Significance of Primary RecordsRekindling the Reading Experience of the Victorian Age Catherine Golden Postscript about the Public Libraries Ruth Perry The texts that make up “Signiﬁcance

Significance ofPrimary Records

Statement on the Significance of Primary RecordsModern Language Association of America

IntroductionG. Thomas Tanselle

What Is the Future of the Print Record?J. Hillis Miller

Traces of a Lost WomanSusan Staves

The Diaries of Queen Lili‘uokalaniMiriam Fuchs

Manuscripts on Microfilm: The Disturbing Case of ProustAnthony R. Pugh

Twentieth-Century Undergraduates and an Eighteenth-CenturyEdition of Diderot’s Encyclopédie

Manon Anne Ress

On the Importance of Judging Books by Their CoversGregg Camfield

Rekindling the Reading Experience of the Victorian AgeCatherine Golden

Postscript about the Public LibrariesRuth Perry

The texts that make up “Significance of Primary Records” werefirst published in Profession 95 (New York: MLA, 1995) 27–50.

Statement on the Significance ofPrimary Records

Modern Language Association of America

The Modern Language Association of America applaudstwo developments aimed at ensuring the future accessi-bility of texts from the past. One is the organized effortto microfilm the texts of nineteenth- and twentieth-century books containing acidic paper that is now, orwill become, brittle; the other is the systematic trans-ference of printed and manuscript texts of all periodsto electronic form. Everyone who cares about the pastshould be grateful to the library world for the way ithas responded to the challenges of textual preservation.Frequently, however, discussions of these developmentsimply that, once reproductions exist, many of the arti-facts from which they derive need no longer be con-sulted or saved. In this climate of opinion, the MLAbelieves that it is crucial for the future of humanisticstudy to make more widely understood the continuingvalue of the artifacts themselves for reading and research.The advantages of the new forms in which old textscan now be made available must not be allowed toobscure the fact that the new forms cannot fully substi-tute for the actual physical objects in which those ear-lier texts were embodied at particular times in the past.

Without broad public perception of the significanceof this point, sizable portions of certain classes of tex-tual artifacts face destruction. The MLA is expressingno opinion about the relative desirability of differentforms of dissemination for future writing; rather, it isstrictly focusing on the future study of texts thatappeared in the past in handwritten or printed form onpaper or parchment. By outlining the theoretical rea-sons for the importance of physical evidence in textualartifacts, the MLA wishes to promote awareness of theissues and to stimulate practical recommendations fortaking action on them.

Texts are inevitably affected by the physical means oftheir transmission; the physical features of the artifactsconveying texts therefore play an integral role in theattempt to comprehend those texts. For this reason, theconcept of a textual source must involve attention tothe presentation of a text, not simply to the text as adisembodied group of words. All objects purporting topresent the same text—whether finished manuscripts,first editions, later printings, or photocopies—are sepa-rate records with their own characteristics; they all carrydifferent information, even if the words and punctua-tion are indeed identical, since each one reflects a dif-ferent historical moment. Any such record may bea primary source, but an object that is primary as asource for one purpose is not necessarily so for another.A primary record can appropriately be defined as aphysical object produced or used at the particular pasttime that one is concerned with in a given instance.

Physical evidence in manuscripts and printed matteris indispensable in two ways. First, physical clues (suchas the structure of the folded sheets in a book) revealfacts about how an item was produced—facts that canin turn lead to the discovery of textual errors and con-tribute to a knowledge of contemporary textual, print-ing, and publishing practices. This kind of evidence hasprimarily been used by analytical bibliographers andscholarly editors. Second, elements of a book’s physicaldesign (such as paper quality, page size, textual layout,choice of letterforms, and arrangement of illustrations)can be significant indicators of how the text thus dis-played was regarded by its producers and how it wasinterpreted by its readers. This category of evidence is

SIGNIFICANCE OF PRIMARY RECORDS

PROFESSION 95

currently being used by those investigating the historyof reading and the social influence of books.

