PART I: A Continuing Discussion on Research Libraries in the 21st Century


The information landscape of early twenty-first century higher education is characterized by ubiquitous, digitized, indexed online access to content. Researchers and students begin, and often end, their quest for information online. Results of research can be and increasingly are published without traditional publishers or conventional formats. Access to these results, and to the cultural and scientific record that constitutes the primary resource base for research and teaching, is, however, narrowed by the increasingly exclusive use of licensing instead of selling. This is but one contemporary paradox among many.

What are the critical functions of the research library in this changing landscape? How should we be rethinking the research library in a dynamic, swiftly changing landscape dominated by digital technology? To explore this question, CLIR convened a meeting of librarians, publishers, faculty members, and information technology specialists on February 27, 2008, in Washington, D.C. To prepare for the discussion, CLIR invited eight of the participants to share their perspectives on the future library in brief essays. The essays were circulated before the meeting and are presented in part II of this volume.

Part I of this report begins with an overview and a summary of key meeting themes, or topic threads. The next section summarizes participants’ views on what a reconceived research library in the 21st century might look like and what its core functions may be. Next is a discussion of the key challenges to achieving that vision. Part I ends with recommendations to provosts, presidents, deans, faculty members, and library directors about how to realize a reconceived library.

Prologue to a Fundamental Rethinking:
Context and Topic Threads

The breadth of discussion underscored that the future of the research library cannot be considered apart from the future of the academy as a whole. Trends that will influence this future are already evident; foremost among them are a distinct rise in cross-disciplinary research and collaborative projects in the humanities as well as in the sciences, and a concomitant increase in research that involves scholars as well as graduate students and undergraduates.

Researchers are asking new questions and are developing new methodological approaches and intellectual strategies. These methods may entail new models of scholarly communication-for example, a greater reliance on data sets and multimedia presentations. This, in turn, has profound consequences for academic publications: it is difficult to imagine traditional printed books and journals adequately capturing these novel approaches. With the predicted rise in new forms of scholarship, the promotion-and-tenure process, which favors print publications, especially in the humanities, will need to be rethought. As these methods of communication change, the procedures, skills, and expertise that libraries need to manage them will change as well. As cross-disciplinary work increases, it will be necessary to reassess the organization of higher education-its departments, schools, and centers. The research library in the 21st century will thus be profoundly influenced by the transformation of scholarship and research as well as by changes in the traditional organizational structures of a university.

The following topic threads reveal a range of perspectives and questions on the transformation of libraries. Although they are presented as discrete topics, the discussions reflected their interdependence.

Culture of libraries: inhibiting change? Libraries are by nature conserving institutions, and this is what we entrust them to do. But how do we balance a conservative, risk-averse nature with the need to respond to a changing environment? We need to think more deeply about what we want our institutions to conserve. Change will require collective action, and such action will be impossible unless people are closer in spirit with regard to risk. We need to experiment and develop opportunities for work in new sectors or new alignments with different organizations. There is a cost to not taking risk-a danger that libraries will become stuck in a niche that becomes smaller and smaller. As one participant observed, “We could be eradicated in the early stages if we are not a player.”

New alliances with students. The Web allows us more autonomy as information creators and consumers. Fewer students today have direct encounters with the library; consequently, they are unaware of a vast amount of useful scholarly information or how to find it.

The library has an opportunity to engage students in new ways-a point that Stephen Nichols explores in his essay “Co-teaching: The Library and Me.” The undergraduate population is key to the future working of research library. Faculty members and librarians need to involve undergraduates in using the data sets that are being brought online. Whether exposing students to research materials in the stacks, engaging them in the use of new online data sets, or supporting them in other ways, the library has a potentially huge role in undergraduate and graduate teaching and learning-a function that is tightly aligned with the mission of the university.

Redefining the library workforce. We tend to think of technology as the enabling factor in the new information environment, but the human aspect is just as important. “Technology needs to be addressed as something that enables human abilities for research and learning,” observed one participant. People will enable the collaboration with other departments, organizations, and professions that will be critical for the 21st-century library.

With this observation comes a serious challenge: How do we repopulate organizations-universities or libraries-in which half of the workforce will retire in the next decade? Libraries must think about staffing in new ways. Hiring only staff with the master’s of library and information science (MLIS) degree is unlikely to bring in the breadth of skill and experience that is needed. Nevertheless, the relative merits of the MLIS should not be our sole focus, cautioned one participant. “We have not been able to translate new ideas, such as co-laboratory or curation, into our normal workflows. We need to think about how we allocate resources. We need to take the expertise we have and think about new positions and new ways to connect with faculty.”

