Friday, November 30, 2007

Against attribution rights

I had the privilege earlier this week of visiting DePaul as a Hosier Distinguished Scholar. I gave a talk about attribution rights, the subject of a forthcoming piece in the Utah Law Review – I had intended to write a proposal to grant some sort of attribution right, but the more I wrote, the less well it worked, so I ended up writing an argument against legal enforcement of attribution norms.

Here are some notes from my talk:

I frame the question as one of when law should intervene to enforce valid social and moral norms. This makes me think about the heartbalm torts, including alienation of affection, breach of promise to marry, and criminal conversation. These were largely eliminated in the first half of the 20th century as legislatures and courts decided that human relationships were too complicated to reduce to write and wrong, and that the torts were being abused by the undeserving – gold-diggers, in that case. These two arguments are in some tension, but together proved persuasive in the decline of the hearbalm torts. I make pretty much the same two arguments against moral rights, and you could fairly question me on that.

The heartbalm torts offer a story of the de-legalization of morality. Another possible analogy might be employment at will, where we have moved from near-complete legal freedom to much less freedom; this can be described as the legalization of morality. Moral rights might be like that – proponents argue that attribution rights should be added to what authors can get through contract.

I identify three sets of arguments for attribution rights. First, there are arguments from authorial high-protectionists. The object of concern: authors. Potential bad actors: owners to whom copyright has been transferred and follow-on authors who may distort/violate a text. Goal: protect the author’s special relationship to the text.

Second, copyright low-protectionists. The objects of concern: audiences and works, not authors. We want to feed authors the minimum possible to encourage them to run in their cages, generating new works. Potential bad actors: copyright owners, who may suppress access and/or extract monopoly rents. Goal: Access. Copyright low-protectionism helps authors insofar as they seek dissemination of their works and credit for them, not control. Low-protectionism also sometimes encompasses compulsory licensing, providing compensation but not control.

Third, trademark-style consumer protectionists. The object of concern: consumers, not in their special role as audiences, but simply as market participants. Potential bad actors: Deceptive marketers, whether initial authors, copyright owners, or non-owner publishers, who may induce consumers to spend time or money inconsistent with their actual preferences. Goal: Enabling rational, undeceived choice.

Can an attribution right be defended on these grounds? I conclude that the second, copyright low-protectionism, is empty – it endorses attribution as a substitute for things it doesn’t like nearly as much, especially control. Copyright low-protectionists are author’s rights advocates, but (once a minimum of copyright protection exists sufficient to incentivize creation) only when authors’ rights don’t interfere with access.

For the others, attribution rights founder in two places: First, the separation of ownership and authorship in the American system. Because copyright is transferrable, and because so many significant works are works for hire, attribution would either create huge practical problems – the same problems that pop up in every joint authorship case – or would have to be limited to a subset of works that are more likely to have single identifiable authors. The practical problems of attribution in multiple-creator works exist even in highly developed and concentrated fields like Hollywood moviemaking, in which 1/3 of films go to arbitration about proper screenwriting credit and everyone involved ends up feeling cheated. Legal rights would just increase the number of battlefields; if experts can’t make the system work predictably after decades, juries and judges are unlikely to do better (and it’s no surprise that a case-by-case system is unpredictable, just like substantial similarity determinations). Among the quirks of creative production is that people routinely disagree about how much they contributed to a work or a project; ask any group, and you’ll get estimates of contribution that add to well over 100%.

The limiting-works solution, by contrast, is bad because it would further entrench the devaluation of joint work and creative production in fields like moviemaking. That’s a theoretical objection, but I think it’s one that ought to resonate with authors’ rights advocates, and of course consumer protectionists are generally thinking about deception in precisely the types of mass-market works that are often works for hire, like (Stephen King’s) Lawnmower Man, so a work-limited right isn’t all that helpful.

Second, any attribution right that doesn’t allow suppression of a work is going to be ineffective a lot of the time. You can’t make people notice attribution (at least not without destroying a bunch of aesthetic value) and you can’t make them care. This is an obvious problem for consumer protectionists, and also I think a serious problem for authors’ rights advocates. If there is an important connection between author and text that ought to be preserved, ineffective attribution doesn’t further the artist’s right, which is one that goes to interpretation by the audience.

For more, please read the paper!

No comments: