English: The library of Sofia University. Imag...
English: The library of Sofia University. Image by Anastas Tarpanov at http://atphoto.bg (Photo credit: Wikipedia)

In Part 1 of this post, I sketched out the increasingly complex environment in which libraries now function and gave examples of formerly-straightforward questions that no longer have obvious answers. In this environment, the role and function of the library have become much less self-evident than they were throughout the centuries during which the answers to those questions seemed obvious. Even in an information environment characterized by serious structural turmoil, however, it seems clear that all libraries are still expected to fulfill two fundamental functions for their sponsoring institutions.

First, with their general or circulating collections and public services, they support the day-to-day intellectual activities of their sponsoring institutions. A municipality establishes a public library in order to help its citizens accomplish their private and public tasks: schoolwork, personal research, recreational reading, etc.; a college establishes an academic library to support the learning, teaching, and research that it expects of its community; a corporation creates a library in order to facilitate the work of its employees; etc.

Second, with what we traditionally call “special collections,” libraries create some kind of permanent (or at least functionally permanent) documentary record. The importance and centrality of this function will vary greatly from library to library, depending on the needs of the sponsoring institution—a fact that can lead to controversy if there’s a gap between what the host institution desires (and is willing to pay for) and what librarians believe the institution should desire and pay for.

How do different libraries balance and provide for these two basic functions in light of the increasingly complex and unpredictable information environment in which they do their work?

It depends. I propose the following basic taxonomy of library types and functions.

Library of Cultural HeritageA library that is charged with preserving the cultural and intellectual heritage of an entire country, region, or (perhaps) ethnic group. While LCHs may offer indispensable resources to researchers, they are concerned primarily with preservation and curation, and secondarily with facilitating research. There may be no circulating collection at all, or circulation may be highly restricted so as to maximize the collection’s stability and permanence. In an LCH, there will be little if any differentiation between general and special collections, since the whole collection has more in common with what most large libraries would designate as special collections than with a typical research library’s circulating collection. An LCH will tend to create digital collections (by making and cataloging images of documents and realia from its holdings) rather than purchase access to digital collections created by others. (Examples: Library of Congress; British Library)

Research Library of RecordA library, typically funded by a large university or in some cases by a large municipality, with a broadly inclusive and relatively stable circulating collection. Its goal is to meet the needs of a broad range of students and scholars by documenting, as thoroughly as possible, the intellectual heritage of its host culture. An RLR will be characterized by a very generous collection budget and staffing that includes bibliographic experts in a variety of disciplines, many of them in the humanities. Access to the circulating collection may be somewhat constrained (limited to enrolled students and faculty, for example) even if the sponsoring institution is a public entity. A significant portion of its working collection will consist of online access to hosted digital content. An RLR will invariably have deep special collections as well as a comprehensive circulating collection. (Examples: Harvard University, Oxford University, New York Public Library, Princeton University)

Research Library of UtilityA large and reasonably comprehensive library that is nevertheless characterized by a changeable circulating collection to which books are added and from which they are withdrawn according to the expressed and demonstrated needs of the curriculum, physical space limitations, and the research agendas of its constituency. Although the collection’s makeup will reflect the curriculum and research foci of the institution, it will also include selected documents from a broad range of academic fields. Like an RLR, an RLU will provide access to extensive hosted online collections and will include a special collections area, one which is selectively deep according to regional interests and other areas of specialty. The majority of medium-to-large university libraries would fall under this category. (Examples: University of Michigan, University of Denver, Texas A&M University)

Academic Library of UtilityAn academic library with a targeted, dynamic collection. The focus of this library is specifically to facilitate day-to-day study, teaching, and research on campus. The collection is less a permanent fund of content than it is a node for the accessing of content, some of which will be “owned” by the library but much of which is only made visible by the library and “purchased” upon use or request. (In some cases, “purchase” might mean “acquire access permanently for the entire constituent community,” while in others it means “acquire access either temporarily for the community or permanently for the requestor.”) Here the term “collection” itself becomes ambiguous. (Examples: community college libraries, liberal-arts college libraries, school libraries)

Public Library – With the exception of a few public libraries that are also RLRs (such as the New York Public Library), public libraries almost exclusively serve communities defined by political geography rather than academic mission. Rarely is it their mission to support in-depth research; instead, they usually support recreational reading, casual or personal research, primary and secondary education programs, and community space needs.

