In my last post, I presented what I consider a likely scenario for the future of research universities and their libraries. Eventually, most of them could either go away or devolve into focused research institutions, but will cease to be “research universities” in the sense we have used that term for the past century or so. They won’t attempt to cover the universe of knowledge, and their libraries–if such still exist–could become information centers focused exclusively on the needs of the moment with no regard for the future. A true research library cannot take into account merely the desiderata of current researchers who happen to be working or teaching at a given moment on a given campus, but must instead consider what’s important to save in the world and preserve that heritage for future generations who will be doing their own historical research. Along with the treasures, research libraries must also collect a lot of trash, because trash and treasure are curiously shifting terms, and historical researchers often discover hidden treasures in what people at the time would have considered unimportant ephemera.

The problem with doing this is the cost. Collecting material from all over the world and cataloging and preserving it is a very expensive endeavor, which is why relatively few such libraries might still be around in a few decades. The model of the research library collecting broadly and deeply spread during the middle decades of the 20th century, when material was relatively inexpensive. This was the period when state universities started to become research universities and began to build very impressive collections, collections, I should add, that allow current researchers to do research they never could have done otherwise, and that preserved our and others’ cultural heritages for future generations to discover. This model is sustainable only if there is concern for the future. However, a large amount of library funding, especially in many public universities, is at the whim of people who notoriously have no consideration of the future beyond their next election. Because politicians usually have no concern for future generations, they cut funding of the tools and institutions we need to build a better future and preserve our world heritage. Thus, when state budgets get tight, they cut educational funding, and this has been going on steadily for several decades. The recent recession just caused a more rapid drop in higher education funding than usual.

In an era of declining funding for higher education and a lack of concern for the future, including future research, we will probably be seeing more things like the “patron-driven acquisition” model discussed in this article. The article profiles a librarian who is supposedly “part of a wave of librarians testing a different and, they say, more efficient mechanism for purchasing library materials: patron-driven acquisition. The idea is that the library users help determine what to buy. For instance, a purchase decision might be based on how many times an e-book is accessed via an online catalog.” This is allegedly “a fundamentally rational method of acquisition” because it doesn’t use resources to purchase and preserve material nobody uses. As the librarian notes, “A big reason for [more libraries exploring this option ] is we’ve all experienced pretty significant budget cuts, and when the money gets tighter, it gets harder and harder to justify spending money on materials nobody wants.” The question not even considered is, how will we know what users 50 years from now will want from the collections we’re building now? That’s the sort of question that research libraries have to consider.

Depending on how extensive patron-driven acquisition is, it seems to me like a good idea. According to the article, Purdue has such a system, but only devotes a small percentage of their funding for it. “Purdue now spends 5 to 7 percent of its book-buying money this way, she says, and expects to increase that to 10 to 15 percent.” For getting contemporary scholars some of the material they need for their research, the method no doubt works quite well. However, if such an acquisitions model were extended to a significant portion of the budget, or even all the budget, and material acquisition was determined solely by current research needs and then discarded when their use drops, it would be difficult to consider such a library a research library because of a complete lack of regard for the needs of future researchers of the now discarded past. They will have to rely on the few strong research libraries left.

There are potential developments that would render this problem irrelevant. For example, the United States could develop a national digital library, as recently called for by Robert Darnton. The ideal digital library would be the universal library that scholars have been dreaming of for centuries. If every document contained in American libraries were digitized and fully available, there would be little need for research libraries of the sort we have now. In Darton’s words, “We can equip the smallest junior college in Alabama and the remotest high school in North Dakota with the greatest library the world has ever known.” This is remarkably similar rhetoric to H.G. Wells’ 1938 World Brain, in which, buoyed by a giddiness about the latest information technology, he predicted “microscopic libraries of record, in which a photograph of every important book and document in the world will be stowed away and made easily available for the inspection of the student….  The time is close at hand when any student, in any part of the world, will be able to sit with his projector in his own study at his or her convenience to examine any book, any document, in an exact replica” (54).

The irony of his prediction should be clear to any librarian. Because of copyright laws, we are not even allowed to microfilm or digitize his 60-year-old book and make it freely accessible to the world. Allowing the Alabama junior college student and the high school student in remote North Dakota to search inside books they can never read isn’t much help. There is good reason to speculate that we’ll have a national digital library with significant accessibility of content created within the last century when Disney decides Mickey Mouse is no longer profitable. Right now, even successful preservation efforts like Hathi, LOCKSS, and Portico can acquire and preserve digital content that will only be accessible after some rather unlikely trigger events.

Or there is the chance that the future will be the rather fanciful one conceived in this article: The User-Driven Purchase Give Away Library: A Thought Experiment, which takes patron-driven acquisition even further. The vision is that libraries buy only the books patrons want and then give them to the patrons. It’s predicated on the assumption that Google Books and the Hathi Trust will have most books digitized and preserved, and that everything else will be digitized and available at least as a license.

Possibly for the most popular content, but it seems unlikely that all the collections of major research libraries will be digitized in a decade, as this thought experiment envisions, unless we’re just talking about monographs published in the United States. If current trends persist, documents formerly printed on paper and sold to libraries could become digital content licensed and controlled in ways that will make it impossible to preserve for future research, whether by a research library or a national digital library. And even if Google Books and the Hathi trust have digital copies of most books, and even if those books are preserved, they will still not be accessible to many unless libraries can purchase and control them. And even if, somehow, the licensing agreements work out well for libraries and long term preservation as well as short term access are achieved, this is still only a portion of what research libraries actually collect.

Sustainable cooperative collection development would also mitigate the problem. Research libraries collect all sorts of material from all over the world, including parts of the world where print publishing is still the norm and might well be for decades. However, by divvying up the world and working collectively, consortia of research libraries could collect and preserve just about everything anyone in the future could possibly want. I suspect regional cooperative collection would be the best alternative, because access to much of this material would probably be restricted to a physical location for a long time to come. The ability to participate in these projects would distinguish the research libraries from others, but the great thing about this model is that regional consortia of even poorly funded research libraries could still develop a robust and diverse regional university library system upon which scholars everywhere in the region could depend.

These are all possible futures for research libraries, and not necessarily dark ones. I believe the ultimate goal of American libraries, as a system, is a universal library accessible to all, and to some extent we have achieved that for academic libraries through resource-sharing. While I hold out little hope for an Infotopia in which such a library exists digitally for everyone, it is still worth working towards, and ultimately is a measure of the success of all libraries, but especially research libraries. Nevertheless, the future of research libraries depends on the relationship between the future and research libraries. Research libraries that cease now to think about the needs of the future will cease to be research libraries in that future. Creating a national digital library, or a universal library of any sort, would be an appropriate goal for libraries collectively, and especially for research libraries collectively, but it requires thinking beyond the needs of the moment. It requires us to think about what scholars and students decades from now will either be thankful for what we have done, or regret what we failed to do.