It was a tough summer for the Google book search settlement. The proposed deal, which would give Google Inc. the right to scan and display millions of copyrighted books, was lambasted as anticompetitive by rivals (Microsoft Corporation, Amazon Inc., and Yahoo! Inc.), the head of the U.S. Copyright Office, and the governments of Germany and France (to name just a few). In September the U.S. Department of Justice urged the federal district court judge hearing the case to reject the settlement as constructed, because of its potential violations of antitrust, copyright, and class action law. Even digital rights advocates, who usually cheerlead for new open forms of information dissemination, gave it hell. Brewster Kahle, founder and director of the nonprofit Internet Archive in San Francisco, called the proposed settlement “a new and unsettling form of media consolidation.”

Despite this opposition, a deal still seems likely. The Justice Department could have effectively killed the agreement, but instead tempered its criticism of the settlement with praise for its aspiration, and even suggested possible fixes for some of the problems it identified. As of press time, lawyers for litigants Google, The Authors Guild, and the Association of American Publishers were scrambling to revise the settlement to address the Justice Department’s concerns, and Manhattan federal district court judge Denny Chin had agreed to indefinitely postpone a scheduled October fairness hearing to give negotiators time to make changes.

Given the scope, size, and sheer audacity of the proposed agreement — Google stands to emerge as the biggest library and bookseller in the world — the controversy doesn’t come as much of a surprise. Lawyers who have studied the terms say it is the most creative use of the class action mechanism they have ever seen.

If replicated, some say, it could change how businesses use class action law to resolve disputes, and could even lead to collectivized solutions to copyright disputes in other media — music, photos, and journalism, for example. Fred von Lohmann, a lawyer for the digital rights group the Electronic Frontier Foundation, says the proposed settlement may encourage stakeholders to “stop worrying about control, and to start worrying about remuneration.”

Google book search began in the fall of 2004 as Google Print, a limited program that would scan and make some in-copyright books available to readers through a new search engine. Google quickly expanded the project, striking deals with the University of Michigan Library, the Harvard University Library, and the New York Public Library to scan and digitize their entire book collections. In exchange, Google promised to turn over a digital copy of each library’s book holdings to that institution. In the next decade, Google said, it hoped to accumulate an archive of 15 million volumes, both in and out of copyright, searchable with a keystroke on its ubiquitous platform. (By way of comparison, the Library of Congress, which boasts the biggest book collection in the world, houses about 32 million volumes.)

The two major professional associations in the publishing world, the Authors Guild and the Association of American Publishers (AAP), immediately protested, and demanded that Google stop its scanning operation. The scanning violated copyright law, they said. Google briefly paused its project, but in the summer of 2005, Eric Schmidt, Google’s chief executive officer, told book executives at a meeting that Google didn’t need their permission. While the entire database of books would be searchable, Google would limit the amount of text that an end user would see to a few lines at a time, he said. This, he said, was allowable under the “fair use” provisions of copyright law.

The line in the sand was drawn. In September 2005 the Authors Guild filed a class action lawsuit against Google in federal court, soon joined by five major publishers and the AAP. Copyright lawyers and digital rights advocates licked their lips. Here, at last, was a showdown between motivated and well-funded parties that could help settle a murky area of copyright law: What, exactly, constitutes fair use in the digital realm?

But a curious thing happened on the way to the courthouse. Instead of litigating, the parties began talking about how a broad licensing agreement might work. Allan Adler, vice president of legal affairs at the AAP, says his organization quickly realized that “a judicial resolution of this suit wasn’t going to take us very far.” If the authors and publishers were to win in court, a tantalizing new possibility — the reincarnation of millions of out-of-print, mostly forgotten books in digital form — wouldn’t happen. If Google were to win, it would still be barred from doing more than scanning and displaying snippets of text. Here was a chance, the parties saw, to craft a licensing deal that would give Google the rights to scan, display, and sell millions of unavailable copyrighted works, with most of the profits returning to the copyright holders.

There was another motivation for the settlement. Lawyers familiar with the talks say the book publishing industry had watched in horror as the music business waged a scorched-earth campaign against file-sharing sites like Napster, only to see their profits plunge and antipathy to their tactics grow. They didn’t want to follow the same path.

