Some Aspects of the Relationships Between Special Collections and Academic Theory

A Paper Presented at the Rare Book and Manuscript Section Annual Preconference, Bloomington, Indiana, 23 June 1995

by Daniel Traister
Department of Special Collections, Van Pelt-Dietrich Library
University of Pennsylvania, Philadelphia

In this paper, I want to consider a few aspects of the varied relationships between special collections and academic theory. This is not a paper about theory. I do not provide brief introductions to the varieties of theories that concern academics in many disciplines these days. I want, instead, to recommend reconsideration of our practice as special collections librarians--not a matter easily separable from theory, of course--and to recommend also that we do so from a perspective perhaps more informed by theory than seems to me to have been our ordinary wont.

The paper has three sections. I begin by commenting on what I take to be our general attitude to theory. Second, I offer a few examples of the ways in which new theoretically-based questions raise, not threats to our relevance or existence, but rather opportunities. My examples are drawn from literature, the subject I studied before I became a librarian and what I continue to teach alongside my work as a librarian. Third, I propose that our work as librarians is a more theoretically-based practice than we may ordinarily suppose and, as a corollary--and as I have also argued elsewhere--that our practice might benefit from, perhaps even change in the wake of, interrogation from points of view informed by various theories now current in the academic disciplines, and by a heightened self-consciousness about our own professional implication in theory.


In the first part of my paper, I comment on our corporate attitude toward theory and theory's impact upon the disciplines our collections exist to serve.

A heightened, often highly critical attention to what had long passed as the theoretical bases of many academic disciplines, particularly in the humanities and the social sciences, increasingly calls into question the ways in which these disciplines are practiced and taught in the modern university. This development is no secret; it isn't even news, and hasn't been for the better part of a quarter of a century.

What is true in the disciplines, however, is not necessarily true of professions. Professions may define themselves (at least partially) in terms of their relationship to a body of theory, but their practitioners nonetheless tend to be indifferent to theory except insofar as it clearly relates to and has an impact on practice. Librarians are typical professionals in this respect. Our cliches about library school vs. library practice are one indicator of this typicality. So is our general mistrust of academic theory. We regard it as airy wind by comparison with the tangible stuff we work with, stuff we pay attention to an value, and whose life and usefulness we seek to conserve. In simple fact--granting for the moment that any category such as "simple facts" can be imagined--the tangibility of the sheer stuff with which a librarian deals is hardly illusory. We should not pay so much for or to conservators if it were.

On the other hand, of course, the tangibility of our printed books, manuscripts, and other documents is only an inert fact about them. It is the only "fact" about them about which we can feel confident; the rest is mere hypothesis. We value such objects because we think--hypothesize--that they provide "evidence," "information," and alternative books that help our readers to decide between ways of reading various kinds of histories, or, if we make a distinction here, various kinds of fictions, or whether to read new (or just different) histories or fictions. We acquire, describe, make accessible, shelve, fetch, return, and preserve and conserve such materials by the shelf-mile. If they don't do this sort of thing for students, teachers, scholars, and writers, then what are we doing wasting our lives in these miles of dusty shelves, worrying about the stuff we keep on them?

Tangibility is something to be reckoned with. As Richard Terdiman notes, "normally objects have an intimate relation to remembrance." Of the objects we call books, manuscripts, and papers--that very tangible stuff which it is the business of librarians to preserve for the sake of those students and scholars who need it--this relation is especially characteristic. In our line of work, Mallarme's crack--that "everything in the world exists in order to end up as a book"--stops being a crack very quickly: he's right. Somewhere on our shelves, everything in the world does sit, gathering dust while it decomposes and making us worry about how we can possibly hold on to it long enough so that its ideal reader--wherever, whenever, that reader may be--will find it.

What I find to be our collective professional doubts about theory, therefore, arise first from our inclination to be interested in tangible objects of the sort we are trained to care for and try to preserve. In addition, however, we are as a group further inclined to doubt theorized approaches to the subjects we have been trained in because of how we are trained to treat their library and archival remains. Bibliographers, catalogers, and those who describe manuscript collections at something below the collection level: all of us are trained to look at trees, not forests. We describe what we see; we dislike the non-evidential; we learn--from Bowers if from no one else--that we can make no decision about a broken "a" in a line of type in a printed book until we have seen every other copy of that edition and examined the "a" in each of them. And after we have done so, we leave the text to the literary critics, whom we continue to distinguish from real "scholars."

