Glass half empty, authorship all there

I've seen several posts about the NYT story on glass artist Chihuly suing others, including a former collaborator, for copyright infringement. Among the issues, which also include the copyrightability of certain styles, is the fraught question of who is an author when the person whose name is on a work does not execute the work, and indeed leaves many decisions up to the person performing the physical acts of creation. There's an obvious feminist take here about the mind/body distinction.

Michael Madison suggests joint authorship, at least when Chihuly gave his collaborators lots of leeway by writing things like "Here's a little sketch but make whatever you want." I'm not sanguine about the chances of joint authorship, which in the 9th Circuit (among others) requires that the person who was generally considered the author before the dispute erupted to have intended to treat his collaborator as a joint author. And the reason that Chihuly's name is on the works -- he's the core of the business, the author-figure on whom a large industry is founded -- is good reason to believe he didn't intend that, no matter what the collaborator thought. If Chihuly contributed any copyrightable expression at all, he will probably get 100% ownership.

What if he didn't contribute copyrightable expression, only general ideas? The works might still be works for hire, or he might have a license of some sort. Let's assume, as a thought experiment, that Chihuly's assistant creates a work for hire; copyright then vests initially in the hiring party. What happens when the assistant leaves Chihuly's employment and creates a new work that is substantially similar to the old one?

It seems to me we have a variant on the underlying dispute in Fogerty v. Fantasy, where the jury decided that John Fogerty didn't infringe Run Through the Jungle with the similar song The Old Man Down the Road. I've long thought the best way to explain that case -- and perhaps this is a way to understand the old Maltese Falcon case as well -- is to say that any substantial similarity isn't the result of the artist copying his earlier work, but of both works springing from some ur-work that isn't fixed but only exists, like a deeper well, in the artist's mind. Or at least, to preserve authors' freedom to transfer copyright in specific works but still continue to use a distinctive style, we have to treat them as people who are not copying their own prior works.

The twist in my Chihuly hypothetical is that, from the law's perspective, the assistant was never the "author" of the first work, unlike Fogerty, who sold the rights his initial authorship gave him. Should that make a difference? Does the author have a special access to some Platonic ideal of an ur-text? If so, is "work for hire" a fundamentally illegitimate concept? On the other side, what are we to make of authors whose audiences believe that later works in a series show poor characterization -- that the Vampire Lestat they know wouldn't do what Anne Rice says he did? If it's not infringing for Anne Rice to write a sequel to her vampire books for a new publisher, why is it infringing for someone else to do so?