From Hugh Stephens Blog:
In my year end blog post looking back at significant developments for copyright and creators in 2022 and looking forward to topics that will be at the top of the agenda in 2023, I identified questions over the legality of a contrived and unproven concept, so-called “Controlled Digital Lending” (CDL), as one of the big issues likely to be clarified this year.
Back in June 2020, four major publishers (Hachette Book Group, HarperCollins Publishers, John Wiley & Sons and Penguin Random House), all members of the American Association of Publishers (AAP) filed suit against the Internet Archive for “Systematic Mass Scanning and Distribution of Literary Works”. The Internet Archive (IA), an organization that brands itself as “a non-profit library of millions of free books, movies, software, music, websites, and more”, provides a number of services including archiving the internet through its “Wayback Machine”, archiving television programs and audio recordings, and digitizing documents and books, both those in the public domain and others still protected by copyright. Although registered as a US non-profit, it was founded and still led by Brewster Kahle, a multimillionaire entrepreneur and digital “guru”.
The publishers’ lawsuit was provoked by the IA’s decision to create a self-described “National Emergency Library” in March 2020, during the peak of the COVID pandemic. The Emergency Library expanded the untested theory of Controlled Digital Lending (CDL), championed by the Archive, by eliminating even the pretence of limits on the numbers of digital copies of books that could be borrowed, in effect eliminating all wait times. As I noted in an earlier blog post (“Are Authors the Enemies of Authors and Publishers?”), COVID was the pretext used by the Archive for pushing the envelope on CDL. Under the IA’s interpretation of CDL, a digital scan of a book can be substituted for the original work by a lending library as long as the library holds the requisite number of physical copies. As is the case with legitimate lending of licensed e-books, there is no need for the borrower to physically collect the work; it is all done digitally including terminating the loan once the book is due.
The issue of scanning a book without authorization in order to provide a substitutable digital version is clearly at odds with the law, especially when it comes to US case law which has been very clear and consistent on this point. As authors and publishers point out, it is a form of copying that destroys the licensing market for e-books. Despite this obvious fact, this has not stopped advocates, like the Internet Archive, from claiming that the practice is somehow fair use under US law. Until COVID hit, the Archive purported to follow what it characterized as normal lending rules by allowing only as many digital copies into circulation as it physically held in its inventory, with a digital copy having to be “returned” before a new copy could be loaned out. However, in reality it exercised no actual controls, simply asserting that it was following the correct “own to loan” ratio. Then, with the arrival of COVID, the Archive dropped all pretence of controls and announced that it was suspending the normal practice of maintaining a wait list and would allow unlimited digital copies to go into circulation.
While a few initially misguidedly lauded the Archive for taking measures to assist consumers who were self-isolating because of the pandemic, it didn’t take long for authors and publishers to push back, pointing out that this unilateral move was a case of giving away someone else’s property without consultation or permission. Although the possibility of litigation had been simmering for some time, the IA’s declaration of the “National Emergency Library” was the precipitating event leading to the filing of suit by the publishers. The IA then ended the Emergency Library program prematurely and Kahle appealed to the publishers to settle the dispute in the boardroom rather than the courtroom. The hypocrisy of this appeal was not lost on authors and publishers who had been trying for years to engage the Archive in meaningful discussions. Both sides moved for summary judgment, and those motions are currently pending in front of the judge.
Link to the rest at Hugh Stephens Blog and thanks to R. and others for the tip.
PG recalls this type of “controlled digital lending” theory having appeared in various guises on past occasions.
As the OP suggests, it won’t work.
Fundamental copyright law requires the author’s approval before anyone can make a copy of the author’s creation, be it words or paintings or anything else.
When a library purchases a physical book and lends it to someone, the library has entrusted its copy to one of its patrons. If the patron doesn’t return the book or loses it or if the patron’s dog eats the book, if the library wishes to allow other patrons to borrow the book, the library’s only legal option is to purchase another physical book from the author or publisher or locate a helpful friend of the library who has purchased a copy of the book and ask if the helpful friend will donate her/his/their physical copy of the book to the library.
Every digital lending scheme PG has examined involves making a copy of the book each time someone wishes to read it. If the library scans one of its physical books, it’s making a copy without permission. It doesn’t matter whether the library destroys the physical book after scanning it. The library has still made a copy.
