Pamela Samuelson played a lead role in voicing academic authors' concerns over the Google Books settlement. That advocacy made an impact: Judge Denny Chin cited her writing in his ruling rejecting Google's deal with authors and publishers, who were represented by the Authors Guild and the Association of American Publishers.
In an interview with The Chronicle on Wednesday, Ms. Samuelson, a copyright expert and professor of law at the University of California at Berkeley, shared her take on what the judge's decision means—and where we go from here.
Q. Is this a good ruling?
A. It's the only ruling really that the judge, I think, could have made. The settlement was so complex, and it was so far-reaching. With the Department of Justice and the governments of France and Germany stridently opposed to the settlement, it seems to me that the judge really didn't have all that much choice. So the ultimate ruling, that the settlement is not fair, reasonable, and adequate to the class, is one that I think was inevitable.
The thing that surprised me about the opinion was that he took seriously the issues about whether the Authors Guild and some of its members had adequately represented the interests of all authors, including academic authors and foreign authors. That was very gratifying because I spent a lot of time crafting letters to the judge saying that academic authors did have different interests. Academic authors, on average, would prefer open access. Whereas the guild and its members, understandably, want to do profit maximization.
Q. So did the ruling address the concerns you raised in general?
A. Yes. The adequacy of representation was one of the key points. I also raised issues of the scope of the settlement in relation to the issue in litigation. Many of the things that the settlement would do are copyright reforms that I think are good. The question is, Can you do this through a class-action settlement? One of the things that was very pleasing to me about the judge's ruling is that the judge also said changes this far-reaching to the default rules of copyright law have to be done through Congress.
The settlement would grant Google about five different licenses that ordinarily, to get that broad a license, you'd have to get it from Congress. It's a license to scan all the books and to store them. A license to make nondisplay uses of them for purposes such as improving search technologies and automated translation tools. It would grant a license for nonprofit researchers to engage in "nonconsumptive" uses—so research uses for academic purposes. It would grant Google a license to give "library digital copies" of the books scanned from library collections back to those libraries and allow the libraries to make certain kinds of uses of the works. And it would give Google a license to commercialize all of the out-of-print books in the corpus. It's really quite extensive.
If Congress was going to grant licenses like this, it wouldn't just grant them to Google. Part of what the Justice Department came to recognize is that the licenses that Google would get from the settlement would create barriers to entry to any other firm, because no one else could get those licenses. That's something that really fed into the antitrust analysis in the case. The settlement would give Google a de facto monopoly over the orphan books [unclaimed works whose copyright owners aren't known or can't be found] that would make a subscription service that it could offer unreachable by any subscription service that anyone else might offer. Google could have millions and millions of books that no one else could reach.
Q. What does the judge's ruling mean for privacy concerns raised about Google Books?
A. He decided that the privacy objections by themselves were not a reason to reject the settlement. The concerns that were raised about privacy issues were nevertheless ones that he thought were very serious. And he indicated that he hoped that any further iteration of a settlement would deal with those issues more seriously.
The way the settlement was drafted, it called for Google to engage in extremely extensive monitoring of access to books. Now you could say that one of the reasons they needed to do that was because, if they're going to pay specific authors for specific books that might be read, let's say, in the institutional-subscription corpus, then they've got to know whose books are being read.
But as we all know, Google basically also wants to know everything that we look at and everything that we read, and they would be engaged in profiling and serving up ads. There were virtually no privacy guarantees for users in the settlement agreement. Efforts to persuade Google to adopt a set of principles were only partially successful, and then they were only willing to say, 'Well, OK, we'll sort of agree to do this.' But they weren't willing to do anything that would bind them.
One of the things the judge noted is these are things they could adopt for a while and then abandon. Libraries have been very, very careful over time about protecting the privacy interests of their user base. And Google was not willing to make commitments to essentially accomplish an equivalent level of protection. When we're talking about a corpus of books that millions of people in the U.S. would be using, not to have any serious privacy commitments here really was distressing.
Q. What does the ruling mean for academic authors?
A. There are a couple of paths that can happen from now going forward. One path is that academic authors can communicate with Google about their interest in making their books available on an open-access basis. That would be something that would allow more of their books to be more widely available.
Second, I'm planning to be working with a group of academics to try to put together a legislative package that would accomplish some of the positive goals that the Google Books settlement raises as possibilities. Much greater access to out-of-print books: I think that goal is really commendable.
