Risk-averse publishers and copyright chilling effects

By Eric Crampton 16/10/2013


When Sky didn’t like that I’d pointed the National Business Review’s readers to Hola, they sent a note to NBR threatening to get the lawyers involved. And so the NBR pulled the line mentioning Hola. Pretty minor in the grand scheme of things, but irritating. It isn’t illegal to link to Hola, or to mention it. But Sky has deeper pockets than the NBR, so why take it to a court test if you don’t need to?

But what happens when this kind of risk-aversion is systemic? Josh Gans reports on the case of academic publishers and estates. Joseph Thomas has been having problems in getting his biography of Shel Silverstein published. Fair use exclusions allow extensive quoting from published work for scholarly purpose. So you don’t really need to seek an estate’s permission to quote from published work. You’d need permission to access their proprietary archives, but not to copy a song lyric or a stanza from a poem in discussing the author’s work. But that doesn’t matter where a defense could prove costly for a publisher with shallow pockets. Unless they can be pretty sure that they’d win and that the court would award costs, why take the risk? Here’s Gans, from whom I quote extensively:

Thomas had been researching a biography of Silverstein for which the only other treatment is a likely biased account. However, to do justice to Silverstein’s work, Thomas has to quote liberally from him and include pictures and what have you. Academic publishers don’t do the legwork on getting copyright permissions so Thomas had to do that. He was shut down by Silverstein’s estate. Basically, he could not quote anything ever. Thomas then took to the pages of Slate and mentioned the sort of things that Silverstein’s estate might be worried about revealing.
My lawyer friends will tell me that copyright law actually protects Thomas and his publishers here. “Fair use” would likely have allowed pretty much anything that Thomas wanted to quote without him having to ask for permission. But the publishers wanted permission asked for regardless. The problem there was that when Thomas received a ‘no’ the publishers took that as a ‘no’ even though it was just a ‘no permission’ and not a ‘no the law says you can’t do this.’ So Thomas can’t get his book published.
In reality, while it is somewhat ‘not giving’ for Silverstein’s estate to not grant permission one can actually see why they wouldn’t be happy seeing the biography published. But that is precisely the reason we have “fair use” because we recognise that copyright holders may want to block more than just copying. I have been lucky in my books. In Parentonomics, I asked both Malcolm Gladwell and Orson Scott Card permission to use lengthy quotes and they gratefully gave it and wished me luck. The only time I was asked to pay was when I wanted to use a picture of Coke can in our chapter on product differentiation in my adaptation of Mankiw’s Principles of Economics. Coke wanted to charge us for permission. I promptly switched the picture to Pepsi and that picture has been presented to almost 100,000 Australian students now.
The problem this time around is academic publishers or publishers in general and what they are willing to do. They appear willing to let a controversial work not be published rather than feel the ire of a law suit that is likely to fail. Thomas certainly isn’t in a position to fight a legal claim but it is a shame that publishers do not see that as part of their role. I guess when it comes down to it, the returns to publishing certain books are too low to add a legal bill to costs. But all in all it points to a broken system.

Returns to publishers aren’t going to be rising anytime soon. Further, that would be a specific fix to a more general copyright problem: expected penalties for bogus claims are pretty slim. And so automated bots can issue DMCA notices with high false-positive rates and impose costs on ISPs and websites without particular sanction. Were there sanctions for filing false reports, or for bringing spurious copyright claims, we might expect fewer of them. And we might expect publishers to start being willing to take the punt where they think the author is in the right.

Many years ago, I tried getting a project through Canterbury’s Ethics Review Board. They wound up killing the project. Not directly, but after the almost year-long process where they agonized over that our project involved deception, followed by finding out we’d have to go through the same rigamarole at the coauthor’s university, we just abandoned it. But here was the project. Imagine this circa 2005.

We identified a set of academics around the world, some at universities with law schools, some without, some public, some private, who’d be willing to host a copy of Thoreau’s essay On the Duty of Civil Disobedience on their faculty website. We’d contacted some of them, but didn’t see great problems in getting a decent set of these up. We then were going to send out bogus copyright take-down notices to their universities from a newly formed Henry David Thoreau Society which also was to have a website with that work, among others. The 1849 work was clearly in the public domain everywhere. We wanted to see how many universities would cave in immediately, how many would at least have their webmaster get in touch with the named academic, and whether places with law schools as repositories of free legal advice might have been more willing to bear potential legal costs by telling us to get stuffed. And all the other correlates (academic seniority of the faculty member’s website, type of university, legal jurisdiction and so on) would have been pretty interesting too.

I still spit whenever somebody says Ethics Review Board. I mostly now just avoid doing anything that needs anybody’s approval. Who needs the hassle? So I can get why publishers might want to avoid copyright court hassles. But the net effect is pretty cruddy.

We need more judges like Otis Wright. Explicit penalties for folks using bogus legal threats to chill protected speech could be helpful.