Restricted LibraryThe working libraries of government agencies, commercial companies, nonprofit organizations, hospitals, and research institutes would all fall under this designation. In the past, the library profession has used the term “special library” to refer to such libraries, but the word “special” may be too vague to be useful – it has come to have a specific meaning to librarians, but communicates little to those outside the profession. In practice, what seem to define all of these libraries are the following characteristics:

  1. Each holds a collection and offers services that are tightly focused on the day-to-day working needs of the organization’s members or employees;
  2. The library’s purpose is to further the corporate goals of a sponsoring organization;
  3. The library is open only to those formally affiliated with the organization.

Thus, what I’m calling a Restricted Library is characterized primarily by restriction in three senses: in its collecting scope, in its mission, and in the access it offers.

Two Caveats:

First: as difficult as it may be to do so in practice, these categories should be decoupled from considerations of prestige. There is no more or less honor in being an excellent Research Library of Record than in being an excellent Academic Library of Utility. Excellence should be measured by each library’s success at doing what its constituents have charged it to do.

Second: this rubric of classification deliberately ignores the important organizational category of “archive.” Why? Because an archive is not a library. Unlike all of the library types in the taxonomy above, an archive serves little or no access function, nor is its purpose to facilitate the day-to-day functioning of its corporate parent; its primary mission is to preserve and safeguard documents that contain important, sensitive, or legally-required information. This is not to say that an archive necessarily prevents all access—some, such as the National Archive of the United States, offer researchers more access, while others, such as the corporate archives of privately-held companies, offer little (if any) access to the public. But in all cases, an archive’s primary purpose is to preserve a record for purposes of corporate memory, rather than to promulgate public access to information resources. (The special collections of a research library should not be confused with an archive in this sense.)

Now, the big question: why bother to classify libraries in this way?

Speaking from inside the library world, I think the need for this kind of taxonomy becomes clear every time one of us says “We should/shouldn’t do X because we’re a library and that’s what libraries do.” I can’t count the number of times I’ve watched librarians and staff argue past each other because they don’t share a common understanding of what kind of library they work in. The radically changed scholarly environment in which we work makes it increasingly important that our strategic conversations be built on an explicitly shared understanding of library taxonomy—while recognizing that every library is, in some important ways, unique.

I think vendors tend to be more conscious of the fundamental differences between categories of library; they have a very strong financial incentive to understand their customers, whereas the incentives for librarians to understand their own libraries is more… mixed. (There’s a topic for another post.) But perhaps a simple system like this will be helpful to vendors and publishers as well.

I have no illusions that everyone will agree with the taxonomy proposed above. If a large enough number of people agree that it’s potentially useful, then I count on feedback from Scholarly Kitchen readers and others to help revise and refine it.

Enhanced by Zemanta
Rick Anderson

Rick Anderson

Rick Anderson is Associate Dean for Collections and Scholarly Communication in the J. Willard Marriott Library at the University of Utah. He speaks and writes regularly on issues related to libraries, scholarly communication, and higher education, and has served as president of NASIG and of the Society for Scholarly Publishing.

View All Posts by Rick Anderson


21 Thoughts on "(Re)Defining the Library, Part 2: How?"

And of course, the personal library, e.g. Papers for the Mac. In meeting the needs of users, it is important to support facile and reliable interfaces to the personal digital libraries that many of us now maintain. Unfortunately the library is often the intermediary left out and users must interface directly with diverse publishers, each with their own standards, GUIs, etc.

How does the collection of “rare books” fit into your categorization? One can imagine that almost all categories might have occasion to purchase an item that would be considered rare from time to time, and no doubt they dominate in special collections. The major rare book collections seem to exist, though, in your RLR category. Do they regard collecting rare books as an end in itself, or do they try to justify their collections in relation to specific strengths of the academic program? Sometimes alumni or other “friends of the library” will donate their personal collections to libraries, so it is not all a matter of a library choosing to seek out such books. I think of the fabulous Robert Taylor collection of British literature at Princeton, for example. (I had the wonderful experience of taking a course about rare book collecting taught by Taylor himself in the very room of the library where his collection was housed.)

