"You are at least watching here a filmmaker with a vision, with a style, making bold choices. I'd rather watch that any day of the week than something else that maybe hits more conventionally satisfying notes. "
-- Adam Kempenaar, Filmspotting #436 in 2013
Putting auteurism in a nutshell
"The second premise of the auteur theory is the distinguishable personality of the director as a criterion of value. Over a group of films, a director must exhibit certain recurrent characteristics of style, which serve as his signature. The way a film looks and moves should have some relationship to the way a director thinks and feels. This is an area where American directors are generally superior to foreign directors. Because so much of the American cinema is commissioned, a director is forced to express his personality through the visual treatment of material rather than through the literary content of the material."
-- Andrew Sarris, Notes on the Auteur Theory in 1962
"Paradoxically, however, the personalities of modern directors are often more obscure than those of classical directors who were encumbered with all sorts of narrative and dramatic machinery. The classical cinema was more functional than the modern cinema. It knew its audience and their expectations, but it often provided something extra. This something extra is the concern of the auteur theory."
-- Andrew Sarris, The American Cinema in 1968
Criticisms of the Auteur Theory, or Auteurism, of which there are legion, tend to make a few simple mistakes, of which this is certainly not a complete list.
1. They get the causality backwards. For example, this formulation from the wikipedia entry:
Auteur theory holds that a director's film reflects the director's personal creative vision, as if they were the primary "auteur" (the French word for "author"). In spite of—and sometimes even because of—the production of the film as part of an industrial process, the auteur's creative voice is distinct enough to shine through all kinds of studio interference and through the collective process.
The correct formulation should be: Auteur theory holds that if the director's personal creative vision is distinct enough to shine through studio interference and the collective process, then that director can be considered an auteur.
This simple misunderstanding explains the logic behind the "what about screenwriters?" objection. Not only is auteurism not necessarily confined to directors, it quite often recognizes the work of producers (Selznik, Thalberg, Bruckheimer), screenwriters (Hecht, Goldman, Kauffman), actors (Lloyd, Marxes), even production designers (Menzies). Jonathan Rosenbaum even posits four different auteurs for Taxi Driver, the director, the screenwriter, the star and the film's composer. The point is not that the director necessarily is the source of a given film's creative vision, it's that when a film does show evidence of a personal creative vision, often, but not always, the source of that vision will be the film's director.
2. They assume the theory is a definitive statement.
Auteurism is an analytical approach to film history. It's not the only one, and it's not the only interesting or valuable one. Because an auteur's personal creative vision can often be obscured by the collaborative process, commercial or generic demands, studio interference, or various other noise, the best way to find evidence of the presence of an auteur is to watch as many of their films as possible. Auteurism is inductive, always in search of more evidence and never satisfied.
To make a simplified example: An auteurist does not deductively assert "Johnnie To is an auteur. Johnnie To directed this set of movies. Therefore everything about these movies reflects the personal creative vision of Johnnie To." Instead, an auteurist takes the set of movies and compares all the elements within them. Say there are nine films, all directed by Johnnie To. An auteurist would note that Set One is three films co-written with Wai Ka-fai, Set Two is three collaborations with director/choreographer Ching Siu-tung, and Set Three is three written and directed by To himself. Johnnie To's personal creative vision would be found not just in Set Three, but in certain elements found in Set Three that also pop up in the films within Sets One and Two, whereas the stamps of the other two potential auteurs could be inferred from the absence of certain elements in the sets of films they were not involved in. If cartoonish stunt-work only appears in the Ching Siu-tung films, then that is evidence for Ching's auteurist signature. If complex plot twists leading to a spiritual epiphany occur only in the Wai Ka-fai films, then that is evidence for Wai's auteurist signature. And if character doubling, game-playing and images marked by bright white lights within dark shadowy spaces occur in all three films, then that might be evidence for Johnnie To's personal creative vision.
3. They assert that the Theory commits the intentional fallacy.
This inductive approach is how auteurism avoids the intentional fallacy. It proceeds first from the evidence of the film(s) to develop a theory of auteurial personality, not from a theory of auteurial personality to analysis of the film. It is impossible to truly know what is in the mind of anyone else, so intention is irrelevant. It doesn't matter why Kenji Mizoguchi uses so many long takes, why he thinks he uses so many long takes or why he says he uses so many long takes. It's the role of the film critic to come up with a theory of what, if anything, Mizoguchi's long takes mean, and what effect, if any, they have on the film and/or on the viewer. And it takes an auteurist to note that Mizoguchi uses a lot of long takes in the first place.
These studies can be exhaustive and exhausting. For an example of the kind of evidence that can be accumulated through studying an auteur's career, check out the website maintained by Mike Grost. Here's his page on Raoul Walsh. This is the raw material of auteurism, not speculative psychologizing of personal biography.
4. They say it amounts to snobbery. This is a two-pronged objection.
A. It elevates the art house above the mainstream.
Sometimes this may be the case, but this is hardly the necessary consequence of auteurism. It's an easily refuted objection, given that the original auteurists were denigrated as "Hitchcocko-Hawksians" for their elevation of mainstream genre filmmakers like Howard Hawks and Alfred Hitchcock to the status of great artists, to the same level, or higher, than prestige filmmakers like Ingmar Bergman or Stanley Kramer. This is how Vertigo, a film received, when it wasn't panned outright, as a mainstream genre picture of little interest to "serious" filmgoers, over 50 years came to be recognized as one of the greatest films ever made (and is now enshrined as such for the next 10 years by the Sight & Sound poll). The purpose of the theory is to discover artistry within the mainstream or without, whether high-, middle- or low-brow. This is why the name of the recent movement Vulgar Auteurism strikes me as redundant. By seeking personal creative visions within the works of mainstream action and genre cinema, these critics aren't creating a new, 'vulgar' form of the theory, they're just being auteurists.
B. It is esoteric and obscurantist
It is true that an auteurist will often value the lesser well-known, less "successful" films of a auteur's career. This is not necessarily out of belligerence, but rather because it is often in these lesser works that the auteur's personal creative vision becomes more evident, and because autuerism takes it for granted that demonstrating a personal creative vision is a value in and of itself. Thus can a film fail to meet all the conventional standards of "success" and still have value if it shows that personal vision. Which is exactly the point that Mr. Kempenaar made at the beginning of this post.