This past weekend, just before the hurricane, I attended In Re Books, a conference about law and the future of the book convened by James Grimmelmann at the New York Law School. Playing the role of Luddite intruder among the futurologists, I gave a talk about the hazard that digitization may pose to research and preservation. Though there were a few librarians, leaders of nonprofits, and even writers present, most of my fellow conference attendees were lawyers who specialize in copyright, and I discovered that copyright lawyers see the world rather differently than do the writer-editor types with whom I usually rub shoulders. They don’t expect publishing as I know it to be around much longer, for one thing. I thought I’d try to write up my impressions of the time I spent in their company. Please keep in mind that I’m not a lawyer myself. I’m just a visitor who went to the fair.
A specter was haunting the conference: the ghost of the settlement that Google Books tried to make with the Authors Guild several years ago. That settlement, slain by Judge Denny Chin in late 2011, had attempted to obtain digital rights to what are known as orphan works, books that are protected by copyright even though the author or publisher who holds the copyright can no longer be found. The settlement had proposed to set up a collective licensing system that would charge for digital access to all books under copyright, parented and ophaned. Proceeds from orphan works, it was suggested, might be shared with findable authors, if no actual rightsholder could be found and if anything was left over after the rights management organization was done paying for itself. The proposal was far from perfect. Why should Google get to sell orphan works and nobody else? Why should the profits from orphan works go to people who didn’t write them? It turns out that the death of the agreement is not much lamented by the copyright lawyers. When Minda Zeitlin, president of the American Society of Journalists and Authors, asked, “Is there anyone better to represent dead and unfindable authors than living and findable ones?” the retort from Pamela Samuelson, a copyright law professor at the University of California at Berkeley, was sharp: “I’m a better representative of an author like me,” Samuelson said, her implication being that an academic author aims in publishing to further knowledge and build a reputation, not make money. Roy Kaufman, who works at the Copyright Clearance Center, a collective licensing agency founded in 1978 in response to the disruptive technology known as the photocopier, was at pains to distinguish his employer’s system from the one advanced by Google Books and the Authors Guild. The Copyright Clearance Center is opt-in and nonexclusive, he assured the audience. His message was studiously non-threatening: mass digitization could involve rightsholders. Maybe it could take the form of collective licenses arranged between social-media networks and publishers. Facebook, for example, could pay the New York Times for articles and photos that its users posted.
Kaufman’s support for collective licensing, however cautious, was atypical. Most at the conference were against it. Samuelson thought it inadvisable in general, as did Matthew Sag, of Loyola University Chicago, who justified his dislike by pointing to the failures and subsequent reboots of a compulsory licensing system recently set up in the United States for the webcasting of music.
What, if anything, will take the ghost’s place? At the conference, a leading contender was the idea that fair use might solve the orphan works problem—an idea recently advanced by Jennifer Urban of the University of California at Berkely. Fair use, as I wrote in a review-essay for The Nation earlier this year, is an exception to copyright written into American law in 1976. It’s because of fair use that a reviewer doesn’t need to ask permission before he quotes from a book, and it’s because of fair use that an Obama campaign commercial can quote a Romney speech, or vice versa, without paying for it. In the last few years, courts have been more and more generous in how they define fair use, perhaps because Congress seems so unlikely to help sort out the tangles in copyright. In a recent case between the Authors Guild and a digital books repository called Hathi Trust, for example, a court found that three of the four things that Hathi wanted to do with digital texts were fair use: data-mining, indexing, and providing access to the blind. America’s 1976 copyright law specifies four factors to consider in determining fair use—the nature and purpose of the new use, the nature and purpose of the original work, the amount taken, and the impact on the original creator’s income—but in the last couple of decades, judges have focused on whether a new use is “transformative” of the old content it borrows from. Whatever purpose Thomas Pynchon had in mind when he wrote Gravity’s Rainbow, for example, he probably didn’t imagine computerized search of his novel along with a myriad of others in order to find patterns of word usage. That’s a completely new use, a transformation of the purpose of his words unlikely to interfere with the money he expected to make from his novel, so the judge in the Hathi Trust case found it fair.