Sandy, these are good questions – and as you might expect, the answers are complicated. Actually, the answer to the first part of your question is simple: rare book collections are invariably part (though almost never the entirety) of the special collections area in a library. The way they fit into my proposed taxonomy is as a subset of content within each library.

The answers to your subsequent questions are more complicated. Very often these collections are built deliberately around themes that are relevant to the library’s sponsoring institution. For example, the University of North Carolina at Greensboro (which used to be called Greensboro Women’s College) has built an impressive collection of detective novels by women authors or featuring women protagonists. Similarly, the University of Nevada, Reno, has extensive special collections related to Basque studies. But sometimes a library holds a collection of rare books not because those books were deliberately sought out and the collection carefully built around a particular area of interest to the institution, but because they were given to the library by a valued friend or significant figure to whom saying “no, thanks” would have been unwise. There are some fascinating and even downright puzzling anomalies in just about every library’s special collections.

Good point, Bill. Can you offer some other examples of this kind of library?

In Portland, we have tool libraries where we can check out various yard & home repair implements. These libraries would fit into this category.

Except that in the case of a tool library, the word “library” applies in only the loosest and most colloquial sense (whereas it applies more traditionally in the case of a genealogy library). I think in the context of my proposed taxonomy, it probably makes sense to stick with the more traditional definition of a library as a collection of information resources rather than of other stuff.

My thought on “special purpose libraries” is that there probably exist libraries that are very restricted in the materials in their collection but are not restricted in the user population (thus failing a criterion in the “restricted” library definition). My impression is that the Portland tool library is a 21st century institutional replacement for the mid-20th century neighborly “Can I borrow your hedge clippers for an afternoon?” Certainly, the Portland tool library has a very distant relation to scholarly publishing. Etymologically, the term library should be restricted to books. That some people choose to apply the term to collections that do not include books is a consequence of freedom of speech. On the other hand, some non-book collections have supported scholarly activities, and there may be a loose relation between publishing and propagating. In 1962-63 I worked in what I was told was the largest fruit fly library in the world. I maintained various species, subspecies, and varieties in the genus Drosophila. From time to time, the library filled requests from researchers around the world.

Looking at a library from a user operation standpoint, a library is characterized by the organized structure of its collection that allows rapid retrieval of any item if you know its “call number” and by a set of catalogs (indexes) that allow learning the call number if you know the normal name of the item (author, title in the case of books) or by other characteristics (subject index). Non-literary “libraries” (seed libraries, virus libraries, bacterial libraries,…) seem to share these attributes of providing user access to the desired material.


  • Bill Everett
  • Sep 13, 2012, 9:47 AM

By no means do I intend to restrict anyone’s free-speech rights. If people find it helpful to refer to a collection of tools, or fruit flies, or anything else as a “library,” then they should feel free to do so. Heck, you have the right call a pile of waffles a “library” if you’d like. I’m only suggesting that for the purposes of this taxonomy and in the context of this conversation, it probably makes sense to assume that by “library” we mean something more along the lines of a collection of information resources and services.

  • Rick Anderson
  • Sep 13, 2012, 10:38 AM

The Portland Tool Library does provide reference services in advice on how to tackle projects, they have a lending/circulation program along with freely distributed guides on known or typical projects undertaken. Much like early Benjamin Franklin “libraries” that often extended to many objects beyond books. In the history of public libraries, many loaned works of art and other cultural artifacts. Just because these are objectsmake it no less a “library” if you’re looking at changing that definition or are you just proposing categorization changes here?

