Gripes over Google Books go technical

Remember the Google Books settlement? After a storm of criticism over the initial draft, Google--and the author and publisher groups that initially sued the search giant--withdrew a proposed settlement hours before a scheduled hearing last fall, promising a significant revision.

That revision was announced on November 19. A final fairness hearing on the revised settlement (the Amended Settlement Agreement, or ASA) is scheduled for February 18.

A number of parties have objected to the revised settlement, most notably the U.S. Department of Justice, which filed a "statement of interest" (PDF) on February 4. The department still believes that the judge should not approve the ASA, but the direction and rhetoric of its statement suggests a considerably muted set of complaints.

The bulk of the government's objections are now addressed to the manner in which the deal has been constructed--specifically, the use of class action litigation to break the legal logjam of U.S. copyright law. (The government also continues to raise antitrust "concerns," though these are relegated to a few pages of objections at the end of its filing.)

Before looking at the class action issues, it's important to understand the nature of the copyright problem. There are millions of out-of-print books still protected by copyright, and negotiating separate deals with the rights holders would be complex and expensive. But that's not the most serious issue. As the Justice Department acknowledges, any company hoping to offer digital access to works published in the last century faces the initial, and perhaps impossible, challenge of actually identifying who currently holds the right in the first place. As it notes, "for many works, especially out-of-print works, rights clearance may not be possible as a practical matter."

Source of the problem
What the government agency doesn't mention is that the source of this "orphan works" problem is the government itself. Over the last hundred years, legislators have repeatedly extended the "limited" copyright monopoly and applied those extensions retroactively. But given the short commercial life of most printed books, many rights holders made no provision for inheritance or transfer of the remaining term of their copyrights--terms that may have even been extended after their death. The result is that millions of out-of-print books with no real hope of returning to print have gone into a kind of limbo. They are still protected by copyright, but no one knows who owns the rights.

The orphan works problem is an unintended consequence of changes to copyright law. Largely at the behest of rights holders for in-print works, Congress keeps extending the term of copyright for everyone, inadvertently expanding the population of orphan works. And though legislation to solve the orphan works problem floats around Congress every term, there's never been serious risk of anything passing.

As long as these works remain out-of-print, the orphan problem is only a modest inconvenience. Now that digital technology raises the possibility of a new commercial life for some of these works, however, it has become a hard stop. The rightful heirs to authors and publishers are unknown and largely indeterminate. And even if the rights holders could be identified, written agreements between authors and publishers--which may or may not have indicated who would retain the rights that would now translate to digital publication--are long gone.

'A bridge too far'
The ASA would largely solve the orphan works problem, for which the government believes the parties "should be commended." The Justice Department, however, still won't endorse the solution. Its particular objection now is the use of the class action to fix the broken copyright system. "Despite this worthy goal," the department wrote to the judge, "the United States has reluctantly concluded that use of the class action mechanism in the manner proposed by the ASA is a bridge too far."

To understand why the government objects to this particular legal technique, it's worth remembering how Google got into this quagmire in the first place. Starting in 2004, the company entered an agreement with several major university and research libraries to scan their decaying books, with the intention of making the contents searchable. The scanning project included works protected by copyright, as well as those whose copyrights had expired (no need to negotiate with anyone for these works). For in-print books, Google made separate agreements with the rights holders--in most cases, the publisher rather than the author.

For out-of-print but still copyrighted works (many of which are orphaned), Google planned only to make the works searchable, believing that activity constituted a "fair use" under copyright law--an exception to the rule that copying a work requires permission of the rights holder. The company was nonetheless sued by trade associations of authors and publishers, who argued that the scanning and searching of the works was not a fair use but was, in fact, a massive copyright infringement.

Class actions are typically used in cases like this, in which a large and possibly indeterminate group suffers damages that are too small to make separate lawsuits cost-effective. Here, the class action was brought by the trade groups, who claimed to be standing in for all rights holders affected by Google's scanning and searching activities.

Before the fair-use defense could be tested, however, the parties entered negotiations to settle the case. While there has been considerable objection to whether the trade groups are good representatives of the class, it's hard to imagine better negotiators, short of including everyone involved, which is impossible. In any case, for rights holders who don't agree, or who prefer to negotiate with Google privately or not at all, opting out of the litigation has been offered using a variety of mechanisms, including online.

Besides lingering objections to the adequacy of the class representatives, the Department of Justice objects to the scope of the ASA. Somehow, a case about copyright infringement and fair use turned into an agreement to make millions of works available in digital form. While the government "recognizes that the parties to the ASA are seeking to use the class action mechanism to overcome legal and structural challenges to the emergence of a robust and diverse marketplace for digital books," the government's principal objection now is to a more technical question: whether the "broad" scope of the ASA complies with the theory and practice of federal class action law.

Broad is not necessarily bad. Under the Federal Rules of Civil Procedure and court cases that interpret them, class actions often begin as simple questions of law that evolve into far-reaching settlements. These settlements may clear away longstanding legal problems and even realign major relationships that restructure dysfunctional industries.

But a class action is a kind of hammer, and not every complicated legal problem looks enough like a nail to employ it. Here, the parties have not only gone beyond the issues of the original lawsuit, but they have also crafted a settlement that in some sense legislates an orphan works solution that Congress failed to craft. Is that too much innovation for a class action? The Department of Justice "reluctantly" concludes that it is.

Are they right? The law of class actions is complex and has evolved substantially over the last few decades. There are many details and conflicting judicial decisions that District Judge Danny Chin is obliged to consider. But in principle, I believe that the elegance of the solution to an otherwise unsolvable problem offered by the ASA makes it a good candidate for approval. (Elegant, not perfect--but no agreement involving millions of people could ever be perfect.)

From opt-out to opt-in?
What is disappointing about the Justice Department's filing, however, is the absence of useful guidance on resolving the government's remaining objections. Details are lacking, and in some cases, the suggestions are clearly impractical. For example, the government half-heartedly calls on the parties to change the ASA from an opt-out to an opt-in agreement. That is, the government suggests that rights holders who want to be bound by the terms of the ASA be required affirmatively to sign on (or perhaps--it's unclear--just to the "forward looking" aspects of the ASA).

The opt-in approach, of course, would defeat the most beneficial aspect of the ASA, which is to rescue from oblivion the millions of orphan works Google has already scanned and millions more to come. Since those rights holders are unknown, there's no way for them to opt in. "That, however," the Justice Department says, "is a judgment better suited for legislative consideration, rather than one for courts to make in the context of approving a settlement."

In other words, the solution to orphan works should be left to Congress, the same Congress that created the problem in the first place and seems unlikely ever to solve it. Meanwhile, the availability of orphan works in cheap digital forms is held hostage to the broad desire for control by large entertainment companies that don't ever want to give up government-granted monopolies on their vaults of in-print works. (A 2003 Supreme Court case held that Congress' most recent term extension was still within the constitutional requirement of a "limited" duration for copyright.)

Despite the continued objections from the Department of Justice to the ASA, what's most interesting in last week's filing is the government's shift to more technical objections. Reading between the lines, my sense is that the Department of Justice is considerably less worried about the ASA than it was the original settlement.

How Judge Chin will weigh that concern, and indeed whether the specifics of class action law will determine how he rules on the ASA's approval, remain to be seen.