Thank you Bernie for inviting me to close the second Preserving America’s Print Resources summit meeting by taking stock of our progress in North America.[1] Even as I will raise questions about roles and responsibilities during the course of my talk, I am reminded that many in this room have been working on this issue for many years. Let’s recognize the print preservation leaders, so many of you here today, who are heroes of our generation’s efforts to ensure the integrity of our cultural and intellectual heritage. Our community is especially fortunate to have the benefit of Bernie’s and CRL’s ongoing focus on preservation issues during the transition from print to electronic formats for journals and government publications.[2]

We are not the first generation of library leaders to fret about preservation. Others within our memory have also felt an inability to secure the resources needed to keep up with our universities’ demands on our space for collections and service provision. That arose most clearly from the Depression to the early postwar years, when all manner of innovations were pursued to remove little used collections from library shelves to free up space. But it has been hundreds of years since the last full format transition, from manuscript to print; and now we are finding our way through the transition from print to digital for many types of materials. You are heroes not because you have succeeded in finding sustainable models to ensure the preservation of our cultural and intellectual heritage—we can only reasonably hope that future generations will celebrate that—but because you have taken up the challenge.

I am going to be as careful as I can to distinguish between preservation objectives and roles—where we band together to safeguard our intellectual heritage for the long-run— and to distinguish them against other types of objectives and roles—such as managing down print, collaborative print management, improved access to collections, and so forth. I see one of the biggest risks to our work as the conflation of other important practices and roles for preservation.

So, I would like to close our summit today by suggesting two matters that require additional consideration.

First, we may have developed a strong network for managing down print but whether that will yield long-term preservation goals is quite another matter.

Second, some trust networks work well for collections management and others for preservation—some for formats in use and others for those superseded—so mission and organizational design choices are tightly interlinked.

While these issues may not be new, the collaborative approaches that we have developed for print management of scholarly journals following their digitization are now sufficiently established that I hope we can now turn some of our attention to these longer-term issues.[3]

Stated preferences

Economists talk about the difference between stated preferences and revealed preferences. Stated preferences are views and opinions. But revealed preferences are how we actually utilize our scarce resources.[4]

The stated preferences of most of our organizations, and probably of all of us individually here today, is that more resources be devoted to preservation generally, and specifically to the preservation of print artifacts following their digital availability. We can recount the many reasons why doing so matters to the integrity of the cultural record and the future of scholarship, from authenticating the digital version[5] to preparing for its re-digitization should that be needed due to catastrophe or changing standards[6] to making available images that may not reproduce as well as we might wish[7]—even to saving the marginalia that can serve as the primary source for scholarship.[8]

Indeed, the library profession holds preservation as a core value,[9] giving us confidence that colleagues, even those whose primary responsibilities may lay elsewhere, still surely must share our wish for greater investments in preservation.

But these are stated preferences. We reveal our preferences in the investments that we make through the organizations that we lead.


When I spoke at the OCLC/CIC/OSU summit on Regional Print Management last year, I challenged us to think critically about the nature of the possible format migration for monographs and other books. I expressed real doubt that it makes sense to redesign our print collections management and sharing frameworks for these materials until we have a better sense of whether and how reading of long-form materials might migrate to e-versions.[10]

For print journals, preservation and access will not be linked together

With regard to periodicals, and certainly for scholarly journals, there is simply no doubt. With certain well understood exceptions, access and reading behaviors have shifted to the new format. Print is superseded. As a consequence, for print journals, preservation and access will not be linked together.

Our regional safety net

In North America, fantastic innovation happens at a regional or system basis. The Midwest Interlibrary Center was a shared off site facility that allowed state universities to manage down their collections in a responsible way, depositing a last copy in Chicago. It developed an approach to shared ownership that was at the time unprecedented. Ohio developed an early network of statewide shared facilities for all its higher education institutions, not just its public universities. The University of California system developed a thoughtful and sophisticated approach to distinguishing between ownership and governance that met a variety of needs of its universities. And in just the past few years, the Western Regional Storage Trust focused intensely on completing journal runs through careful data analysis rather than on governance and ownership, while the Association of Southeast Research Libraries developed a Cooperative Journal Retention program that allows universities to step forward to make their own commitments to one another with extremely low overhead. To name just a few of the fantastic regional and systems innovations in the realm of collaborative print management and retention. Strong partnerships and trust networks emerge almost naturally at a regional level in our geographically vast country.[11]