Third, what we emphasize and value in the materials we care for are most often the apparent products of "high" culture, materials, that is, produced under the patronage of, for use by, or in some now-contained way for the entertainment of, those who run with society's dogs, not its hares. I don't want to overemphasize this point. A good deal of high or "official" literature includes, even expresses, oppositional, otherwise contained points of view (which we know, I should point out, through the work of theoretically-informed readers). None the less, it is usually not the demotic, the popular form--the peasant's poems, the woman's novel, the slave's or the prisoner's account, the labor union songsheets, or the original manuscripts of '50s rhythm-and-blues songs--that we trot out to show visitors or cuddle ourselves (if our collections have them at all). We are, all of us, quite heavily invested in the maintenance of "the canon" in terms of what we collect, what we value in what we collect, and what we believe we should continue to collect.

Finally, of course--and the objection to theory most commonly expressed by those librarians with whom I have spoken about it--is that its pursuit diverts both student and research attention away from primary sources and documentary research. We have the stuff, the things; but the theorists are just concerned with words about the things. They don't read books any more, they just read theory.

In summary, special collections training socializes us to value the tangible, the physical object, whose physical existence we are taught to protect as well as we can, as opposed to the intellectual, which we leave to take care of itself. (Such an attitude only extends a value basic to American librarianship: we are, in general, very chary about asking too closely what our readers plan to do with the materials we set before them.) Second, and closely related, our library or archival training encourages greater attention to individual objects than to classes of materials. Third, we tend to emphasize in our collections those holdings reflective of high, often conservative cultures, whose values it is awfully easy to find not only reflected but also validated by such works. Alterations of theories of relative value and significance would threaten the stability--the steady development--of the collections we and our predecessors have worked to build. Fourth and finally, we worry about changes in academic attitudes that seem likely, by rendering us a sideshow, to threaten our funding and, hence, our existence.


In the second part of my paper, I want to sketch just a few areas where theoretically-based work raises, not threats to our relevance or existence, but rather opportunities for mutual support and growth. If I am right in supposing that we tend, by and large, to mistrust, even to fear, the intrusion of theory into the work of our academic colleagues, then perhaps even a brief discussion of a few such examples may be as important as my effort--which will conclude this paper--to demonstrate that we rely on theory just as our academic colleagues do and stand to benefit from self-conscious theory-based approaches to our institutional practices.

Let me begin, then, with the biggest bugaboo of all: that theory is simply a covert form of Marxism, its development an attempt to resurrect through a back, academic door a social and economic theory that has in practice collapsed and gone the way of the dinosaur (a metaphor I have never heard used by anyone who knows that dinosaurs survive, quite nicely, as birds). Perhaps it is true that we need pay, should pay, no attention to Marx or his modern theoretical followers. Curiously, however, a renewed interest in such topics as the history of books and printing is, in important ways, the result of Marxist scholarship and its emphasis on materialism. An interest in the material means by which texts get transmitted, since the documentary evidence for those means tends to survive in libraries and in special collections departments, has focused new attention on older books and manuscripts, even, indeed, on galleys, proofs, editorial correspondence, and, in general, the material processes by which words and ideas get disseminated.

This set of concerns dovetails nicely with new interest in the concept of authorship itself. Who (if anyone) is responsible for words and ideas that get disseminated? How many mediators lie between their initial expression, subsequent elaboration, and ultimate appearance? Are these processes individual or collaborative? What pressures exist to contain, repress, or silence words that appear potentially dangerous, or, on the other hand, to disseminate as broadly as possible what seems to support a society's elites and its power structures? These questions also have sent (contrary, perhaps, to our initial expectations) ideologically-informed readers into special collections departments to look, perhaps in fresh ways, at our materials. Proliferating studies of copyright and authorial ownership indicate an arena of contact between law, publishing history, government regulation, and authorship, the latter in this respect seen as an outgrowth of the perception (or is it the manufacture?) of a concept of authorial rights and the consequent creation of the author as a legal entity and of an author's work as a form of property.

These are not questions relevant only to the modern era. How authorship is to be constructed in, for example, the sixteenth or the seventeenth centuries, when much literature circulated in manuscript, has brought new scrutiny to the ways in which writers, readers, editors, and printers collaboratively created, and might continue collaboratively to create, the texts we read, teach, and study as if "Wyatt," "Donne," "Marvell," and "Rochester" were unequivocally and univocally responsible for what appears under their names. Even in the dark era when I was in school we knew such assumptions to be untenable; but no one really wanted to face their alternatives. It is exhilarating to find that scholars are now asking such questions and beginning what is likely to prove a long process of answering them; answering them, I might add, in ways that a non-print/computer-based dissemination technology is likely to affect in ways I can only begin to predict, in the wake of looking at the work being done at the University of Virginia with a late nineteenth-century writer, Dante Gabriel Rosetti, who poses some analogous problems.