Not only do editions differ from one another, butalso copies within an edition (of any period) often varyamong themselves; as a result, every copy is a potentialsource for new physical evidence, and no copy is super-fluous for studying an edition’s production history.Furthermore, since the shape, feel, designs, and illus-trations of books have affected, and continue to affect,readers’ responses (some of which have been recordedin the margins of pages), access to the physical forms inwhich texts from the past have appeared is a funda-mental part of informed reading and effective class-room teaching; if that access is to be as widespread as itcan be, the number of available copies of past editions,held in libraries of all kinds, must be as large as possi-ble. The existence of community libraries along withacademic libraries has been, and will continue to be,essential for bringing historical embodiments oftexts—and the sense of the past they impart—to awide readership. The loss of any copy of any edition—from the earliest incunables to the latest paperbackreprints (regardless of whether its text is consideredinteresting or consequential at the present time)—diminishes the body of evidence on which historicalunderstanding depends.

There is an obvious practical consideration that alsosupports the retention of textual artifacts (handwrittenas well as printed) after their texts have been copied:the fact that the accuracy and stability of reproductionscan never be guaranteed. For this reason, the preserva-tion of the sources of photographic or electronic repro-ductions would seem a prudent course even if thosereproductions were the equals of the sources; but sincethey cannot possibly be, a concern for maintaining ourinheritance of textual artifacts is not simply desirablebut imperative.

It is clearly unrealistic to expect that all currentlysurviving manuscripts and printed books can be saved.They are subject to the same vicissitudes as every otherphysical object, and their survival depends both on thematerials out of which they are made and on the na-ture of the events that befall them. But the attitudesthat people hold about them can be instrumental ineither mitigating or exacerbating the destructive effectsof these factors. As more people come to see the impor-

tance of primary records, more use will be made ofthem in reading and teaching, and more constituencieswill join together in the search for ways of financingartifactual preservation, storage, and access. Morerecords will then be saved because there will be widersupport for the allocation of resources to this purpose.Decisions about priorities for preservation will stillhave to be made, by individual as well as institutionalowners of material, but those decisions will be reachedin a framework that recognizes the artifactual value ofevery object. An appreciation of the significance ofphysical evidence also necessitates the adoption of stan-dards for the creation and identification of reproduc-tions, in order to minimize the damage done to primaryrecords by the processes of reproduction and to maxi-mize the usefulness of the reproductions.

Readers find themselves turning continually to re-prints or reproductions of some kind. As they welcomethe benefits conferred by new technology for creatingreproductions, they must remember the distinctivelimitations of every form of reproduction and the con-tinuing need for the artifactual sources on which thereproductions are based. Not only do those artifactsprovide the standard for judging the reproductions;they also contain, in their physicality, unreproducibleevidence that readers (scholars, students, and the gen-eral public) need for analyzing and understanding,with as much historical context as possible, the writ-ings that appeared and reappeared in them. If weapproach the electronic future with these thoughts inmind, we will be more rigorous in our demands of newforms of textual presentation and more vigilant in ourprotection of the artifacts embodying the old forms.Both these actions are necessary to ensure the continu-ation of productive reading, teaching, and scholarship.

The Modern Language Association of America rec-ommends that representatives of library, conserva-tion, and scholarly organizations form a task groupto promote continued thinking and cooperativeactivity leading toward (1) the maximum retentionand preservation of textual artifacts, as well as arefining of the selection criteria necessarily entailed,and (2) the use of responsible procedures in the cre-ation and identification of photographic and elec-tronic reproductions based on those artifacts.

