THIS BOOK IS DEDICATED TO MIKE HAUSMAN.

They are most happy who have no story to tell.

—Anthony Trollope, He Knew He Was Right

I would like to thank my editor, Dawn Seferian,

for her great patience; and Rachel Cline, Scott

Zigler, Catherine Shaddix, and Elaine Goodall for

their help in the construction of this book.

PREFACE

This book is based on a series of lectures I gave at the film school of

Columbia University in the fall of 1987.

The class was in Film Directing. 1 had just finished directing my

second film, and like the pilot with two hundred hours of flying time,
I was the most dangerous thing around. I had unquestionably
progressed beyond the neophyte stage but was not experienced
enough to realize the extent of my ignorance.

1 offer the above in mitigation of a book on film directing written by a

fellow with scant experience.

In support of the proposition, however, let me suggest this: that the

Columbia lectures dealt with, and endeavored to ex-

PREFACE

plain, that theory of film directing I had concocted out of my rather

more extensive experience as a screenwriter.

There was a newspaper review lately of a book about the career of a

novelist who went to Hollywood and tried to succeed at writing
screenplays. He was deluded, the reviewer said, in this pursuit—how
could he have hoped to succeed as a screenwriter when he was nearly
blind!

The reviewer exhibited a profound ignorance of the craft of

screenwriting. One does not have to be able to see to write films; one
has to be able to imagine.

There is a wonderful book called The Profession of the Stage

Director, by Georgi Tovstonogov, who writes that a director may fall
into one of the deepest pits by rushing immediately to visual or
pictorial solutions.

This statement influenced and aided me greatly in my career as a

stage director; and, subsequently, in my work as a screenwriter. If
one understands what the scene means, and stages that, Mr.
Tovstonogov was saying, one will be doing one's job for both the
author and the viewer. If one rushes, first, into a pretty, or pictorial,
or even descriptive staging, one may be hard-pressed to integrate
that staging into the logical progression of the play. And, fiirther,
while so hard-pressed, and while working to include the pretty
picture, one will undoubtedly become wedded to its eventual
inclusion, to the detriment of the piece as a whole.

This concept was also stated by Hemingway as, "Write the story, take
out all the good lines, and see if it still works."

My experience as a director, and as a dramatist, is this: the

piece is moving in proportion to how much the author can leave out.

A good writer gets better only by learning to cut, to remove the

ornamental, the descriptive, the narrative, and especially the deeply
felt and meaningful. What remains? The story remains. What is the
story? The story is the essential progression ojincidents that occur to
the hero in pursuit of his one goal.

The point, as Aristotle told us, is what happens to the hero . . . not
what happens to the writer.
One does not have to be able to see to write such a story. One has to
be able to think.

Screenwriting is a craft based on logic. It consists of the assiduous

application of several very basic questions: What does the hero want?
What hinders him from getting it? What happens if he does not get
it?

If one follows the norms the application of those questions will

create, one is left with a logical structure, an outline, from which
outline the drama will be constructed. In a play, this outline is given
to the other part of the dramatist—the ego of the structuralist hands
the outline to the id, who will write the dialogue.

This conceit is analogous, I think, to the case of the structuralist

screenwriter who gives the dramatic outline to the director.

I saw and see the director as that Dionysian extension of the

screenwriter—who would finish the authorship in such a way that (as
always should be the case) the drudgery of the technical work should
be erased.

PREFACE

I came to film directing as a screenwriter, and saw the craft of

directing as the joyful extension of screen writing, and taught the
class, and offer this book accordingly.

David Mamet Cambridge, Massachusetts Spring 1990

DIRECTING

^0^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

FILM

STORYTELLING

The main questions a director must answer are: "where do I put the
camera?" and "what do I tell the actors?"; and a subsequent question,
"what's the scene about?" There are two ways to approach this. Most
American directors approach it by saying, "let's follow the actors
around," as if the film were a record of what the protagonist did.

Now, if the film is a record of what the protagonist does, it had better
be interesting. That is to say, this approach puts the director in a
position of shooting the film in a novel way, an interesting way, and
he or she is constantly wondering, "what's the most interesting place
to put the camera to film this love scene? what's the most interesting
way I can shoot it plainly? what's the most interesting way that I can
allow the actor to

DAVID MAMET

behave in the scene in which, for example, she proposes to him?^^

That's the way most American films are made, as a supposed record
of what real people really did. There's another way to make a movie,
which is the way that Eisenstein suggested a movie should be made.
This method has nothing to do with following the protagonist around
but rather is a succession of images juxtaposed so that the contrast
between these images moves the storj forward in the mind of the
audience. This is a fairly succinct rendition of Eisenstein's theory of
montage; it is also the first thing I know about film directing,
virtually the only thing I know about film directing.

You always want to tell the story in cuts. Which is to say, through a
juxtaposition of images that are basically uninflected. Mr. Eisenstein
tells us that the best image is an uninflected image. A shot of a
teacup. A shot of a spoon. A shot of a fork. A shot of a door. Let the
cut tell the story. Because otherwise you have not got dramatic
action, you have narration. If you slip into narration, you are saying,
"you'll never guess why what I just told you is important to the
story." It's unimportant that the audience should guess why it's
important to the story. It's important simply to tell the story. Let the
audience be surprised.

The movie, finally, is much closer than the play to simple

storytelling. If you listen to the way people tell stories, you will hear
that they tell them cinematically. They jump from one thing to the
next, and the story is moved along by the juxtaposition of images—
which is to say, by the cut.

People say, "I'm standing on the comer. It's a foggy day. A bunch of
people are running around crazy. Might have been

ON DIRECTING FILM

the full moon. All of a sudden, a car comes up and the guy next to me
says ..."

If you think about it, that's a shot list: (1) a guy standing on the
comer; (2) shot of fog; (3) a full moon shining above; (4) a man says,
"I think people get wacky this time of year"; (5) a car approaching.

This is good filmmaking, to juxtapose images. Now you're following

the story. What, you wonder, is going to happen next?

The smallest unit is the shot; the largest unit is the film; and the unit
with which the director most wants to concern himself is the scene.

First the shot: it's the juxtaposition of the shots that moves the film
forward. The shots make up the scene. The scene is a formal essay. It
is a small film. It is, one might say, a documentary.

Documentaries take basically unrelated footage and juxtapose it in

order to give the viewer the idea the filmmaker wants to convey.
They take footage of birds snapping a twig. They take footage of a
fawn raising his head. The two shots have nothing to do with each
other. They were shot days or years, and miles, apart. And the
filmmaker juxtaposes the images to give the viewer the idea oi great
alertness. The shots have nothing to do with each other. They are not
a record of what the protagonist did. They are not a record of how
the deer reacted to the bird. They're basically uninflected images. But
they give the viewer the idea of alertness to danger when they are
juxtaposed. That's good filmmaking.

Now, directors should want to do the same thing. We should

DAVID MAMET

all want to be documentary filmmakers. And we will have this

advantage: we can go out and stage and film exactly those
uninflected images we require for our story. And then juxtapose
them. In the editing room, one is constantly thinking: "I wish 1 had a
shot of. . ." Well, you've got all the time in the world before the film is
shot: you can determine what shot you are going to require later, and
go out and shoot it.

Almost no one in this country knows how to write a movie script.

Almost all movie scripts contain material that cannot be filmed.

"Nick, a young fellow in his thirties with a flair for the unusual." You
can't film it. How do you film it? "Jodie, a brash hipster, who's been
sitting on the bench for thirty hours." How do you do that? It can't be
done. Other than through narration (visual or verbal). Visual: Jodie
looks at watch. Dissolve. It is now thirty hours later. Verbal: "Well, as
hip as I am, it has surely been a trial to've been sitting on this bench
for the last thirty hours." If you find that a point cannot be made
without narration, it is virtually certain that the point is unimportant
to the story (which is to say, to the audience): the audience requires
not information but drama. Who, then, requires this information?
This dreadful plodding narration that compromises almost all
American filmscripts.

Most movie scripts were written for an audience of studio executives.

Studio executives do not know how to read movie scripts. Not one of
them. Not one of them knows how to read a movie script. A movie
script should be a juxtaposition of uninflected shots that tell the
story. To read this script and to "see" the movie will surely require
either some cinematic ed-

ON DIRECTING FILM

ucation or some naivete—neither of which is going to be found in the

studio executive.
The w^ork of the director is the work of constructing the shot list
from the script. The work on the set is nothing. All you have to do on
the set is stay awake, follow your plans, help the actors be simple,
and keep your sense of humor. The film is directed in the making of
the shot list. The work on the set is simply to record what has been
chosen to be recorded. It is the plan that makes the movie.

I don't have any experience with film schools. I suspect that they're
useless, because I've had experience with drama schools, and have
found them to be useless.

Most drama schools teach things that will be learned by anyone in

the normal course of events, and refrain from insulting the
gentleman or gentlewoman student of liberal arts by offering
instructions in a demonstrable skill. I suppose that film schools do
the same. What should film schools teach? An understanding of the
technique of juxtaposition of uninflected images to create in the
mind of the viewer the progression of the story.

The Steadicam (a hand-held camera), like many another

technological miracle, has done injury; it has injured American
movies, because it makes it so easy to follow the protagonist around,
one no longer has to think, "what is the shot?" or "where should I put
the camera?" One thinks, instead, "I can shoot the whole thing in the
morning." But if you love that morning's work at dailies (screenings
of the footage you're shooting on a daily basis), you'll hate it when
you're in the editing room. Because what you're seeing in dailies is
not for your amusement; it should not be "little plays." It should be
uninflected, short

DAVID MAMET

shots that can eventually cut, one to the other, to tell the story.

Here's why the images have to be uninflected. Two guys are walking
down the street. One of them says to the other guy . . . Now you,
reader, are listening: you are listening because you want to know
what happens next. The shot list, and the work on the set, should be
no more inflected than the cuts in the little story above. Two guys
walking down the street. . . one guy starts to talk to the other. . .

The purpose of technique is to free the unconscious. If you follow the

rules ploddingly, they vsdll allow your unconscious to be free. That's
true creativity. If not, you will be fettered by your conscious mind.
Because the conscious mind always wants to be liked and wants to be
interesting. The conscious mind is going to suggest the obvious, the
cliche, because these things offer the security of having succeeded in
the past. Only the mind that has been taken off itself and put on a
task is allowed true creativity.

The mechanical working of the film is just like the mechanism of a

dream; because that's what the film is really going to end up being,
isn't it?

The images in a dream are vastly varied and magnificently

interesting. And most of them are uninflected. It is their
juxtaposition that gives the dream its strength. The terror and beauty
of the dream come from the connection of previously unrelated
mundanities of life. As discontinuous and as meaningless as that
juxtaposition might seem on first glimpse, an enlightened analysis
reveals the highest and the most simple order of organization and,
so, the deepest meaning. Isn't that true?

ON DIRECTING FILM

The same should be true of a movie. The great movie can be as free of
being a record of the progress of the protagonist as is a dream. I
would suggest that those who are interested might want to do some
reading in psychoanalysis, which is a great storehouse of information
about movies. Both studies are basically the same. The dream and
the film are the juxtaposition of images in order to answer a
question.

I recommend, for example. The Interpretation of Dreams by

Sigmund Freud; The Uses of Enchantment by Bruno Bettelheim;
Memories, Dreams, Reflections by Carl Jung.
All film is, finally, a "dream sequence." How incredibly
impressionistic even the worst, most plodding, most American movie
is. Platoon really is not any more or less realistic than Dumbo. Both
just happen to tell the story well, each in its own way. In other words,
it's all make-believe. The question is, how good make-believe is it
going to be?

WHERE DO you PUT

THE CAMERA?"

CONSTRUCTING A FILM

(A COLLABORATION WITH

STUDENTS IN THE COLUMBIA

UNIVERSITY FILM SCHOOL)

STUDENT: MAMET:

MAMET: Let's make a movie out of the situation weVe in now. A

bunch of people are coming to a class. What's an interesting way to
film this? From above. Now, why is that interesting?

STUDENT: It's interesting because it's a novel angle and it gives a

bird's-eye view of everybody coming in, sort of accentuating the
numbers. If there are a number of people coming in, you may want to
suggest that that's significant.

How can you tell if this is a good way to film the scene? There are any
number of ways to film it. Why is "from above" better than any other
angle?

MAMET:

DAVID MAMET
How are you going to decide what's the best way to shoot it?

STUDENT:" It depends what the scene is. You could say the scene is
about a really tempestuous meeting and have people pacing around a
lot. That would dictate a different scene than one in which the
tension is underlying.

MAMET: That's exactly correct. You have to ask, "what is this scene
about?" So let's put aside the "follow the hero around" way of making
movies and ask what the scene is about. We have to say our task is
not to follow the protagonist around. Why? Because there are an
infinite number of ways to film a bunch of people in a room. So the
scene is not simply about a bunch of people in a room; it's about
something else. Let us suggest what the scene might be about. We
know nothing about the scene other than it's a first meeting. So
you're going to have to make an election as to what this scene is
about. And it is this election, this choosing not "an interesting way"
to film a scene (which is an election based on novelty and basically a
desire to be well-liked) but rather saying, "I would like to make a
statement based on the meaning of the scene, not the appearance of
the scene," which is the choice of the artist. So let's suggest what the
scene might be about. I'll give you a hint: "what does the protagonist
want?" Because the scene ends when the protagonist gets it. What
does the protagonist want? It's this journey that is

ON DIRECTING FILM

11

going to move the story forward. What does the protagonist want?
What does he or she do to get it—that's what keeps the audience in
their seats. If you don't have that, you have to trick the audience into
paying attention. Let's go back to the "class" idea. Let's say it's the
first meeting of a series of people. A person, in the first meeting,
might be trying to get respect. How are we going to address this
subject cinematically? In this scene the subject wants to earn the
instructor's respect. Let's tell the story in pictures. Now, if you have
trouble addressing this thing, and your mind draws a blank, just
listen to yourself telling the story to a guy next to you in a bar. How
would you tell that story? STUDENT: "So this guy comes into the
class and the first thing he does is sit right next to the professor and
he started to look at him very carefully and . . . and listen very
carefully to what he's saying and when the professor dropped his
prosthetic arm, he reached down and grabbed it and gave it to the
professor." Well, yes. This is what the writers do today, the writers
and directors. But we, on the other hand, want to keep everything
that's "interesting" out of the way. If the character is not made to be
interesting, then the character can only be interesting or
uninteresting as it serves the story. It's impossible to make a
character "interesting in general." If the story is about a man who
wants to earn the respect of the instructor, it's not important that the
in-

MAMET:

DAVID MAMET

structor have a prosthetic arm. It's not our task to make the story
interesting. The story can only be interesting because we find the
progress of the protagonist interesting. It is the objective of the
protagonist that keeps us in our seats. "Two small children went into
a dark wood ..." Okay; somebody else? You're writing the film. The
objective is to earn the respect of the instructor.

STUDENT: "A guy in film class, who arrived twenty minutes early,
sat at one end of the table. Then the class came in with the instructor,
and he picked up his chair and moved it, trying to sit near the
instructor, and the instructor sat on the other side of the room."

MAMET: Good. Now we've got some ideas. Let's work with them a
little bit. A fellow arrived twenty minutes early. Why? To earn the
respect of his instructor. He sat at one end of the table. Now, how can
we reduce this to shots?