We need new career paths for people who want to work in academic libraries, and we need the means to support them. Some libraries use short-term project funding to hire staff with the needed new skills, but find it difficult to retain these talented individuals once a project ends. We also need to accommodate the work styles of a distributed staff. Many who are drawn to work in a library may not wish to be tied to one location. Accommodating a distributed staff would also allow libraries to consider sharing positions with other institutions and to approach certain problems collectively.

New kinds of engagement with faculty. Digital scholarship provides new opportunities for collaboration between faculty and librarians. Libraries have faculty-like expertise that is valuable in many areas of scholarship-for example, in documentary and edition work. Libraries also have great potential to contribute to data curation. At Johns Hopkins University, for example, the library’s digital curators create the intellectual data model that enables digital objects to be organized and programmed.

At the same time, libraries will need to become more aware of the data problems within various disciplines and what is being done to solve them. This means, for example, that libraries should be involved in the experimental or developmental stages of such work and should help shape solutions. Extensive work in metadata, for example, is now going on outside of libraries. Libraries would also benefit from greater awareness of the protocol work going on in a Web-scaled world. “Libraries have often created specialized tools for access to metadata and the associated content, such as Z39.50 and OAI-PMH,” noted one participant. “If instead, we always used Web-wide tools and protocols, we could let the Web do what it does best-massive-scale, pervasive tools, just enough complexity to get the job done-while we focus our scarce resources on well-focused, open-sourced, agile, lightweight, loosely coupled services that would make the work of our local communities more efficient and effective.”

In the future, data curation will not be static. It will have to encompass the informatics that go into making data accessible in new ways on a continuous basis. It will not mean bringing data to a state of normalcy and then allowing those data to be used. It will require understanding and investing in the research that is going into the use of that data over time. “We need to pay attention to what people have already done because we are going to have to help people use that data in different ways that creators don’t care about,” said one participant. How will libraries embed themselves on that scale?

Identifying the library’s competitive advantage. Several participants noted the keen-almost “Darwinian”-competition for resources within institutions. Can we move from the need to survive to something better? Can we change how we go about our work, rather than just continue to seek more money?

Unless libraries take action, participants cautioned, they risk being left with responsibility for low-margin services that no one else (including the commercial world) wants to provide. An analogy is the U.S. Postal Service (USPS). Its innovative, high-margin services, such as international and overnight delivery, have been taken over by private firms, leaving the USPS largely with its lowest-margin-of-return function: domestic mail delivery. When the broad digital availability of books erodes the comparative advantage of large research collections, where will the library’s comparative advantage lie? As Paul Courant notes in his essay “The Future of the Library in the Research University,” that advantage could be found in ensuring the “bibliographic” integrity of digital scholarly materials or in developing new tools and services that exploit information technology. During the discussion, he cautioned that in the digital world, universities must think carefully about getting into the business of preserving, migrating, curating, taking care of, and buying more servers for all the world’s information, since many of these functions can be outsourced. Two areas in which the library has an interest and can deliver unique value are advocating for preservation and setting standards for quality control. The library should take responsibility for ensuring that mechanisms for preserving digital records exist, and that those mechanisms function as efficiently as possible.

The library’s relationship with the commercial sector. There was vigorous debate over the nature of the research library’s relationship with commercial entities. Several participants noted that business models and return on investment, rather than the public good, could drive decisions that are at odds with the university’s fundamental mission. The library has a “social contract” with the university and cannot abdicate responsibility for basic roles, such as keeping primary data. As an example of what is at stake, one participant noted the closure of the Environmental Protection Agency research libraries under the guise of reorganization. In the wake of the closures, which began in 2006, a significant body of scientific data and information has become unavailable to researchers and the public.

Other participants argued that partnerships with commercial entities will be necessary-and indeed are already common-for libraries. At the very least, libraries will have to “co-evolve” with the parts of the commercial sector that license information so that libraries can ensure ongoing access for scholars. It is also clear that the library community alone does not have the capacity to do software engineering at the level needed in increasingly complex Web environments. Some of the most interesting areas for future library work are being developed by commercial entities. If libraries fail to partner with commercial entities to provide new services, will libraries fall behind and become irrelevant? We must be careful not to focus simply on identifying things libraries do that others don’t. University presses made the mistake of attempting to carve out a niche that they alone could fill, and over time this has diminished their function.