28 • Statement on the Significance of Primary Records

Introduction

G. Thomas Tanselle

I

The material gathered here results from the activitiesof the MLA Ad Hoc Committee on the Future of thePrint Record, which during the past two years has con-sidered how best to publicize the importance of preserv-ing textual artifacts after the texts in them have beenreproduced in microfilm, electronic, or any other form.The committee, which (despite its name) understandsits charge to encompass manuscript as well as printedmaterial, consists of Shelley Fisher Fishkin (English,University of Texas, Austin), Phyllis Franklin (ex officio,MLA), Everette E. Larson (Hispanic Division, Libraryof Congress), Philip E. Lewis (French, Cornell Univer-sity), J. Hillis Miller (English, University of California,Irvine), Ruth Perry (English, Massachusetts Institute ofTechnology), Alice Schreyer (Rare Books, University ofChicago Library), Philip Stewart (French, Duke Uni-versity), and G. Thomas Tanselle, chair (John SimonGuggenheim Memorial Foundation).

The origins of the committee go back to a paper thatPhyllis Franklin, executive director of the MLA, read atthe June 1992 convention of the American LibraryAssociation. In preparation for that paper, she took aninformal survey of 319 people (members of the MLAand of other associations in the American Council ofLearned Societies), asking them to comment on thewidespread belief that reproductions of texts can sup-plant the originals. Of the 169 respondents, 94.5%affirmed the importance of preserving primary records,noting the inadequacy of reproductions for biblio-graphical and textual research and for studies that focuson the materiality of texts. In the light of this unam-biguous response to her survey and of the interestaroused by her paper, she asked the Executive Councilof the MLA to consider what the MLA might do tofurther the cause of the preservation of textual artifactsin an age that has seen considerable discarding and

deaccessioning of materials once they are reproduced.At its February 1993 meeting the council establishedan Ad Hoc Committee on the Future of the PrintRecord, charging it to develop an association statementon the subject and to determine other ways of publi-cizing the issues.

The main activity of the committee thus far hasbeen to prepare the “Statement on the Significance ofPrimary Records” printed here, which is the product offour meetings (8 Oct. 1993, 4 Mar. and 14 Oct. 1994,and 31 Mar. 1995), correspondence and telephoneconversations between meetings, and the considerationof many letters of suggestion from interested persons inthe library and scholarly worlds. The draft of the state-ment that emerged from the 14 October 1994 meetingwas given wide circulation on the Internet, by mail dis-tribution to members and other interested persons, ashandouts at the San Diego convention last December,and in the Spring 1995 MLA Newsletter. In addition,one of the three convention sessions held by the com-mittee (sess. 499, on 29 Dec.) was an open hearingentirely devoted to discussion of the draft statement.The committee took all the responses to the draft intoaccount and at its 31 March 1995 meeting produced aconsiderably revised document for submission to theMLA Executive Council. On 19 May 1995 the coun-cil formally adopted this version as an official state-ment of the association. Besides being printed here, the“Statement on the Significance of Primary Records”will be circulated widely to newspapers and magazinesas well as to scholarly and library associations. The


The author is Vice President of the John Simon GuggenheimMemorial Foundation.

PROFESSION 95

committee hopes that it will be endorsed by othergroups or used by them as the basis for statements oftheir own and that it will thus serve to promote furtherdiscussion and action.

The pages that follow present a number of relatedpapers stimulated by the committee. All but the lastare drawn from two of the sessions that the committeesponsored at the 1994 convention to acquaint the MLAmembership with the relevance of the committee’s workto research and teaching. The first of these sessions wasa forum entitled “The Importance and Challenge ofPreserving Research Materials in Their Original Forms”(sess. 254, on 28 Dec.), which was divided into twoparts. Part 1 consisted (after a brief introduction fromme as chair of the committee) of two major papers:one by Paul Mosher, director of the University of Penn-sylvania Library, on the electronic future as seen from alibrarian’s perspective, and the other by J. Hillis Miller,a former president of the MLA and a member of thecommittee. Miller’s paper is printed here and providesan excellent illustration of how the physical presenta-tion of a text affects reading and why a reprint can be aprimary record for studying a critic’s response.