Shot of him coming in, shot of the classroom, shot of him sitting,
shot of the rest of the class coming in. Good. Anybody else?
STUDENT: A shot of a clock, a shot of the moment when he comes
in, hold on this until he decides where he's going to sit, a shot of him
waiting alone in the empty room, a shot of the clock, and a shot of
many people coming in.

MAMET: Do you need a shot of the clock? The smallest unit with
which you most want to concern yourself is

STUDENT:

MAMET:

the shot. The larger concept of the scene is to win the respect of the
instructor. This is what the protagonist wants—it's the
superobjective. Now, how can we figure out the first beat of the
scene? What do we do first?

STUDENT: EstabUsh the character.

MAMET: The truth is, you never have to estabHsh the character. In
the first place, there is no such thing as character other than the
habitual action, as Mr. Aristotle told us two thousand years ago. It
just doesn't exist. Here or in Hollywood or otherwise. They always
talk about the character out there in Hollywood, and the fact is there
is no such thing. It doesn't exist. The character is just habitual action.
"Character" is exactly what the person literally does in pursuit of the
superobjective, the objective of the scene. The rest doesn't count.

An example: a fellow goes to a whorehouse and comes up to the

madam and says, "what can I get for five bucks?" She says, "you
should have been here yesterday, because ..." Well, you, as members
of the audience, want to know why he should have been there
yesterday. That's what you want to know. Here, however, we tell the
story, full of characterization.

A fellow, trim, fit, obviously enamored of the good things of life but
not without a certain sombemess, which might speak of a disposition
to contemplation, goes to a gingerbread gothic
whorehouse situated on a quiet residential street, somewhere in a
once-elegant part of town. While . walking up the flagstone steps . . .

This is one of those American movies we make. The script and the
film are always "establishing" something.

Now, don*t jou go "establishing" things. Make the audience wonder

what's going on by putting them in the same position as the
protagonist.

As long as the protagonist wants something, the audience will want

something. As long as the protagonist is clearly going out and
attempting to get that something, the audience will wonder whether
or not he's going to succeed. The moment the protagonist, or the
auteur of the movie, stops trying to get something and starts trying to
influence someone, the audience will go to sleep. The movie is not
about establishing a character or a place, the way television does it.

Look at the story about the whorehouse: isn't that how most
television shows are formed? A shot of "air," tilt down to frame a
building. Pan down the building to a sign that says, "Elmville General
Hospital." The point is not "where does the story take place?" but
"what's it about?" That's what makes one movie different from
another.

Let's go back to our movie. Now, what's the first concept? What is
going to be a building block that is necessary to "achieve the respect
of the instructor"?

ON DIRECTING FILM

1S

STUDENT: MAMET:

STUDENT: MAMET:

STUDENT:

. . . The guy arrives early?

Exactly so. The guy arrives early. Now, the way you understand
whether the concept is essential or not is to attempt to tell the story
without it. Take it away and see if you need it or not. If it's not
essential, you throw it out. Whether it's a scene or a shot, if it's not
essential throw it out. "The guy says to the madam ..." Well,
obviously you can't start the whorehouse scene like that. You need
something before that. "A guy goes to a whorehouse and the madam
says ..." In this example the first building block is "a guy goes to a
whorehouse."

Here's another example: you have to walk to the elevator in order to

get dovsaistairs. In order to get down, you have to go to the elevator
and get in there. That's essential to get downstairs. And if your
objective is to get to the subwaj and you begin in an elevated floor of
the building, the first step will be "to get downstairs."

To win the respect of the instructor is the superob-jective. What

steps are essential? First, show up early.

Good. Yes. How are we going to create this idea of earliness? We

don't have to worry about respect now. Respect is the overall goal. All
we have to worry about now is earliness; that's the first thing. So let's
create the idea of earliness by juxtaposing unin-flected images. He
starts to sweat.

DAVID MAMET

MAMET: STUDENT:

MAMET: STUDENT:

MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:
STUDENT: MAMET:

STUDENT:

Okay, what are the images?

The man sitting by himself, in a suit and tie, starting to sweat. You
could watch his behavior. How does this give us the idea of earliness?
It would suggest that there's something he's anticipating.

No, we don't have to worry about anticipating. All we have to know in

this beat is that he's early. Also, we don't have to watch behavior. An
empty room.

Well, there we go, that's one image. A shot of a man by himself in an

empty room juxtaposed with a shot of a group of people coming in
from outside.

Okay, but this doesn't give us the idea of earliness, does it? Think
about it. They could all be late.

Let's express this in absolutely pristine, uninflected images requiring

no additional gloss. What are the two images that are going to give us
the idea of earliness?

A guy is walking down the street and the sun is rising and the street
cleaners are going by and it's dawn and there's not a lot of activity on
the street. And then maybe a couple of shots of some people waking
up and then you see the guy, the first man, come into a room and
other people are in there finishing up a job that they were doing,
maybe finishing the ceiling or something like that.

ON DIRECTING FILM

17

MAMET:

STUDENT:
MAMET:

STUDENT: MAMET:

STUDENT:

MAMET:

STUDENT:

Now, this scenario gives the idea of early morning, but we've got to
take a little bit of an overview. We have to let our little alarm go off
once in a while, if we stray too far off the track; the alarm that says,
"Yes—it's interesting, but does it fulfill the objective?" We want the
idea of earliness so that we can use it as a building block to winning
respect. We do not absolutely require the idea early in the morning.
Outside the door you could have a sign saying "Professor Such-and-
such's class" and giving the time. Then you could have a shot of our
guy obviously sitting by himself with the clock behind him. Okay.
Does anybody feel that it might be a good idea to stay away from a
clock? Why might we feel that? Cliche.

Yeah, it's a little bit of a cliche. Not that it's necessarily bad. As
Stanislavsky told us, we shouldn't shy away from things just because
they are cliches. On the other hand, maybe we can do better. Maybe
the clock ain't bad, but let's put it aside for a moment just because
our mind, that lazy dastard, jumped to it first and, perhaps, it is
trying to betray us.

So you have him coming up, and he's in the elevator, nervous and
maybe looking at his watch. No, no, no, no. We don't need this in
there, do we? Why don't we need this? Maybe a small clock . . . ?

DAVID MAMET

MAMET:

MAMET: ... He doesn't even have to look nervous. This gets down to
what I tell the actors too, which we'll discuss later. You can't rely on
the acting to tell the story. He doesn't have to be nervous. The
audience will get the idea. The house has to look like a house. The
nail doesn't have to look like a house. This beat, as we described it,
had nothing to do with "nervousness"; it is about being early, and
that is all it is about. Now, what are the images here?

STUDENT: We see the guy come down the hall and he gets to the
door and is trying to rush in and he finds that it's locked. So he turns
and looks for a janitor in the hall. The camera stays with him. How
do you know he's looking for a janitor? All you can do is take
pictures. You can take a picture of a guy turning. You can't take a
picture of a guy turning to look for a janitor. You've got to tell that in
the next shot. Can you cut to a janitor?

Now the question is, does a shot of a guy turning and a shot of a
janitor give you the idea of earliness? No, it doesn't. The important
thing is alwajs apply the criteria. This is the secret of filmmaking.

Alice said to the Cheshire Cat, "which road should I take?" And the
Cheshire Cat said, "where do you want to go?" And Alice said, "I
don't care." And the Cheshire Cat said, "then it doesn't matter which
road you take." If, on the other hand, you do care where you're going,
it does matter which

STUDENT: MAMET:

ON DIRECTING FILM

19

STUDENT:

MAMET:

STUDENT:

MAMET: STUDENT: MAMET: STUDENT:

MAMET:
road you take. All you have to think about now is earliness. Take a
look at the idea about the locked door. How can we use this, because
it's a very good idea. It's already more exciting than a clock. Not
more exciting in general, but more exciting as applied to the idea of
earliness.

He comes to the door and it's locked, so he turns, he sits and waits.

Now, what are the shots? A shot of the man coming down a hall.
What's the next shot? A shot of a door, he tries it, it's locked, it
doesn't open.

He sits down? That's it.

Does this give us the idea of earliness? Yes? What if we combine

them all. Start with the sun rising. The second shot is of a janitor
mopping in the hall, going dovm the hall, and as he goes down,
there's someone sitting in front of the door and the guy gets up and
points to the door and the janitor could look at his watch and the guy
points to the door again and the janitor looks at his watch and shrugs
and unlocks it.

Which sounds cleaner? Which gives us more clarity in this instance?

The toughest thing in writing and directing and editing is to give up
preconceptions, and apply those tests you have elected are correct for
the problem.

We do that by applying ourselves to our first

principles. The first principle, in this case of the scene, being it's not
a scene about guys coming into a room, it's a scene about trying to
win the respect of the instructor; the second small principle being
this beat is about earliness. That's all we have to worry about,
earliness.

Now, we have two plans here. Which is simpler? Always do things

the least interesting way, and you make a better movie. This is my
experience. Always do things the least interesting way, the most
blunt way. Because then you will not stand the risk of falling afoul of
the objective in the scene by being interesting, which will always bore
the audience, who are collectively much smarter than you and me
and have already gotten up to the punch line. How do we keep their
attention? Certainly not by giving them more information but, on the
contrary, by withholding information—by withholding all
information except that information the absence of which would
make the progress of the story incomprehensible.

This is the kiss rule. K.I.S.S. Keep it simple, stupid. So we have three
shots. A fellow is walking down the hall. Tries the handle of the door.
Close-up of the door handle being jiggled. Then the fellow sits down.
STUDENT: 1 think you need one more shot if you want to show his
earliness. He opens up his briefcase, pulls out a handful of pencils,
and starts sharpening them.

RECTING FILM

21

MAMET:

STUDENT: MAMET:

STUDENT: MAMET:

STUDENT: MAMET:

Okay now, you're getting ahead of yourself. We've finished our task,
right? Our task is done when we've estabhshed the idea of earliness.

As Wilham of Occam told us, when we have two theories, each of

which adequately describes a phenomenon, always pick the simpler.
Which is a different way of keeping it simple, stupid. Now, you don't
eat a whole turkey, right? You take off the drumstick and you take a
bite of the drumstick. Okay. Eventually you get the whole turkey
done. It'll probably get dry before you do, unless you have an
incredibly good refrigerator and a very small turkey, but that is
outside the scope of this lecture.

So we've taken the drumstick off the turkey— the turkey being the
scene. We've taken a bite off the drumstick, the bite being the specific
beat of earliness.

So let us posit the identity of the second beat. We don't have to follow
the protagonist around, do we? What's the next question we have to
ask? What's the next beat?

Exactly so. What's the next beat? Now, we have something we can
compare this next beat to, don't we?

The first beat.

Something else, which will help us to figure out what it's going to be.
What is it? The scene? The objective of the scene: exactly. The
question the

DAVID MAMET

STUDENT: MAMET:

STUDENT: MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

answer to which will unerringly guide us is, "what's the objective of

the scene?" Respect.

To win the respect of the instructor is the overall objective of the

scene. That being the case, if we know the first thing is to arrive
early, what might be a second thing? A positive and essential second
beat, having arrived early. In order to do what. . . ? To earn the
respect of the instructor. Yes. So what might one do? Or another way
to ask it is why did he arrive early? We know to win the respect of the
instructor is the superobjective. He might get out the instructor's
book and brush up on the instructor's methodology. No. That's too
abstract. You're on too high a level of abstraction. The first beat is
earhness. So on the same level of abstraction, what might be the
second beat? He was early in order to do what? Prepare.

Perhaps in order to prepare. Anyone else? Now, don't we have to deal

with the locked door? He has an obstacle: the door is locked; he has
to respond to that obstacle.

Forget about the protagonist. You have to know what the protagonist
wants because that's what the film is about. But you don't have to
take a picture of it. Hitchcock denigrated American films, saying they
were all "pictures of people talking"—as, indeed, most of them are.

ON DIRECTING FILM

23

STUDENT: MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

STUDENT:

You tell the story. Don't let the protagonist tell the story. You tell the
story; jou direct it. We don't have to follow the protagonist around.
We don't have to establish his "character." We don't need to have
anybody's "back story." All we have to do is create an essay, just like a
documentary; the subject of this particular documentary being to
win the respect of The first essay is on earliness; what's the second
thing?

Could it be to wait?

To wait? What's the difference between to wait and to prepare?

The protagonist is more active. In which? The second. In terms of

what?

In terms of his action. It's stronger to have the actor do something.

I'll tell you a better test. To prepare is more active in terms of this
particular superobjective. It's more active in terms of to win the
respect.

This class is about one thing: learning to ask the question "what's it
about?" The film is not about a guy. It's about to win the respect of.
The beat is not about a guy coming in. It's about earliness. Now that
we've taken care of earliness, let's say the next beat is to prepare. Tell
the idea of to prepare as if you're telling it to somebody in a bar. So
this guy was sitting on a bench waiting, waiting.

DAVID MAMET

MAMET:

STUDENT:

MAMET:

STUDENT: MAMET: STUDENT: MAMET:

just waiting. And he pulled out of his briefcase a book written by the
professor. Now, how do you shoot that? How do you know it's a book
written by the professor? We could have the name of the professor
on the door, and in the same shot see the name on the book.
But we don't know that he's preparing for the class. You don't have to
put in all this literary narration—see how narration weakens the
film? You do have to know the beat is about preparing. It's a very
important distinction. We don't have to know it's preparing for the
class. That's going to take care of itself. We do have to know it's
preparing. The boat has to look like a boat—the keel does not.

We don't need waiting. Waiting is trying to reiterate. We've already

got earliness. We took care of that. All we have to do now is
preparing. Listen to yourselves when you describe these shots. When
you use the words "just," "kind of," and "sort of," you're diluting the
story. The shots shouldn't be just, kind of, or sort of anything. They
should be straightforward, as straightforward as the first three shots
in the movie.

He starts to comb his hair, straighten his tie. Does this fall under the
heading o( preparing? It's like grooming. Preparing could be
preparing physically or it could

ON DIRECTING FILM

2S

MAMET:

be preparing for the subject matter at hand— to wm the respect of.

Which is going to be more specific to the scene? What is going to be

more specific to the overall superobjective, to win the respect of the
instructor? To make oneself more attractive, or to prepare?
STUDENT: He pulls out his notebook, reads through it very fast,
then thinks, no, then he goes back and looks at a certain page.

Now, this falls afoul of one of the precepts we have been discussing,
which is: tell the story in cuts. We're going to adopt this as our motto.

Obviously there are some times when you are going to need to follow
the protagonist around for a bit; but only when that is the best way to
tell the story; which, if we are dedicated in the happy application of
these criteria, we will find is very seldom the case. See, while we have
the luxury of time, here in class or at home making up the
storyboard, we have the capacity to tell the story the best way. We
can then go on the set and film it.

When weVe on the set, we don't have this luxury. Then we have to
follow the protagonist around, and we'd better have ourselves a
Steadicam.*

*The Steadicam is no more capable of aiding in the creation of a good

movie than the computer is in the writing of a good novel—both are
labor-saving devices, which simplify and so make more attractive the
mindless aspects of a creative endeavor.

DAVID MAMET

So what weVe trying to do is find two or more shots the juxtaposition

of which will give us the idea oi preparing.

STUDENT: How about: this guy has a three-ring binder. And he

takes a little piece of white cardboard and rips off the perforated
edges, folds them in half, puts them into the little plastic tabs that
divide the pages in the three-ring binder.