There was extensive discussion of the library’s relationship with publishers. Libraries must position themselves to retain their intellectual advantage. As one participant noted, “Any functions that don’t require human intellect will default to commercial interests.”

Expanding the idea of collaboration and collective action. The library’s traditional position at the center of campus reflects its function as a crossroads for intellectual activity. Although students, teachers, and researchers increasingly obtain information electronically, the library retains that time-honored position. And in fact, the library’s role has become more compelling, given that many of the current challenges in scholarly communication stem from the need to resolve cross-discipline issues in sharing digital resources. Libraries are uniquely situated to work at the nexus of disciplines. But doing this work effectively requires new types of outreach and collaboration. “We need new alignments for moving into new sectors,” observed one participant. The definition of “community” must be broadened. Libraries must have the capacity to engage in new ways with the disciplines and to interact more broadly with faculty, publishers, and even commercial interests.1

Libraries could play an essential role in helping organize information in such fields as bioinformatics; they could also help create data structures that favor interoperability among disciplines or institutions. Collective action will also be needed to resolve issues relating to copyright law. Sometimes, collaboration makes sense for budgetary reasons-for example, to save money on housing collections through the creation of shared print repositories. In the future, the economic viability of libraries is likely to increasingly depend on their ability to forge alliances with the larger community. At the same time, while the potential advantages are numerous, participants acknowledged that there is often a tension between collaboration and self-interest, and that more models for effective collaboration are needed.

Need for experimentation. Participants expressed much enthusiasm about the library as a laboratory-or co-laboratory-for digital scholarship, a theme that Richard Luce explores in his essay “A New Value Equation Challenge: The Emergence of e-Research and Roles for Research Libraries.” Increasingly, humanist scholars are creating work with dynamic processes that will need a home. Could the library provide such a home, and might this be a way for librarians to connect with faculty? This would require rethinking how resources are allocated. It can be difficult to convince provosts to hire new staff simply because they are needed for experiments. It can be harder still to sustain momentum and retain innovative staff with short-term project funding. It can also be challenging to bring people together in an organization that doesn’t foster innovation. Agreeing to solve collective problems collectively could free up funds for innovation. Can we convince provosts to take a small risk to generate potentially great success?

The fragility of academic publishing. If scholarship is becoming more collaborative in the digital world, at what point is it fixed? Is publication secondary evidence, and is the process somewhere in the middle? What is the role of the commercial publisher in such a scenario? The connection between the university press and the university is fragile. Many publishers are carrying on in old modes; for example, they are focused on converting print to digital. But the potential utility of collections goes beyond just having them in digital form; it extends to the ability to layer intellectual value on top of raw text. We can build new types of research environments with digital publications, and create knowledge that can be reused and reconstituted. Publishers need to add value: the academy demands this. Participants debated whether publishers or academic departments are better positioned to add the intellectual layer even as they acknowledged that publishers, in response to changing markets, seem to be doing this work now. Nevertheless, libraries are moving from being consumers of information to being creators; the research tool Zotero, developed by the Center for History and New Media at George Mason University, is one example.

Related Issues

Peer review. Peer review is fundamental to the advancement of knowledge: it attests to the validity and authority of new thinking within the context of a rich intellectual tradition. But traditional forms of peer review, which have focused exclusively on publication, are no longer sufficient. New models are needed to judge scholarly output that now includes databases, Web sites, and other forms of digital scholarship. Such models might take into account the reception of a work as revealed through citations, links, usage, and commentary on the Web. There is potential value in exposing a work to a much broader and more diverse group of users than would be possible with traditional peer review. (There is also the value of having peers on the Web review a digital resource designed to be shared over the Web.)

At the same time, some participants expressed concern about the risks of interpreting Web hits, rankings, or links as endorsement of quality. Some suggested that peer review on the Web may not reflect serious scholarly values. For example, even a site that is ostensibly peer reviewed, such as Wikipedia, does not source things rigorously. We now have to be able to judge the quality of peer reviewing.

Given the more collaborative nature of digital scholarship, how do we recognize the range of people who are involved in the process of scholarly assurance? Should we, for example, think about extending peer review to the scanner who discovers a miscollated paper? If each contributor to the process is reviewed, the review or quality becomes transparent and builds confidence. More peers will be involved because there is more that needs vetting if things are to be safely transmitted from one community of scholars to another.

