing the treatment for There Will Be Blood, it would probably be more like this:

“There’s a landscape. A harsh, unforgiving landscape. Rocks. Dirt. Punishing sun. Texas. Or Hell. And there’s a hole in that landscape. A wound. And deep in that wound, silent but for the steady pound of his pick, there is a man. Filthy, strong, mustachioed. Deep in a wound in the Earth, the man slams his pick, a tool of destiny, against the flinty walls of this hole, this grave he has prematurely dug in order to bury his soul. Sparks fly from the rocks, each one a symbol of the life of man, which flares only for an instant before being forever snuffed out.”

And so on.

Again, while the treatment should be readable, the most important thing about it is that it gets across your story points in a voice that gets across the feeling of the movie. The hard part is that it has to get across all the story, scene by scene, all the way through, with no “and then there’s this cool action set-piece I’ll figure out later” thrown in.

Now, what’s the point of this exercise? Why not just write the goddamn screenplay? Wouldn’t that just take less time?

Well, perhaps. But if you’re anything like me, what you will have at the end of your process is a screenplay no one will want to read. Because you haven’t worked out the story ahead of time as a treatment.

The point of all this pre-work work is to iron out all your plot points, character arcs and whatnot so that, when you sit down to write your screenplay, you’ve already done all the work and you can enjoy the process of writing.

Some people, I guess, can just sit down at their computer and open up Final Draft and just go ahead and start in writing their screenplay, and “feel” where it should go next, and those people can allow themselves to wander and surprise themselves and come up with something new and startling and original and amazing, something they wouldn’t have come up with if they had sat down ahead of time to think things out.

Two things:

1. I am not one of those writers.

If I start a screenplay working like that, what happens is I have a great idea for an opening sequence, then I get to page 25 or so and I realize that the great opening sequence isn’t going to work because it contradicts something that happens later in the act, but I’m loath to go back and kill my great opening sequence because I loved it so dearly when I was writing it and it turned out so nice and, well, what if I just typed up some brilliant bullshit to cover up the fact that I started without knowing where I was going?

2. I find that if I have done the difficult work of ironing out my story before I write my screenplay, I am able, once I sit down to do that, to do that creative thing, where I take chances and just “let ideas come” and “fool around” with the ideas and so forth. Once I have the tracks laid and nailed in place, I find that I can make the train engine fancy or plain or asymetrical or goofy and know that it will still get to the station on time.

There is no set length. I’ve read treatments as short as two pages and as long as 42. The ones that are two pages long, I’ve found, favor sensation over logic and leave out a lot of crucial stuff. (My favorite sentence in a treatment ever, written by a very successful writer/director, for a project that didn’t happen, was “And did I mention the radioactive sharks?” I guess you had to be there.)

Here’s the answer to your anxiety of the blank screen: go back to lesson 1: All Writing Is Rewriting. Don’t try to write the whole screenplay in one go, write it in incremental steps.  You rewrite as you go along.

Start with the cool idea. Don’t worry about making it artful, just worry about making it clear.

Here’s a cool idea: a man’s wife is murdered, and he is accused of the crime.

So, you write that down. Nothing else, just that. And maybe that’s your work for the day.

And just keep that file open on your desktop. Call it MURDER GUY or something.

Don’t worry about making it a big hit screenplay yet, just worry about getting ideas down on paper.

Ideas come to you over the days: the man accused of the crime is a surgeon. The actual man responsible for the crime has a prosthetic arm. There’s a spectacular bus crash. There’s a showdown at a dam. There’s a chase through a hotel laundry. So forth.

You write down all those ideas as they occur to you. In no particular order. Maybe you’ll use all of them, maybe you won’t. Carry a little notebook in your pocket so you can jot them down as they occur to you. I ride my bike around Santa Monica in order to think and can often be seen stopped in the middle of a parking lot, jotting little things down in my little trustynotebook.

And the cool thing about computers is they don’t care what order you write stuff down in, they’ll let you painlessly organize it later.

Okay. So you’ve got your cool idea for a movie, and you know who your protagonist is and what he wants. Congratulations! You’ve completed the most important part of writing a screenplay.

Now: who will be working against your protagonist? Well, there’s the guy who really killed the doctor’s wife, and there’s a US marshal who is out to get the doctor whether he’s innocent or not, and there’s the guy who hired the guy who killed the doctor’s wife, and there’s everybody else in the world who thinks the doctor is guilty and should be in jail. That’s a formidable array of adversaries and you’re well on the way to writing a big hit blockbuster.