In the spring of 2006, executives and lawyers began e-mailing various proposals about how a comprehensive settlement might work, say lawyers familiar with the negotiations. The authors were most interested in getting paid for their out-of-print works. The publishers, meanwhile, wanted to ensure nothing could be done with in-print books without their permission. Google wanted a deal that would incorporate the most troublesome class of books: in-copyright, out-of-print books, for which the rights holders cannot be determined. These so-called orphan books encompass works that were published anonymously, or, more commonly, where the author has died and the rights have been passed on to an unknown heir. (No one knows exactly how many in-copyright books are “orphans” in the United States. Some estimates peg the number at about 1 million.)

In October 2008, after more than two years of talks, mostly held at Google’s New York headquarters, the parties to the lawsuit inked an agreement that would give Google permission to continue with its scanning project. In exchange, Google agreed to pay $120 million to authors whose books have already been scanned, to pay the other side’s lawyers (a hefty $45 million), and to set up an independent book rights registry, which would track rights and distribute revenue from Google ad sales. (Google would receive 37 percent of ad revenue, and rights holders, 63 percent.) Under the proposed deal, Google can also display up to 20 percent of the out-of-print books it scans — much more than the snippets it originally proposed — and may also sell these books for download, with a price set by the page. (Readers will still have to go to traditional retailers for bound copies.) Proceeds from orphan-works sales will be held in escrow for five years, and then that money, too, will go to the registry.

Google’s primary negotiators were in-house lawyers David Drummond, the chief legal officer, and Alexander Macgillivray, who recently left to become general counsel for Twitter, Inc. On the other side of the table were Paul Aikin from the Authors Guild; Richard Sarnoff, president of Bertelsmann AG’s digital media operations; and Macmillan chief executive John Sargent, as well as their lawyers, Michael Boni of Philadelphia’s Boni & Zack for the Authors Guild and Jeffrey Cunard of Debevoise & Plimpton for the AAP. (Neither side would comment on the record for this story.)

The deal solved two problems — the orphan-works issue and the logistical challenge of tracking down individual rights holders — by requiring publishers and authors who don’t want to participate to “opt out.” This is where class action law and copyright law collide headlong. The opt-out, of course, is standard under class action law. But no one has ever used an opt-out to sidestep copyright law, which requires a rights seeker to affirmatively ask for permission to use a work. For this reason, lawyers in the digital rights community say that the settlement, if approved, could be the biggest thing to happen to copyrights in a generation.

Andrew Bridges, a Winston & Strawn copyright lawyer, says Google used the class action procedure as a “speedy and creative alternative” to two other mass book licensing paths that would have failed. The first path would have been to strike individual deals with rights holders, a prospect that Bridges says is “impossible at the scale at which Google is operating.” The other path would have been to hope for a legislative solution. Bridges says this would be the preferred option, except that the political climate for copyright legislation “lets very few things happen that make so much sense.” Congress has been debating what to do with the orphan-works issue for years, for example, but has never taken action.

“I give Google an enormous amount of credit,” says Randal Picker, a University of Chicago Law School intellectual property professor. “They had a smart legal strategy that drove the business strategy.”

A deal would alter the copyright universe. A decade of combat over the meaning of copyright in the digital age has produced few winners. Basic questions about how to value and protect content remain unsolved. How do content creators, for example, cope with the rapid technological advances that have freed intellectual content from physical media such as books, compact discs, and newspapers?

No one has suffered more casualties than the music industry, which has gone after infringers with furious desperation. From a litigation standpoint, the industry has won every battle, beginning earlier this decade when it sued file-sharing companies Napster, LLC, and Kazaa BV out of business and continuing through this year, when the Recording Industry Association of America won a $1.92 million verdict against Jammie Thomas-Rasset, a Minnesota woman who had shared music over the Kazaa network. From a business perspective, the strategy has proved both ineffective — a recent study in the United Kingdom found that the iPod of the average 14- to 24-year-old held 842 illegally copied songs — and a public relations fiasco. (Richard Marx, whose song “Now and Forever” was one of those that Thomas-Rasset downloaded, issued a statement in which he said, “I’m ashamed to have my name associated with this issue.”) In response to the criticism, the RIAA has said that it doesn’t intend to pursue new cases.

