Jessica Silbey

Do you ever wonder how it is that libraries can lend books repeatedly, while copyright owners (e.g., book authors) are granted the exclusive right to distribute their copyrighted works? Or how publishers make money selling books at retail prices when a person can resell books (or buy used books) for much less (hello Amazon Used Books for under a dollar!)? The reason is because of copyright’s “first sale” doctrine, 17 U.S.C. § 109, codifying the common law’s exhaustion principle, which says owners of lawfully made copies are allowed to dispose of those copies without regard to copyright law. In other words: the right to distribute is “exhausted” with the first sale. This means we can resell, lend, or give away our books. What we can’t do is make copies of them.

And that is where the concept of “digital lending” runs headlong into copyright law. When libraries buy books, they can lend them without restrictions. When libraries lend e-books, those e-books come with contractual provisions limiting their lending in substantial ways because reading an e-book requires making a copy, which is not exhausted by the first sale doctrine. As the Supreme Court has said (in the patent context, which has a similar first sale principle): “exhaustion applies only to the particular item sold, and not to reproductions.”¹

But libraries, which Professor Rub aptly describes as “among the most cherished institutions in our society,” are increasingly engaging in e-book collection and e-lending to accommodate patron preferences and needs. E-lending enhances libraries’ core mission providing access to books and information and is especially effective when in-person patronage is challenging, such as in rural areas and in cases of disability or a pandemic. But e-book lending is substantially more expensive for libraries and license restrictions allow publishers to retain much more control over uses, disrupting the first sale balance that has kept libraries flourishing for so long. Professor Rub’s article describes this imbalance, the legal puzzles it raises, and then proposes some novel solutions.

In the background and haunting the article’s discussion is the case of Hachette et. al. v. Internet Archive, currently pending before the U.S. Court of Appeals for the Second Circuit. That case concerns a multi-publisher lawsuit against a different kind of library: the Internet Archive, a non-profit digital-only library, which describes itself this way: “Like a paper library, we provide free access to researchers, historians, scholars, people with print disabilities, and the general public. Our mission is to provide Universal Access to All Knowledge.” The Internet Archive engages in what it calls “controlled digital lending,” (CDL) whereby instead of an in-person library patron borrowing a print book (thus making that book unavailable to all other patrons), the Internet Archive patron receives a digital copy of the book (which the Archive scanned for this ). While the Internet Archive patron has access to the digital scan, no other patron does. CDL aims to mimic the scarcity model that is the basis of print lending.

The problem, of course, is that CDL does not precisely mimic the model of print lending, as Professor Rub explains. First, book scans are copies, and making a copy is an exclusive right of the copyright holder that is not exhausted upon purchase of a lawful copy of the book. (Whether it is fair use under certain conditions is a good question that the article addresses at some length, as does this White Paper by David Hansen and Kyle Courtney.)

Second, CDL scans do not degrade the way print books do, and so libraries do not have to replenish their stock of digital collections as they do for print books, which are often replaced after significant use. Publishers claim to rely on library replacements to turn profits. Without such repurchases, publishers assert that their revenues fall as does their support for authors. (For whether publishes complaints about their continued viability in the digital age can be fully credited, I urge readers to consider The Publisher’s Play Book: A Brief History of Publishing Industry’s Obstruction of the Library Mission.)

Third, CDL is instantaneous, whereas print lending has significant friction that slows access. Friction means libraries lend print copies less frequently than e-books (or scans), and thus to serve demand, libraries will buy more print copies when they don’t have to license more e-books. These differences lead to conclusions drawn by Professor Rub and the many others he cites that digital first sale does not exist. It also means that the Internet Archive’s CDL practice is at substantial risk of being held unlawful.

That doesn’t leave libraries without options, however. Professor Rub’s article is largely devoted to saving (if not empowering) libraries and CDL practices, because as he explains, libraries all over the country face a serious financial crisis. Public libraries – the model created by Benjamin Franklin in 1790 in my neck of the woods (Franklin, Massachusetts) – are unusual around the world because they are free and open to anyone. They serve many functions beyond lending books, including anchoring communities as places to convene and learn. But their budgets are threatened by diminishing tax revenue and exploitative pricing of e-book licenses (which Professor Rub describes in detail). Moreover, state laws, such as the Maryland Library Ebook Fairness Law, which would have required book publishers to offer public libraries reasonable licenses to ebooks, have been successfully challenged as unconstitutional.

So what are some of the solutions? Professor Rub provides many possible avenues. One suggestion is to “replicat[e] the physical world” by creating friction in digital lending so that CDL resembles print borrowing. Although CDL would still involve unauthorized copying, such conditions of friction ostensibly minimize market harm. Given the public benefits of distribution, the library practice would comfortably fall within fair use.

A second option is to adopt a “digital public lending right” (ePLR) akin to the Public Lending Rights in EU countries that collect royalties for lending books (because the EU does not have the equivalent of the first sale doctrine). An ePLR would allow libraries full digital exhaustion rights as they enjoy with print books, but publishers would be entitled to collect royalties determined by a public governance body to offset revenue loss.

A third way to achieve digital lending is to support market segmentation along variables of time, usage, and wealth. For example, digital books might be unavailable for new titles (similar to home movie streaming which is often delayed until after theaters show new releases). Or public libraries could offer preferential digital access to patrons whose income is below a certain wealth threshold and who are less likely to allocate their scarce resources to purchasing books (or ebooks).

My favorite of his suggestions focus on state-run public libraries (including public university libraries). These suggestions encourage states (1) to use their market power to demand favorable licensing terms from publishers, and (2) to exercise sovereign immunity if sued for copyright infringement stemming from their CDL practices. Given how important libraries are for state government and communities, encouraging states to get in the fight to save them makes a lot of sense.

Each of these suggested solutions (some of which Professor Rub acknowledges) pose problems, but they are worth considering to save our most “cherished institutions.” Some of my favorite quotes about libraries explain why: “Libraries are full of ideas – perhaps the most dangerous and powerful of all weapons.” (Sarah Maas, from Throne of Glass) And “the only thing that you absolutely have to know, is the location of the library.” (Albert Einstein) Whether you like your books in paper or electronically (and there is some interesting empirical research on just that question), libraries should provide both and they need help doing so. Professor Rub’s article significantly advances the conversation and hopefully recruits more lawyers and legislatures to the cause of saving the free public libraries.

1. Bowman v. Monsanto, 569 U.S. 278 (2013).