Okay. So you’ve got a protagonist with a very strong want and a set of strong antagonists to oppose him. Here is where you sit down and rough out an outline.

Shouldn’t be more than a few lines. If you know who your protagonist is and what he wants, it shouldn’t be too hard to figure out a basic three-act structure for his story.

(In case you haven’t guessed, I’ve chosen for my outline model The Fugitive, Jeb Stuart and David Twohy’s masterful 1993 thriller.)

So let’s say that Act I is the doctor’s wife’s murder, the doctor’s trial and incarceration and his miraculous escape from that. That’s some freakin’ first act! You take a wealthy, successful man, kill his wife, take away all his money, his reputation, his freedom and damn near his life, all in the first 40 minutes of the movie. This is what we in the business call a “reversal of fortune,” and it’s one of the story tricks that help make hit movies.

Okay, so you just write down:

ACT I: Dr. Kimble’s wife is murdered, he is accused, arrested, tried, convicted, sent to prison, but somehow escapes.

So, all of Act I’s energy flows in one direction: Making Dr. Kimble a fugitive of justice. We could call this little 40-minute movie “Dr. Kimble Becomes A Fugitive.”

Now then, if you’ve seen The Fugitive recently, you know what happens next. Dr. Kimble runs for a few minutes, but then he realizes that he cannot run forever, and in fact, he has something he absolutely must do that will require him going back to his life and sneaking around. What must he do? He actually announces it at the end of Act I: “Find That Man!” he remembers telling someone, and so he, and the narrative, literally do a 180 and head in the opposite direction, and all of Act II, the next 40 minutes of the movie, are another expertly paced little narrative called “Dr. Kimble Finds The One-Armed Man.” Every scene is about Dr. Kimble’s pursuit of the One-Armed Man, every roadblock is there to stand in his way of Finding The One-Armed Man.

But all you need to write down for now is: ACT I: Dr. Kimble Becomes A Fugitive, ACT II: Dr. Kimble Finds The One-Armed Man.

What’s left? What’s left is Dr. Kimble Finds The Man Responsible For His Wife’s Death. And all of Act III of The Fugitive, every single scene, is dedicated to Dr. Kimble’s step-by-step figuring out who is The Man Responsible For His Wife’s Death.

So that’s just three sentences, but it is, honest to God, the bulk of the work of writing a screenplay. The problem, of course, is knowing what those three sentences are. Once you do, everything else can begin to fall into place.

Once you have those three sentences, then you can start filling in your empty spaces. Don’t open your screenwriting program, you’re not there quite yet. Just concentrate on the next step. Take your time, there’s no rush. Keep jotting things down as they occur to you.

The next step is figuring out your broad strokes. How do you begin to figure out your broad strokes? You go back to your beginning: who your protagonist and what does he want?

Who is responsible for Dr. Kimble’s wife’s murder? Well, Dr. Kimble is a wealthy surgeon — it’s probably someone in the medical profession. Who would stand to lose so much that they would sanction a murder to keep their secret? Probably someone in the pharmaceutical business. How is Dr. Kimble going to find the one-armed man? Well, he’s a doctor, he would understand something about the manufacture and distribution of prosthetic limbs, or at least he would know how to gain access to hospital records.

You see? By simply knowing who your protagonist is and what he wants, everything else begins to fall into place. Who is the US marshal pursuing Dr. Kimble? Well, if Dr. Kimble is single-minded in his goal, so should the US marshal. If Dr. Kimble is filled with moral outrage regarding his wife’s death, well, let’s make the marshal morally abstentious — make it that he literally doesn’t care if the doctor is guilty or not. That makes him a formidable adversary and will most likely win the actor who plays the character an Oscar.

Everything the protagonist does, everyone he meets, everyone who’s against him, everyone who helps him, should be, in some way, related to who he is and what he wants. This relationship can be literal, consonant, assonant, thematic or opposite, but the more you link every possible thing in the script to the protagonist and his goal the stronger your script will eventually be.

But we’re still not ready yet. What you need to do first is build your three-sentence act description into a three-page outline, a document that will give you a rough idea of the order of your scenes. This document should be very loose and informal — don’t worry about sentence structure or wording or anything like that, no one is ever going to see this document except you.

Break each act down into smaller sequences:

ACT I: Dr. Kimble’s wife is murdered, Dr. Kimble is brought in for questioning, Dr. Kimble is suspected of murder, Dr. Kimble is arrested, Dr. Kimble is tried, Dr. Kimble is convicted, Dr. Kimble is sentenced, Dr. Kimble is put on a bus and sent to prison, the bus is involved in a spectacular crash with a train, Dr. Kimble escapes and becomes a fugitive.

