From DB:

Every so often we’re told that the real “author” of a film is the screenwriter. What do we make of this?

I think I first heard this idea in 1966. It was already old then. Publishers had for some time brought out books of screenplays, treating them as literature (e.g., the James Agee scripts) or as the ur-form of the film in the days before home video (John Gassner’s Twenty Best Film Plays). Responding to Andrew Sarris’ auteur theory, Richard Corliss wrote The Hollywood Screenwriters in 1972. The book was organized in a fashion parallel to Sarris’ trailblazing The American Cinema, ranking screenwriters and suggesting their recurring thematic concerns. But it had nothing like the impact of Sarris’s account of directorial differences. Once Corliss became a critic for Time, he did what everybody else did: attributed the goodness or badness of a film primarily to the director and barely mentioned the screenwriter.

Ever since then, the notion has resurfaced periodically, usually with an air of thunderous discovery. David Kipen subtitled his book The Schreiber Theory boldly, too boldly: A Radical Rewrite of American Film History. The idea is recycled in Joe Eszterhas’s new book The Devil’s Guide to Hollywood: The Screenwriter as God! (St. Martin’s Press). Beyond the scurrilous gossip and thumbnail advice he offers, Eszterhas claims that the screenwriter is the key creative individual in movie making. He denounces the auteur theory, which assigns artistic control primarily to the director, and he constantly asserts that no great movie was ever made without a great screenplay.

Eszterhas does grant that when a director writes the script him- or herself, that justifies the auteur theory. But when the director works from a script by another hand, the prime mover isn’t in doubt.

You’re the storyteller, not the director. . . . If anybody in your presence refers to the director as a “storyteller,” deck him (or her).

In musical terms, you are the composer; the director conducts the orchestra.

This idea poses a lot of problems.

Many films need no script. Consider Ballet Mécanique, Oskar Fischinger’s work, Tony Conrad’s Flicker, Ernie Gehr’s Serene Velocity, Andy Warhol’s Haircut, Lewis Klahr’s Two Minutes to Zero Trilogy, and many, many others in the avant-garde or experimental filmmaking tradition. Eszterhas’s claim would apply only to fictional mass market cinema—that is, the movies at the multiplex. But even then there are difficulties.

First, in principle we could have a mainstream movie that was completely unscripted. Maybe the directors and actors and crew cooked up, day by day, a series of scenes in which the dialogue was improvised. Maybe they shot at a 100:1 ratio and carved out of the footage a recognizable, ordinary film. We’d then have something that we’d accept as a passable movie without there ever having been a script.

Granted, this is very unlikely to happen, and it’s probably even more unlikely to result in a good movie. Perhaps the closest existing example is the working method of Wong Kar-wai, who writes his scenes day by day and shoots them more or less off the cuff. Gradually, he compiles enough material to assemble a film. The practice can cause anxieties for his collaborators, and there’s certainly a lot of wasted time and footage, but the results certainly pass muster. Even if you don’t buy the example, it remains possible in principle to make an acceptable mass-market film without a script.

There are more complicated theoretical issues too.

If we all agree that a movie consists of images and sounds, not words on a page, then Eszterhas’ analogy with music fails. In the Western concert tradition, the score captures key features of the identity of a piece. A musical score is written in a more stringent notational system than the language of screenplays. There isn’t a one-to-one relationship between what’s written in the script and what shows up on the screen. The sounds and moving images aren’t specified by the screenplay to the degree that performances of Sibelius’ Fifth Symphony are specified by the score.

The screenwriter writes: “An office.” But the director and the cinematographer and many others must specify how that office looks, the style of the furniture, the way light strikes the surfaces, the position from which the camera sees the action, and hundreds of other factors that shape the image. No one could write all these details into a screenplay. Even if he or she tried, any item could always be construed in a different way when the image is created. Is the desk the shade of brown that the screenwriter envisioned? Are the lampshades the same size?

No prose description of any chunk of reality can uniquely specify the layout of objects and all the patterns of light, color, and shape. By the same logic, no screenplay can exhaustively pre-plan the movie image.

What about dialogue? Much the same holds true as for the image. Even if an actor says the writer’s line word for word as written, it can be delivered in many different ways. And of course facial expression, stance, gesture, and all the other physical components of performance are visual, and can’t be singularly specified any more than our office setting.

Here’s a thought experiment. Give four directors the same script. I think that we intuitively believe that the result would be four substantially different movies, even if all scenes were retained, no new ones were added, and not a line of dialogue was altered. Consider as a rough but intriguing example Michael Mann’s TV movie LA Takedown (1989) redone as Heat (1995). There are enough similarities to allow us to see how different two realizations of a scripted scene can be, even in the hands of the same director.

So the words on the page may be transmuted into images and sounds in an indefinite number of ways. Mass-market film production is a process of constant revision: the shooting transforms the script, and then editing and sound work transform the footage.

Someone could reply that the film script is more like the script for a theatre piece. A play can be realized in many different versions, or productions, but we consider the playwright the true author. And we do call a script a screenplay, a play for the screen. So forget the analogy to music. Maybe the film director is like the stage director, and the screenwriter is the true author. We talk of Shakespeare’s Macbeth in, say, Jonathan Miller’s production, so why don’t we speak of Joe Eszterhas’s Showgirls in Paul Verhoeven’s production?

Again, though, we don’t need a script to put on a theatre piece. There are improvised plays, standup comedy routines that grew organically without ever being committed to paper, and commedia dell’arte performances that rely on tradition, not playscripts. Further, for many traditions, the unique identity of a play rests in its text, which must be followed more or less strictly. You can stage Macbeth on the flying trapeze or underwater, and it’s still Macbeth. But if you rewrite the dialogue or add scenes and characters, you’re not doing Shakespeare’s Macbeth any more. Staging a play doesn’t revise the script, as filmmaking does, but rather embodies it.

A movie script can’t uniquely identify what image or noise or passage of music should be presented on the screen. A play script can’t determine every detail of production either, but in many theatrical traditions the script has become the sine qua non of the production. What is inviolate is the text, the speeches; the director has some leeway in handling the stage directions.

A screenplay isn’t a blueprint or road map, let alone something like a musical score. It’s not even as integral as a play text. At best it’s an approximation, a set of suggestions. To borrow from Pirates of the Caribbean: It’s more in the nature of guidelines, really.

Making a film in the mass-market cinema is a series of constant revisions and rethinkings. One script may be revised by another screenwriter, and the result of that may be revised by a team of other writers. By the time The Script reaches the shooting phase of production, it’s often already a collective product. The revision process continues right to the end. Anything in the shooting script may be tossed out if something more fitting is discovered in the course of shooting, editing, and sound design.

In many cases the person supervising such discoveries is the director. This isn’t to deny writers an important role in the creation of a film. All participants contribute something, and scriptwriters contribute a lot. But it seems that in movies made the Hollywood way, the role of the director is in principle the crucial one in coordinating all the inputs and creating something coherent.

In practice, of course, the role may be filled by many individuals. For any particular case, the director may not have the ability or authority to fill the role, and other individuals can assume it. In such cases, we could say that the producer took over when the director left the project, or the star grabbed control of the scene from the director, and so on.

In fact, we do tend to say such things. Which suggests that deep down, there’s a little bit of auteurist in all of us.