  • Jill Emery
  • Sep 13, 2012, 11:30 AM

What I’m suggesting is that we stick to a more narrow definition of “library” for the specific purposes of this proposed taxonomy. Technically the word “library” implies “books” – the word literally means “a collection of books.” Over time, most libraries have come to include lots of documents that aren’t books, and to offer services that aren’t strictly about information access. But the organizations included in my proposed taxonomy are all organized around a core mission of providing access to information resources. A collection of hammers and hedge trimmers doesn’t fit the definition of a “library” in any etymological or traditional sense (even if it also offers instructions on how to use the tools, and even if it echoes a service offered long ago by Benjamin Franklin). Nor does a collection of fruit flies. That doesn’t mean there’s anything wrong with calling those collections “libraries”—I’m only suggesting that if my taxonomy is expanded to include all organizations and services that could conceivably be called “libraries,” it might quickly become useless for its intended purpose, which is to help us think about the differences between “libraries” in the more traditionally-understood sense.

  • Rick Anderson
  • Sep 13, 2012, 12:04 PM

Rick, I think Bill has a point here. Places like the Vesterheim Norwegian-American Museum have book collections and archives. It is a cultural heritage institution to be sure, but I think they see themselves as much a research center as a Norwegian heritage museum. The Family History Library would probably fit best as a LCH in your taxonomy, but you would have to bend the rules. It is not heritage they are preserving; the book collection doesn’t circulate, but the films do; it does have a stable collection and it is intended to be permanent; the focus is very narrow and while they are digitizing from their own holdings they are also acquiring new digital content. Funding is private, but the library is open to the public. In fact, they are working hard to make it all accessible via the Internet.

I think your taxonomy needs either another category or some broader definitions.

  • Vern Reed
  • Sep 13, 2012, 9:14 PM

Vern, you may be right. The challenge in a case like this is that no matter how the taxonomy is formulated, there will always be particular libraries that fit no category very well, or that straddle them.

  • Rick Anderson
  • Sep 14, 2012, 9:06 AM

Great post Rick! One of the best commentaries on the various roles that libraries play that I have ever read.

Nice post, Rick. Mission drift among academic libraries is problematic and often prevents us from better-meeting our constituencies and institutions needs. The challenge, as you point out, is getting people to understand that meeting the needs of users of an ALU is no less important than meeting the needs of a user at a RLR.

One thing to consider, though, is it seems like there needs to be another category either between RLU and ALU or below ALU. Your description of the ALU collections seems to categorize comprehensives and liberal arts schools nicely, but what I think would differentiate them from community college and school libraries is the fact that they almost always have special collections. These will (generally) be of a different scope and scale compared to RLU’s, but they may be no-less significant to the cultural and scholarly record.

Good point, Doug. Can you suggest a way of sharpening that distinction in the taxonomy?

Nice post … Some random things that occurred as I read …

1. Your own library is an RLU?
2. How do you see the distribution (staying in the US to make it easier)? How many RLRs are there, i wonder? It would be interesting to think how the distribution might change over time as there is some shift in the configuration of HE. RLUs becoming ALUs. Or maybe another category is needed
3. I guess that while many National Libraries are focused on the cultural/scholarly record of their own country, the BL and some other national libraries aspire also to be RLRs.
4. I also wondered what sort of flattening you would see when you go outside the US, as most countries have a much smaller HE system. I was prompted to ask about particular countries but that might get into tricky conversations 🙂 About whether particular institutions are actually RLRs, RLUs, or ALUs. Of course, in some cases a part of the response might be that responsibilities are discharged within a group or system of provision.
5. Which prompts … You don’t talk much about the systemwide responsibilities for the scholarly record that various libraries might take on, even as they don’t see it as central to their own mission. Look at the West project for example one of whose benefits might be to provide a mechanism for the diffusion of responsibilities among a wider group, while acknowledging different levels of contribution? Other examples occur.
6. Where would IRLAs fit? http://irla.lindahall.org/ Sometimes restricted maybe, sometimes not.
7. It does have an academic flavor. I guess some might argue for more differentiation in the Public category?

