Juan Carlos Ulate / Reuters

When Yale recently decided to relocate three-quarters of the books in its undergraduate library to create more study space, the students loudly protested. In a passionate op-ed in the Yale Daily News, one student accused the university librarian—who oversees 15 million books in Yale’s extensive library system—of failing to “understand the crucial relationship of books to education.” A sit-in, or rather a “browse-in,” was held in Bass Library to show the administration how college students still value the presence of books. Eventually the number of volumes that would remain was expanded, at the cost of reducing the number of proposed additional seats in a busy central location.

Little-noticed in this minor skirmish over the future of the library was a much bigger story about the changing relationship between college students and books. Buried in a slide deck about circulation statistics from Yale’s library was an unsettling fact: There has been a 64 percent decline in the number of books checked out by undergraduates from Bass Library over the past decade.

Yale’s experience is not at all unique—indeed, it is commonplace. University libraries across the country, and around the world, are seeing steady, and in many cases precipitous, declines in the use of the books on their shelves. The University of Virginia, one of our great public universities and an institution that openly shares detailed library circulation stats from the prior 20 years, is a good case study. College students at UVA checked out 238,000 books during the school year a decade ago; last year, that number had shrunk to just 60,000.

Before you tsk-tsk today’s kids for their lack of bookishness, note that the trend lines are sliding southward for graduate students and faculty members, too: down 61 percent and 46 percent, respectively, at UVA. Overall, across its entire network of libraries, UVA circulated 525,000 books during the 2007–08 school year, but last year there were only 188,000 loans—nearly 1,000 fewer books checked out a day. The Association of Research Libraries’ aggregated statistics show a steady decrease of the same proportion across its membership, even as student enrollment at these universities has grown substantially.

Maybe students aren’t checking the books out but are still consulting them regularly within the library? This also does not appear to be true. Many libraries also track such in-house uses, by tallying the books that need to be reshelved, and the trends are the same. At my library at Northeastern University, undergraduate circulations declined 50 percent from 2013 to 2017—before we decided to do our own book relocation—and our logged number of books removed from shelves but not checked out also dropped by half.

These stark statistics present a conundrum for those who care about libraries and books. At the same time that books increasingly lie dormant, library spaces themselves remain vibrant—Snell Library at Northeastern now receives well over 2 million visits a year—as retreats for focused study and dynamic collaboration, and as sites of an ever wider array of activities and forms of knowledge creation and expression, including, but also well beyond, the printed word. It should come as no surprise that library leadership, in moments of dispassionate assessment often augmented by hearing from students who have trouble finding seats during busy periods, would seek to rezone areas occupied by stacks for more individual and group work. Yet it often does come as an unwelcome surprise to many, especially those with a powerful emotional attachment to what libraries should look like and be.

What’s happening here is much more complicated than an imagined zero-sum game between the defenders of books and library futurists. The decline in the use of print books at universities relates to the kinds of books we read for scholarly pursuits rather than pure pleasure, the rise of ebooks and digital articles, and the changing environment of research. And it runs contrary to the experience of public libraries and bookstores, where print continues to thrive.

Unlike most public libraries, the libraries of colleges and universities have always been filled with an incredibly wide variety of books, including works of literature and nonfiction, but also bound scientific journals and other highly specialized periodicals, detailed reference works, and government documents—different books for different purposes. Although many of these volumes stand ready for immersive, cover-to-cover reading, others await rarer and often brief consultations, as part of a larger network of knowledge. Even many monographs, carefully and slowly written by scholars, see only very sporadic consultation, and it is not uncommon for the majority of college collections to be unused for a decade or more. This is as it should be: Research libraries exist to collect and preserve knowledge for the future as well as for the present, not to house just the latest and most popular works.

But there is a difference between preservation and access, and a significant difference, often unacknowledged, in the way we read books for research instead of pleasure. As the historian Michael O’Malley humorously summarized the nature of much scholarly reading and writing, “We learn to read books and articles quickly, under pressure, for the key points or for what we can use. But we write as if a learned gentleman of leisure sits in a paneled study, savoring every word.” Or as he more vividly described the research process, academics often approach books like “sous-chefs gutting a fish.”

With the rapidly growing number of books available online, that mode of slicing and dicing has largely become digital. Where students or faculty once pulled volumes off the shelf to scan a table of contents or index, grasp a thesis by reading an introduction, check a reference, or trace a footnote, today they consult the library’s swiftly expanding ebook collection (our library’s ebook collection has multiplied tenfold over the past decade), Google Books, or Amazon’s Look Inside. With each of these clicks, a print circulation or in-house use of a book is lost. UVA’s ebook downloads totaled 1.7 million in 2016, an order of magnitude larger than e-circulations a decade ago. Our numbers at Northeastern are almost identical, as scholars have become comfortable with the use of digital books for many purposes.

I’ve seen my own book usage change over time. When I was a graduate student studying Victorian history at Yale, the university’s towering collection in Sterling Library, next door to Bass (then called Cross Campus Library), allowed me to find and leaf through relevant books easily. Now almost all of the texts I consulted for my dissertation are available online in repositories such as HathiTrust, which stores digitized books from research libraries, many of them freely available for download since they were published before 1924, the cutoff for public-domain works. If I were doing the same scholarly project today, I would likely check out only a small subset of books that I needed to pay careful attention to, and annotate others digitally in my PDF reader.