There, that’s ten narrative beats, that’s a goodly number for an act. It’s still just the broad strokes. And you do that for the other two acts as well. And you can try to work in the the cool beat ideas you’ve had. The bus crash — hey, maybe that’s how Dr. Kimble gets out of having to go to prison. The showdown at the dam, maybe that’s the scene when Dr. Kimble first meets his chief antagonist. The chase through the laundry, maybe that’s the final struggle between Dr. Kimble and his adversary. So forth.

The point of all this is that you don’t want to sit down and sweat through forty pages of screenplay and then suddenly realize you’ve painted yourself into a corner. Because if you sit down and sweat out forty pages of screenplay before you realize you’ve painted yourself into a corner, the temptation will be to try to come up with some absurd piece of dramaturgy that will miraculously allow you to move on and not have to go back and fix the first forty pages of your screenplay. The end result will be that, instead of spending a little time at the beginning of the process thinking out your outline, you will find yourself at the end of the process with a screenplay that doesn’t work.

You’re still not ready to start yet. Next, The Treatment.

The “guy” Mr. Katzenberg refers to is, of course, The Protagonist. The reason for Mr. Katzenberg’s mounting anxiety and anger toward my presentation is that I am wasting his time. I am describing the movie we’re making in every way but the way that matters. Because structurally, the only thing that matters in a screenplay is What The Protagonist Wants. There is nothing else.

(Mr. Katzenberg repeats his question to me many times during my work with him, so many times that I finally write it down on a postcard and stick it up over my desk. And if you are an untested screenwriter reading this journal, I advise you to do the same.)

Simply put, What The Protagonist Wants is the reason the movie is happening. Charles Foster Kane wants love on his own terms, Sheriff Brody wants to rid Amity Island of the shark, Henry Hill wants to be a gangster, Michael Corleone wants to distance himself from his family, Roy Neary wants to meet the aliens, Indiana Jones wants to recover the Ark of the Covenant, Luke Skywalker wants to get the hell off Tatooine. The movie is nothing more or less than the protagonist pursuing his goal and the things that get in his way. The stronger the protagonist’s drive, the better the story will be. The more formidable his opposition, the better the story will be.

And that is all a screenplay is. The Protagonist pursues his goal, and forces get in his way. And either the Protagonist gets what he wants or he does not, and sometimes, during the pursuit of the goal, the Protagonist’s goal changes. So Michael Corleone starts off wanting to distance himself from his family and ends up becoming the family patriarch. Luke Skywalker starts off wanting only to get off Tatooine and ends up saving the galaxy. And in some of the best movies, the protagonist’s goal changes so much that, by the end of the story, we are left with a profound, exhilarating sense of life as it is lived.

Is it formula? It is not. It is storytelling. This is how it works. There are millions of possible variations to this idea, but this is how it works. When a movie gets boring, it’s because the moviemakers have strayed too far from the protagonist’s pursuit of his goal. If a movie is uninvolving from the get-go, it’s because the screenwriter has failed to invest his protagonist with sufficient enough ardor in pursuit of his goal. Or worse yet, he has failed to give his protagonist a goal at all. The antagonists are unfocused, the protagonist gets off on a tangent, the big musical number (or action sequence) stops the show but does not raise the stakes.

Somewhere in the back pages of this journal I referred to screenplay structure as a boat. You’re building a boat. If you follow the rules, your boat will float. If you are proficient in your skills, your boat will sail. If you are remarkably talented, your boat will zoom across the water, win the race, impress everyone and bring you millions of dollars. If you don’t know what you’re doing, your boat will sink. And if you are an “artist” with some brilliant “new idea” about what a “boat” is, you will have a work of art that is not a boat.

Why does it have to be this way? Why is this rule so ironclad? Why does it work? I don’t know why it works. I’ve learned through practice and experience that it does and that’s good enough for me.

Let’s go back to that meeting again about the movie with the talking ants. Mr. Katzenberg asks me “What Does The Guy Want?!”

What do I do? This is what I do: I stammer and look at my notes and say “well, he wants to change society,” or “he wants to find a better way to live” or “he can’t help but think that somewhere there is a better world.” All these, it turns out, are the wrong answers. The protagonist’s goal cannot be vague, ideological or symbolic. It must be concrete and physically attainable. John Connor may ultimately fight for freedom, but his goal in Terminator 2 is to get his mom out of the hospital and destroy the evil robot from the future.