Hi, Lorcan —

1. Yes.
2. It seems to me that the vast majority of American university libraries are RLUs, with only a small handful of RLRs. Not all of them would agree with me, of course, nor would everyone agree on where I draw the line. We’d probably all agree that Harvard’s library is in a different class from the U of Utah’s. But is the University of Michigan library more like Harvard’s or more like Utah’s? How about U Washington? The lines are going to be fuzzy.
3. This may well be true, and in some cases they may be both simultaneously. I should also acknowledge that the BL and LC are not really equivalent insitutions — LC is not really a national library, but truly is the library of Congress and its mission and policies reflect that. It does, however, also serve an explicitly pseudo-archival function at the national level. (By “pseudo-archival” I mean that it’s not technically an archives, but is intended to serve as a sort of archive of the country’s intellectual heritage in addition to serving the research needs of Congress.)
4. Good point — I should probably have acknowledged that my proposed taxonomy only really works (if at all) for large and fully-developed countries. Probably not even for all of them, especially as you get out of the Western hemisphere and into large countries with very different traditions of education and librarianship.
5. That’s true; I really didn’t adress that issue. It’s probably a good topic for a separate post. Data management in libraries is another such topic.
6. I think I would lump IRLAs in with Restricted Libraries, since they are almost invariably connected with private organizations, have tightly focused collections, and tend to offer pretty limited access to the general public. Thus they fit the three “restriction” criteria I laid out in my post. (Unless I fundamentally misunderstand them, which is possible.)
7. I was actually a little selfconscious about that, and am now wondering whether it might be useful to create two separate taxonomies: one for academic libraries and another for publics.

You are also denying a space for completely born digital libraries such as Digital Public Library, California Digital Library, Texas Digital Library, & the Internet Archive ( some archival material but with their book program–becoming more digital lending oriented). These are all new cultural entities that produce both current & archival born digital artifacts. Most are collaborative entities that exist beyond a single institution. Born digital libraries may need their own category.

Yes, maybe so. It’s complicated by the fact that (as you point out) these are often second-order entities that exist only as collaborative projects of primary institutions. At what point does a collaborative digital archive become a separate entity that can usefully be considered a “library” separate from the libraries whose digital collections make up its content? That’s a tough one.

Rick, your comment in this post about archives reminded me of an exception that may prove your rule. The Daily Telegraph recently ran an article http://www.telegraph.co.uk/history/9441669/We-must-draw-on-our-historians.html about the refurbished library in the Foreign and Commonwealth Office in London, UK, which combines printed materials with archives on open shelves, intended primarily as a working tool for the FCO’s historians – although Foreign Secretary William Hague is quoted as saying that the historians “were languishing in a basement and now the light is shining on their books. It is intended to be a signal to the whole Foreign Office to use them…” As the item is about the need for history to inform modern policy-making, I think he means making good use of the books and papers as well as of the historians themselves.

There’s an interesting history to the FCO library space. After the creation of the FCO in the late 1960s it housed first the legal library and then the main library FCO collection, which although obviously a Restricted Library also had elements of your RLR about it and would occasionally admit researchers as a library of last resort. Five years ago the historical collection passed to a college of London University where it has become an RLR (minus the online element). An important part of the collection’s value lies in the selection of materials that was made by the Office’s librarians at various dates since the founding of the Office in 1782 – or perhaps even earlier for a few of the original items inherited from an earlier owner. in 2007 the remaining collection at the FCO then became very much a Restricted Library with the characteristics you describe, drawing extensively on electronic resources. Now William Hague’s initiative has promoted the historians and their collection into the refurbished former library space, combining published and archive materials as a research resource. It looks like a library, which is I guess why it’s being called one!

This I think is an interesting example of a new way of thinking of archives as a current, active element within a hybrid form of library. The new FCO archive has an active access function and it supports the day to day functioning of the corporate parent. It probably does contain sensitive material, and certainly does contain material that is important to the FCO, even though much of this is probably also held at TNA (The National Archives). Its existence doesn’t negate your taxonomy (phew) but does suggest that some libraries will not fit easily into a single category.

Some of the current radical thinking about libraries will undoubtedly produce new ways of working that will further stretch the boundaries and hint at new categories or sub-categories. A lot of that thinking is being driven by corporate management and accountants rather than librarians, because lately they tend to have a different idea of “what libraries do” (or ought to do) and we are better at arguing among ourselves than at showing those corporate suits the value of what libraries do.

Leave a Comment