Sag and Samuelson favored Urban’s idea, which was also mentioned by Doron Weber of the Alfred P. Sloan Foundation, which funds the Digital Public Library of America. Since I hadn’t read Urban’s paper, I asked Sag what kind of transformation lay behind her deployment of fair use. There wasn’t any, he explained, to my surprise, and now that I look at the paper, I see what he means. Urban thinks libraries and universities should be able to provide digital facsimiles for their patrons to read—exactly the same use for which the books were originally published. She also frankly admits to wanting the right to reproduce entire works, not just samples or snippets of them. But she argues that such use would be fair nonetheless, based on the four factors conventional in fair-use analysis. She maintains that libraries and universities are nonprofit institutions, who would be offering access to the texts as a noncommercial service for such public-spirited purposes as research and preservation. (For this part of her argument to hold water, would a university library need to open itself to the public in a general way? Right now the services of a university library, however worthy, are for the most part bestowed only on its own students and faculty, and their character is not purely altruistic.) And she argues that the orphanhood of an orphaned work is more important than previous analysts have seen: “Orphan works,” she writes,
represent a clear market failure: there is no realistic possibility of completing a rights clearance transaction, no matter how high the costs of that transaction, because one party to the transaction is missing.
Therefore market harm, the fourth factor of fair-use analysis, is nugatory, in Urban’s opinion. The trouble with her argument here, I think, is that it’s impossible to know whether a so-called orphan work is really an orphan or merely a work whose parents haven’t shown up yet. If the parents do exist, the market harm to them is real, and it would be as wrong for a court to give the value of their work to Urban’s university library as to give it to Google or a third-party author. Urban seems to be transferring the copyright rather than carving out an exception to it, and I’m afraid that only Congress, in its capacity as the sovereign power of the United States, has the authority to dispose of someone else’s copyright, in an act of eminent domain. Without any claim of a transformation, it seems unlikely to me that Urban will convince a court to define fair use so broadly that it includes reproducing whole works for much the same purpose that they were originally published. But this is just my opinion. The copyright lawyers seem excited by her idea, and as yet no one knows how far it will go. It’s up to the courts. As Jule Sigal, of Microsoft, noted in his presentation, the orphan-works problem has passed through the Age of Legislation (2005-2008) and the Age of Class Action (2008-2011), and we are now living in the Age of Litigation.
The other big new idea at the conference was that the first-sale doctrine might be extended to e-books. That sentence will sound like gibberish to the uninitiated, so let me back up and explain. The first-sale doctrine is a legal concept that limits the control that copyright affords. Specifically, it limits copyright control to the period before an item under copyright is first sold. Once you buy an ink-on-paper book, for example, you’re free to re-sell the book on Ebay at a fraction of the cost. Or give it to your boyfriend. Or take an X-acto blade to it and confuse people by calling the result art. You don’t have the right to sell new copies of the book, but you’re free to do almost anything else you like with the specific copy of the book that you bought. Without the first-sale doctrine, used bookstores would be in constant peril of lawsuits.
Two speakers at the conference told the story of Bobbs-Merrill v. Straus, the 1908 case that established the first-sale doctrine. On the copyright page of the novel The Castaway, the publisher Bobbs-Merrill set the retail price at one dollar and threatened to sue discounters for breach of copyright. Macy’s sold the book for eighty-nine cents anyway, triggering a lawsuit, and the court ruled that copyright afforded Bobbs-Merrill control over the book’s price only up to the moment when Bobbs-Merrill, as a wholesaler, sold copies to Macy’s, which then became free to set whatever retail price it wanted. Ariel Katz, of the University of Toronto, noted that the story is usually told as if the case involved an attempt at what’s known as “vertical” price-fixing—that is, an attempt by a wholesaler to fix the prices charged by independent retailers further down the supply chain. But Katz maintains that it was actually a story of “horizontal” price-fixing—that is, an attempt at collusion in price-fixing by companies that are supposed to be in competition with one another, wholesalers in collusion with wholesalers, and retailers with retailers. The Straus brothers who ran the Macy’s department store were “retail innovators,” Katz explained, who sold a wide variety of goods, including books, at steep discounts, thereby angering publishers and traditional booksellers. The members of the American Publishers Association publicly swore to refuse to supply retailers who discounted the retail price of books, and the American Booksellers Association publicly swore to boycott any publishers who didn’t toe the American Publishers Association’s line. It was the Straus brothers who first went to court, accusing the publishers and booksellers of antitrust violations, but the outcome of this first case was ambiguous: the court ruled that publishers could only set the prices of books that were under copyright. It wasn’t until the 1908 case that the court limited price-setting even of copyrighted books to the period before their first sale.