MAMET: This is an interesting idea. Let's say it in shots: he takes his

notebook, he takes out a piece of paper, which is one of those little
tabs. We cut to the insert (a tight shot on his hands). He's writing
something on the tab. He sticks the piece of paper in the plastic
thing. Now we cut back out to the master (the main shot of the
scene). He closes the notebook. This is all uninflected, isn't it? Does
this give us the idea oipreparing? I'll ask you another question: which
is more interesting—if we read what he's writing on the tab or if we
don't read what he's writing?

STUDENT: If we don't.

MAMET: Exactly so. It's much more interesting if we don't read what
he's writing. Because if we read what he's writing, then the sneaky
purpose of the scene becomes to narrate, doesn't it? It becomes to
tell the audience where we are. If we don't have any sneaky purpose
in the scene, then all that beat has to be about is preparation. What's
the effect of this on the audience?

STUDENT: It arouses their curiosity.

MAMET: Exactly so, and it also wins their respect and thanks,
because we have treated them with respect, and have not exposed
them to the unessential. We want to know what he's writing. It's
obvious that he was earl^. It's obvious that he is preparing. We want
to know: early for what? preparing for what? Now we've put the
audience in the same position as the protagonist. He's anxious to do
something and we're anxious for him to do something, right? So
we're telling the story very well. It's a good idea. I have another idea,
but I think yours is better.

My idea is that he shoots his cuffs and that he looks down at his cuff,
and we cut to an insert and we see the shirt has still got the tag on it.
So he rips the tag off. No, I think yours is better, because it goes more
to the idea of preparedness. Mine was kind of cute, but yours has
much more to do with preparedness. If you have the time, as we do
now, you compare your idea to the objective, and as the good
philosophers we are, as followers of the ways of both the Pen and the
Sword, we choose the way that is closer to the objective, discarding
that which is merely cute or interesting; and certainly discarding that
which has a "deep personal meaning" for us. If you're out on the set,
and you don't have any leisure at all, you may choose something
simply because it's a cute idea. Like mine about the cuffs—

DAVID MAMET

STUDENT:

MAMET:

STUDENT: MAMET: STUDENT: MAMET:

in your imagination you can always go home with ^ the prettiest girl
at the party, but at the party sometimes that is not true.
Now let's go on to the third beat. What's the third beat? How do we
answer that question? Go back to the main objective, to win the
respect of the instructor.

Absolutely. Now: let's approach this differently. What's a bad idea for
the third beat? Waiting.

Waiting is a bad idea for the third beat. Preparing is a bad idea for
the third beat. Yes, because we already did it. It's like climbing the
stairs. We don't want to climb a stair we've already climbed. So
preparing again is a bad idea. Why play the same beat twice? Get on
with it. Everybody always says the way to make any movie better is
bum the first reel, and it's true. All of us have this experience almost
every time we go to the movies. Twenty minutes in, we say, ''whj,
they should have started the movie here. " Get on with it, for the love
of Mike. Get into the scene late, get out of the scene early, tell the
story in the cut. It's important to remember that it is not the
dramatist's task to create confrontation or chaos but, rather, to
create order. Start with the disordering event, and let the beat be
about the attempt to restore order.

We're given the situation: this fellow wants such and so—he has an
objective. That's enough chaos

for you right there. He has an objective. He wants to win the respect
of his teacher. This fellow lacks something. He's going to go out and
get it.

Entropy is a logical progression toward the simplest, the most

ordered state. So is drama.* The entropy, the drama, continues until
a disordered state has been brought to rest. Things have been
disordered, and they must come back to rest.

The disorder is not vehement in this case, it's fairly simple: someone
wants a guy's respect. We don't have to worry about creating a
problem. We make a better movie if we worry about restoring order.
Because if we worry about creating problems, our protagonist's going
to do things that are interesting. We don't want him to do that. We
want him to do things that are logical.
What's the next step? What's the next beat going to be about? We're
talking in terms of our particular progression. The first beat being to
arrive early. The second one being preparation, to prepare. And the
third one being? (Always thinking in terms of the superobjective of
the movie, which is to gain the respect of That's your test. That's the
litmus test: to gain the respect of.) STUDENT: To introduce himself?

*I know the dictionary defines entropy as a progression toward the

most disordered state—but on this point, I take issue with that most
excellent book.

DAVID MAMET

MAMET:

STUDENT: STUDENT: MAMET:

STUDENT: MAMET:

STUDENT:

MAMET:

Mayhap the beat is about greeting. Yes, what do we

call that kind of greeting?

Acknowledgment. . . ?

Ingratiation . . . ?

To ingratiate, to pay homage to, to acknowledge,

to greet, to make contact. Which, of all these, is

most specific to the superobjective to gain the respect

op

1 think homage.
All right, then. Let's make up a little photo essay

about homage here. The deeper you can think, the

better it's going to be. Deeper in the sense of writing

means "what would it be like to me?" Not "how

might anyone pay homage?" but "what does the

idea of homage mean to me.^" That's what makes

art different from decoration.

What would be real homage? The professor arrives, and our guy goes
to shake his hand.

Okay. But this is like the watch, isn't it? Earliness— watch. Homage—
handshake. There's nothing wrong with it, but let's think a little bit
deeper, because we might as well, now that we have the luxury of
time.

What would be a lovely way to show homage, a way that really means
something to you? Because if you want it to mean something to the
audience, it should mean something to you. They are like you— they
are human beings: if it don't mean something

ON DIRECTING FILM

31

to jou, it ain't going to mean something to them. The movie is a

dream. The movie should be like a dream. So if we start thinking in
terms of dreams instead of in terms of television, what might we say?
We're going to have a little photo essay, a little documentary about
homage.

STUDENT: When you say a dream, you mean it doesn't have to be

believable in the sense that someone would actually do it in real life?
MAMET: No, I mean ... I don't know how far we can stretch this
theory, but let's find out, let's stretch it till it breaks. At the end of
Places in the Heart, Robert Benton put a sequence that is one of the
strongest things in an American movie in a long time. It's the
sequence where we see everyone who was killed in the film is now
alive again. He's created something that is like a dream in this. He is
juxtaposing scenes that are discontinuous, and that juxtaposition
gives us a third idea. The first scene being everyone's dead. The
second scene being everyone's ahve. The juxtaposition creates the
idea of a great wish, and the audience says, "oh my God, why can't
things be that way?" That's like a dream. Like when Cocteau has the
hands coming out of the wall. It's better than following the
protagonist around, isn't it?

In House of Games, when the two guys are fighting about a gun in
the doorway and we cut away to a shot of the sidekick, the professor
character, looking on, then you hear the gunshot. That's pretty good

DAVID MAMET

filmmaking. It wasn't great filmmaking, maybe, but it was a lot better

than television. Right? It gives us the idea. They're fighting; you cut
to the guy looking. The idea is what's going to happen and we can't
do anything about it.

It conveys the idea of helplessness, which is what the beat is about.

The protagonist is helpless: we get it without following her around.
We put the protagonist in the same position as the audience —
through the cut —by making the viewer create the idea himself, in his
own mind, as Eisenstein told us. How about if the student presents
something to the professor? Some kind of special present. Or he
bows when the guy comes in, and offers him a chair? No, youVe
trying to tell it in the shot. We want to tell it in the cut. How about
this—the first shot is at the level of feet, a tracking shot of a pair of
feet walking. And the second shot is a close-up of the protagonist,
seated, and he turns his head quickly. What does the juxtaposition of
the two things give us?

Arrival. And?
Recognition.

Yeah; it's not quite homage, it's attentiveness or attention. At least,

it's two shots creating a third idea. The first shot has to contain the
idea of where the feet are. The feet are a little bit distant, right? With
the idea that the feet are distant and the fellow hears

STUDENT:

MAMET:

STUDENT: MAMET: STUDENT: MAMET:

RECTING FILM

33

STUDENT: MAMET:

STUDENT: MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

them anyway, what does the juxtaposition of these two things give
us? Awareness.

Awareness; perhaps not homage, but awareness or great attention,

which might just sneak up on homage. What about if we had the long
shot of the feet coming down the corridor and then a shot of our guy
standing up? It shows a Httle bit more homage, in that he's standing
up.

Especially if he were to stand up in a humble way. He doesn't have to

do it in a humble way. All we have to show is him standing up. He
doesn't have to stand up any way at all; all he has to do is stand up.
The juxtaposition of that and the shot of the other guy far off gives
the idea of homage. How about when the guy stands up he bows his
head?

It doesn't really tell any more. And it's more inflected, which is to say
worse for the purpose of filmmaking. The more we "inflect" or "load"
the shot, the less powerful the cut is going to be. Anyone else?

A shot over the protagonist with a notebook. He looks up, stands up,
and runs out of the shot. A shot of our hallway and the door in the
hallway, which has a glass window to it. Protagonist runs into the
shot and opens the door just as a man walks in the other direction.
Yes. Good. I see you like that. Two questions we

might ask ourselves—one question is does that convey the idea of

homage? and the other is do I hke it? If you ask the second question,
you say, well, heck, I don't know if I like it or not. Am I a fellow with
good taste? Yes. Does this have as much good taste in it as I think 1
have in myself? Gosh, I don't know. Tm lost.

The question you do want to ask is does it convey the idea of

homage? If it does convey the idea of homage, then go on to the next
step: do I hke it? There is the inner ability Stanislavsky called the
"judge of yourself," which one might characterize as a certain
amount of artistic good taste. That's going to function anyway
because we all have good taste. It's the nature of the human being to
please. We all want to please one another. Nobody doesn't want that.
There's no one who doesn't want to succeed. What we're trying to do
is make our subconscious work for us by making that task at which
we can succeed very simple and very technical so we don't have to
throw ourselves on the mercy of either our good taste or the cinema-
going public.

We want to have some test that allows us to know when our job is
done without relying on our good taste. That test here is does it
convey the idea of homage? Feet way off, man stands up. I think it
does. Let's go on to the next beat. What's the next beat after homage?
What's the first question we want to ask?
ON DIRECTING FILM

35

STUDENT:

MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

What's the superobjective? Good. What's the answer? To win the

respect of the professor. So after showing homage, what's the next
beat? To impress.

It's a tad general. It also rather reiterates the superobjective. To

impress, to win respect. They are too similar. One part at a time. The
boat has to look like a boat; the sail doesn't have to look like a boat.
Make each part do its job, and the original purpose of the totality will
be achieved—as if by magic. Make the beats serve the scene, and the
scene will be done; make the scenes, in the same way, the building
blocks of the film, and the film will be done. Don't make the beat do
the service of the whole, don't try to reiterate the play in the scene.
It's like **would anyone like a cup of coffee because I'm Irish," right?
It's how most acting is done today. "I'm so glad to see you today
because, as you'll find out later, I'm a mass murderer." Anyone? The
next beat?

To gain acknowledgment? That's also rather general. To please?

You can't get more general than that. To show affection?

To gain respect by showing affection? Maybe; what else? To show

self-confidence? Be dynamic. See, these things you suggest really

DAVID MAMET

STUDENT: MAMET: STUDENT: MAMET:

could come at more or less any point, and they will betray us into a
circularity more appropriate to the epic than to the dramatic form.
But what will be the next essential thing to come after showing
homage? To blow jour own horn?

Would you do that to gain someone's respect? No.

You can ask yourself the question thusly: what would / like to do in
the best of all possible worlds to earn someone's respect? It's a
question of what you might do in your wildest imagination, not what
you might do because you are bound by the strictures of polite
behavior. We don't want our movies to be bound by that. We'd like
our movies to be greatly expressive of our fantasy life.

There's another question we probably need to ask at this point. We

might ask ourselves when are we going to be done? so we will know
when the movie is done. We could go on trying to gain respect
indefinitely. So we need a cap. Without a cap, the essential problem
of the throughline, which is to gain respect, really can lead into a
never-ending spiral, which is capped only by our good taste. So
perhaps we need a throughline with a more positive, that is to say a
more definite end than to gain respect.

For example, getting a reward. Reward being a simple and physically

identifiable sign of respect. On this level of abstraction, the reward
could be, for example, what?

ON DIRECTING FILM

37

STUDENT: It could be that he wants the teacher to do him a favor.

MAMET: Okay; anybody else?

STUDENT: He wants the teacher to give him a job.

MAMET: Yes; anything else?

STUDENT: The teacher gives him a pat on the back.

MAMET: That's not as specific as the first two, is it? I take it that
you're speaking rhetorically. In which case the pat on the back is
similar to gain his respect in this: it is deficient in that it lacks a cap
or an objective, so that one is unsure when one is finished. It's going
to make our task a lot easier if we always know both where we're
going and when we're finished. If the job is the objective, then when
that job is given or when that job is absolutely denied, the scene will
be over.

Or perhaps we could say the reward the student requires is this: he

wants the teacher to change a grade. Then, when the teacher changes
the grade, the scene wall be over; or if the teacher categorically
refuses to change the grade and no hope is left, then the scene would
be over. So we could say that the throughline of the scene in that case
is to get a retraction. Then that's what everything in the scene would
be about.

What's the first thing that's done to get a retraction? To show up

early, right? What's the second thing? To prepare. The third beat is
to pay homage. It's going to be a lot easier to find out what the fourth
beat

DAVID MAMET

is for to get a retraction than to gain his respect, because now we

have a specific test for knowing when the movie will be over; we
know where we have to end up, and we can find a beat that will lead
us to that end. Does anybody know what a MacGuffin is? STUDENT:
It's Hitchcock's phrase for a little invented device

that will carry the action. MAMET: Yes. In a melodrama—

Hitchcock's movies are melodramatic thrillers—a MacGuffin is that
thing which the hero is chasing. The secret documents . . . the great
seal of the republic of blah-blah-blah . . . the delivery of the secret
message . . . We, the audience, never really know what it is. You are
never told more specifically than "it's the secret documents."

Why should you be? We'll fill in for ourselves, unconsciously, those
secret documents which are important to us.

In The Uses of Enchantment, Bruno Bettelheim says of fairy tales the

same thing Alfred Hitchcock said about thrillers: that the less the
hero of the play is inflected, identified, and characterized, the more
we will endow him with our own internal meaning— the more we will
identify with him—which is to say the more we will be assured that
we are that hero.

"The hero rode up on a white horse." You don't say "a short hero rode
up on a white horse," because if the listener isn't short he isn't going
to identify with that hero. You don't say "a tall hero rode up

on a white horse," because if the listener isn't tall, he won't identify

with the hero. You say "a hero," and the audience subconsciously
realize tbej are that hero.

The MacGuffin is that thing which is important to us —that most

essential thing. The audience will supply it, each member for himself.
Just so in the objective to get a retraction. It's perhaps not necessary
to know at this point a retraction of what.

The actor doesn't have to know it. A retraction of a grade, a

retraction of a statement, a retraction of a reprimand. It's a
MacGuffin at this point. The less the objective is inflected, the better
off we, the audience, are. The less the hero is described to us, the
better off we are.

Step four, anybody? We know where we're going, and we know who's
going with us. We know who we love, but the devil knows who we'll
marry. To get a retraction. Tally-ho, then, me hearties.

STUDENT: You have to ask for the retraction.