Part 2 of the forum, organized by Ruth Perry underthe title “Object Lessons,” presented six speakers whooffered personal testimony, with specific examples,regarding the way physical evidence has been impor-tant in their own work. Three were selected for inclu-sion here to represent the range of situations dealtwith: Susan Staves, on the eighteenth-century play-wright and novelist Elizabeth Griffith; Miriam Fuchs,on the diaries of Queen Lili‘uokalani; and Anthony R.Pugh, on Proust’s manuscripts.

The second of the committee’s sessions at the con-vention was entitled “Teaching in the Library: AWorkshop on Using Primary Materials in the Class-room” (sess. 459, on 29 Dec.). Shelley Fisher Fishkin,who presided, had chosen ten short papers thatdescribed specific instances in which teachers put pri-mary records to successful use in the classroom. Aswith part 2 of the earlier session, three of the papershave been chosen to reflect the rich diversity of thatsecond session: Manon Anne Ress, on Diderot’s Ency-clopédie ; Gregg Camfield, on Mark Twain; and Cather-ine Golden, on Victorian serialization.

The final piece is by Ruth Perry and entitled a post-script because it is an extension of a point made in thestatement. Her paper calls attention to the essentialrole that public libraries play in intellectual and cul-tural life and to the fact that the existence of manylibraries is currently threatened. The committee did

not include this topic in the statement because it wasfocused on the role of primary records in understand-ing the past; the value of the intellectual exchange thathas traditionally taken place in public libraries, thoughcertainly a valid point, is a separate concern. (Thecommittee, by the way, has emphasized throughoutthat it takes no position about whether printed or elec-tronic forms are more desirable for the disseminationof new writing; its concern has been solely with theimportance of artifacts in reading—the importance,that is, of reading texts in the physical forms they tookat the historical moment one is studying.) The state-ment does, however, make clear that public as well asacademic libraries have performed a great service inbringing the historical forms of texts to a broad public;and it is this point that the Perry paper builds on.

II

Because these papers focus on showing, throughexamples, the practical uses of primary records, it isperhaps in order here to provide—as background forthe concisely expressed theoretical points in the “State-ment on the Significance of Primary Records”—someof the comments I made at the opening of the 28 De-cember forum. In pointing out that reproductions oftexts cannot entirely supplant the original forms ofthose texts, the committee is not making any criticismof the current programs for microfilming brittle booksor for creating databases of electronic texts. Obviouslya microfilm of a book is better than no book at all, andelectronic texts are searchable and manipulable in waysthat printed texts are not. The committee’s aim is notin the slightest to disparage new developments butsimply to make more widely understood the fact thatno reproduction of a text can ever be a fully adequatesubstitute for the original, since every reproductionnecessarily leaves something unreproduced. Besides,there is always an uncertainty attaching to a reproduc-tion; the user of one at any point may wonder whetherthe original was accurately rendered, and the only wayto find out is to examine the original. The use of origi-nals as the ultimate check on the accuracy of reproduc-tions is simply an illustration of what it means to useprimary evidence. Even if there were no other reasonfor needing access to originals, this one is sufficient.

But there are other reasons that are rooted in the sig-nificance of artifacts and the relation of form to con-tent. All artifacts—not just books—can be studied asphysical objects to discover two major classes of his-

30 • Significance of Primary Records: Introduction

torical information that can influence the interpretationof any visual or verbal symbols present on the objects.One class relates to their production history, to thetechniques of their manufacture; the other focuseson their postproduction history, on the implications oftheir physical appearance once the objects were created.Those scholars pursuing the first class of informationexamine objects for clues that reveal details about howthe objects were made. In the study of printed books,this pursuit is called analytical bibliography, and it canprovide information about typesetting, proofreading,and presswork—information that is essential not onlyto printing and publishing history but also to textualhistory and textual criticism (the genealogy of texts andthe evaluation of their correctness, according to what-ever standard of correctness is chosen). In the latter halfof the twentieth century, an age of scholarly editing inmany fields, editors have increasingly recognized thatone of the foundations of their work is analytical bibli-ography—which in turn requires an understanding thatprinted books are like manuscripts in offering primaryevidence for textual study and in regularly presentingvariant texts, since even copies from a single editioncan be expected to contain variations (a point that ap-plies to books of all periods).