If a library acquires an ebook, absent permission from the owner of the ebook’s copyright, the library can’t make a digital loan to a patron via an electronic copy while still maintaining a copy of the ebook on the library’s computer system or anywhere else.
Just because electronic copying is very simple to do and happens trillions of times per day on computers around the world doesn’t mean that making an electronic copy of a bunch of organized electrons that are protected by copyright law is not a violation of copyright law.
Please note that that the copyright law includes an exclusion for Fair Use of the copyrighted material. Fair Use does not constitute a violation of copyright law.
Here’s a reasonable general definition of Fair Use from The Stanford Libraries:
In its most general sense, a fair use is any copying of copyrighted material done for a limited and “transformative” purpose, such as to comment upon, criticize, or parody a copyrighted work. Such uses can be done without permission from the copyright owner. In other words, fair use is a defense against a claim of copyright infringement. If your use qualifies as a fair use, then it would not be considered an infringement.
Of course, as with all laws, Fair Use contains more than a few gray areas. However, making a complete copy of a creative work and lending, selling, etc., the copy to one or more others is not protected under the Fair Use exception to general copyright law.
(Personal note – PG believes all of his excerpts of copyright-protected material in TPV constitute fair use. He asks himself if that is the case for at least one or more items he posts each week. His judgement has been questioned on only one occasion, many years ago, but the questioner stopped complaining after PG sent his analysis of fair use for the challenged post.)
Hi PG,
Good analysis, thanks for sharing. I’m not always comfortable relying on Hugh Stephens, he has a very defined agenda at times, and he slips in some bias in some of his analysis, but most of it is okay in this one, no obvious flags. In your part, I’m good with just about all the elements, except a small head-scratcher for me on one paragraph:
“If a library acquires an ebook, absent permission from the owner of the ebook’s copyright, the library can’t make a digital loan to a patron via an electronic copy while still maintaining a copy of the ebook on the library’s computer system or anywhere else.”
I follow some library management discussions online, and it seems like some libraries that choose to manage all their files themselves may have copies still in their system after a loan, depending on what they are doing in the first place and from which company they acquired it. Some acquire 10 digital copies, some acquire 1 digital copy and maintain control for up to 10 borrows. It shows up in online piracy discussions sometimes because if the library has only one with multiple loans licensed, then everybody borrows the same one…if it shows up on a pirate site, hard to figure out who borrowed and leaked it. If it is separate files, there is a smaller pool to examine. Of course, true pirates usually strip that info anyway. Of course, if someone has a copy out, they can’t just issue an extra one on their own initiative if they haven’t licensed another copy. 10 licenses, 10 loans.
Yet from the digital librarians’ discussions, it seems to me that they still have the original on their system, just locked. There’s no “file” coming back from the user, just a removal of 1 of 10 locks when a copy comes back in. Yet in addition to the 10 copies they have out, and the one in their system, they often have a backup copy as part of their ongoing records management in case of crash or hacking. And if their backup approach is solid, they likely have a full backup plus a major backup plus incremental backups. They could actually have an active copy plus 3 more backups in their system of every file.
Many libraries have gotten around all of this headache by simply using an external service like OverDrive. OD and others license out copies as part of the loan system, the libraries have much reduced overhead to manage, and no concerns about copies, copyright violation, etc., it’s all with the OD overlords. 🙂 As far as I can tell, it seems like most of the fee-for-service tools maintain separate copies officially (10 copies for library X), but they also appear to be doing backups and retaining copies. Lots of people have borrowed files and it’s become corrupted, yet miraculously the original is perfectly safe and ready to be loaned out again.
One thing that I wonder about is directly linked to the original premise, however misused, in the approach cited in the OP. SSome of the recent posts out there, including Stephens, seems to miss is that there is often a completely legal and valid archival function built into a lot of libraries. University libraries in particular are often co-housed with the university libraries, and often have rare book sections, etc. While one would be hard-pressed to argue the latest Janet Evanovich book needs to be preserved in an archive, some universities ARE often fighting two competing principles — respecting the copyright of a long out-of-print book that still has viable copyright vs. the ravages of time slowing eating away the pages that the book is made of. I know a few archivists have wondered if there could be an exemption made whereby, for example, a copy of a book could be made by scanning even when in active copyright, BUT the actual copy would have to be locked until it is eventually released into the public domain. It’s preserved for posterity, while still respecting the copyright term. That argument all goes to crap though if the book is still in print OR if there is a licensable ebook version…but the Gutenberg archive spends a lot of time and effort to resurrect dead texts through scanning. There does seem to be an issue there, just not an easy nut to crack for anything prior to the creation of ebook versions.