MAMET: Good. Now, wasn't that a breath of fresh air? The

invigorating infusion of fresh air that this direct and blunt beat
brings into this discussion is the same breath of fresh air that it will
bring into the film. Now we have: to show up early, to prepare, to pay
homage, and to ask for as the four beats of the story to get a
retraction.

DAVID MAMET

STUDENT:

MAMET:

Don't you think showing up early and preparing are the same as
paying homage?

You are saying that these may be subsumed under the larger beat to
pay homage? I don't know. I have a question about to prepare, which
we may come back to. Now you see that the process weVe going
through here is re-forming the large to better understand the small,
and re-forming the small to better understand the large (working
from the su-perobjective to the beats, and reworking from the beats
to the superobjective, et cetera), until we come up with a design that
seems to fulfill all of our requirements. Then we'll put that design
into action and we will shoot it.*
Now, we may find, as I found a little bit in my first movie and to a
greater extent in my second, that after we've shot it we have to refine
it further— which phenomenon scientists call the Jesus Factor,

*The process we are going through in this room is the exploration of

the dynamic between the moment and the objective. It is this
dynamic that, in this discussion, in film, in the theater, gives both the
moment and the entirety strength—in the beautiful drama, each
moment serves the purpose of the superobjective, and each moment
is beautiful in itself. If the moment only serves the superobjective, we
have plodding narrative pseudodrama, good only for object-lesson or
"message" plays. If the moment only stands for itself, we have only
self-indulgent or "performance" art. The effort that the dramatic
artist spends in analysis frees both him and the audience to enjoy the
play. If this time is not spent, the theater becomes the most dreadful
of marriage beds, in which one party whimpers "love me," and the
other pouts "convince me."

ON DIRECTING FILM

41

a technical term meaning "it works correctly on paper but for some
reason doesn't work when we get it on its feet."

That happens sometimes. All you can do then is try to learn from it.
The answer is always there. Sometimes it requires more wisdom than
we possess at that instant—but the answer is always there.
Sometimes the answer is: "Fm not smart enough to figure it out yet,"
and we must remember that the man said, "A poem is never finished
—only abandoned."

All right, enough lovemaking. We got our three beats and looked at
the throughline and said, "perhaps this throughline is not very good."
We reformed the throughline away from to gain respect and decided
that it was to get a retraction. Now we can look back at the beats, and
we may say that perhaps preparing is out of order. Perhaps paying
homage is what that beat is about. I don't know. Let's forge on a little
bit and see if the fourth step gives us some more clues.
Do we have to decide what the end result is? You mean does the hero
get the retraction? Who's interested to know if he gets the retraction
or not? Anyone? STUDENT: I'd want to know, because then we can
do something with the response of the teacher to the homage. Does
the teacher know why the guy's there? Is he suspicious of—

STUDENT MAMET:

DAVID MAMET

MAMET: No; no, no, forget the teacher; let*s stick with the
protagonist. We stick with the protagonist, and that ' will tell us the
story. Because the story is his story. We're here not to create disorder
but to create order. What's the inherent disorder? "The other guy has
something I want." What does the other guy have? The power to
issue a retraction. When is the story over? When the hero^et5 it. The
disorder is inherent in the story. What we're trying to do is create
order. When the hero either gets a retraction or finds that he cannot
have a retraction, order will be restored. The story will be over, and
there will be no further reason to be interested. Up to that point,
what we're trying to do is bring about that blessed state of bliss in
which there is no story. For as Mr. Trollope told us, "they are most
happy who have no story to tell."

Let's go on. Let's be jolly, jolly scientists and take one step at a time.
The next step we've suggested is to ask for. What are some
alternatives?

STUDENT: To plead his case.

MAMET: To plead his case. Now, as you see,^we're suggesting two

stories of two different lengths. Why? To plead his case is eventually
going to have to contain to ask, right? And this is what determines
the length of the healthy story—it is determined by the least number
of steps absolutely essential to secure the hero's objective. Who likes
which beat better— to ask for

ON DIRECTING FILM

43
STUDENT: MAMET:

STUDENT: MAMET:

or to plead his case? On what basis can we determine which is better

for the story? On the basis of why he's asking for the retraction? No.
We don't care why. It's a MacGuffin he's asking for. Because he needs
it. But we don't know anything about it. I don't think we need to.
Anybody think we need to? What you're talking about is what the
ilHterate call the "back story." You don't need it. Remember that the
model of the drama is the dirty joke. This joke begins: "A traveling
salesman stops at a farmer's door"—it does not begin: "Who would
think that the two most disparate occupations of agriculture and
salesmanship would one day be indissolubly united in our oral
literature? Agriculture, that most solitary of pursuits, engendering
the qualities of self-reliance and reflection; and salesmanship, in
which ..." Does the protagonist have to explain why he wants a
retraction? To whom is he going to explain it? To the audience? Does
that help him get it? No. He must only do those things that help
him^et a retraction. All he has to do is^et a retraction. The guy says
to the girl, "That's a lovely dress"— he does not say, "I haven't been
laid in six weeks." Now, the question is: on what basis can we decide
which is better in this beat—to plead his case or to ask? My feeling is
to plead his case is better. Why? Because I'm having a good time and
I'd like the

DAVID MAMET

Story to go on a little bit longer. I don't think I have any better basis
than that, and I think that that's all right. But I'd better check,
because I know that I have a capacity for self-delusion. So the
question I ask myself is, "does it run afoul of any of the rules we've
discussed to use to plead his case rather than to ask at this point?" 1
check my rules, and my answer is No, so I'll choose the one I like.

STUDENT: Since pleading is more inflected, isn't that an attempt to

be more interesting?
MAMET: I don't think so; and I don't think it's either more or less
inflected. 1 think it's just different. I think it's a choice. You could say
to plead his case; you could say to present his case. By the way, we
didn't say these beats had to be uninflected. We said that the shots
had to be uninflected. Paying homage may or may not have certain
inherent psychological overtones. We talked about to plead, to ask, to
plead his case, to present his case. Each of these is going to call up
associations in the actor. It is these personal, immediate
associations, by the way, that both induce the actor to act and keep
him in line with the intentions of the author. This is what brings the
actor to the play—not those gyrations of emotional self-abuse that
hack teachers have fobbed off as preparation.

STUDENT: How about to bargain, or to bribe?

MAMET: What about these ideas, in terms of the structure?

ON DIRECTING FILM

45

Let's talk about to bargain, because that's a little bit simpler.

STUDENT: The problem is that we started with a different

throughline. Bargaining wouldn't work with gaining respect, but it
might be a way to get a retraction.

MAMET: This is a problem you're going to run into a lot in dramatic

structure. Because if you are creating it, either creating a film of your
own or taking someone's film and trying to find the inherent
dramatic structure in it, no angel is going to come down to you and
say, "this is the throughline." What's going to happen is exactly this
process of wondering and revising—to work every time either to
create or to discern a throughline.

We've decided now that to get a retraction is the throughline of the

scene. We are on to the beat following to prepare. Perhaps this next
beat is to present the case. So this is now our new beat. What a relief
to get on to this new beat. What self-respect we must feel for taking
upon ourselves the onus of this task so as to save the audience the
trouble. To present the case.

Our task now is to find a series of uninflected shots that will give us
this idea: to present the case. The student wants to present the case
to the teacher. Now, where are we going to find a clue? We have four
beats. We're working on the fourth beat. What is going to be our clue
to the answer of the shots?

DAVID MAMET

STUDENTi MAMET:

STUDENT:

MAMET:

STUDENT: MAMET:

STUDENT:

MAMET: STUDENT:

Some helpful hint we might find to the answer of

presenting the case.

How we prepared?

Exactly so. The previous beat will provide a clue.

It was to prepare. The beat that we thought, in terms

of the new throughline, might possibly be dorky

may, in effect, offer us quite a hint. So let's go back

and look at our shot list for to prepare. It would be

nice for the sake of cleanliness if we knew whether

there was something we were wasting in there. Some

extra step, which weakened to prepare but might

strengthen to present the case. Like Indians of yore,

we want to use all parts of the buffalo.

The shot where he opens the notebook, has the

little strip with the cardboard things, rips them,

writes on one, puts it in the tab holder.

Good. Now, what are the shots for to present the

case?

The presentation of the notebook in some way.

What are the actual shots? A guy comes into a room,

a guy in the room approaches the desk. Our criterion

is that a juxtaposition of shots will-give us the idea

we require in this instance, to present the case. We

have to know what weVe taking a picture of

Start with a shot of a desk with nothing on it and

the notebook is pushed in.

What's the next shot?

The reaction from the teacher. Either approval or

disapproval.

ON DIRECTING FILM
47

MAMET:

MAMET: No. All this beat has to be about is presenting the case. We
don't need the teacher's reaction here.

STUDENT: If the first shot were a presentation of the book and the
second shot the teacher looking down, wouldn't the juxtaposition of
those two shots present the case to him?

Maybe the first shot is the empty desk and a book is placed into it,
and the second shot is the teacher at the desk looks down at the book
and he also looks up, and we cut to a shot of the student. 1 think we
need the student there because he presents the case.

But couldn't we recognize the notebook from scene two? We know

it's the same student we saw preparing, so we don't need a shot of
him. The book is identification enough? Yeah; we know it's the
student's book. The book stands for the student.

Very good. Of course, you're right. I got caught up in the idea of

following the protagonist around. Good. Now, this brings us to the
application of the principle of the throughline to the plastic elements
of production.

What music is playing? What time of day or night is it? What do the
costumes and the sets look like? At one point you mentioned
someone reading a magazine. You say a magazine: what magazine?
I'm not overstating the case; because somebody makes these
decisions, and that person is called the direc-

STUDENT:

MAMET: STUDENT:

MAMET:

tor? The prop person is going to say, "what should the notebook look
like?" and what are you, the > director, going to say? First off, what is
the untutored person going to say? "Golly, the scene is about to get a
retraction, so what kind of notebook does a person who wants to get
a retraction carry?" If this seems dorky, if this seems overstated to
you, look at American movies. Because that's the way all American
movies are made. "Hi, how are you today because I just got back
from Vietnam." In Hollywood, a committee of thugs wants to make
sure that each word, moment, shot, prop, sound, et cetera, in a movie
wall stand for and, in effect, advertise the film. This committee is
called "producers," and they are to the arts what the ducking stool
was to jurisprudence.*

What answer do we give to the prop person who says "what's the
notebook look like?" What are you going to say?

*Natural, creative exuberance and self-confidence are wonderful

things in an artist. They are inhibited from growing into arrogance
not through the content but because of the process of education.
Even the minimally serious artist is humbled constandy by the
screaming demands of craft.

Those who style themselves "producers" have not had the benefit of
any such education, and their arrogance knows no bounds. They are
like the white slave owners of old, sitting on their porches with their
cooling drinks and going on about the inherent laziness of the Negro
race. The "producer," having never had a run-in with the demands of
a craft, sees all ideas as basically equal and his own as first among
them, for no reason other than that he has thought of it. This notion
is easier to fathom if one thinks

ON DIRECTING FILM

49

STUDENT: Doesn't it depend on what the objective is or isn't?

MAMET: No, because you can't make a "retraction notebook" any

more than you can act what room you just came out of—though there
are, to their shame, schools of acting that purport to teach such.
What should the notebook look like—this "retraction notebook"?
STUDENT: Put a label on the cover?

MAMET: The audience won't read it. It's like a sign. The audience
doesn't want to read a sign; they want to watch a motion picture, in
which the story is advanced through the cut.

STUDENT: They don't have to read it. It's a black folder, white label,
looks like a book report.

MAMET: Why should it look like a book report? I mean, it's not a
bad idea that it should look like a book report, but why is it a good
idea that it look like a book report? Prop person says, "what does it
look like?"

back to the period of early adolescence and to, perhaps, the critique
of an English teacher who said of our efforts: "I don't understand" or
"It is unclear," of which correction one thought: "The old fool. . . /
know what I meant."

I have a great deal of pride and, I suppose, a large admixture of

arrogant pride. 1, in my generally losing contest with these self-styled
"producers," many times console myself by thinking that after society
falls apart, I will be able to eke out at least my meals and shelter by
putting on plays that may make people laugh; but that these
"producers" would have to wait until I and those like me went to
work before they could eat.

Yes, that is how I see "producers." They are "let me take that cow to
the fair for you, son."

DAVID MAMET

STUDENT: MAMET:

STUDENT: MAMET: STUDENT: MAMET:

STUDENT: MAMET:

STUDENT: MAMET: STUDENT: MAMET:

What's the correct answer? What does it do? What

does the report do?

It presents the case.

Right. Now, what's the shot hst for presenting the

case?

The open book on the desk.

What's the next shot?

The face of the teacher.

What is not the next shot? The face of the student,

right? So, therefore, how does the book look?

Prepared.

No, you can't make the book look prepared. You

can make it look neat. That might be nice, but that's

not the most important thing for your answer to

the prop person.* Think about the shot list and the

objective to present the case. To make it prepared, to

make it neat, to make it convincing, the audience

ain't going to notice. What are they going to notice?

That it's the same book they've seen already.

So what's your answer to the prop person?

Make it recognizable.
Exactly so! Good! You've got to be able to recognize it.

That is the most important thing about this report.

This is how you use the principle of throughline to

*The audience will accept anything they have not been given a reason
to disbelieve. So the report must look, minimally, neat, for if it did
not, the audience might question the sincerity of the hero's desire.
The neatness of the report is an antiseptic rather than a creative
consideration.

answer questions about the set and to answer questions about the
costumes. The book in general is not important. What's important is
what it does in the scene. The most blatant thing it does in the scene
is present the case. Since we aren't going to see a shot of the student,
it's got to present the case for him. That shot of the uninflected book
has to present the case. Since we know that it has to be uninflected,
the answer cannot be "it's a prepared book." The answer cannot be
"it's a contrite book." The answer must be "it's got to be the same
book we saw in shot two." In choosing the book, you are teUing the
audience that thing without which they cannot understand the
movie. In this case, it's the essential element of the shot. That
without which the beat will not survive is that it's the same notebook
as in the previous beat. It is essential to the telling of the story.

Every time you make a choice as a director, it must be based on

whether the thing in question is essential to telling the story. If we
don't need the shot of the student, then we'd better be jolly, jolly sure
that they understand that it's the same book.

The audience is only going to look at the most overriding thing in the
frame. You must take charge of and direct their attention. It's also
the principle of magic: What is the single important thing? Make it
easy for them to see it, and you're doing your job. You don't have to
make it a book about getting jour retraction. You do have to make it
the same

DAVID MAMET
STUDENT: MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET: STUDENT:

MAMET: STUDENT:

notebook. So our beats are to show up early, to prepare,

to pay homage, to present the case. What were the shots

for to show up early?

He arrives and tries the doorknob.

No. I hope you don't think I'm being picayune, but

it's very useful to think of the film in exactly the

same way the audience is going to perceive the film.

What they're going to see in the first shot is a man

walking down the hall. What are the shots?

Man walks down hall, shot of a hand on a doorknob,

same man sits down on a bench.

Perfect. Now, why did all those Olympic skaters

fall down? The only answer I know is that they

hadn't practiced enough. Practice with these tools

until you find them boring, then practice some more.

Here is a tool—choose your shots, beats, scenes,

objectives, and always refer to them by the names

you chose.

What are the shots for to prepare? Man takes the notebook out, rips
out a tab, writes something down on a tab, puts the tab in the plastic
thing, closes the tab. Good. To pay homage?