Promotion and tenure. Many of the questions relating to promotion and tenure are related to, or extensions of, those relating to peer review. What is the cultural product of a merit society in the 21st century? Is it a traditional journal, or is it a set of processes in social solidarity and knowledge sharing? The traditional efficient, closed system of publications as credentials made promotion-and-tenure decisions clear; it also influenced research funding. In the humanities especially, promotion-and-tenure guidelines are no longer clear-cut. Some institutions now weigh digital scholarship and digital publication more favorably than they used to, but many do not. This inconsistency contributes to the appearance of fragmented, uncoordinated projects in the digital environment and makes it more difficult to bring coherence to electronic resources that preserve and make accessible our cultural heritage. In this way, promotion-and-tenure policies have a direct impact on the quality and utility of digital resources.

International context. Digitization appears to be driving the desire to cooperate abroad. Funding agencies in Europe, for example, want to see partnerships because institutions are often competing for public money. The rationale for collaboration abroad might therefore be viewed as an extension of the rationale for collaboration at home, although rights and regulatory environments are different. Will forging more international collaborations help forge new scholarly relationships as well?

Reconceiving the Library:
A Vision for the 21st Century

Having created a shared context during their initial discussion, participants focused on the following major questions: If we could define and design a library in the 21st century, what would it look like? What would its core functions or role be? What academic mission or parts of academic missions would it support?

The discussion suggested that some core library functions will remain consistent with the library’s traditional roles in support of the university’s mission and the public good. These roles include preserving, with an emphasis on resolving the challenges of digital preservation and conservation; maintaining special collections and repositories; curation; and teaching research and information-seeking skills. Many believe that these traditional roles have become even more important as the economics of information have changed.

Participants suggested that the library of the 21st century will be more of an abstraction than a traditional presence. Rather than reaching a consensus on the future, participants offered a range of perspectives:

  • The 21st-century library will mirror basic changes in how scholars work and will evolve in step with new scholarly methodologies and the scholarly environment. Working at the nexus of disciplines and across boundaries, libraries will have the flexibility, expertise, and organizational capacity needed to be partners in research involving large, heterogeneous data sets. The library will not necessarily be a physical space, and it may not be a collection. It might take the form of a distributed project.
  • The library’s work will be organized according to the interests of a broader number of stakeholders. It will depend on mechanisms that ensure the quality of digital resources and make it possible to share them. The new library will be organized to work collectively on common problems; this may include federating collections or staff or coordinating collection management decisions. For example, libraries will routinely make decisions about keeping print and digital resources so that each institution does not have to retain everything.
  • The library will exploit the potential of embeddability, enabling users to interact with information at progressive levels of value-added functionality-from a simple link, to automated metadata harvesting, to “actionable” data.
  • The library will be a laboratory for understanding how a new generation of faculty and students do their work, and for supporting experimentation and innovation in processes that enhance e-research across many communities.
  • Librarians will have deep experience in intellectual problems such as the structure and construct of information, the delivery of information, and the specialized needs of information communities. Institutions will share expertise.
  • The library will play a critical role in ensuring the authentication and persistence of digital information, including Web-based information, that is important to future scholars.
  • Library staff will be more distributed than at present. Librarians might do some of their work in spaces that are traditionally academic spaces, and faculty might use wired or smart classrooms in the library.

One participant suggested re-envisioning the library by turning the organization inside out.2 Look at where the fringe activity is now, he said, and think about how it would look at the center of library functions. At the center could be investment in metadata-making material available to the scholarly community in a systematic way. Activities such as management of print archives and rationalizing print collections are at the periphery now. What if they were at the center? Multimedia collections are relatively weak, as is the ability to describe them. Suppose they were at the center? Scholarly communication and the creation of digital resources would be put at the center. Print and journal material, now central, would be at the edges.

Challenges And Constraints

What will it take to realize the vision of the next library? Meeting participants discussed the need for a new service paradigm that supports roles rather than functions, and process rather than product. Professional communities-librarians, faculty members, publishers, technical staff-must be less insular. The importance of engendering collaboration, and the difficulty in doing so, figured prominently in the discussion as well as in the essays. Implicit in the discussions was the fundamental challenge of how long it takes to effect change. Participants noted several challenges raised in the earlier discussion:

  • Libraries tend to be risk-averse organizations; to remain relevant, they must be willing to experiment and innovate.
  • A sense of ownership-for example, of staff or collections-has kept libraries from engaging in truly collective work. Among faculty, competition for grants often induces self-interested behaviors, rather than collaboration.
  • Adherence to traditional hiring practices, including, in some cases, restricting hiring to individuals with the MLIS degree, makes it difficult for libraries to attract or retain staff with special expertise, such as a disciplinary background to connect teaching, research, and collections.
  • Half of today’s librarians will retire in the next decade. We need new career paths for people who want to work in academic libraries, and we need the means to support them.
  • Experimental or innovative projects are frequently supported by special grants. Libraries invest significant time and training in special project staff, but have trouble retaining them once the grant ends.
  • At this time, we do not know who will be responsible for analyzing and interpreting various kinds of dynamic information resources and making them available to the public. If this responsibility falls to the academy, how will the library address it?
  • Production of data and metadata on a very large scale for broad use needs a high level of organization. We do not have institutions that can deliver that organization.
  • The traditional separation between libraries and commercial entities needs to be reconsidered. There is enormous potential in productive collaboration between libraries and for-profit corporations.
  • As more information is digitized, print repositories will become increasingly important. We are currently stymied by the lack of effective print repositories and models for organizing them.
  • Libraries’ ability to share digital information and keep it usable is limited by a range of issues relating to copyright.


Transformations in scholarly communication and in the organization of higher education will demand new ways of doing business-not only within the library but throughout the academy as well. Research libraries will need broad institutional support as they seek to meet the demands of this new environment. On the basis of the discussion and the essays, CLIR proposes the following recommendations for higher education leadership.

  1. In collaboration with library professionals, professors, and information technologists, administrators in higher education need to develop a rigorous research agenda that will explore the influences that are transforming education so that they may better respond to and manage change.
  2. The research library should be redefined as a multi-institutional entity. The current model of the library as a stand-alone service provider to the university is obsolescent. Exploiting digital networks and emerging digital libraries and research environments, many libraries should deaccession duplicate copies of printed books, form coalitions that minimize costs for collection development, and consider sharing staff on a consortial, federated basis. Collaboration can generate savings that the library can allocate to other activities supporting teaching and research.
  3. Collaboration should undergird all strategic developments of the university, especially at the service function level. Greater collaboration among librarians, information technology specialists, and faculty on research project design and execution should be strongly supported. Areas of immediate concern include mechanisms of scholarly publishing, institutional repository development and sustainability, data curation broadly defined, and digital resource development. Any research project, digital resource, or tool that cannot be shared, is not interoperable, or otherwise cannot contribute to the wider academic and public good should not be funded.
  4. Institutions need to support environments, within and external to libraries, that not only promote but demand change. More funds should be allocated for experimental projects and new approaches; staff with nontraditional or new areas of expertise must be hired.
  5. Higher education communities, working with research libraries, need to define what models of scholarly communication represent a valid cultural product. Currently, the printed book and journal article take precedence, but the digital environment entails a more nuanced understanding of scholarship as a process in social solidarity and sharing of information. Criteria for promotion and tenure need to be reassessed. Finally, peer review requires similar study. It may prove essential for all aspects of the scholarly process-data sets, research background, Web commentaries, links, and other manifestations of the digital age that are made available and sustained over time.
  6. Instruction and delivery mechanisms should be designed according to what we know of human learning and discovery. The functions of libraries must be aligned with the core mission of research and education at the institutional level. We need to create professional and practice layers that enhance research and teaching across disciplines.
  7. University administrators and librarians should consider creating new training and career paths for professionals going into the area of scholarly communication. New leadership programs need to be developed that reflect the rise in collaborative research and that integrate support services such as those provided by research libraries into the process and methodologies of research.
  8. Institutions should use studio and design experiences as the basis of a new library school curriculum. Students of library and information sciences should learn to participate in the design and delivery of information resources that serve the scholarly community. Academic librarians should be engaged in the process through project provision and supervision.
  9. Higher education needs to articulate not only the benefits it conveys to university and college students but also the value it provides to the public. The popular conception of higher education has been influenced by critics who dismiss its perceived high costs and the impracticality of its curriculum, by those who are intent on taxing the larger endowments, and by those who want federal intervention to lower tuition costs. The cultural, social, and technological advancements that higher education can foster are lost in this impassioned rhetoric.


1 Participants acknowledged that the problem of departmental insularity permeates academic institutions, but that it may be less severe in libraries than it is elsewhere in the academy

2 This idea was initially raised in spring 2007 at a meeting of associate university librarians at The Andrew W. Mellon Foundation.