A third possibility is that, if this matter goes into litigation, I think academic authors will probably offer support to Google in its fair-use defense, because we are the kind of people who think that if you scan my book in order to index it and make little snippets available, that's actually a good thing. That's going to promote more access to my books, and that's what I want as an academic.
Q. What do you think of the prospects for legislative change?
A. It would require a lot of energy, and a lot of coalition building. But I think that there's some possibility of it, actually. I'm not wildly optimistic about it. There is this amazing vision of access to knowledge that a lot of people are in favor of. If that's true, then we ought to be able to come up with something that would make it all work.
All of the major parties have been in favor of orphan-works legislation. Because of the settlement, for the last two and a half years, that legislation has been on hold. Right now, if a book is an orphan, or you think it's an orphan, you can't make it available to anybody because the copyright owner could come out of the woodwork, and then you could get sued, and statutory damages would be awarded against you, and that would be bad.
The legislation that Congress has been considering, and that the Copyright Office recommended, was that once you make a reasonably diligent effort to locate a copyright owner, then you should be able to use the work if your diligent effort doesn't find that copyright owner. So go ahead and use the work, and make free use of it actually, and if the copyright owner shows up later, then maybe you have to take it down. But statutory damages and other remedies that otherwise would ordinarily apply, would not apply.
I have some tweaking that I want to do to that particular approach, but nevertheless it seems to me that that's more consistent with the utilitarian principles of copyright law than the settlement approach, which would have charged profit-maximizing prices for orphan books through the end of their copyright term, even though there's no copyright owner out there who actually deserves and needs the compensation that Google would be providing.
Q. Who might lead the coalition to push for legislative change?
A. Obviously Google will have an interest in thinking about this. The AAP [Association of American Publishers] and the Authors Guild were in support of orphan-works legislation. Most of the technology companies were in favor of it. Libraries and academics were in favor of it. It's just that it was taking awhile, so Google just said, "Hey, we'll solve the orphan-works problem ourselves."
Q. Could there be a meaningful settlement under the "opt-in" model described by the judge? He said many concerns would be resolved if the settlement made rights-holders opt in, by asking to have their works included in the Google Books project, rather than forcing them to opt out.
A. It's hard to say, because trying to read the tea leaves about what the details would be is very difficult at this point. From the standpoint of the objections of most of the authors—not necessarily academic authors, but other authors—an opt-in regime is actually respectful of copyright. But it doesn't solve the orphan-works problem. All the orphans would be out of the settlement. Then if you want to make it available, you've got to come up with a fair-use argument.
Let's say a library makes a reasonably diligent search for the copyright owners of certain books, and they can't find the copyright owners to get a rights clearance. I can make an argument that making orphan books available for nonprofit educational purposes, after you have reason to believe it's an orphan, is fair use. So while I think it would be better to do this through legislation, I'm not willing to give up on the idea that within the existing framework, there's a way to at least address some of the orphan-work problems.
Q. What about all the research that had been planned for Google Books?
A. That's one of those details that's very important in terms of any revised settlement. The question that the judge didn't address is whether it would need to be an opt-in regime for everything, or only an opt-in regime for the commercialization [which refers to selling ads against the books, selling the books themselves, and putting together an institutional subscription to the corpus]. There are five licenses.
Suppose that Google said, "I'm willing to make it an opt-in regime for the commercialization of the out-of-print books, but I want to be able to scan the books, I want to be able to make nondisplay uses of them, I want to be able to authorize nonconsumptive research, and I want to be able to make library digital copies available to my library partners without any compensation to the rights-holders." So, as to that, it's an opt-out regime; as to commercialization, it's an opt-in regime.
Q. What happens next?
A. They've got two options. One is to go back to litigation. The other is to come up with a new settlement. Now the statements that the publishers and the Authors Guild made are very clear that they want to actually reach another settlement. And the judge is encouraging another settlement.
Whether Google will be willing to settle on different terms is a question that's quite open right now. Before the judge, in February of last year, Google's chief lawyer basically said that this opt-in deal, which the Justice Department was urging, was not acceptable to them, that unless it was an opt-out regime—that is to say that they get to commercialize the books unless the author shows up and says, "Don't do this"—they weren't willing to settle the case. Now that may have been something that they've been willing to say before the judge because they want the judge to feel like he's got to approve it. But they might actually change their mind.