Why must the protagonist’s goal be physically attainable? Because movies are made of pictures. A movie is not a novel, it can’t get inside a character’s mind very efficiently. What movies do best is record physical activity: the man runs, the car leaps off the bridge, the dinosaur attacks, the man and woman kiss, the building explodes, the spaceship glides silently across the vast reaches of nothingness. Serious movies about characters thinking deep thoughts are not going to capture a very big audience, but the dumbest movie in the world about people outrunning orange fireballs and large metal objects flying through the air will capture an enormous audience.

This is not to say that a movie cannot be about serious things. Ingmar Bergman made some of the greatest movies ever made about very serious things indeed, but his movies work because, beneath his experiments in formalism, he has a remarkably strong, even old-fashioned, sense of drama.

And if you can find a movie about subjects more serious than the ones in The Godfather, let me know.

The protagonist’s pursuit of his goal can be drawn clumsily or with great subtlety and sophistication. It can be boldly stated from the first scene (“All my life I wanted to be a gangster”) or it can remain mysterious and unsettling to the end (I’m looking at you, Daniel Plainview). It can be done so elliptically as to confound (remind me to tell you about the structure of 2001 some time) or it can be hammered home with a big iron mallet (“Let My People Go”).

Can there be a movie with a passive protagonist, where the protagonist doesn’t want anything in particular, and things just kind of happen to him? Yes. I can’t think of one off the top of my head, but I’m sure there’s one out there somewhere.

Oh wait, I’ve thought of one that comes close: Bambi. I can’t tell you why a movie with no plot and a passive protagonist can be such a classic narrative and a crushingly emotional experience, but Walt Disney (Walt Disney!) pulled it off somehow.

(I often imagine Walt Disney finally becoming unfrozen one day and showing up at the studio that bears his name, and everyone there is so glad to see him and they ask the master if he has any great new ideas for movies and he says “Yeah, how about a 2 1/2 hour plotless movie that celebrates the art of symphonic music and a 61-minute cartoon about a baby elephant who learns how to fly while he’s in the middle of an 8-minute-long alcohol-induced hallucination?”)

(Perhaps we could say that Bambi wants To Learn. He wants to learn the names of things, how to behave, how to be safe, how to have fun, so forth, and in the end he learns a few lessons he would have rather not learned, and finally, through experience, achieves Wisdom. Boy, that movie blows me away.)

Even The Dude wants something — to solve the mystery of the missing girl. It takes him 90 minutes to arrive at this desire, but he finally gets there. And I would say that if there is one solid reason why a movie as brilliant as The Big Lebowski failed at the box office, The Dude’s lack of ambition would be it.

Can there be movies with multiple protagonists? Yes there can. As a rule, they don’t do as well as movies with single protagonists. Pulp Fiction would be the exception to this rule. Hannah and Her Sisters is another one.

The key to analyzing a movie’s structure is to identify the protagonist (not always as easy as it appears to be) and then trace that character’s path through the narrative. The protagonist’s path through the narrative is the meaning of the movie. I can’t think of an exception to this rule.

When I get done writing the ant movie, I sit down and watch all my favorite movies again. Now that I have the key to analyzing structure, somehow they’ve all become different movies. Things that once seemed weird or mysterious or confounding now seem obvious and baldly stated. Narratives that were once gorgeous and sweeping now seem as dry and clinical as a schematic. For a period of time, all movies are ruined by this process, I’m not seeing a movie anymore but a structure with pictures hung on it, but finally I am able to absorb this idea into my gut and enjoy these movies again, not just for their screenplays but for the moviemakers executions of their screenplays. If you have an interest in writing movies, I suggest you submit to this process.

There are many many books out there about screenwriting that have all these terms, dozens of them it seems, for all these different beats that every successful screenplay supposedly has, and I’ve tried reading a few but I can’t make any sense of them. On the other hand, I found Robert McKee’s Story to be hugely illuminating and useful. McKee gets a lot of stick from the screenwriting community and I’m not quite sure why. What Story did for me was not promote formula but identify tools. There are names for all the different parts of stories just like there are names for all the different parts of a boat, and up until reading the book I was just kind of fumbling around in my toolbox grabbing hold of whatever felt right, sticking my boats together in whatever way pleased me, whereas after reading Story I was able to look at my work and see where I had built well, where I had patched over a hole with a bit of shiny metal, where I had forgotten to attach a tiller, et cetera.

Also, I found David Mamet’s On Directing Film extremely helpful.