Reception theory looks for evidence in special collections departments: here, too, it is useful to locate actual printed books owned, or manuscripts written by, let us say, women; marginal annotations by the early readers of political, economic, or literary works; letters in which African-American readers during the 1920s respond to a writer associated with the Harlem Renaissance, such as Langston Hughes--or to a writer writing about the Harlem Renaissance, such as Carl Van Vechten. The records of the Book-of-the-Month Club, currently under study by a person who has already written about Harlequin romances, may tell us much about how modern books were chosen for mass market distribution and about their value as cultural--and as literal!--coin. The ways in which books are marketed and audiences segmented along a variety of divisive and cohesive lines is something that reception theorists find evidence for in the materials contained in departments and institutions such as ours.

Many other theorized approaches simply to my own field, literature, could be cited, and the ways in which they have turned out to rely on, not to have abandoned, primary resources indicated. No matter how long the list, we should still have missed the most obvious and significant arena of potential cooperation between new theorists and librarians: collection development.

The arena of the collectible has expanded enormously. All we need is ears to hear this Good News--and, of course, wallets to buy or donors to provide what will satisfy our expanding institutional appetites. The introduction into academic respectability of radically oppositional writers--ideologically and politically threatening--as well as of writers who emerge not only from political- but also from economic-, race-, class-, sex-, gender-, generic-, and geographically/regionally-rooted margins, or the just outright excluded, means that materials with potential interest and significance abound (in almost all price ranges, to be blunt about money issues); and the sheer variety of potentially useful material presents library collection officers, in as well as out of special collections, with a unique opportunity to work together with concerned faculty and students to strengthen collections in ways that will improve research and teaching potential in fields where people are actually doing work. Self-consciousness about theory (that is, about theories) will, in short, help keep our collections (as well as ourselves) alive.


The third part of my paper suggests that, even if we generally distrust theory, we are--far more than we may think--implicated in theory in our own practices as librarians. Our practices might therefore benefit from, perhaps even change in the wake of, the sort of self-conscious critical interrogation which has been, as I take it, the intention of other speakers at this conference also to encourage.

I begin with a simple--although not uncontroversial--example, our generally unquestioned relationship to a disintegrating theory of textual transmission and its related practice of textual scholarship. I mentioned earlier an approach to the objects in our care that we have learned from Fredson Bowers and his epigones. They have taught us to look carefully at the objects we care for, to describe what we see, and to dislike and distrust what we cannot evidence. Of course, these are not criteria I want to disparage or discourage. That they derive from and are applied according to theory, however, is a point we would do well to recall.

We are curators and catalogers, to be sure, not bibliographers. None the less, we know that we can make no "real" decision about the significance of, for instance, a broken type in a book we are examining until we have seen every other copy of that edition and considered the condition of that sort in each so as to build up a sense of the "ideal" and the range of documented variations from that ideal. Usually, we really can't do this: there's too much work to be done, too little time to do it in, and too many places we might have to go to in order to see the copies we would wish to see. Occasionally, when such questions seem of real interest, we write letters of inquiry to colleagues: we have all written and received queries of this sort. At some point, we finish this and countless other tasks like it. The book goes off, cataloged, to its proper location on the shelf; or, edited, the text goes off in its "perfected" form to be published for the edification of literary critics.

Yet the sort of approach to physical books and their description that Bowers championed also represents a theory, just as much as the work of the literary critics. I can demonstrate this point here only by recalling his famous essay, not on a broken piece of type, but on the substantive "sullied/solid" crux in Hamlet ("Oh, that this too too what? flesh would melt," etc.). What dis Bowers think he had proved in that essay, and why did he think so? He thought he offered a purely bibliographical demonstration of why "sullied" is the right word. I can imagine no modern reader who would agree. Bowers sought--and, not too surprisingly, found--bibliographical justification for a reading of this crux that supported a New Critical and Calvinist approach to the play. A bibliographer's version of Maynard Mack's "The World of Hamlet," it is, in 1995, just as unthinkable as Mack's essay. Modern textual scholars respond to this crux in two main (and completely different) ways. First, as exemplified by the Riverside edition, a modern text may read "sallied," which Bowers rejected, even though that word does appear in the early texts, as "sullied" does not. Second, a modern Hamlet may (and more modern Hamlets increasingly will) read "too much griev'd and sallied," "too too sallied," and "too too solid"--presenting, that is, all three variant readings which appear in early texts.

