Does the Google Books Settlement set the stage for a congressional intervention?
Last month, in a decision by Denny Chin, a New York federal district court threw out a settlement negotiated between Google and representatives of book authors and publishers, who had sued to stop Google from scanning and indexing published books to return as search results. New York Law School professor James Grimmelmann's been following the case closely, maintaining a comprehensive website tracking the litigation, writing extensively about it, and hosting a symposium about it. We talked to him about the future of the Google Books litigation, the possible role of Congress in the ongoing relationship between Google and book publishers and authors, and the notion of the Great American Digital Library.
William McGeveran: Judge Chin has taken more than a year to write this decision; were there any surprises in it?
James Grimmelmann: At first, its length and tone. I'd been expecting a treatise on copyright, antitrust, and civil procedure. My students and I identified 76 different issues raised by objectors. When it took Judge Chin over a year to issue the opinion, I assumed it was because he was drafting a doorstop.
But no. It's 48 pages, and that's in double-spaced Courier, which makes it, what, 10 normal pages? He writes only briefly on most of the major issues. He cites very few cases; this is not an exhaustive analysis of each and every thrust and parry. He's also unafraid to leave issues unresolved. He takes a strong position that he doesn't have the power to approve forward-looking settlements like this, but when it comes to the meaty copyright, antitrust, and privacy issues—on which so much ink has been spilled—he doesn't give definitive answers.
That makes it sound like a weak, waffling opinion. But the more I think about it, the savvier I think it is. He didn't destroy the settlement; he just quietly deflated it. He resolved a one-of-a-kind case in a way that doesn't warp the legal system for anyone else, he gave the parties a reasonable way forward with a much less ambitious settlement, and he let the many people who've been paying close attention know that their voices have been heard.
McGeveran: Well, the settlement got so far away from the original issue in the case, and so much more ambitious. To start, Google was digitizing books solely to allow it to return "snippets"—small excerpts of text—in response to search queries. That’s why the authors and publishers sued. But the proposed settlement contemplated Google becoming a digital publisher of last resort for out-of-print books, and forcing copyright holders to opt out if they didn’t want that. As you say, Judge Chin “quietly deflated” that broader vision of online books, but the Super Card Catalogue that Google tried for in the first place would still be an extraordinary boon.
Grimmelmann: Search engines radically improve access to information. If you want to know something, then finding out where you can learn about it is half the battle. I can understand copyright owners' complaints when the search engine substitutes for actually consulting the original; a search engine that gives its users 10-page excerpts for free would be devastating to short-story writers. But it's scandalous to think that copyright owners might have a right to hold up the creation of good indices. They should be beating down Google's doors demanding to be indexed, and the sooner the better. Of all of the complaints that could be hurled against search engines, I think this one is near the bottom of the list.
McGeveran: Maybe the parties will go back to that narrower fight now, and maybe Google will win. I always thought Google had a decent argument under existing copyright law, specifically the fair use doctrine, that copying for the purpose of indexing is legal.
Grimmelmann: I think cases like Perfect 10 and Kelly, allowing search engines to display thumbnail images, provide a good legal foundation, but I wouldn't say those precedents are firmly embedded in our legal landscape yet. Some people claim that you can presume implied consent to index for something that was posted online, but not for books.
For me, the third factor listed in the statute is a pretty strong argument in favor of fair use: “the amount and substantiality of the portion used in relation to the copyrighted work as a whole.” The portion shown to the user is tiny in relation to the whole book, which cuts in favor of fair use.
McGeveran: Of course, Google might agree with us that the fair use defense for snippets is strong (although not a slam dunk), but also might not want to go through the cost and delay and uncertainty of litigating. As you say, Judge Chin leaves the door open for a less sweeping settlement based on copyright holders opting in to a book-selling arrangement with Google rather than requiring them to opt out of it. But the instant analysis after the ruling all seemed to reflect a conventional wisdom that Google won't make that more modest deal.
Grimmelmann: I have to say, the conventional wisdom seems hung up on the idea that "opt in" must mean this exact settlement, but with everything switched from an opt-out to an opt-in basis. That precise outcome is pretty unlikely. This settlement was designed to rest on an opt-out base. Inverting that exact structure is like balancing a pyramid on its point.
But that doesn't mean that some settlement with an opt-in approach to selling complete books is hard to engineer, or unlikely. Here's how something like that might work. First, the parties would start with the conduct at stake in the original lawsuit—Google's scanning and searching—and settle that, on the usual opt-out basis, in exchange for some compensation. Then, Google offers authors and publishers the book sales and the institutional subscription, if they choose to participate. That could be accomplished through Google's existing Partner Program, or it could be through a settlement, but either way, it would be an opt-in offer.
McGeveran: And I suppose that would get around many of the procedural issues that the court really focused on (in contrast to any substantive issues like copyright or antitrust law).