Shot of the man looking and getting up out of the frame. A shot of the
man running to a glass door. He opens the door, a man walks
through. Good. Next beat? To present the case. An empty desk. The
notebook

RECTING FILM

53

MAMET:

STUDENT: MAMET: STUDENT: MAMET:

STUDENT:

MAMET: STUDENT:

MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

put on the desk, and a shot of the man sitting at the desk, looking
down at it.
Good. Let's finish it now. How do we reach a conclusion?

The teacher could start considering the book. What is the beat we are
trying to dramatize here? Judgment.

Okay, the idea is judgment. Consideration is a different way of saying

it. But the teacher considering the book doesn't really have any
weight of montage behind it. It's basically expository. A guy picks up
evidence and looks at it and makes up his own mind. Not good
storytelling, as Aristotle told us. The character shouldn't "just get the
idea." Why is the next hcdit judgment, if all the way through, the
beats are about the student and the teacher? Don't you want to
follow the course of the student and not the teacher? What's your
idea?

I saw the beat as taking a stand. He's presented the case, and you cut
to him standing there, and he's not going to take no for an answer.
And you cut back to the teacher looking up at the kid. What are some
other ideas for the next beat? Receipt of the retraction. Yeah, that's
an idea. To be denied. That's not really the beat; that's the result.
That's

DAVID MAMET

the end of some other beat. The student/protagonist has to be

working toward completion.

STUDENT: At this point in the story, you are going to expect the
response of the professor. The next logical beat after presenting the
case is judgment, judging the case. When that beat is over, he has or
hasn't gotten the retraction. We don't have to follow the student to
complete the throughline, do we?

MAMET: No.

STUDENT: But it's the kid's job to get a retraction.

MAMET: Yes, it is. But it doesn't have to be a picture of the kid. We

want to know what happens next in terms of the throughline, not in
terms of what the protagonist does. What was our last shot in the last
beat?

The professor looking down at the book. The professor's looking

down. Cut to a shot of a bunch of kids in the doorway. A new kid
comes, and they all look over to one side or another. We cut to their
point of view of the empty classroom with the kid sitting there and
the professor looking at him. To get to the idea of judgment. Now
we're ready for the resolution. We see the professor in a long shot, he
opens the book, he looks down to his right, we cut to the desk
drawer, we see him open the desk drawer and he takes out a stamp
pad. You see him stamp the book. And you cut to a shot of the kid,
who is smiling, and he picks up the book, and we cut to a shot of the
kid's hand closing the

STUDENT: MAMET:

book, and then from the back of the classroom you see the kid go to
his seat and the professor stand and call the rest of the class, and
they go in, and they sit downi. All right?

STUDENT: What if he didn't get the retraction?

MAMET: I don't know. It's our first movie. Let's make it a happy
ending, what the hell. And now we're done, and that was excellent
work.

COUNTERCULTURAL

ARCHITECTURE AND DRAMATIC

STRUCTURE

I was a student in the turbulent sixties in Vermont at a coun-

tercultural college. In that time and place, there flourished
something called a school of Countercultural Architecture. Some
people back then thought that traditional architecture had been too
stifling, and so they designed and built a lot of counterculture
buildings. These buildings proved unlivable. Their design didn't
begin with the idea of the building's purpose; it began with the idea
of how the architect "felt."

As those architects looked at their countercultural buildings over the

years, they may have reflected that there's a reason for traditional
design. There's a reason that doors are placed in a certain way,
there's a reason that sills are made a certain way.

All those countercultural buildings may have expressed the

intention of the architect, but they didn't serve the purpose of the
inhabitants. They all either fell down or are falling down or should be
torn down. They're a blot on the landscape and they don't age
gracefully and every passing year underscores the jejune folly of
those countercultural architects.

1 live in a house that is two hundred years old. It was built wdth an
axe, by hand, and without nails. Barring some sort of man-made
catastrophe, it will be standing in another two hundred years. It was
built with an understanding of, and a respect for, wood, weather, and
human domestic requirements.

It's very difficult to shore up something that has been done badly.
You'd better do your planning up front, when you have the time. It's
like working with glue. When it sets, you've used up your time. When
it's almost set, you then have to make quick decisions under
pressure. If you design a chair correctly, you can put all the time into
designing it correctly and assemble it at your leisure. In fact, the
ancient chairmakers—which is to say chairmakers up until about the
turn of the century— used to make their chairs without glue because
they correctly understood not only the nature of joints but the nature
of woods. They knew which woods would shrink and which would
expand with age, so that these woods, when correctly combined,
would make the chair stronger over time.

I recognized two things in finishing up my second movie. When

you're doing the movie, after you finish with the shot list but before
you start shooting it, you have a period called "preproduction." In
preproduction, you say, "you know what would be a good idea? To
really make the audience understand that we're in a garage, what
about a sign that says 'garage.' "

ONDIRECTINGFILM S9

So you meet with your art department and you talk a lot about signs
and you make up a lot of signs. 1 made two movies and I made up a
lot of signs. You never see the signs in a movie— never. You just
never see them. They are after-the-fact attempts to shore up that
which was not correctly designed. Another handy but useless
"reminder" tool is the process of looping, or ADR (Automatic
Dialogue Reading—dialogue recorded and inserted after the movie
has been shot), to communicate to the audience information the film
lacks. For example, dubbing words into somebody's mouth when we
see his back on the screen. To wdt: "oh, look, here we go down that
staircase that we're trying to get to the bottom of." That never works
either. Why? Because all that the audience cares about is what is the
thrust of the scene —what does the hero want? More precisely, what
is the essential aspect of the shot? They aren't there to look at signs,
and they won't look at them. You can't force them to look at them. It
is the nature of human perception to go to the most interesting thing;
and just as we know in terms of the dirty joke, the most interesting
thing is what happens next in the story that you promised the
audience you were going to tell them. You can't make them stop and
look at that sign. They don't care to indulge you by listening to your
looping, so you'd better do your work beforehand.

That work is done in understanding the nature of the materials and

using that understanding in the design of the film. That's basically
what a film is; it's a design. You know, all these personally felt
statements of people who try to put a lot of garbage into the shot and
pan around a bunch to show how moved they are by their chosen
subject: these are just like

countercultural architecture. They may be a personal statement, but

they don't serve the turn of the inhabitants or, in this case, the turn
of the viewers who would like to know what happens next. You tax
the audience every time you don't move on to the next essential step
of the progression as quickly as possible. You're taxing their good
nature. They may indulge you for political reasons—which is what
most of modem art is about. Political reasons being, "dammit, I hke
those kinds of bad movies" or "I hke that kind of countercultural
statement. I am one of that group, and I endorse the other members
of this group, who appreciate the sort of things this fellow is trying to
say." The audience can endorse the triviality of modem art, but they
can't like it. I suggest you think about the difference between the way
people talk about any performance artist and the way they talk about
Gary Grant. And to you lovely enthusiasts who will aver that the
purpose of modem art is not to be liked, I respond, "oh, grow up."

The job of the film director is to tell the story through the
juxtaposition of uninflected images —because that is the essential
nature of the medium. It operates best through that juxtaposition,
because that's the nature of human perception: to perceive two
events, determine a progression, and want to know what happens
next.

"Performance art" works, as it's the nature of human perception to

order random images in favor of an overriding preconception.
Another example of this is neurosis. Neurosis is the ordering of
unrelated events or ideas or images in favor of an overriding
preconception.

"1 am," for example, "an unsightly person": that's the over-

ON DIRECTING FILM

riding preconception. Then, given any two unrelated events 1 can

order them to make them mean that. "Oh, yes, I understand. This
woman came out of the hall and did not seem to notice me and
rushed into the elevator and quickly pushed the button and the
elevator closed because I am an unattractive person." That's what
neurosis is. It is the attempt of a disordered mind to apply the
principle of cause and effect. This same attempt takes place
subconsciously in the viewer of a drama.

If the lights go out and the curtain goes up, the overriding idea is "a
play is taking place"; "someone is telling me a story."
The human brain, understanding that, will take all of the events in
the play and form them into a story just as it forms perception into
neurosis. It is the nature of human perception to connect unrelated
images into a story, because we need to make the world make sense.

If the overriding idea is that a play is taking place, then we will form
the images that we see between the time the curtain goes up and the
time the curtain comes down into a play whether or not they have
been structured as one. Just so with the movie, which is why bad
filmmaking can "succeed." It is our nature to want to make sense of
these events—we can't help it. The human mind would make sense of
them even if they were a random juxtaposition.

This being the nature of human perception, the smart dramatist will
use it to his or her advantage and say, "well, if the human mind is
going to do all that anyway, why don't / do it first? Then I will be
going with the flow rather than battling against the tide."

If you aren't telling a story, moving from one image to

another, the images have to be more and more "interesting" per se. If
you are teUing a story, then the human mind, as it's working along
wdth you, is perceiving your thrust, both consciously and, more
important, subconsciously. The audience members are going to go
along with that story and wall require neither inducement, in the
form of visual extravagance, nor explanation, in the form of
narration.

They want to see what's happening next. Is the guy going to get
killed? Is the girl going to kiss him? Will they find the money buried
in the old mine?

When the film is correctly designed, the subconscious and the

conscious are in alignment, and we need to hear what happens next.
The audience is ordering the events just as the author did, so we are
in touch with both his conscious and his unconscious mind. We have
become involved in the story.
If we don't care what happens next, if the film is not correctly
designed, we may, unconsciously, create our ovsoi story in the same
way that a neurotic creates his ovsti cause-and-effect rendition of the
world around him, but we're no longer interested in the story that
we're being told. "Yes, I saw that the girl put the kettle on the fire and
then a cat ran out on stage," we might say of "performance art." "Yes,
I saw, but I don't quite know where it's going. I'm following it, but I
am certainly not going to risk my unconscious well-being by
becoming involved."

That's when it stops being interesting. So that's where the bad

author, like the countercultural architect, has to take up the slack by
making each subsequent event more diverting than the last; to trick
the audience into paying attention.

The end of this is obscenity. Let's really see their genitals, let's really
endanger the actor through stunts, let's really set the building on fire.
Over the course of a movie, it forces the filmmaker to get more and
more bizarre. Over the course of a career, it forces a filmmaker to get
more and more outre; over the course of a culture, it forces the
culture to degenerate into depravity, which is what we have now.

Interest in a film comes from this: the desire to find out what
happens next. The less reality conforms to the neurotic's view, the
more bizarre his explanation must become, the end of which
development is psychosis—"performance art" or "modem theater" or
"modem filmmaking."

The structure of any dramatic form should be a syllogism— which is

a logical construct of this form: If A, then B. A play or movie
proceeds fi-om a statement: "if A" (in which a condition of unrest is
created or posited), to a conclusion: "then B" (at which time entropy
will once again rear its corrective head, and a condition of rest will
have been once again achieved).

For example, as we've seen, if a student needs a retraction, he will

pursue a series of actions that will lead him to the retraction or to an
irrevocable denial of the retraction. And then he will be at rest; a
condition of entropy will have been achieved.
This entropy is one of the most interesting aspects of our life as a
whole. We are bom, certain things happen, and we die. The sexual
act is a perfectly good example. Things are called into motion that
did not heretofore exist and that demand some form of resolution.
Something is called into existence that did not heretofore exist, and
then the unrest that this new thing

creates has to be resolved, and when it's resolved, the life, the sexual
act, the play, is done. That's how^ you know when it's time to go
home.

The guy solved his problem at the whorehouse. The guy lost all his
money at the racetrack. The couple was reunited. The bad king died.
How do we know this is the end of the story? Because the rise to
power of the bad king was the problem that we came to see solved.
How do we know that when they kiss it's the end of the movie?
Because it's a movie about the boy not getting the girl. The solution
of the problem posited at the beginning of the experience is the end
of the story. That's also how we know the scene is over, isn't it?

We said that the scene is the correct unit of study. If you understand
the scene, you understand the play or movie. When the problem
posited by the scene is over, the scene is over. A lot of times in
movies you want to get out of the scene before the problem is over
and have it answered in the next scene, as a matter of fact. Why? So
that the audience will follow you. They, you will remember, want to
know what happens next.

To get into the scene late and to get out early is to demonstrate
respect for your audience. It's very easy to manipulate an audience—
to be "better" than the audience—because you've got all the cards. "I
don't have to tell you anything; I can change the story in midstream!
I can be whatever I want. Go to hell!" But listen to the difference
between the way people talk about films by Werner Herzog and the
way they talk about films by Frank Capra, for example. One of them
may or may not understand something or other, but the other
understands what it is to tell a story, and he wants to tell a story,
which is
the nature of the dramatic art—to tell a story. That's all it's good for.
People have tried for centuries to use drama to change people's lives,
to influence, to comment, to express themselves. It doesn't work. It
might be nice if it worked for those things, but it doesn't. The only
thing the dramatic form is good for is telling a story.

If you want to tell a story, it might be a good idea to understand a

little bit about the nature of human perception. Just as, if you want
to know how to build a roof, it might be a good idea to understand a
little bit about the effects of gravity and the effects of precipitation.

If you go up into Vermont and you build a roof with a peak, the snow
will fall off. You build a flat roof, the roof will fall dovsni from the
weight of the snow—which is what happened to a lot of the
countercultural architecture of the 1960s. 'There may be a reason
people have wanted to hear stories for ten million years," the
performance artist says, "but I really don't care, because / have
something to say.^^

The film business is caught in a spiral of degeneracy because it's run

by people who have no compass. And the only thing you can do in
the face of this downward force is tell the truth. Anytime anyone tells
the truth, that's a counterforce.

You cannot hide your objective. No one can hide. Contemporary

American films are almost universally sloppy, trivial, and obscene. If
your objective is to succeed in the "industry," your work, and your
soul, will be exposed to these destructive influences. If you
desperately crave acceptance by that industry, you will likely become
those things.

The actor cannot hide his or her objective, neither can the

playwright, neither can the film director. It a person^s objective is

truly—and you don't have to do it humbly, because you1l get humble
soon enough— to understand the nature of the medium, that
objective w^ill be communicated to the audience. How^? Magically.
1 don't know how. Because it wdll. It just can't be hidden. In addition
to what you will or will not learn about the medium through your
desire to understand it, that desire itself will be manifested.

1 carve wood sometimes. It's magical how the wooden object creates
itself. One becomes enthralled by and very observant of the grain of
the wood, and the piece tells you how to carve it.

Sometimes the piece is fighting back against you. If you're honest in

making a movie, you'll find that it's often fighting back against you
too. It's telling you how to write it. Just as we found in the "got a
retraction" movie.

It's very, very difficult to do these very, very simple problems.

They're fighting back against you, these problems, but the mastery of
them is the beginning of the mastery of the art of film.

THE TASKS OF THE

DIRECTOR

WHAT TO TELL THE

ACTORS AND WHERE TO

PUT THE CAMERA

I've seen directors do as many as sixty takes of a shot. Now, any

director who's watched daiHes knows that after the third or fourth
take he can't remember the first; and on the set, when shooting the
tenth take, you can't remember the purpose of the scene. And after
shooting the twelfth, you can't remember why you were bom. Why do
directors, then, shoot this many takes? Because they don't know
what they want to take a picture of. And they're frightened. If you
don't know what you want, how do you know when you're done? If
you know what you want, shoot it and sit down. Suppose you are
directing the "get a retraction" movie. What are you going to tell the
actor who does that first beat for you? What do we refer to; what is
our
compass here? What is a simple tool to which we may refer to answer
this question?