Our safety net is getting better, but it is not good enough

Our safety net is getting better, but it is not good enough, we all seem to agree. CRL’s Print Archive and Preservation Registry (PAPR) and to some extent OCLC’s WorldCat as well allow us to begin to analyze our achievements and identify gaps. There are a broadening number of periodicals for which libraries have committed some retention and a growing number of libraries committing to retain various periodicals. But objectives and commitments vary and disclosure is limited. As the Center for Research Libraries has recently reported, “many archiving programs provide relatively little certainty that archived titles will be preserved intact and complete, and will be available for the long-term.”[12]

Pure preservation is hard to fund

Resources are limited and preservation is rarely a first priority of a library’s parent organization—its university or other funding body.

While it is difficult to draw lines about who takes responsibility for print preservation, we know it is not always easy to fund pure preservation especially in the context of a format migration. The Brittle Books Program[13] and the Federal Depository Library Program[14] each offers a fairly good example of the challenges we have experienced. Comparatively few libraries have maintained fully realized library preservation programs,[15] and relatively few institutions elect to invest in shared digital preservation services. Priorities must therefore be established. Indeed they already have been.

Our revealed preferences

Libraries’ underlying rationale in developing and participating in regional or system-level print retention programs is to enable them to free up space locally.[16] To do so, libraries seek to reassure their constituencies that everything being deaccessioned is being retained somewhere reasonable and that access can be assured at least for the foreseeable future. In these cases, an investment in collaborative management of print weighs more towards a political than a preservation objective.

Let me be clear: by reviewing our motives in making these investments and considering their implications, I mean to offer no criticism. The effort to free up space yields cost reductions, and opportunities to reinvest in library spaces, that ultimately result in real value to the university and real strategic benefit to the library. This work should be celebrated.

But given this set of objectives, it is entirely understandable, and even appropriate, that many of our retention commitments are time-limited. While there is always the possibility of renewal, there are no guarantees. And therefore we must also acknowledge the risks this approach imposes on us collectively.

Essentially all our libraries face the same incentives and resource constraints

The reason is that essentially all our libraries face the same incentives and resource constraints. The resource savings generated by managing down print collections is large enough to generate investments for guaranteeing remote print access for a time. But when resource reallocations inevitably take place in the future, a library will need to weigh participation in shared print retention initiatives right alongside other expenses, looking at the value that each yields.[17]

Certainly, the current configuration of commitments is not intended to be maintained indefinitely. Time-limited retention commitments, even when there are many of them, cannot be relied upon for assurance that a given periodical will be available in print form in the long-term future. In this sense they are not really preservation at all.[18]

We will see soon enough how libraries and their parent universities choose to reinvest, if at all, in collaborative print management initiatives, once the short term advantages of retaining guaranteed access to the print original have passed. Through these reinvestments, we will express our revealed preferences.

The historian who will one day take up the burden of reviewing our generation’s work may find in retrospect that our efforts ensured the wide preservation of print versions of digitized materials. But our efforts so far can hardly assure us looking out to the future that this will necessarily be the case.

Another approach

Building up preservation from these regional print management efforts faces limits because it is linked to the fading need for access to the print format. Another approach is to link preservation of the underlying print version to the provision of access to the digitized version. It is good to link print preservation to the digital preservation agent and access provider. Good, because the value of the digitized will continue to grow relative to that of the print artifacts. So, if incentive structures are organized to ensure a linkage between digital and print, it should improve the prospects for sustaining print preservation over time.

JSTOR’s print repository investment for its digitized scholarly journals is probably the best known example. JSTOR partnered with the University of California and Harvard University, utilizing exacting page-level validation, principally to ensure its ability to re-digitize its holdings as needed. LLMC offers up a somewhat similar example, securing original materials in appropriate conditions based on its primary business of providing access to digitized versions.

The dark or dim archives that JSTOR and LLMC have spearheaded seemed right at the time, but as libraries are now turning their attention to managing down their print collections, they are more concerned about how access to the print originals has been limited by the “dark archives” approach. As we turn our attention to the longer-term print preservation conundrum, the model of a “tax” on digital access to support preservation surely merits further consideration.