(As Katz pointed out, it isn’t obvious why publishers and booksellers should have been willing to collude in fixing prices, and he proposed an economic explanation that I wasn’t quite able to follow. He suggested that the price-fixing was an attempt to solve a challenge first discovered by Ronald Coase: if you sell a durable good and you’re a monopolist, you soon find that your monopoly isn’t as profitable to you as you’d like it to be, because you’re in competition with yourself—that is, you’re in competition with all the durable goods you’ve already sold, which suppress demand. The only way to keep prices from falling is to convince consumers that you’ll never let them fall. Katz argues that the limit to booksellers’ shelf space helped publishers make credible their promise never to lower prices, and that in the digital world, where shelf space is unlimited, no similar promise will be as credible. He ran out of time before explaining in detail how this mechanism would work, and as I say, I didn’t quite follow. I also wasn’t quite certain that books qualify as durable goods. Most people, once they’ve read a book, prefer to read a new one instead of re-reading the one they just finished, a fact that suggests that books are more like loaves of bread than refrigerators. But I may be missing something.)
Aaron Perzanowski, of Wayne State University, framed the story of Bobbs-Merrill v. Straus in the context of a common-law tradition of rights exhaustion—the word exhaustion here having the sense of a thing coming to its natural end. In Perzanowski’s opinion, the right to control price is not the only aspect of copyright that expires when an item under copyright is sold. The owner of a work has purchased the use and enjoyment of it, Perzanowski argued, including perhaps the rights to reproduce the work and to make derivative works. Perzanowski made explicit a further leap that remained mostly implicit in Katz’s talk: Shouldn’t the first-sale doctrine apply to e-books, too? As a contractual matter, e-books are rarely sold, in order to prevent exactly this eventuality. In the fine print, it transpires that what distributors purchase from publishers, and what readers purchase from distributors, are mere licenses. But if courts were to recognize readers of e-books as owners, the courts could grant readers the right to re-sell and a limited right to reproduce what they had purchased. Jonathan Band, of Policy Bandwidth, in his assessment of recent legal victories won by university libraries on the strength of fair-use arguments, noted that he saw the first-sale doctrine as likely to be important in future disputes over digital rights. Libraries, he said, felt that they had already purchased the books in their collection and ought to be able to convey them digitally to their patrons.
Extending the first-sale doctrine to e-books might make libraries happy, but it would horrify publishers. Right now, only two of the six largest American publishers allow libraries to lend all of their e-books, and one of those two sells licenses that expire after twenty-six check-outs. Librarians sometimes become quite indignant over the limitations and refusals. “Are publishers ethically justified in not selling to libraries?” one asked at the conference. A recent Berkman Center report, E-Books in Libraries, offered some insight into publishers’ reluctance:
Many publishers believe that the online medium does not offer the same market segmentation between book consumers (i.e., people who purchase books from a retailer) and library patrons (i.e., people who check out books from a public library) that the physical medium affords.
When was the last time you checked out a printed book from the library? My own impression is that gainfully employed adults rather rarely do. (At least for pleasure reading. Research is a different beast.) Maybe they prefer to buy their own books for the sake of convenience, which ready spending money enables them to afford. Or maybe it’s to signal their economic fitness to romantic partners, or to broadcast their social status more generally. But whatever the reason, the fact is that publishers don’t sacrifice many potential sales when they sell printed books to libraries, because library patrons by and large aren’t the sort who purchase copies of books for themselves. The case seems to be different with e-books, though, especially if patrons are able to check them out from home. E-book consumers signal their economic status by reading off of an I-pad XII instead of a Kindle Écru; the particular e-book that they’re reading is invisible to the person on the other side of the subway car, so it might as well be a free one from the library. That means that e-book sales to libraries cannibalize sales to individual consumers. Publishers have tried charging libraries higher prices for e-books. They’ve tried introducing technologically unnecessary “friction,” such as a ban on simultaneous loans of a title, or a requirement that library patrons come in person to the library to load their reading devices. The friction frustrates library patrons and enrages librarians, and even so, it hasn’t been substantial enough to reassure the publishers who are abstaining from the library market altogether. If the future of reading is digital, the market-segmentation problem raises a serious question about the mission of libraries. In his remarks at the conference, the writer James Gleick, a member of the Authors Guild who helped to negotiate its late settlement with Google Books, said that he doubted that every lending library needed to be universal and free, and that he wished the Digital Public Library of America, which is still in its planning stages, were trying to build into its structure a way for borrowers to pay for texts under copyright. The challenge of bringing e-books into public libraries turns out to be inextricable from the larger problem of how authors will be paid in the digital age.
I’ll try to report what the lawyers think of that larger problem in a later post.
UPDATE: Part two here.