First of all, you (and the writer of the article) is getting the auteur theory wrong. That’s not unsual: the basic principle as stated by Andrew Sarris has been completely misconstrued for years. Sarris did not say that the director is not automatically the “auteur” of the film; he specifically gave examples of films that had no auteur (e.g, “Casablanca”), and spoke of films where the script was more important than the director (he called these “filmed but not directed”).
What he did say was that great directors put their stamp on their films. The films deal with common and similar themes. For instance, Hitchcock often deals with ordinary men put into extraordinary situations; Orson Welles often deals with extraordinary people in ordinary situations. Howard Hawks deals with competent people. In addition there are technical similarities: Hawks usually films people at eye level; Eisenstein uses montage almost exclusively; Ophuls uses camera movement, and Hitchcock uses both fairly equally.
Also, since film is a visual medium, it’s the visual element that’s foremost. A screenwriter can write the script, but the director determines what it looks like. As an example, consider Hank Quinlan’s entrance in Touch of Evil. You could write that scene (“Quinlan exits the police car. He is extremely fat, mountainous and blubbery”), but it’s the visual that is memorable.
But, you say, Welles wrote the script. That’s different. Many other directors don’t.
That’s a misunderstanding coming from reading the credits. The top directors (Hitchcock, Ford, Hawks) may not be credited with writing the script, but they are involved in picking the material, hiring the writer, accepting the script, and adding their own lines and touches to it on set. There’s the famous story of Raymond Chandler trying to write a script for Strangers on a Train. Chandler said Hitchcock kept telling him how to write it, and how the entire situation was impossible: no one would believe that Bruno had not planned in advance to meet Guy on the train. But, if you see the film, it’s quite clear – all visually – that it’s a chance meeting (they show two pairs of shoes – Bruno’s going first onto the train and sitting down and Guy arriving several minutes later, so it’s clear the meeting was random). But the key element here was that Hitchcock was working with the writer on the script in its formative stages. He never took credit, but he was involved. (Nowadays, nearly all film scripts are rewritten by the director.)
Indeed, one of the better confirmations of the auteur theory was a book a few year ago that tried to do for film writers what Sarris had done for directors in his landmark (and also misunderstood) book The American Cinema. The author charted screenwriters careers, movie by movie in the same way Sarris had done with directors. And you know what? Whenever a top screenwriter worked with a good director, the result was a good movie, and whenever the same screenwriter worked with a weak director, the result was at most, a so-so film.
Yes, film is a collaboration, but who makes the final decision on the set? The cinematographer may decide to shoot a certain way, but it’s the director who ultimately says, “that looks good.” Same with the actors: the director decides which take is the best (and usually tells the actors what he expects). Same for editing: a great editor can put the film together, but the director gets the final decision.
So, ultimately, the director is the one most responsible for the final product. And an auteur director puts a strong stamp of his personality on the film.
But there are auteurs other than the director. Charlie Kaufman is a good example of a writer/auteur. And the one-hit wonders can be directors who aren’t really auteurs (though, not always: Coppola made a bad misstep with One from the Heart and then had trouble getting the control he needed). There are plenty of times where directors are just hired guns with no meaningful input. Someone like Joe Johnston comes to mind – but note Johnston’s October Sky, which clearly shows something of the director’s personality.
(Bob Fosse made a very clever dig at those who think a director is unimportant in All That Jazz: it shows Joe Gideon working obsessively on getting a scene exactly right, and later shows a critic lambasting the film, saving praise for the very scene Gideon keeps recutting and re-editing – because it let the actor perform without the director’s interference.)
But the auteur theory, in its original form, did not say that the director is always the auteur, just that the great directors are.