Pressures to broaden a regional approach

The boundaries between these two approaches, regional print and digital access driven, are somewhat porous. We look at regional models and we wonder, couldn’t this be extended just a little further? On what basis can we exclude our colleagues in New Jersey if we are the Midwest Interlibrary Center? Our colleagues in Texas if we are the Association of Southeast Research Libraries? Our colleagues in Missouri if we are the Western Regional Storage Trust? And we realize that the direct marginal costs of including additional collaborators is a lot lower than the funds they can contribute to the shared effort, making the burden more broadly shareable and therefore ultimately making the whole enterprise more sustainable.

The challenges we have in drawing such boundaries, and indeed the advantages of tearing down walls that have come to seem arbitrary, raise an obvious question. Why organize these activities regionally rather than nationally? And so the Midwest Interlibrary Center became CRL, with more than 200 members in five countries. And the Ohio Computer Library Center became OCLC, claiming more than 72,000 members in 170 countries.

Scale and success

But do we celebrate these achievements or bemoan them? I am not aware of any full analysis of the culture of library or university collaboration on a national basis across its full life-cycle. Such a project is truly needed. I suspect that it would find at least for library collaborations that there is an enormous tension: on the one hand, we have tremendous interest in the potential for increasing scale to yield both efficiencies and improved services; and on the other hand, we feel discomfort when we perceive the corporate or bureaucratic practices that tend to result from an effort to yield efficiencies, and we wonder if our collaborative efforts are operating more like vendors. Here I think most especially of how OCLC is perceived,[19] but I have heard the same characterization applied to JSTOR and Project Muse and sometimes even CRL. Over the past year I have begun to track references to HathiTrust as a vendor as well.

When our collaboration vehicles are forced to compete for market share it poses very different incentives than if we can treat them as a shared resources

We don’t need to concern ourselves with these particular organizations today. Even so, I can’t resist raising a thought experiment for you. If the library community had been able to put all its discovery eggs in the OCLC basket, would libraries’ position with respect to discovery be stronger than it is today? One cannot answer this type of contrapositive question with any great accuracy, of course. But, when our collaboration vehicles are forced to compete for market share it poses very different incentives than if we can treat them as a shared resources, as natural extensions of our libraries.

How would an anthropologist understand the way our academic library community praises collaboration, scales up successful regional collaborations, and then evaluates them on an ROI basis while writing them off as too remote and corporate? As collaborative planners for print preservation, we should probably ask where this pattern of building up only then to tear down will continue to repeat itself again.

A preservation strategy

I am sure that by now I have lost some of you. You might be thinking, “My library was never responsible to anyone else for what we preserved. The practices we are pursuing today are more thoughtful and stronger than anything we did in the past.”

Nearly 15 years ago, CLIR’s Task Force on the Artifact in Library Collections termed our traditional approach to preservation as “preservation through proliferation.” Proliferation was a preservation strategy that involved the acquisition of many print copies of the same material at dozens or hundreds of libraries, many of which retained them over time. As a self-organizing strategy, preservation through proliferation is imperfect but at the same time achieved great things.[20]

Today’s efforts to manage print collections on a collaborative basis are an enormous step forward in securing new opportunities for our libraries to serve scholars and students. But they have ended print preservation through proliferation. We now have a managed print strategy, comprised of a set of managed print initiatives, which do not constitute a preservation strategy.

Our library leaders—you and your colleagues—face a delicate balancing act in working to find sure footing for the academic library in a digital environment even while universities are changing around us. In many respects, the wonderful collaboration vehicles that we have created to help us manage down print collections find this balance. At the same time, Nicholson Baker’s critique of the transition from newsprint to microfilm, although overwrought in a variety of ways, pointed to an essential principle. Even if access needs appear to be addressed to a significant degree and improved in many ways by the new format, our desire to be bold in leading change should be tempered with a healthy dose of caution.[21]

Many of our print management initiatives are understood to be of limited duration. After we have convinced ourselves, and our faculty members, that there are essentially no residual print access needs for digitized journals, this framework will, I hope, somehow, be restructured into some form of national grid for print preservation.

But as I have shown, our history here in the United States shows that we should be skeptical of our ability to sustain national-level efforts to preserve little-used print materials. The only possible exception to that may be redirecting some funds used to pay for digital access to support long-term preservation of print, as LLMC and JSTOR have demonstrated, but even these initiatives face pressures.

Thus the question we face is whether we can achieve some kind of informal North American coordination around preservation—not just shared print. I would like to suggest that, for the preservation-minded, this is a worthy challenge for our next decade.