Thought-provoking post, as always…thanks!
Not a lawyer nor do I play one anywhere but the situation you posit has been more or less litigated in the HADITHI TRUST DIGITAL LIBRARY case of 2012:
https://en.m.wikipedia.org/wiki/Authors_Guild,_Inc._v._HathiTrust
The ruling determined Fair Use.
“HTDL’s main objective is the long-term preservation of the collection. Member libraries may order replacement copies of works if “(1) the member already owned an original copy, (2) the member’s original copy is lost, destroyed, or stolen, and (3) a replacement copy is unobtainable at a fair price.”[3] The HTDL main functionality is full-text search. When search results are found in works in the public domain, the work is displayed online, and so are works for which the copyright holder has granted permission. For other works, only page numbers and the number of search results per page are shown. In addition, the HTDL makes its collection available to students with print disabilities by offering them secure system access for screen readers. The collection of works available to print-disabled students through HTDL is often larger and easier to navigate than those offered through most university student disability student services offices.[1]”
A more general case that to my knowledge has *not* been litigated is the digital copying of a pbook for personal use without *distribution*. Such may or not be deemed Fair Use or a deminimis violation but since it hasn’t been litigated…
The key aspect of the IA violations is precisely the distribution on the digital copy, whether controlled or not, because it substitutes for the original copyrighted item, thereby decreasing its value and harming the owner of the copyright.
The HTDL exists within the very limited carve outs of academic and accessibility use which is far from the general availability space championed by the IA and its apologists.
Scanning and distributing a Stephen King volume (even on a one-to-one basis) competes with the commercial product because ebooks and pbooks are *not* the same product obtainable via the same license. (One is a mass of dead tree pulp while the other is a limited non-transferable license to access.) In fact, throughout the history of ebooks the publishers have studiously kept the licensing and distribution of both products distinct and separate, even in the rare case of print+digital bundles.
Pure non-distributive archiving by libraries would seem to be covered by existing Fair Use via precedent but I’m sure somewhere out there somebody will use it to generate billable hours sooner or later.
Although you are not a lawyer, F., you could certainly do a better job of lawyering than some lawyers I know.
It’s been some time since I’ve read the Hathi opinions (it’s about twenty years old), but, as I recall, the final result fell into the no harm/no foul category. If the Hathi Trust’s practice was not depriving the author/publisher of any sales/royalties, there wasn’t a violation of copyright law.
If I have my iPad set to back up its electronic contents for safekeeping should the iPad be damaged or destroyed, restoring those contents to another iPad isn’t really a harm to the owners of copyrights for any stored intellectual property on the damaged iPad.
People have been backing up their computers, including licensed software stored thereon, for a very long time without Microsoft/Apple making any objection to the best of my knowledge.
Thanks for your detailed comment, P.
With regard to libraries lending digital books, it’s my understanding that they do so under a license from the book’s publisher (which typically obtains the right to do almost anything via its publishing contract with the author).
The licensing terms of the publisher/library (or publisher/big group of libraries) contract permits the library to make and loan a specified number of copies of the ebook files of a book to patrons.
Per part of your comment, if a library has a license to loan up to ten copies of an ebook and it has already loaned copies to ten patrons, the ebook will be unavailable to any patron until either one of the ten patrons who have borrowed the ebook either “returns” the ebook to the library or the due-date of the ebook loan passes and the ebook becomes unavailable to the earlier borrower.
I can think of a lot of ways the borrower might be able to circumvent the termination of an ebook loan, but have not experimented with any because I respect the copyright owner’s rights which the library is enforcing.
As far as Hugh Stephens is concerned, if you put two attorneys in a room to discuss a non-trivial legal issue, you’ll likely generate three or more opinions about the correct resolution of that issue. That’s why we need judges, who will determine which side is right, and appellate judges if someone complains that the first judge made the wrong decisions. You eventually run out of appellate judges and the last decision stands.