Modern texts "increasingly will" provide such alternatives, as I have just said, because of what has been happening in the once relatively staid world of Shakespearian textual studies in the past fifteen years, as well as to the entire theory--descending from Pollard, McKerrow, and Greg, through Bowers, to Tanselle--of how printed texts get transmitted. This is a much more complex matter than my example of the crux; once again, I can, in this context, only allude to a large body of (admittedly still controversial) scholarship. Suffice it to say that almost nothing of what we once learned as "fact" about this subject is stuff that people now in the field of textual scholarship believe as confidently as they once believed it--if, in fact, they believe it at all. This change is difficult for us, since those "facts" support the screed special collections librarians usually trot out when we sing our own raison d'≖tre, that is: "we are the repository for 'the evidence' out of which new textual studies will produce 'definitive' texts for students and researchers." The Greg-based model took principles of Biblical textual scholarship and modified them for the study of Elizabethan texts generally and Shakespeare's texts specifically; but it has broken down almost completely, in part because Greg's school treated printed dramatic texts without reference to many realities of dramatic production and in part because it regarded the plays it considered as "Literature" as if they had been composed by an all-powerful "author"--propositions remarkably difficult to defend. The model is under pressure not only at its source, that is, in the study of the Shakespearian text itself, but also on other fronts, as well. For instance, the textual study of nineteenth- and twentieth-century writers who exist in manuscript as well as printed versions, each displaying major differences from the others, and all displaying the intervention of many hands in addition to an author's own, throws into relief questions that a largely manuscript-less period such as the Elizabethan never prompted Greg and his followers to ask. In fact, of course, even that period was not "manuscript-less"; we merely lacked, by and large, manuscript evidence for the drama produced during it. As time goes on, more scholars are looking at other literary forms where the manuscript evidence is extensive, unlike the situation for the drama, and asking the questions even of Elizabethan and Stuart non-dramatic writers that other scholars now ask about more modern writers--and, increasingly, even about Shakespeare and his stage contemporaries, as well.

Our relationship, as special collections librarians, to a disintegrating theory of textual transmission and related practice of textual scholarship affects, without our always self-aware participation, the ways in which we build our collections: we seek out variant editions, issues, and states, as well as manuscript exemplars, of works we possess--nothing wrong with that, of course--but we do so with a particular editorial model in mind. How apposite that model is for recent as well as older authors or books is not something we often question. How apposite it might be for other kinds of research we tend also not to question, so that we inquire lesss closely than we might into the kinds of variant and alternative appearances of texts other readers might find informative.

We acquire such materials, in addition, largely only for authors whom we expect to be edited at one point or another. This is an even more constricting criterion. A more theoretically-aware stance might open us to a drastic broadening, or at least suggest real reconsideration, of what is worthy of our acquisitive attentions.

For we know, although admitting this knowledge is not always easy, that librarians, and hence libraries, collect and preserve materials far more haphazardly than our users appear to realize. Librarians' decisions about what shall be contained by our institutions are frequently accidental and contingent, dependent upon a host of factors most of which are unaffected by any serious consideration of the significance of what libraries might be thought to exist to preserve. The repositories we build are, in consequence, constructions that only partially represent--not "reconstructions" that faithfully re-present--the vast universe of potentially salvageable memories which print and its surrogates have produced. Despite an ideal of comprehensiveness or universality, we routinely, and without much examination, omit materials which we believe libraries simply cannot contain. We also define other sorts of materials as not part of the universe we try to reflect.

The role of accident in the construction of research libraries is difficult to convey. Let me offer one approach. Who pays attention to the provincial? and how do we define that amorphous category, anyway?

The general lack of interest in the provincial is not new. American literature itself long suffered under this general excuse for its dismissal. Even today no one needs to look as far afield as Australia, New Zealand, India, Africa, or the Caribbean for examples of English-language writers whose works and history--this at many English-language universities with serious pretensions as research-supporting institutions and active and productive Departments of English--students cannot study or teachers teach. Works defined as "regional" or "generic" simply do not get selected for inclusion in library collections. At the library of the university from which Zane Grey graduated, how many westerns are part of the library's collection? or books of any sort published in Salt Lake City or Reno by publishers other than the universities located in those two cities? How many books are stigmatized (and hence uncollected) by library selectors because they can be subsumed under rubrics that consign them to automatic oblivion: "science fiction," "mysteries," "romances," "bestsellers"? In terms of their collecting interests, how many American research libraries find the literature of anglophone Canada (to say nothing of francophone Canada) to be just as distantly compelling as that of Australia? How much of the literature of the radical left or right do we collect? There are exceptions in all of these various categories, of course, but the exceptions are just that: they are unusual. They result from geography (Canadian institutions collect Canadian writers; Nevada institutions Nevada writers); from odd institutional contexts (the existence of a center for study of Australian and New Zealand issues at Penn State); from current events (the Oklahoma bombings surely boosted library sales for Andrew Macdonald's The Turner Diaries) or from odd institutional selectors (with, say, a taste for Australian poetry, science fiction, or radical American writing). The exceptions, in short, result from accidents. The collections which those accidents produce are contingent upon geography, sub- institutional entities with uncertain life spans, and the presence and continued existence of non-institutionalized selection practices and practitioners able, in their acquisitions, to follow up on their own partialities as well as to acquire the obvious, the mainstream, and the central--without which selectors are invited to seek alternative employment.