  1. This issue brief was presented at Preserving America’s Print Resources II: A North American Summit in Berkeley, California, on June 25, 2015. The full conference program is available at .
  2. I was fortunate to receive extremely helpful comments and suggestions on this paper from a number of outside readers. I thank Kimberly Lutz, Deanna Marcum, Jacob Nadal, and Lizanne Payne for reviewing drafts, but the views presented here, and their limitations, are mine alone. I thank Liza Pagano for designing the cover.
  3. Some aspects of the dynamics that I consider here will differ with other types of materials and other parts of the world. I’ve opined on the fact that book collections will work differently than journals, not least because they are used differently and reading uses have not yet migrated to digital formats, at least not among faculty members. See Roger C. Schonfeld, “Stop the Presses: Is the monograph headed toward an e-only future?” Ithaka S+R Issue Brief, December 10, 2013, available at .
  4. This resonates for me with Rick Anderson’s efforts to distinguish between what one will do on any given issue as opposed to what one believes one should do. See “Can, Should, and Will. Pt. 1: Because What Libraries Need Is One More Venn Diagram,” Library Journal, November 29, 2013, available at .
  5. See for example Authenticity in a Digital Environment (Council on Library and Information Resources, 2000), available at .
  6. See for example JSTOR’s stated intention to maintain print archives for re-digitization purposes, .
  7. For a summary of a number of studies touching on these issues, especially for journals, see Ken Ladd, “An Examination of the Failure Rate and Content Equivalency of Electronic Surrogates and the Implications for Print Equivalent Preservation,” Evidence Based Library and Information Practice 5(4), 7-20, available from . For an examination of some of these issues as they pertain to the materials available through HathiTrust, see Paul Conway, “Preserving Imperfection: Assessing the Incidence of Digital Imaging Error in HathiTrust,” Preservation, Digital Technology & Culture 42(1), 17–30, available at .
  8. See for example the Book Traces project, . My notes on this issue are available at .
  9. .
  10. The talk is available here: .
  11. OCLC’s work several years ago focusing on print management at mega-scale really helped us contemplate the value of natural regions. Brian Lavoie, Constance Malpas and JD Shipengrover. 2012. Print Management at “Megascale”: A Regional Perspective on Print Book Collections in North America. Dublin, Ohio: OCLC Research.
  12. Center for Research Libraries, “Print Archiving and Shared Print in North America: A Preliminary Analysis and Status Report,” June 11, 2015, available, at .
  13. The Brittle Books program, funded by the NEH and expected to receive $20 million per year as a major national initiative, ended after only a few years.
  14. The number of libraries that choose to serve as members of the Federal Depository Library Program has been eroding at least in part due to the opportunity to access materials online. Participation statistics are available in the GPO’s LSCM Year in Review, which is available at .
  15. .
  16. CRL’s analysis calls this “space redeemed.” There are exceptions to be sure. Stanford has indicated an objective to increase the size of its print collections. Mimi Calter, “Western Regional Storage Trust: Cooperative Storage for Print Journals,” presentation at the The Fiesole Collection Development Retreat, April 13, 2012, slides available at .
  17. It is economically possible to calculate the recurring savings from freeing up shelf space and other avoided costs, as I explained in JSTOR: A History (Princeton: Princeton University Press, 2004) at pages 6-16 and 182-184. But this cost avoidance is difficult to “put in the bank.” The costs avoided do not become a restricted fund that can be devoted permanently to shared print retention. Rather, the cost avoidances and the expenses of participating in shared print retention are intermingled in the library’s overall budget.
  18. OCLC’s Rick Lugg, an expert who has worked with dozens of libraries and consortia on protocols for managing down print collections, has noted: “These are early days in shared print. There is a strong sense that the print collections management process will be iterative, and that retention needs may look different in 5-10 years, as shared print programs become the norm, and user comfort with non-print formats grows. Basically, we may not need as many copies in the future as we are committing to retain now. So they don’t want to make a formal, public 25-year commitment now, and end up holding something whose value declines over time.” Personal Communication, June 10, 2015.
  19. Remarks to the effect that OCLC is “indistinguishable from any other vendor” are widespread. See for one recent example, .
  20. Stephen G. Nichols and Abby Smith, The Evidence in Hand: Report of the Task Force on the Artifact in Library Collections (Washington DC: Council on Library and Information Resources, 2001), available at , 53, 70.
  21. Nicholson Baker, Double Fold: Libraries and the Assault on Paper (New York: Vintage, 2001).