To give direction to the actor, you do the same thing you do when
you give direction to the cameraman. You refer to the objective of the
scene, which in this case is to get a retraction; and to the meaning of
this beat, which here is to arrive early.

Based on this, you tell the actor to do those things, and only those
things, he needs to do for you to shoot the beat, to arrive early. You
tell him to go to the door, try the door, and sit dovm. That is literally
what you tell him. Nothing more.

Just as the shot doesn't have to be inflected, the acting doesn't have
to be inflected, nor should it be. The acting should be a performance
of the simple physical action. Period. Go to the door, try the door, sit
down. He doesn't have to walk down the hall respectfully. This is the
greatest lesson anyone can ever teach you about acting. Perform the
physical motions called for by the script as simply as possible. Do not
"help the play along."

He doesn't have to sit down respectfully. He doesn't have to turn the

door respectfully. The script is doing that work. The more the actor
tries to make each physical action carry the meaning of the "scene" or
the "play," the more that actor is ruining your movie. The nail doesn't
have to look like a house; it is not a house. It is a nail. If the house is
going to stand, the nail must do the work of a nail. To do the work of
the nail, it has to look like a nail.

The more the actor is giving him or herself over to the specific
uninflected physical action, the better off your movie is, which is why
we like those old-time movie stars so much. They were awfully damn
simple. "What do I do in this scene?"

was their question. Walk down the hall. How? Fairly quickly. Fairly
slowly. Determinedly. Listen to those simple adverbs— the choice of
actions and adverbs constitutes the craft of directing actors.
What's the scene? To get a retraction. What's the meaning of the
beat? To arrive early. What are the specific shots? Guy walking down
the hall, guy tries the doorknob, guy sits down. Good luck will be the
residue of good design. When the actor says, "how do I walk down
the hall?" you say, *i don't know. . . quickly." Why do you say that?
Because your subconscious is working on the problem. Because
you've paid your dues at this point and you're entitled to make what
may seem to be an arbitrary decision but may also be a subconscious
solution to a problem; and you have honored the subconscious by
referring the problem to it long enough for it to cough up the answer.

Just as it's in the nature of the audience to want to help the story
along, to help along good work, that is to say work which is respectful
of its inner nature, just so it's in the nature of your subconscious to
want to help this task along. A lot of decisions that you think are
going to be made arbitrarily are arrived at through the simple and
dedicated workings of your subconscious. When you look back at
them, you will say, **well, 1 got lucky there, didn't 1?" and the answer
will be "yes" because you paid for it. You paid for that subconscious
help when you agonized over the structure of the film. The shot list.

Actors will ask you a lot of questions. "What am I thinking here?"

"What's my motivation?" "Where did I just come from?" The answer
to all of these questions is it doesn 't matter. It doesn't

matter because you can't act on those things. I defy anyone to act
where he just came from. If you can't act on it, why think about it?
Instead, your best bet is to ask the actor to do his simple physical
actions as simply as possible.

"Please walk down the hall, try the doorknob." You don't have to say
"try the doorknob and it's locked." Just try the doorknob and sit
down. Movies are made out of very simple ideas. The good actor will
perform each small piece as completely and as simply as possible.

Most actors are, unfortunately, not good actors. There are many
reasons for this, the prime reason being that theater has fallen apart
in our lifetime. When I was young, most actors, by the time they got
to be thirty, had spent ten years on the stage, earning their living.
Actors don't do that anymore, so they never get a chance to learn
how to act well. Virtually all of our actors in this country are badly
trained. They're trained to take responsibility for the scene, to be
emotional, to use each role to audition for the next. To make each
small and precious moment on the stage or screen both "mean" the
whole play and display their wares, to act, in effect, "sit down
because I'm the king of France." It's not that actors are dumb people.
To the contrary, the job, in my experience, attracts folk of high
intelligence, and most of them are dedicated people; bad actors and
good actors are in the main dedicated and hardworking people.
Unfortunately, most actors don't accomplish much, because they're
badly trained, underemployed, and anxious both to advance their
career and to "do good."

Also, most actors try to use their intellectuality to portray

the idea of the movie. Well, that's not their job. Their job is to
accomplish, beat by beat, as simply as possible, the specific action set
out for them by the script and the director.

The purpose of rehearsal is to tell the actors exactlj the actions called
for, beat by beat.

When you get on the set, the good actors who took careful notes will
show up, do those actions—not emote, not discover, but do what
they're getting paid to do, which is to perform, as simply as possible,
exactly the thing they rehearsed.

If you, the director, understand the theory of montage, you don't

have to strive to bring the actors to a real or pretended state of frenzy
or love or hate or anything emotional. It's not the actor's job to be
emotional—it is the actor's job to be direct.

Acting and dialogue fall into the same boat. Just as with the acting,
the purpose of the dialogue is not to pick up the slack in the shot list.
The purpose of dialogue is not to carry information about the
"character." The only reason people speak is to get what they want.
In film or on the street, people who describe themselves to you are
lying. Here is the difference: In the bad film, the fellow says, "hello,
Jack, I'm coming over to your home this evening because I need to
get the money you borrowed from me." In the good film, he says,
"where the hell were you yesterday?"

You don't have to narrate with the dialogue any more than you have
to narrate with the pictures or the acting. The less you narrate, the
more the audience is going to say, "wow. What the heck is happening
here? What the heck is going to happen next. . . ?" Now, if you're
telling the story with the pictures, then the dialogue is the sprinkles
on top of the ice cream cone.

It's a gloss on what's happening. The story is being carried by the

shots. Basically, the perfect movie doesn't have any dialogue. So you
should always be striving to make a silent movie. If you don't, what
will happen to you is the same thing that happened to the American
film industry. Instead of writing the shot list, you'll have the student
rise and say, "isn't that Mr. Smith? I think I'll get a retraction from
him." Which is what happened to American films when sound came
in, and they've gotten worse ever since.

If you can learn to tell a story, to break dovsoi a movie according to

the shots and tell the story according to the theory of montage, then
the dialogue, if it's good, will make the movie somewhat better; and if
it's bad, will make the movie somewhat worse; but you'll still be
telling the story with the shots, and they can take the brilliant
dialogue out, if need be—as, in fact, they do when a film is subtitled
or dubbed—and a great film, so treated, is injured hardly at all.

Now that we know what to tell the actors, we need an answer to the
one question the crew will ask you again and again—"where do we
put the camera?" The answer to this question is, "over there."

There are some directors who are visual masters—who bring to

moviemaking a great visual acuity, a brilliant visual sense. I am not
one of those people. So the answer I'm giving is the only answer I
know. I happen to know a certain amount about the construction of a
script, so that's what I'm telling you. The question is, "where do I put
the camera?" That's the simple question, and the answer is, "over
there in that place in which
it will capture the uninflected shot necessary to move the story
along."

"Yes, but," a lot of you are saying, "I know that the shot should be
uninflected, but really since it's a scene about respect shouldn't we
put the camera at a respectful angle?"

No; there is no such thing as "a respectful angle." Even if there were,
you wouldn't want to put the camera there—if you did so, you
wouldn't be letting the story evolve. It's like saying: "a naked man is
walking dovm the street copulating with a whore while going to a
whorehouse." Let him^et to the whorehouse. Let each shot stand by
itself. The answer to the question "where do you put the camera?" is
the question "what's the shot of?"

That's my philosophy. I don't know better. If I knew a better answer

to it, I would give it to you. If I knew a better answer to the shot, I
would give it to you, but because I don't, I have to go back to step
number one, which is "keep it simple, stupid, and don't violate those
rules that you do know. If you don't know which rule applies, just
don't muck up the more general rules."

I know it's a shot of a guy walking down a hall. I'm going to put the
camera somewhere. Is one place better than another? Probably. Do I
know which place is better than another? No? Then Til let my
subconscious pick one, and put the camera there.

Is there a better answer to the question? There may be, and the
better answer may be this: in the story board for a movie or a scene,
you may see a certain pattern developing, which

might tell you- something. Perhaps your task as a designer of shots

is, after a point, that of a "decorator," quite frankly.

"What are the Equalities' of the shot?'' I don't happen to think that's
the most important question in making a movie. I think it's an
important question, but I don't think it's the most important
question. When faced with the necessity of a particular election, I'm
going to answer what I think is the most important question first,
and then reason backward and answer the smaller question as best I
can.

Where do you put the camera? We did our first movie and we had a
bunch of shots with a hall here and a door there and a staircase
there.

"Wouldn't it be nice," one might say, "if we could get this hall here,
really around the comer from that door there; or to get that door
here to really be the door that opens on the staircase to that door
there?'''' So we could just move the camera from one to the next?

It took me a great deal of effort and still takes me a great deal and
will continue to take me a great deal of effort to answer the question
thusly: no, not only is it not important to have those objects literally
contiguous; it is important to fight against this desire, because
fighting it reinforces an understanding of the essential nature of film,
which is that it is made of disparate shots, cut together. It's a door,
it's a hall, it's a blah-blah. Put the camera "there" and photograph, as
simply as possible, that object. If we don't understand that we both
can and must cut the shots together, we are sneakily falling victim to
the mistaken theory of the Steadicam. It might be nice to have these
objects next to each other so as to avoid having to move the crew, but
you

don't get any sneaky artistic good out of literally having them next to
each other. You can cut the shots together.

This relates to what I said about acting: if you can cut different
pieces, different scenes together, different lines together, you don't
have to have somebody in every shot u^ith the same "continuous
intention." The same "commitment to and understanding of the
character." You don't need it.

The actor has to be performing a simple physical action for the space
of ten seconds. It does not have to be part of the "performance of the
film." Actors talk about the "arc of the film" or the "arc of the
performance." It doesn't exist on stage. It's not there. The
performance takes care of both. The "arc of the performance," the act
of controlling, of doling out emotion here and withholding emotion
there, just doesn't exist. It's like a passenger sticking his arms out of
the airplane window and flapping them to make the plane more
aerodynamic. This commitment to the arc of the film—it's ignorance
on the part of the actor, ignorance of the essential nature of acting in
film, which is that the performance will be created by the
juxtaposition of simple, for the most part uninflected shots, and
simple, uninflected physical actions.

The way to shoot the car crash is not to stick a guy in the middle of
the street and run over him and keep the camera on. The way to
shoot the car crash is to shoot the pedestrian walking across the
street, shoot the shot of the onlooker whose head turns, shoot the
shot of a man inside the car who looks up, shoot the shot of the guy's
foot coming down on the brake pedal, and shoot the shot underneath
the car with the set of legs lying at a strange angle (with thanks to
Pudovkin, for the

above). Cut them together, and the audience gets the idea: accident.

If that's the nature of film for the director, that's the nature of film
for the actor too. Great actors understand this.

Humphrey Bogart told this story: When they were shooting

Casablanca and S. Z. (Cuddles) Sakall or someone comes to him and
says, "they want to play the ^Marseillaise,' what should we do?—the
Nazis are here and we shouldn't be playing the 'Marseillaise,' "
Humphrey Bogart just nods to the band, we cut to the band, and they
start playing "bah-bah-bah-^/i."

Someone asked what he did to make that beautiful scene work. He

says, "they called me in one day, Michael Curtiz, the director, said,
'stand on the balcony over there, and when I say "action" take a beat
and nod,' " which he did. That's great acting. Why? What more could
he possibly have done? He was required to nod, he nodded. There
you have it. The audience is terribly moved by his simple restraint in
an emotional situation—and this is the essence of good theater: good
theater is people doing extraordinarily moving tasks as simply as
possible. Contemporary playwriting, filmmaking, and acting tend to
offer us the reverse—people performing mundane and predictable
actions in an overblown way. The good actor performs his tasks as
simply and as unemotionally as possible. This lets the audience "get
the idea"—just as the juxtaposition of uninflected images in service of
a third idea creates the play in the mind of the audience.

Learn this, and go out and make the movie. You'll get someone who
knows how to take a picture, or jou learn how to take a picture; you
get someone who knows how to light, or jou

learn how to light. There's no magic to it. Some people will be able to
do some tasks better than others—depending upon the degree of
their technical mastery and their aptitude for the task. Just like
playing the piano. Anybody can learn how to play the piano. For
some people it will be very, very difficult—but they can learn it.
There's almost no one who can't learn to play the piano. There's a
wide range in the middle, of people who can play the piano with
various degrees of skill; a very, very narrow band at the top, of people
who can play brilliantly and build upon a simple technical skill to
create great art. The same thing is true of cinematography and sound
mixing. Just technical skills. Directing is just a technical skill. Make
your shot list.

PIG — THE MOVIE

The questions that you want to ask as a director are the same
questions you want to ask as a writer, the same questions you want
to ask as an actor. "Why now?" "What happens if I don't?" Having
discovered what is essential, you then know what to cut.

Why does the story start now? Why does Oedipus have to find out
who his parents are? This is a trick question. The answer is this: he
doesn't have to find out who his parents are. He has to cure the
plague on Thebes. He discovers he, himself, is the cause of the plague
on Thebes. His simple quest for external information led him on a
journey, which resulted in his discovery. Oedipus is the model of all
tragedy, according to Aristotle.
Dumbo has big ears; that's his problem. He was bom with it. The
problem gets worse, people make more and more fun of him. He has
to try to learn to cure it. He meets little friends along the way who
come to his aid in this classic myth. (The study of myth is very useful
for directors.) Dumbo learns to fly; he develops a talent that he didn't
realize he had and comes to this understanding about himself: that
he's not worse than his fellows. He's perhaps not better, but he's
different, and he has to be himself—when he realizes this, his journey
is over. The problem of his big ears has been solved not by ear
reduction but by self-discovery—and the story is over.

Dumbo is an example of a perfect movie. Cartoons are very good to

watch—are much better to watch, for people who want to direct, than
movies.

In the old cartoons, the artists realized the essence of the theory of
montage, which is that they could do whatever the heck they wanted.
It wasn't any more expensive to draw it from a high angle or from a
long angle. They didn't have to keep the actors late to draw a
hundred people rather than one person, or send out for that very
expensive Chinese vase. Everything was based on the imagination.
The shot we see in the film is the shot the artist saw in his
imagination. So if you watch cartoons, you can learn a great deal
about how to choose shots, how to tell the story in pictures, and how
to cut.

Question: What starts the story now? Because if you don't know what
starts the story, what's the impetus to start the story, then you have
to rely on "back story" or history, all those dread terms that those
swine out in Hollywood use to describe a process they not only do
not understand but don't particularly

ONDIRECTINGFILM SI

care about. The story is not begun because the hero ^'suddenly gets
an idea"—it is brought into being by a concrete external event: the
plague on Thebes, the big ears, the death of Charles Foster Kane.
Thus you start the story in such a way that you bring along the
audience. They are there at the birth. So they are going to want to
know what happens next. "Once upon a time," for example, "there
was a man who owned a farm" or "there were once three sisters."
Just like a dirty joke. That's how the drama is structured—and this
drama, like the dirty joke, is just a specialized form of fairy tale.

The fairy tale is a great teaching tool for directors. Fairy tales are told
in the simplest of images and without elaboration, without an
attempt to characterize. The characterization is left up to the
audience.* In fairy tales, we see that it is simple to know when to
begin and to know when to stop. And if one can apply those simple
tests to the play as a whole, one can apply them to the scene, which is
only a small play, and to the beat, which is only . . . et cetera.