The constructs built, in their acquisition decisions, by librarians who staff libraries (and, what is even more obscure, the employees of specialized library book vendors), as they go about the quotidian task of selecting and acquiring the materials that constitute the library collections and special collections with which present and future scholars work and upon which they will continue to rely, are almost completely unexamined. Each item-by-item decision about what is and what is not worth acquiring and preserving adds to that construct. A vast bureaucracy exists to build the research libraries on which scholars depend. That bureaucracy is almost completely unstudied, not by its own constituents, but rather by those whom it ostensibly serves. This despite the fact that, increasingly, those scholars whom it serves have come to recognize how other bureaucracies and social organizations, especially those with functions which can be broadly grouped together as intellectual or ideological, demand scrutiny and interrogation. Librarians need to learn this lesson, too. Our own studies of collection development tend to be methodological and quantitative. There are other important questions we might be asking, as well--questions to which numbers alone will not provide an adequate answer.

The gap between what is published, or what is available in the out-of-print, used, or rare marketplaces, and what is collected does not lessen, it widens, from year to year--as does the gap between the collected conventional and the uncollected unconventional. This gap has consequences for what our colleagues and students can remember (research) and teach. Every student, teacher, and scholar in the historical humanities today knows that this is a time when study of the scattered, disregarded, unedited, or unrepublished materials that still live in this gap is providing the substance for what is proving to be some of the most exciting (and literally reinvigorating) historical and literary scholarship of our time. Just consider, as one example, the reinvigoration of our sense of what "counts" as "American literature" that has resulted from our changed sensitivity to what women, Americans of color, and the political left have all contributed to that construct. A study by Cary Nelson, Repression and Recovery: Modern American Poetry and the Politics of Cultural Memory 1910-1945, undergirds this paper in many more ways than one. Alan Wald puts his finger on one of Nelson's most important points in his 1991 review of Nelson, when he remarks: "a key lesson of Nelson's book is that partisans of cultural, economic, and political equalitarianism cannot count on the internal dynamics and networks of extant institutions to preserve even the most rudimentary material artifacts of oppositional cultures." Wald doesn't cite libraries specifically in that indictment; but in fact almost all research libraries are perpetuating the gap between what is available and what is collected for future scholars to struggle against.

Accidents that plug small sections of this vast gap are not method; they do not "cancel deficiencies out" in some mystical way. The institutional structures through which libraries collect materials which will ultimately be part of the documentary record of our pasts need analysis. They need it from those who work within and tend to accept as givens the constraints of these structures; they need it as well from those who retain the capacity to be surprised by those structures--perhaps even to find them annoying. Both kinds of analysis need to be informed by modern theory as well as practice, if they are to do anything more than validate the status quo; and both need to be undertaken cooperatively.

This third part of my paper has noted that, whether we know it or not, we are implicated in theory when we adopt as a model the Greg-Bowers-Tanselle theory of textual transmission and build collections with that theory in mind. We tend not to think of this model as a theory and, as a result, are less aware than we should be that its bases are far from firm. We need to ask if the critics of this model have things to tell us about ways we might be collecting, things we might be collecting, that we presently miss. More generally, I have suggested that we rest too contentedly within the unexamined boundaries of our own collecting practices. We would benefit from informed analysis of these practices, from external examiners who are our users but do not necessarily share our assumptions, and from internal examiners who need, now and again, to ask questions that cannot be answered quantitatively only. We need alternative ways of looking at the world of potential collectibles.


If we can encounter new theory sympathetically, we are likely to find ourselves better able to reconsider our practice as special collections librarians from a perspective in accord with the realities of our academic context. Awareness of the potential support and cooperation to be encountered from theory and its practitioners will, in short, enrich our capacity to analyze our practices, enhance our functions, and integrate us more firmly into the academic world of which we are a part.

Return to Daniel Traister's Home Page.

Last update: August 1995.