Volume 6, Number 4, Supplement
Selection for Preservation
Aug 1982

Deciding What to Save

An Address given before the American Association for the Advancement of the Humanities in December 1981

Over the past two decades, a series of distinguished librarians and archivists, as well as several committees, commissions, and conferences have made long and impressive statements about the need to conserve library and archival materials. These have, however, rarely if ever been followed by any concrete actions or programs. The difficulty of making the leap from consciousness-raising to active commitment is due at least in part to the all-encompassing nature of the problem and the lack of any clear agreement on the scope of what we mean when we talk about library conservation. In addition, there has been in the past a total absence of consensus on priorities for dealing with what to many has appeared as a hydra-headed monster. It seems useful therefore to begin by trying to define the nature of the beast in order to develop a strategy for attaching it. In short, just exactly what is the scope of the conservation problem and what segments of it should be tackled first?

Are we talking just about the paper trail left by man in both his individual and corporate character? Or do we also include all those other materials on which our multi-media society records its activities and culture? For example, sound recordings, magnetic tape, videotape, film (both photographic and movie), and newest of all, videodiscs, will probably be of as much interest to the scholars of 2082 as our scientific journals and interoffice memos. Indeed, the whole field of oral history has flourished in recent years on the strength of the conviction that the written record is incomplete, given our predilection for verbal rather than written communication. In addition, more and more of the raw data generated by our society is in machine-readable form and retrievable only from magnetic tape. This is now true of much material which in the past was collected and stored on paper, such as census records, tax data, police records, scientific data and so forth.

Just as we are no longer a purely paper but rather a multi-media culture, so our scholarship has burst through the borders of what used to be clearly defined disciplines in terms of intellectually acceptable areas of scholarly inquiry and in so doing has created a demand for a far broader range of source materials than were previously collected by libraries and archives. Thus, we must consider whether or not priority for preservation should be given to traditional materials of fairly predictable scholarly interest. Should we on the other hand be trying to preserve at least a sampling of our culture across the board? Given the tremendous expansion of the kinds of sources considered appropriate for scholarly research, these may be one and the same thing. Just as scholars are now looking at Victorian dime novels to learn how people felt about everything from family planning to treatment of the insane, so in the future they may want access to the complete recordings of Elvis Presley or the entire corpus of Dallas to give them insights unavailable from more conventional sources.

Similarly, we confront a problem in determining the depth as well as the breadth of the materials which are to be preserved. The Joint Committee on the Archives of Science and Technology is currently trying to arrive at some guidelines for distinguishing among scientific disciplines in terms of the kinds of records which should be saved for each. One of the questions being considered is whether laboratory notes must be saved if the research has been written up in published reports.

Neither has anyone really begun to tackle the question of how much of certain kinds of record material is enough. If we have the local records of the coal miners' union for West Virginia, do we also need them for Pennsylvania? If the Jewish Community in New Haven is known to be extremely well-documented, should Hartford do the same?

The question of scope also involves definition of the time period whose products are in most urgent need of attention. Can we simply pick an arbitrary date when new techniques of manufacturing greatly aggravated the problem of paper deterioration? For example, can we settle on 1830 and just leave everything earlier to local initiative, in the belief that it is not only probably in better physical shape but can also command local resources on the grounds of rarity and artifactual value? Similarly, can we disclaim responsibility for anything published after 1985, in the expectation that by then the work of the Bailey Committee will have prompted a significant number of publishers to use permanent durable paper and improved binding techniques and materials, and that the Library of Congress through the Copyright Office will have implemented a program f or preserving current American imprints?

Then there is the issue of geographic coverage. Are we talking about all the books and serials in our repositories or only holdings of American origin? Is it reasonable to assume that the countries of Western Europe will also soon launch massive retrospective conservation programs to preserve their national heritages? Is it cost-effective, given the fact that libraries in this country arrange materials by subject rather than place of origin, to expend the man-hours needed to weed out the American imprints from all the rest? Would a program to deal with all publications dealing with American history and culture regardless of imprint be easier, cheaper and just as intellectually justifiable to run?

And where would such a program leave the thousands of linear feet of manuscripts and archives, photographic prints and architectural drawings which are also part of our paper-based documentation? Here again, there is a choice to be made. When we say library conservation, do we mean just published materials, that is, essentially books and serials, or do we mean all those other things which man had committed to paper?

The issue of scope also involves choosing between an "elitist" and a "populist" approach. Should our first concern be the rare and the beautiful: the first editions, the letters, both published and unpublished, of great men and women, the manuscripts of great literature, the diaries with eyewitness accounts of significant events? A slightly different way of approaching this issue is to ask whether our emphasis should be on saving those books and manuscripts of great artifactual and aesthetic value as opposed to those of purely intellectual interest.

This leads us immediately to the next hard question, which is how to determine which materials must be retained in their original formats and which will continue to be of intellectual value even if transferred to another medium such as microform or videodisc. Can a single generalization be made which will be valid for all disciplines and fields, or does the character of the presumed ultimate user's research need to be taken into consideration? For example, would a student of literature need to see all the variant editions of a work in the original, whereas a historian might be content with the standard edition on microfilm?

Also related to the above issues surrounding the complex question of what to save is the question of who should make these decisions. Librarians? Archivists? Bibliographers? Scholars? Committees of all interested parties? Here again, would it be simpler, less time-consuming, and therefore less expensive just to take the plunge and begin?

I do think a few choices are warranted on the grounds of expediency if nothing else. First of all, the period from 1835 to 1985 is defensible as that most urgently needing treatment. A nice round number like 150 years is also useful for public relations purposes. Similarly, a nationalistic approach has much to recommend it, particularly in terms of fund-raising. Here, my personal preference is to begin with Americana rather than American imprints. Initially, a program should focus exclusively on paper but should be structured to include other media as information previously found on paper begins to appear in other formats. Although the idea is tempting because it would greatly simplify the task involved, the program should not be limited just to print materials, but should also include original sources. One of the advantages of the subject approach is that it would make it easier to incorporate manuscripts and archives as well as books and serials.

It should, however, be made clear at the very start and repeated at frequent intervals that everything cannot be saved for all time. "Planned deterioration" should therefore be the fate of large amounts of material, particularly record materials, In essence, scholars should be alerted to the fact that only a small amount of non-print documentation will be retained permanently on the premise that its research value can be adequately exploited within the natural lifetime of the materials in question. Of that which is earmarked for permanent retention, a miniscule proportion will be in its original form. The rest will be in facsimile of some kind.

Similar scheduling of print materials also needs to be done, preferably as part of the acquisitions process so that institutions will make a conscious decision when they acquire an item whether to retain it for only five or ten years, for the expected natural life of the item, in facsimile forever, or very rarely in its original form for as long as technology and human care and ingenuity can prevent it from crumbling into dust.

In regard to the collections already on the shelves of institutions, this kind of discrimination can be applied retrospectively within the context of a national resource-sharing program such as that now being developed on a limited scale by the Research Libraries Group. Ultimately, most institutions will have to survey their collections not just to determine the extent of deterioration, but also to identify those areas in which they are sufficiently strong to accept the responsibility of retaining the national preservation copy in whatever format is appropriate.

I will conclude by once again urging that we should not agonize too much over the finer points of definition of scope, but should begin to deal with the most easily grasped portion of the problem in an organized way as soon as possible. Whatever is done will be a big step ahead of the present posture of continuing to wring our hands and do nothing.