Given the difficulties that Google still faces putting together a deal, how likely is it that such a feat could be replicated for the music business? Von Lohmann at the Electronic Frontier Foundation is cautiously optimistic. He says that there has already been some movement toward a collectivized licensing of music. Warner Music Group Corp. is creating a service called Choruss, which will seek to cut deals for blanket downloading licenses to universities, and, eventually, individual users. Two other record labels have joined the project, which is still in the development stage. Any Choruss deal, of course, would probably be a strict licensing arrangement between record companies and the universities. Also, there is some precedent: Rights clearance member organizations ASCAP and BMI already license live and broadcast performances.

Many practitioners don’t share Von Lohmann’s optimism. George Borkowski, a Venable copyright lawyer who represented the RIAA in its copyright fights with Napster, offers a flat “no” when asked whether a group settlement could ever work for the music business. “The sound recording industry opposes blanket compulsory licenses as a way to deal with downloading,” he says.

Winston & Strawn’s Bridges, who often represents online media companies in disputes with the recording industry, agrees that record companies are fundamentally opposed to collective settlements. The music industry is much smaller in terms of the number of people or companies that control the copyrights, he says, and is “more accustomed to combat” than the book publishers and authors. Further, he says, the music industry itself can’t even agree on which entity controls what rights. Depending on the use of the same underlying song, royalties are often collected by different companies, all of which would need to be mollified to get a group licensing deal done.

As Google has learned, blanket compulsory licenses are controversial in the book world, too. The Google settlement is framed as a nonexclusive deal, with other entities free to draft their own agreements with the authors and publishers. But critics say that publishers and authors will have little motivation to sign deals with Google’s would-be competitors. Microsoft Corporation, which abandoned its own digital library project last year, said in a court filing that the settlement “confers on Google a new monopoly by authorizing Google (and Google alone) to engage in the wholesale commercial exploitation of entire copyrighted books.”

UCLA law professor Douglas Lichtman says that Google is trying to make a “sweetheart deal” for itself that can’t be replicated by a competitor through a standard contract. Foreign authors have cried foul, claiming Google doesn’t have the right to scan their works. European litigation over the deal is ongoing. U.S. register of copyrights Marybeth Peters told a congressional committee that the settlement is “fundamentally at odds with the law,” and that it usurped a role occupied solely by Congress.

In its opposition, Justice struck a similar note: “A global disposition to the rights of millions of copyrighted works is typically the kind of policy change implemented through legislation, not through a private judicial settlement.” But, at the same time, the opposition held out hope for the parties: “A properly defined and adequately represented class of copyright holders may be able to settle a lawsuit over past conduct by licensing a broader range of conduct to obtain global ‘copyright peace.’ ” Modify the settlement, the Justice Department seemed to be saying, and it would back the deal.

What needs to be done, in the government’s eyes? First and foremost, address the antitrust issues. As it stands, Google and the authors and publishers have agreed to certain pricing algorithms and discounting agreements that will apply to out-of-print books. Justice takes a dim view of these “joint price-setting mechanisms.” The parties will also need to reassure the government that competitors will also have access to all the scanned books, including the orphan-works collection.

The parties must also resolve a conflict that the government says exists between class members and absent rights holders. Under the current proposal, class members have no financial incentive to track down absent rights holders. The government says that making the deal an opt-in rather than opt-out agreement would solve this concern, but also notes that other solutions might work. The parties, for example, could use proceeds from the sale of orphan books to fund a search for these missing copyright holders.

If not quite a road map to a revised settlement, the Justice letter at least suggests an acceptable finish line. Can negotiators get there? It won’t be easy, but neither was crafting any kind of deal in the first place. The good news, for Google, for the class members, and for those who hope a completed deal would spur other content providers to make their products universally accessible, is that the Google book search settlement’s most powerful critic — the Justice Department — also supports its negotiator’s broader aims. “A properly constructed settlement,” Justice’s filing says, “has the potential to breathe life into millions of works that are now effectively off-limits to the public.” •