Grimmelmann: An ordinary judgment is backwards-looking because it releases the plaintiff's claims based on things the defendant has already done. You punched me yesterday, I sue you today, the court awards me some money against you. If you hit me with your car tomorrow, that's not part of the judgment. The proposed settlement here was essentially forward-looking: it dealt with Google's rights to sell class members' books tomorrow, and next year, and far into the future.
What's more, Google hasn't sold books in this way in the past; it's just been scanning and making an index of the books. Indeed, Google wouldn't try to sell books in this way unless it had the settlement. It's not even a settlement based on the continuation of Google's past conduct. That kind of class-action settlement—one that revolves on the plaintiffs giving up future rights to sue over things the defendant hasn't done yet—is very unusual.
A settlement doesn't make or declare law; it's much closer to a contract between the parties, one of whose terms involves dismissing the case without resolving it. (If you think about using a class action to force millions of people sign a contract, you can see why this settlement would have been worrisome.) Also, the court can't just invent law when there are statutes and regulations in a field of law: the common-law-making power ends where Congress has spoken. And Congress has spoken, a lot, about copyright and many of the other issues in the case.
McGeveran: The opinion emphasizes that a lot. At one point the court says the dispute concerned "matters more appropriately decided by Congress than through an agreement among private, self-interested parties."
Grimmelmann: Well, first, the settlement would have involved truly immense numbers of people: millions of authors, many thousands of publishes, and eventually, everyone in the U.S. who reads. Second, it would have rearranged legal rights going forward. It wasn't a matter of past compensation for a one-time harm that's over and done with. It would have set up a bookstore to last for a century or more. Third, it's an immensely complicated deal, one with a lot of moving parts. And fourth, it would have pushed on the basic policies underneath a lot of different areas of law, such as copyright and antitrust.
Just as a general matter, anything with all of those characteristics sounds like a job for Congress. In a democracy, that's how it's supposed to work: the elected legislature makes broad general rules going forward, and the courts have a narrower role dealing with the application of those rules to specific cases looking backwards. The exact same settlement, while still controversial, would have been much less troubling if it came through Congress, because there's both greater expertise in hashing out compromises and greater accountability.
Of course, our current Congress doesn't exactly inspire a lot of confidence, but that's not a reason to give up on the basic tenets of representative democracy and separation of powers. If we stop trying to make the system work, that's when we're doomed.
McGeveran: In light of all that, and the fact that it’s been about three years of negotiating the settlement and then waiting for the court ruling, I can see Google trying to get Congress to intervene now. A change in copyright law could solve a lot of their problems. Congress could clearly state that digitizing and indexing are allowed. At least for "orphan" works (those with unknown rightsholders—by some estimates, half of all books covered by copyright), Congress could even create a new rule giving everyone (not just Google) the right to copy them. That would sidestep the concern of the Justice Department and potential competitors like Amazon that the proposed settlement gave Google an unfair advantage, possibly even one that violated antitrust law. Everyone would get the same deal. After that, as you say, Google could go ahead with a new arrangement to offer an opt-in to known rightsholders. And finally, the democratic objections you raise would go away too if change came from Congress rather than the courts. You argued almost two years ago that Google should have gone to Congress—shouldn't they go now?
Grimmelmann: Oh, I agree completely. I originally thought that if the settlement were expanded to permit competitors to come in on the same terms, that would solve many of the issues, but as I thought about the class-action issues, I came to the conclusion that institutionally, this is the kind of job for Congress, not the courts. My next article concludes that almost all of the controversies around the settlement arise because of the way it uses a class action to concentrate power in Google. Legislation that didn't force everything through a Google bottleneck would immediately undo many of those concerns.
McGeveran: So, what it looks like we've got coming now is: maybe another attempt at settlement, maybe more complex litigation over the unresolved fair-use issues, and maybe a new copyright law passed by Congress. Most likely some combination.
Grimmelmann: I think the most likely outcome is a new settlement limited to scanning and searching, and Google continues to make its Partner Program available for anyone who opts in. And I think we will see a push for orphan works legislation limited to books, but I don’t know exactly what that looks like and neither do its supporters.
McGeveran: Limited to books because photographers have been the big opponents of orphan works proposals in the past?
Grimmelmann: Right. And I think Google would be happy to leave images aside in order to get the content they really want. Meanwhile the library community—including people like Robert Darnton and his proposed National Digital Library—would be happy enough with that too. They have enough to do with digitizing printed matter like books and pamphlets to keep them busy for a decade at least.
McGeveran: I suppose the devil would really be in the details here. The law would have to say how diligently you need to search for a rightsholder before declaring the book “orphaned” and a host of other issues. Do the people who would support this idea have consensus on any of that?
Grimmelmann: Oh no. And I have a feeling Google is only interested if the diligence requirement is somewhat minimal, otherwise it would be hard to scale.
McGeveran: And the congressional action?
Grimmelmann: Well, let’s just say, not in this session. If you are debating whether to shut the government down, I imagine orphan books fall pretty far down the list.
McGeveran: Combine that with the complexity of another settlement and I guess it’s probably going to be a long time before either the Super Card Catalogue or the Giant Digital Library becomes a reality.