"Once there was a farmer who wanted to sell his pig." How do I know
when I'm finished? When the pig's sold, or when the farmer
discovers that he cannot sell the pig—when the end of that syllogism
has come to pass.

Now, not only do 1 know when to start and when I want to stop, but I
also know what to keep in and what to throw out. The farmer's
interesting encounter with a female swineherd, which has nothing to
do with selling the pig, probably shouldn't

*BetteIheim, The Uses of Enchantment.

DAVID MAMET

be in the movie. In plotting a film, one can also ask: "what am I

missing here?" Am I going from the beginning to the end in a logical
progression? And if not, what missing term will render the
progression logical?

Here's a story: "once there was a farmer who wanted to sell a pig."
Now, how do you open the film? What are the shots? How do you
make up your shot list?

STUDENT: You establish a good farm.

MAMET: Why do you have to establish a good farm? Everybody in
Hollywood always whines, "but we won't know where we are ..." But I
put it to you, ladies and gentlemen, how often in the thousands of
movies that weVe all seen has anyone said, "hey, wait a second, I
don't know where I am"? In fact, quite the contrary is true. You come
to a movie in the middle, turn on the TV in the middle, look at a tape
in the middle, you know exactly what's happening, always,
immediately. You are interested in it because you want to know
what's going on. That's what interests you. What would be better
than an establishing shot of a farm? What will answer the question
"why now?"

STUDENT: The reason he has to sell his pig?

MAMET: The reason he has to sell his pig. What's his reason? The
answer will lead us to a very specific beginning. A beginning specific
to this film —rather than one specific to a film. "Once there was a
farmer who

RECTING FILM

83

STUDENT:

MAMET:

STUDENT:

MAMET:

wanted to sell his pig" leads us to "once there was a farmer who had
to sell his pig." You will find that the study of semantics, which is the
study of how words influence thought and action, will help you
immensely as a director. Notice the difference in those two
beginnings: they lead you down very divergent trains of thought.
They will change the words you use to tell your ideas to the actors.
It's very, very important to be concise. Okay: "once there was a
farmer who had to sell his pig" A vdde shot of pigs in a pasture. And
then the farmer walking across the pasture. The next shot is a For
Sale sign that he is hammering. Into his pig? Into a post.

Uh huh. Exposition in film is like exposition in any art form. If you

explain the joke's punch line, the audience might understand it, but
they won't laugh. The real art, the essential art in choosing the shots,
is not so much to make the audience understand as to invest yourself
in the clear telling of the story. You aren't smarter than thej are.
The/re smarter than you are. You understand the story as well as you
possibly can, and then they will too. Putting up a sign is an easy way
out. That's not always bad per se, but I think we can do better. We
can ask what the character is doing, but better to ask what is the
meaning of this scene? (To help understand this

DAVID MAMET

STUDENT: STUDENT: MAMET:

distinction, may I recommend the "Analysis" chapter of yl Practical

Handbook for the Actor, Bruder, Cohn, Olnek, et al.)

Literally, on the page, as it is written, the farmer has to sell his pig.
What does this mean in this scene? The essence of having to sell
one's pig could be many different things. The essence could be, a
man fell on hard times. The essence could be, a man had to leave his
ancestral home. A man had to take leave of his best friend. A man
had to do his duty. A man had too many pigs.

Well, yes. But you're thinking on a different level of abstraction. The

point is not the pigs, right? The point is what does the pig mean to
the man? A man's business, for example? What might be the
meaning of that? A man's business grew too fast for him. What you
want to dramatize is not the surface, "a man needs to sell his pig/'
but the essence —what selling the pig means in this story.

Why does he need to sell the pig? The more specifically you think
about the nature of the story, the more you can think of the essence
of the scene rather than the appearance of the scene, then the easier
it will be to find the images. It's a lot easier to find specific images for
"a man fell on reduced circumstances" than for "once there was a
man who had to sell a pig"

Jung wrote that one can't stand aloof from the

images, the stories, of the person who's being analyzed. One has to
enter into them.

If you enter into them, they'll mean something to you. If you don't
enter into them, then your subconscious will never work. You'll never
come up with anything that the audience couldn't have thought of
better at home.

It's like the actor who goes home and figures out what the
performance is supposed to mean, then shows up on the stage and
does that performance. The audience will probably understand this
actor, and his performance, but they won't care.*

"Pig for Sale." Why? The problem starts now. The picture starts with
the inception or the discovery of the problem. Most movies start
thusly: "honey, is that damn pig, which we can ill afford to keep, still
eating up the last groceries in the

*Stanislavsky said that there are three types of actors. The first
presents a ritualized and sup>erficial version of human behavior,
this version coming from his observation of other bad actors. This
actor will give the audience a stock rendition of "love," "anger," or
whatever emotion seems to be called for by the text. The second actor
sits vdth the script and comes up with his own unique and
interesting version of the behavior supposedly called for by the
scene, and he comes to the set or stage to present that. The third,
called the "organic" actor by Stanislavsky, realizes that no behavior
or emotion is called for by the text—that only action is called for by
the text—and he comes to the set or stage armed only with his
analysis of the scene and prepared to act moment-to-moment, based
on what occurs in the performance ... to deny nothing and to invent
nothing. This last, organic actor is the artist with whom the director
wants to work. He is also the artist we most admire on stage and in
films. Cuhouslj, he is not the artist most usually
DAVID MAMET

STUDENT:

MAMET:

STUDENT:

MAMET:

house?" The real artistry of the film director is to learn to do without

the exposition; and, so, involve the audience. Let's come up with
some dramatic shot lists that are going to communicate the idea
"why now?"

The letter from the bank arrives? Let's stay away from that.

Start in a graveyard, and the farmer is at the grave, and the next shot
would be the house and it's nearly deserted and there's no food in the
cupboard. When we see the empty cupboard, we might wonder, why
don't they just kill the pig? Here's a different story: A small child is
dressed in rags and playing in the yard and then a shot of the pig
jumping over the fence and attacking the child. Eh? Kid playing in
the yard—she sees something, starts to run away. Second shot, a pig
jumps into the thing, squeak, squeak, squeak, squeak, squeak; and
the third shot is the farmer walking down the

denominated the great actor. Over the years, I have observed that
there are two subdivisions of the thespian's art: one is called Acting,
and the other is called Great Acting; and that, universally, those who
are known as the Great Actors, the Premier Actors of their age, fall
into the second of Stanislavsky's categories. They bring to the stage
and screen an intellectual pomposity. The audience calls them Great,
I think, because it wants to identify with them—with the actors, that
is, not with the characters the actors portray. The audience wants to
identify with these actors because they seem empowered to behave
arrogantly in a protected setting. On the other hand, look at the old
character actors and comedians: Harry Carey, H. B. Warner, Edward
Arnold, William Demarest; look at Thelma Ritter, Mary Astor, Celia
Johnson. Those people could act.
ON DIRECTING FILM

87

road with the pig. Does that tell you a story? Yes. But how can we do
it without showing the pig mauling the child? We don't want to show
the pig mauling the child, because that has one of two results. Every
time you show the audience something that is "real," they think one
of two things: (1) "oh, dash it all, that's fake" or (2) "oh my God,
that's reair Each one of these takes the audience away from the story
you're telling,* and neither one is better than the other. "Oh, he's not
really copulating with her" or "oh my God, he's really copulating with
her!" Both lose the audience. If we suggest the idea, we can shoot it
better than if we show it.

What about we cut from kid playing in the yard to mom's in the
kitchen, she snaps around, and then she's running out, she grabs a
broom, for example. Third shot is dad walking the pig out of the bam.
He's on the road, we see him going on past the gate. He's obviously
going to get rid of the pig.

STUDE^4T: But he could just shoot the pig. Don't we have to show
the empty cupboard to show why he has to sell the pig?

MAMET: If you try to narrate the fact that the family is on the brink
of poverty, you split your focus. You split it between (1) "I need the
money" and (2) "the pig just attacked my daughter." Now the guy has
two

*This is the meaning of the concept "violating the aesthetic distance.

DAVID MAMET

STUDENT STUDENT STUDENT MAMET:

reasons to sell the pig, which is not as good as one reason to sell the
pig. Two reasons are equal to no reasons—it's like saying: "I was late
because the bus drivers are on strike and my aunt fell downstairs."
And so now, what's the idea: once upon a time, a man had to sell a
dangerous pig. The first shot is the child playing in the yard. The
second shot is the pig eyeing her. Cut to mom in the kitchen.

She hears something, she turns, she picks up a broom and runs out
of the house. Cut to a shot of the farmer leading the pig down the
road. Okay.

Here's another possibility. There's an interior of a bam. It's a shot of

the door. DoOr opens, here comes a farmer in work clothes. He
comes in and lays down his hoe, picks up a lantern and lights it. Now
he turns, and there's a tracking shot of him walking past a row of
empty pens to one pen that has a pig in it. He puts the lantern on the
ledge. He reaches down and takes out a small trough and puts it in
front of the pig. Then he comes up and empties a sack of grain into
the trough. He turns the sack upside down and empties it. Then cut
back to the shot of the trough, and only two or three kernels fall into
it. Then next day—that is, exterior day, which shows the audience
that time has passed. We know the bam sequence was at night—it
involved lighting a lantern. This is a shot of an exterior,

RECTING FILM

89

and it is day. It may be picayune to suggest that a filmscript not

contain the description ''the next day"—but as the audience can only
determine that it is *'the next day" from that which they see on the
screen, perhaps it would be a salutarv habit only to describe those
things that the audience is going to see on the screen. The shot of the
farmer leading the pig down the road: how does that worlc^

STUDENT: WeVe not worrying about the baby anymore.

MAMET: That's right. It's a different story. One is the idea of a man
getting rid of danger, a man eradicating danger. The other is a man
brought to straitened circumstances. Yeah, you're right. I like the
dangerous pig more. How do we know when that story is over?

STUDENT: He sells it or he doesn't.

MAMET: So what happens now? "John," the pig's owner, is walking
down along the road with the pig, when he comes to a crossroads,
and as we say in Chicago, he sees a prosp)erous-lc>oking man
walking dovvTi the road. They enter into a conversation, and John
convinces the man to buy his pig. Just as the deal is about to be
concluded, however, what happens?

STUDENT: The pig bites the man.

MAMET: We said the essence of the scene was the man wants to
divest himself of the pig. He's offered a perfect opportunity to sell the
pig. Great, we didn't expect it, we thought we were going to have to
go all the darned way to town and have to take the bus home, with
nothing to read. Now, out of nowhere, comes

DAVID MAMET

STUDENT: MAMET:

this guy, a buyer, a perfect opportunity presents itself, and what

happens? The pig, the dangerous pig, bites the guy. Now, what's this
beat about? Failed attempt.

No, let's describe the beat as a step on the road to the objective of the
scene, which is to divest oneself of a dangerous commodity. You
might say, capitalizing on a golden opportunity. That is the active
thrust of the beat. "Failed attempt" is just the result.

The great thing about this method is this: what did we say the film
was about? A man had to rid his house of danger. That's what you go
out to film. It doesn't matter that all your cinematographers and
assistant directors and producers are pleading with you to show
more of the farm. You'll say to them, "why? It's not a movie about a
farm. You want to see a movie about a farm? Great. You know? Go
see a travelogue. Go look at a map. This is a movie about a man who
has to rid his house of danger. Let's make this movie. The audience
knows what a farm looks like or they don't. That's their lookout. Let's
respect their privacy." So, a man tries to capitalize on a golden
opportunity.
Well, we could start wdth him walking with the pig, and on the
roadside, fixing a broken cart wheel, he sees another farmer. And he
goes over and takes the initiative to talk to that man. Stay with the
shots: our guy walking down the road. He stops because he sees
something. Come to his

STUDENT:

MAMET:

RECTING

LM

91

STUDENT:

MAMET:

STUDENT: MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

point of view, a cart with a broken wheel, two pigs in the back, and a
prosperous farmer is fixing the cart wheel. Now, what would we like
to include? To keep the idea of capitalize, what about if he does
something with the pig? He might walk differently, knowing he was
going to sell the pig.
The idea is capitalize. The verb is not to make a sale but to capitalize.
He can spruce up the pig.

The shot is: he takes out a handkerchief and wipes off the pig's face.
He wants to sell the pig. He might take the handkerchief and put it
around the pig's neck.

1 like this handkerchief. Let's think of something else. How else could
he capitalize? He wipes the pig's face off and he ties the thing around
the pig and he walks over to the guy. What's going to happen now?

Maybe he would help the guy fix the cart wheel. That way he would
get into his confidence. He could do that. That would help him
capitalize. Good.

When he helps the guy, that would make his sale easier.

Yes. We got the shot of him sprucing up the pig, now a shot of him as
he leads the pig over to the farmer, who's just pulling the cart out,
maybe helps him push the cart the last inch up the road, and

DAVID MAMET

the shot of these two guys talking for a couple of

seconds. The new farmer looks down at the

pig'

MAMET:

looks up at the guy, they talk, the new guy reaches in his pocket,
gives our guy some money. It doesn't have to be more intricate than
that. Does that tell the story?

Or else you don't have him putting the hand in the pocket. You have
the two guys talking, blah-blah-blah . . . STUDENT: . . . and the shot
of that pig, with the same look he had before he attacked the little
girl. Exactly so. We have the two guys talking, and they shake hands.
Now we have a shot of the new farmer picking the old pig up and
putting the pig inside his cart. It's an open cart, so we can have a shot
of the pig in the cart, extreme close-up of the pig. We cut to the pig's
point of view, through the bars, of the two guys talking. While the
new farmer puts his hand in his pocket for money, we cut to the shot
of. . .

. . . the pig jumping out of the cart, and the next shot is our farmer
walking down the road with the

pig-Great. Now we are really telling the story of "once

there was a farmer who tried everything he could

to sell a dangerous pig."

So now our guy is back to walking down the

road with the pig. What's the next interchange going

to be? Where's he going to go? Anybody? Let's make

STUDENT:

MAMET:

RECTING FILM

93

STUDENT: MAMET:

STUDENT: MAMET:

STUDENT: MAMET:

STUDENT: MAMET:

sure we follow the rule against circularity. Don't do the same thing
twice. This circularitj, or repetition of the same incident in different
guises, is antithetical to the dramatic form. It is the signature of both
the epic and the autobiography, and the reason both are adapted into
drama with much difficulty and little success.

The slaughterhouse.

We're going to go to the slaughterhouse next. All right. But before we

get there, we want to move the story along. Why was it a golden
opportunity when he saw the farmer on the road? Because he didn't
have to travel. So because he blew that golden opportunitv, now
what?

He has to go all the way into town, after all. And with what time-
honored convention of cinema do we dramatize that? It's night, and
it used to be day . . . ? It's night, and we are at the slaughterhouse.
The dark, inky, Egyptian blackness of night has fallen as only it
knows how to fall. Unencumbered by the roseate glow of the
mercur\^ vapor lamps of the city streets, reflected by the trapped
smog of the inversion layer caused by those internal combustion
engines so favored by today's urban men and women as a means of
powering those machines designed and appointed for their
conveyance. Night, 1 say again, night has fallen. One half of that
circularity,

DAVID MAMET

STUDENT MAMET:

in sum, that never-ending round of day-and-night. Night: for some a

time of sleep, w^hile for others a time of wakefulness, as in the case
of our farmer. Night has fallen.

