July 8, 2024

A Response to Hachette v. Internet Archive from Michelle Wu

On June 28, 2024, the 2nd Circuit Court heard oral arguments in the Hachette v. Internet Archive appeal. We invited Michelle Wu, retired law library director and professor at Georgetown Law, to give this response.

Last Friday, the U.S. Court of Appeals for the Second Circuit held oral arguments in Hachette v. Internet Archive. Below, I discuss three lines of questioning, as well as two repeated claims that misdirect substantive analysis of the issues in the case. 

If Congress had not statutorily protected first sale, would library lending be legal? 

In response to these questions, both parties acknowledged that library lending would be legal even without Congress’ intervention but both pointed to relatively recent authorities as to why. I wish that there had been more time to dig deeper into this question, as library lending has existed for centuries, long before copyright laws existed. It was and would remain legal simply because, with or without legislation, it is a manifestation of the principles of the alienation of property: once someone legitimately acquires an object, that person may use or reuse  it any way they wish. The principle of alienation has existed as long as humans have been alive, and it is considered a natural right, one that cannot be invalidated by government because without it, the productive interpersonal use of property (e.g., dissemination, sales) is impossible. 

Why did this line of questioning matter? 

Because the publishers rest much of their argument on the fact that, by statute, first sale applies only to physical objects, not to digital ones. And if the only authorization that libraries have to lend is that specific statute, then any lending not covered by it would be infringing. By having parties recognize that the rights protected by copyright are broader than those explicitly authorized in legislation, the court creates the opportunity to refocus the discussion onto the purpose of copyright laws. First sale need not apply for library use to meet copyright’s purpose. 

Is CDL’s transformative argument format specific?

This was an eye-opening line of questioning, mostly because it made apparent that something libraries had assumed was obvious might not be obvious to courts or copyright owners. Digitization of a print book makes it more accessible and efficient. It is the fact that the original is in an analog format – one that cannot be easily read by screen readers, conveyed to people in remote places or in isolation, accessed by users even after the physical library is closed, and so forth – that allows CDL to make an argument of transformativeness. In contrast, if a library took a born-digital book, made a copy, and lent that copy, any argument for transformativeness would fail. The digital copy in that case would be no more effective than the original.

Why did this line of questioning matter? 

Other cases had previously acknowledged that wholesale copying or distribution could be transformative, but the district court below had refused to apply their reasoning to CDL, relying instead on the more common statement that a copy that substitutes for the original cannot be transformative. By asking these questions, then, the court signaled its openness to the idea that creating a substitution for the original  is not the only test for transformativeness. Many transformative fair uses produce copies(e.g., home recording), and in those instances, substitution alone is always insufficient to assess transformativeness. 

Is the fourth factor (actual or potential market effect) relevant at all?

In a series of questions, the court showed its awareness of the history of copyright and the cases where market harm arguably made no difference to a case’s outcome like Sony Corp. It seemed willing to consider that the factor given the greatest weight by the district court was a red herring.

Why did this line of questioning matter? 

It reasserted that market harm need not drive the outcome of cases. Every fair use can have a market impact, and even substantial market impact need not result in a finding of infringement. Consider the facts of Sony, a case about the legality of home recording and the devices that enabled recording. In that case, like in this one, publishers alleged that to find fair use would bring great market harm and the destruction of creative industries. 

It is certainly conceivable that home recording could impact the market. After all, someone recording their favorite movies off a cable channel might decline to renew their subscription, instead rewatching their recordings. If those engaged in home recording all canceled their cable subscriptions, there would undoubtedly be lost income. But even if the publishers could have proven that they lost significant revenue from such cancellations, the outcome would not have changed. After all, the purpose of recording was to meet a need (time shifting) that the original method of dissemination could not. Home recording is simply a way to ensure that users could get the full value of legitimate access. For example, it enabled everyone who subscribed to the same cable channel to view its programming, even if some were at work or otherwise engaged at the time programs aired. That purpose – meaningful access to content one has purchased or otherwise has legitimate access to – does not lose its validity simply because content publishers lose money. 

Side note: The aftermath of Sony also shows that commercial actors are extraordinarily bad at predicting harm, only able to see catastrophe. Home recording did not destroy industries but rather, it opened up a completely new revenue stream for commercial actors. 

Two misguided arguments

Next are two topics raised by the court and/or publishers that arguably are undermined by their own facts.  

Redigi. Capitol Records v ReDigi was referenced both by the court and by publishers. It is most often cited as denying the application of first sale to digital copies. 

ReDigi was a case about “resale” of digital music, but the parties to that case and its court conceded that (1) the seller only licensed the music, and (2) the license terms and the technological controls over the music made it such that the seller could redownload the music after “resale”. In other words, reselling the music was a factual impossibility. The music could not be conveyed in a manner where the seller lost access when the buyer gained it.

Format. Unfortunately, both parties and the court stated that there are different markets for the same book based on format (e.g., print book, e-book) without ever addressing the basis for treating these as different markets instead of one. But the definition of copyright – which determines what qualifies for protection – is not format based, so explaining why these are different markets is essential to identifying what data is relevant to market harm. 

It is, of course, impossible for anyone to predict what a court will ultimately decide based on oral arguments. But the questions asked make me hopeful that the appeals court will do two things that the district court did not. One, that it will recognize that copyright is purposeful. That all of its components – statutes, regulations, common law, equity – aim for the same purpose of societal benefit, and that this purpose guides analysis. Two, that fair use is contextual. It is not a systematic checklist of the four factors, and there are no brightline rules. 

If the purpose of copyright and the contexts of CDL and the book market become the focus of inquiry, several things will become clear. 

First, even under CDL, copyright owners are paid for their work. They may not be paid how much they want, but the number of copies of the content used is acquired on a legitimate market. 

Second, if CDL is deemed unfair in the current book market, libraries (and therefore the public) lose the ability to collect and preserve works for current and future users. 

Through the lens of copyright’s purpose, one has to ask if permitting or banning CDL serves authors and the public better. It seems to me that the answer is clear. With CDL, there is not only financial support for authors and publishers but also stable public access to works by current and future readers. Without CDL, there is increased financial support for authors and publishers of bestselling titles, decreased financial support for the vast majority of other authors and publishers, and public access to works is limited only to a small number of current readers. I would ask both authors and readers which of these alternatives do they think serves them best? And I would ask legislators and courts which of these better meets the purpose of copyright?

Continue reading

A graphic of a person holding a smartphone and discovering books inside
Subscribe to our list

Join the community!

Join the community fighting for digital equity and a better future for libraries.