The decline in print circulation also coincides with the increasing dominance of the article over the monograph, and the availability of most articles online. In many fields, we now have the equivalent of Spotify for research: vast databases that help scholars search millions of articles and connect them—often through highly restrictive and increasingly unsustainable subscriptions, but that is another story—instantly to digital copies. (There is also a Napster for research articles, of which we shall not speak.) Very few natural and social scientists continue to consult bound volumes of journals in their field, especially issues that are more than a few years old. UVA recorded nearly 3 million e-journal downloads in 2016, a massive and growing number that is typical of most universities.

In addition, the nature of scholarship is also changing, still with significant reading and writing, of course, but also involving the use and processing of data in a wide array of disciplines. To serve these emerging needs, Northeastern University Library has added full-time specialists in data visualization and systematic review (the process of synthesizing, statistically, exhaustive research from multiple studies), and an entire division dedicated to new forms of digital scholarship.

Our research library, like many others, has also seen a surge in group work rather than the solitary pursuit of the canonical research paper. More classes are assigning team-based projects instead of individual essays, as many urgent problems, such as climate change, call for large-scale interdisciplinary work and multiple perspectives. University libraries have correspondingly seen reservations for collaboration spaces surge. Last year, we had a record 100,000 hours of group-room bookings in our library, meaning that these spaces were occupied constantly from 8 a.m. to midnight.

At the same time—and perhaps this is one of the feel-good stories related to physical collections—there is an increasing use of archives. Many students still find the direct encounter with primary sources thrilling, and instructors and library staff have found creative ways for them to use these special collections. We have doubled our archival holdings in the past five years, focusing on Boston-related materials such as our recent acquisition of millions of photographs and negatives from The Boston Globe, and have greatly expanded our program of teaching with these artifacts.

A positive way of looking at these changes is that we are witnessing a Great Sorting within the library, a matching of different kinds of scholarly uses with the right media, formats, and locations. Books that are in high demand; or that benefit from physical manifestations, such as art books and musical scores; or that are rare or require careful, full engagement, might be better off in centralized places on campus. But multiple copies of common books, those that can be consulted quickly online or are needed only once a decade, or that are now largely replaced by digital forms, can be stored off site and made available quickly on demand, which reduces costs for libraries and also allows them to more easily share books among institutions in a network. Importantly, this also closes the gap between elite institutions such as Yale and the much larger number of colleges with more modest collections.

These trends around research collections are likely to continue. A small number of regional pools of books at a monumental scale—tens of millions of books from scores of universities working together—are already envisioned in the United States, which will ensure preservation and access for future generations and effectively act as gigantic shared libraries, or what David Prosser, the executive director of Research Libraries UK, has called “collective collections.” “Print books are historical artefacts … but some are more valuable artefacts than others,” Prosser has argued. “No library can be completely universal and decisions need to be made about what to collect and where to store material. By looking at collections collectively we can better serve the needs of readers, ensuring that what we have is well looked after (and yes, sometimes that means in ‘remote-storage’).”

Unfortunately, more troubling factors are also at work in the decline of print books within colleges. Statistics show that today’s undergraduates have read fewer books before they arrive on campus than in prior decades, and just placing students in an environment with more books is unlikely to turn that around. (The time to acquire the reading bug is much earlier than freshman year.) And while correlation does not equal causation, it is all too conspicuous that we reached Peak Book in universities just before the iPhone came out. Part of this story is undoubtedly about the proliferation of electronic devices that are consuming the attention once devoted to books.

The sharp decrease in the circulation of books also obviously coincides with the Great Recession and with the steady decline of humanities majors, as students have shifted from literature, philosophy, and history to STEM disciplines—from fields centered on the book to fields that emphasize the article.

When I tweeted about this under-discussed decline in the use of print books in universities, several respondents wondered if, regardless of circulation statistics, we should keep an ample number of books in the library for their beneficial ambience. Even if books are ignored by undergraduates, maybe just having them around will indirectly contribute to learning. If books are becoming wallpaper, they are rather nice wallpaper, surrounding students with deep learning and with some helpful sound-deadening characteristics to boot. If that helps students get into the right mind-set in a quiet, contemplative space, so be it. Maybe they will be more productive, get away from their distracting devices, and perhaps serendipitously discover a book or two along the way.

You can certainly see this theory at work in new library designs in which the number of volumes is more quietly reduced than at Yale, with books lining the walls of study spaces but not jutting out perpendicularly like the old, high-capacity stacks, so as to leave most of the floor open for tables, chairs, and spaces for group work. Perhaps that is the right approach, the right compromise, for some schools and students. Of course, you can also find students who love spaces without books, or who work better with some background noise—alas, whenever you discuss these matters, all students tend to generalize from the study space that works for themselves.

But there is another future that these statistics and our nostalgic reaction to them might produce: the research library as a Disneyland of books, with banker’s lamps and never-cracked spines providing the suggestion of, but not the true interaction with, knowledge old and new. As beautiful as those libraries appear—and I, too, find myself unconsciously responding to such surroundings, having grown up studying in them—we should beware the peril of books as glorified wallpaper. The value of books, after all, is what lies beneath their covers, as lovely as those covers may be.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.