(In other news, this summer will see a film of Asimov's I, Robot starring Will Smith and apparently not based on Harlan Ellison's screenplay, first published in Asimov's and then as a book.)
The post with the reference to a film of Dhalgren is actually a letter from the screenwriter, who is responding to an excellent post in which Silliman figured out the standard three-act structure of commercial films.
Having studied playwrighting and screenwriting at NYU for three years, this is a subject I know well. Being an inveterate contrarian, I rebelled as much as possible against the idea of a three-act structure, which most of the teachers of screenwriting forced us to abide by, and have since come to resent it beyond all reason. Like any formula, it's perfectly capable of producing very interesting films, but is also, in general, a way for writers to become lazy and not think closely about structure beyond how to chop and sculpt their story to fit into the parameters of the three acts. Great writing doesn't come from fitting something into a formula, it comes from thinking both about structure and content and striving to let the two support and inform each other. Sometimes the effect will be a three-act style, sometimes not.
But the audience becomes accustomed to the style most often presented to it, and so films which violate the three-act structure tend to alienate mass audiences, because the audience members arrive to the theatre like strict Platonists, the universal "good movie" structure stuck in their heads, any violation from it befuddling. Any work both teaches an audience how to read it and helps create an audience for it, and the prevalence of the three-act structure has taught millions of people how to read and interpret films, thus blinding many of them to any other possibilities.
Silliman ends his post about structure with two paragraphs which summarize the problems facing anyone who wants to break away from dominant (and dominating) structures of writing:
One problem that any serious post-avant writing confronts is that, over time, readers come to understand the landmarks to any new terrain. What was comically misidentified in the 1970s becomes instantly recognizable just 25 years later. In order to keep it new, the writer (me or you or whomever) must go beyond the exoskeletal components of structure to create a sense of liveliness internally -- through word choices, sentence juxtapositions, the underlying logic. I obviously have a serious bias towards building in devices -- like the "new sentence" -- that block or at least slow the integration of the text, the point at which it moves from the first of my three mock stages into the moving machinery one. Even as a reader, I am more apt than not to avoid reading the title until the very end of the poem & oftentimes not even then. I’ve gone through entire books without taking note of a title. I simply find them too confining. And I guess that my own titles have a tendency to point anywhere but the text.Isn't the challenge to the SF writer to awaken readers by confronting them with new information -- new imaginings and ways of imagining? Traditional structures, then, may only hamper writers from communicating with readers, because the traditional structure will lull a reader into missing everything important in the writing. (Of course, using traditional structure to play around with readers' expectations can be a lot of fun, too, and one that many SF writers are quite skilled at -- think of so many stories by Fredric Brown which rely on a reader's intuitive grasp of where a story is going and how it has been set up to then create a surprise ending.)
The logic behind all this isn’t newness for the sake of novelty, some sort of attention deficit approach to contemporary meaning, but rather to maximize the reader’s (& my own) attentiveness to detail. That’s what gets lost when a reader gets too comfortable with the landmarks of the poem -- why, for example, it’s so very hard to write a good haiku -- just as it’s what gets lost when you get too familiar with a landscape or city. Slushing around Chicago in the snow last weekend was a great reminder of just how awake one feels when confronted with so much new information.
Or it may be that for the majority of stories, a traditional structure is the one which best serves the tale. What does that say about the stories we want to tell?