Now, our farmer walks into tov^i, walks wearily into town, it being
night, and walks up to the slaughterhouse. Anybody? What if it's
locked?

The slaughterhouse is locked, and then what? Do it in shots.

STUDENT: Shot of the road at night with the farmer and pig.
Another shot of the slaughterhouse. Takes the pig over there. Shot of
the farmer at the slaughterhouse door, which is locked.
Yes. What idea do we seem to be dramatizing here? Last chance to
sell the pig?

Let's call this beat the end of the weary quest. It's not that it's his last
chance; it's that the story is over. Now we're getting good luck as the
residue of good design, we're getting some extra mileage out of
having been assiduous and following the form. What's the extra
mileage? It's night because it took him a long time to get to the
slaughterhouse. It took him a long time because he didn't get a ride
on the truck. He didn't get a ride because the pig bit the driver. That
same dangerous pig about whom we are now composing a story—so
that even the night is a function of the throughline. The extra mileage
is that the slaughterhouse is locked. Now

MAMET:

STUDENT:

MAMET:

ON DIRECTING FILM

95

STUDENTT: MAMET:

STUDENT: MAMET:

we have a raking shot from around the comer of the front of the
slaughterhouse, and we see that the hght is on, and we see in the
office, the Httle office, we see the Hght go off. A guy comes out the
office door, turns the key, and walks off screen left, as the farmer
comes up from the right and tries the door. So it's the end of the
weary quest. How do we know it's a slaughterhouse? How do we
know it's a slaughterhouse? There's a big pen full of pigs behind it.
We don't have to know it's a slaughterhouse. We have to know it's
where he wants to go. It's the end of the quest. There's a building
with a pen with a lot of pigs in it, and he's walking toward it.
"The end of the quest" does not, however, mean the end of the story.
End of the weary quest is only the title of this beat. Every turn takes
us to the next. That's why it's a good story. Oedipus wants to end the
plague. He finds that this plague hit because somebody killed his
father, and he finds that he's the guy. Any good drama takes us
deeper and deeper to a resolution that is both surprising and
inevitable. It's like Turkish taffy; it always tastes good and it always
sticks to your teeth.

Do you need the guy leaving the slaughterhouse? I think so. But it's
the same question as "where do you put the camera?" At some point,
you, the director, are going to make some decisions, which may seem
arbitrary but which in fact may be based

DAVID MAMET

STUDENT: MAMET:

STUDENT:

MAMET:

STUDENT:

MAMET:

on a continually emerging aesthetic understanding of the story. My

answer to your question is: "I think so." End of the weary quest.

What tool are we going to use to help us determine what happens

next? The throughline.

And we know the throughline is he wants to rid himself of a

dangerous pig. So he sits down and waits.

He could sit down and wait at the slaughterhouse. He could tie up

the pig at the slaughterhouse and go down the block to the bar. Sits
down and has a drink, and the farmer from before comes in and
starts a fight. We come back to the pig, he*s tugging on the rope and
he breaks free and runs into the bar and saves our guy.
Now we get a little bit of extra bang for our buck! We got interested
in our story and the quiddities and oddities of the story; and what
suggested itself was a possible ending to our story. And the reason
that we laugh at our ending is it contains the two essential elements
that we learned of from Aristotle, surprise and inevitahihty.

Aristotle uses rather different words, as he's talking about tragedy

rather than drama: he calls the two fear and pity. Pity because of the
fate of the poor guy who got himself in such a jam; and fear because,
in identifying with the hero, we see that it could also happen to us.

ON DIRECTING FILM

97

STUDENT:

MAMET:

STUDENT:

The reason we identify is that the writer left out the narration. We
only saw the story.

We can identify with the pursuit of a goal. It's much easier to identify
with that than with ''character traits.*'

Most movies are written, "he's the wacky kind of guy who ..." But
then we can't identify with that person. We don't see ourselves in
him because we aren't being shown his struggle but instead are
shown those idiosyncrasies that divide us from him. His "knowledge
of karate," his wacky habit of yodeling to call his dogs, his peculiar
partiality to antique cars . . . how interesting. It's a good thing that
the people in Hollywood have no souls, so that they don't have to
suffer through the lives they lead. Who would like to suggest another
ending? I was just thinking that perhaps the pig has to fight one
more person.

As Leadbelly says about the blues, he says in the first verse use a
knife to cut bread, and in the second verse use a knife to shave, and
in the third verse use it to kill your unfaithful girlfriend. It's the same
knife, but the stakes change, which is exactly the way a play or movie
is structured. You don't want to use the knife in the first verse to cut
bread and in the second verse use it to cut cheese. We already know
it can cut bread. What else can it do? But shouldn't we elaborate on
the pig's danger somehow, at this point, to raise the stakes?

DAVID MAMET

MAMET: We don't have to get him in trouble. We've got to get him
out of trouble. Remember, our task is not to create chaos but rather
to create order out of a situation that has become chaotic. We don't
have to worry about making it interesting; all we have to worry about
is getting rid of the pig.

Let's complete this story in a happy, peppy manner that is both

surprising and inevitable or, at the very least, pleasing, or, at the
very, verj least, internally consistent. We're sitting on the steps with
the pig. It's nighttime. The slaughterhouse is locked.

STUDENT: Well, the next shot is it's daylight and there's a guy
walking up the steps to open up the front door of the slaughterhouse
and can you guess what's going to happen then? He's going to sell the
pig.

MAMET: And then the movie's over. Okay.

STUDENT: How about: it's morning, he wakes up, he thinks

something is missing or he feels for his wallet and it's gone. Then we
cut to the pig sitting there peacefully and then another shot, a guy
lying there dead wdth our guy's wallet in his hand. The pig saved his
wallet.

MAMET: So the pig redeems himself and he can set the pig free.
Setting the pig free fulfills his original purpose, doesn't it? If the
original purpose is to rid himself of danger.

STUDENT: Why didn't he set the pig free before?

MAMET: All right. Good. You have found a very important logical
lacuna in our film. He is trying, throughout,

to rid himself of danger. After the first sequence, when the pig
attacks the little girl, we, as you point out, need a second sequence,
which we might call "the easy solution to a difficult problem." In this
sequence, the farmer is leading the pig away. Shot of the pig,
abandoned, on a hillside. Shot, pig's point of view, of the farmer
walking away.
Shot of the farmer approaching his house. The farmer stops. Shot,
his point of view, of the pig, back in his appointed stall. And then we
resume the story, and the next sequence, after 'The Easy Solution," is
"Capitalizing on a Golden Opportunity."

Good. I think this discovery makes it a better movie. By the way, I

have always found that these piddling points inevitably reveal most
important information when they are explored. They are, I think, like
the minor or half-forgotten points in dreams. One is tempted to
brush past them and think of them as unimportant. But no step in
the logical progression is unimportant. And I know, from my own
experience, that persistence in these "small" points will be rewarded.

Here's another possible ending. It's dawn. Your sound department is

torturing you to okay the inclusion of the sound of birds chirping,
ladies and gentlemen. Oh, well. You see the same guy from the office
open up the slaughterhouse and see the pig. He opens up and leads
the pig into the pen.

Our guy wakes up, there's no pig. He goes in, he wants his pig. The
owner says, "how am I going to know what pig is yours, bobbity,
bobbity, bobbity." Our man is obstreperous. The owner of the
slaughterhouse gets in a fight with our farmer and is going to brain
him to get him to stop bothering him about the pig. We cut to a shot
of the pig, our proverbial pig shot, looking back through the fence at
our guy. We know it's our pig because it's wearing the handkerchief it
acquired in the "golden opportunity" beat, heinl Next we have a shot
of the slaughterhouse owner turning, and then a shot of our guy,
walking back down along the road wdth the pig. Close-up: our guy
stops, turns.

Angle: the pig, who is looking back down the road. Hold on this. The
farmer starts walking the pig back in the direction the pig is looking.

Cut to our proverbial pig shot, our pig looking at something. Shot of
our farmer paying some money to the slaughterhouse owner. Back to
the pig shot, back to the slaughterhouse owner taking the money,
slaughterhouse owner moving gingerly past our pig.
Our pig looking through the bars. Shot, his point of view: the
slaughterhouse owner entering a pen in which is one single pig. He
starts to lead this pig out.

Now. Final sequence. Our farmer walking down the road with two
pigs. Shot of the farmhouse. The wife comes out. Shot, her point of
view: our farmer

leading home the two pigs. Shot of the barnyard fence. The gate is
swung open, the two pigs enter. Shot of the farmer looking on. Shot
of the two pigs kissing. Fade out, fade in. Shot of a pig suciding many
litde piglets. Shot of "our" pig, with the handkerchief around his
neck, riding the little girl around on his back. Shot of our farmer
looking on. What a pig. That's the movie, perhaps. He solved his
problem. He didn't get rid of the pig, he got rid of the danger. Now,
you can look back over the shot list and ask, "what have I left out?"
As you have devoted yourself consciously, honestly, and gently to the
story, you will have created a certificate of deposit, if you will, in your
subconscious, on which you can draw for simple answers to the
question of "where shall 1 put the camera?"—such questions also
being aided by your reference to your list of objectives: a man tries to
rid himself of danger, a man takes the easy solution to a difficult
problem, a man tries to capitalize on a golden opportunity, a man
comes to the end of a long quest, a man tries to regain possession, a
man rewards a good deed. That is the story the director must tell—
the internal story of the hero's persistence in a difficult world.
Anybody with a Brownie can take a picture of a "pig"

CONCLUSION

It's always up to you to decide whether you are going to tell the story
through a juxtaposition of shots or whether vou are not. It's not
always up to you to decide whether or not that process is going to be
interesting. Any real technique is going to be based on things within
your control. Anvthing that is not based on things within your
control is not a real technique. We would like to learn a technique of
directing and anal\-zing as concrete as that of the shoemaker. The
shoemaker will not say when the harness breaks, *'golly, you know, I
did it in the most interesting way I knew how!" Stanislavsky was once
having dinner with a steamboat captain on the Volga River and
Stanislavsky said, "how is it that among all the major and minor
paths of the Volga River, which are so many and so dangerous,

you manage to always steer the boat safely?" And the captain said, "I
stick to the channel; it's marked." And the same thing is true here.

How is it that, given the many, many ways one might direct a movie,
one might always be able, with economy, and perhaps a certain
amount of grace, to tell the story? The answer is: "stick to the
channel; it's marked." The channel is the super-objective of the hero,
and the marker buoys are the small objectives of each scene and the
smaller objectives of each beat, and the smallest unit of all, which is
the shot.

The shots are all you have. That's it. Your choice of the shots is all
you have. It's what the movie is going to be made up of. You can't
make it more interesting when jou get to the editing room. And also
you can't rely on the actors to take up the slack. You can't rely on
them to "make it more interesting." That's not their job either. You
want them to be as simple as you are in your choice of the shots.

If you're correct in the small things, the smallest of which in this case
is the choice of a single uninflected shot, then you will be correct in
the larger things. And then your film will be as correct and as
ordered and as well-intentioned diS you are. It can never be more so,
but it can be less so if you desire to manipulate the material, or hope
that God wall intervene and save you, which is what most people
mean when they talk about "talent."

You might want the shoemakers' elves to save you, but how
wonderful it is not to need the elves to save you. Especially under
conditions of great stress, you have to know your trade. And there is
a trade to screenvmting and there is a trade to

directing a movie. They're very much the same trade. If you pay the
price, you can learn that trade. If you persevere, that analytical
method of thinking will become easier for you. The problems of the
individual films will not get easier—they only get easier for hacks.
The task is always the same. Stick with it until you solve it. It's not
your job to make it pretty. It will be as pretty or unpretty as God
intended. It's your job to make it correct according to your first
principles.

We, just like the protagonists in our movies, have a task. In

completing the task, we have to go from one thing to the next as
logically as possible. Our work is like mountain climbing. It's
frightening sometimes and it's usually arduous, but we don't have to
climb the whole mountain all at once. All we have to do is make a
foothold here, figure out what that beat is or what that shot is or
what that scene is; and when we're completely secure here, reach
until we get the next foothold that is absolutely secure. Dramatic
analysis is a bit like plotting out a compass course over rough
territory. When we get lost, or get confused, terrified, tired,
frightened, all of which will happen to you if you do get the chance to
direct a movie, all we have to do is refer to our map and compass.
The analysis is not the movie, any more than the map is the terrain—
but the right compass and analysis will enable you to navigate both.

The more time you have invested, and the more of yourself you have
invested in the plan, the more secure you will feel in the face of
terror, loneliness, or the unfeeling or ignorant comments of those
from whom you are asking a whole bunch of money or indulgence.

Someone once asked Daniel Boone if he had ever been lost.

He replied, "I was never lost, but I was once a mite bewildered for
three days."

It's good, as the Stoics tell us, to have tools that are simple to
understand and of a very limited number—so that we may locate and
employ them on a moment's notice. I think the essential tools in any
worthwhile endeavor are incredibly simple. And very difficult to
master. The task of any artist is not to learn many, many techniques
but to learn the most simple technique perfectly. In doing so,
Stanislavsky told us, the difficult will become easy and the easy
habitual, so that the habitual may become beautiful.
It is the pursuit of an ideal that is important. This pursuit will lead to
a greater possibility of the unconscious asserting itself, which is to
say, the greater possibility of beauty in your work. The Navahos, I am
told, used to weave flaws into their blankets to let the devils out.

Some contemporary artist said, "well, we don't have to weave in the

flaws. We can try to weave perfectly. God will see that there're
enough flaws in them anyway; that's human nature."

The application of these principles, in my experience, will help you to

weave as perfectly as is humanly possible—which is to say not very
perfectly at all.

Keep giving yourself over to the simple task. This dedication will give
you great satisfaction. The very fact that you have forsworn the Cult
of Self for a little while—the cult of how interesting you and your
consciousness are—will communicate itself to the audience. And they
wall be appreciative in the extreme and give you the benefit of every
doubt.

Is it possible to "do everything right" and still come up with

a bad movie? To "do everything right" means to progress according

to philosophically correct principles step by step such that your
evaluation of your own effort is honest and you are happy that you
have fulfilled the specific task at hand. Is it possible to do that and
come up with a bad movie? What's the answer to that? Well, it
depends on your definition of bad. Once again, a tool that the Stoics
would advise us to use is this: if, before going into battle, you asked
an omen of the gods and they told you that you were about to lose,
would you not be bound to fight in any case?

It's not up to you to say whether the movie is going to be "good" or

"bad"; it's only up to you to do your job as well as you can, and when
you're done, then you can go home. This is exactly the same principle
as the throughline. Understand your specific task, work until it's
done, and then stop.
hiisril on a sriit's of rli (loliiiiihia I ni\ri sil>. On

lakiii*:.
Nru ^

Miililztr IMzr ami llir rs ( jirlr \uar<l. \\v is

(>i iwo 4'<Mlrriions oi (vssa\

I'lmiiiMl, aiK

aii<l ilirtMhMl llir iiion irs Hinisr iif(iaftu's a lliinns i.lmniiv. \\v
li\«'s in Massarhnstlls

JKINC; IM N(;i IN

m?

soiiStitcl. Ntx% Voik. N.V MMMt II IS. A.

This book made available by the Internet Archive.