The second major approach to artifacts concentrateson the sensuous—primarily visual—characteristics ofobjects. Every object, whether or not it was intendedby its producers to have a utilitarian function, can belooked at for whatever aesthetic value it may possess.The historically oriented form of this investigation,when applied to books, not only attempts to show howtheir visual and tactile features (such as typography,layout, leaf size, and binding) reflected cultural trendsbut also tries to understand how those features haveaffected the responses of readers over the years. Suchresearch is clearly relevant to the history of reading andof the spread of ideas—that is, to the broad field oftencalled histoire du livre, the history of the effect ofprinted books on society, which has attracted a greatdeal of attention in the last several decades.

Of the two approaches to books as artifacts, the firstdeals with hidden evidence, with details not normallynoticed by readers; the second treats of features thatreaders were meant to notice and that do in fact influ-ence, to one degree or another, their interpretations ofwhat they have read. The first produces evidence forreconstructing the texts that authors (or others)intended; the second looks at the texts that actuallyappeared and their physical settings. (A discrepancybetween intended and published texts—that is, be-

tween works and documents—is always to be ex-pected, since the medium of verbal communication,language, is intangible and any tangible representationof it may distort what was intended, even in thoseinstances where visual effects were part of what wasintended.) The two approaches are thus complemen-tary. Both illustrate the ways in which the reading ofphysical evidence is involved in the interpretation oftexts; both show why the historical study of printedtexts rests on the examination of the actual artifacts inwhich they have appeared.

It follows from these points that the books in exist-ing book stacks should never be abandoned, becausethey will remain crucial as the original sources forfuture study of works transmitted in printed form.There can be no book in which the format and otherphysical features are unrelated to the process of readingand understanding the book’s text. But a recognition ofthis fact does not stand in the way of an enthusiasticacceptance of the developing technology for the elec-tronic dissemination of texts. After all, even thosescholars who understand that microfilm and xero-graphic copies do not fully substitute for originals havegladly used them as convenient interim tools. Theavailability of printed texts in electronic form is anadvance greater in degree but not different in kind: itaccomplishes in a far more sophisticated way the samefunction that xerography has fulfilled, making textswidely accessible and more easily manipulable at theprice of removing them from their original physicalcontexts. All scholars should welcome the day whenthey can sit in their studies and call up on their termi-nals an enormous array of texts without the cumber-some process of interlibrary loan or the ordering ofxerographic copies. But they should also realize whatevidence they are thereby missing and why recourse tothe originals can never be rendered irrelevant, howeverinconvenient it may happen to be. Many discussions ofthe future of libraries speak of access replacing owner-ship; but when it is understood that access to physicalevidence is an essential kind of access and that printedbooks must therefore be preserved in multiple copies,the questions of ownership and care remain significant.

The theoretical content of the “Statement on theSignificance of Primary Records,” in short, is that textsand their settings are not separable; that all thecharacteristics of the artifacts conveying texts are poten-tially relevant to the act of careful reading; that thosecharacteristics can differ even among copies of individ-ual editions; and that there is a consequent need to pre-serve as many copies of printed editions as possible in

G. Thomas Tanselle • 31

order to maximize both the quantity of evidence avail-able and the access to that evidence. The usefulness oftextual reproductions is not in question, but it has nobearing on the rationale for the preservation of artifacts.

III

Those who wish to read further on this topic mightturn to several of the papers in the published proceed-ings of the Houghton Library fiftieth-anniversary sym-posium (ed. Wendorf), especially the papers by NicolasBarker, Werner Gundersheimer, Alexandra Mason,David McKitterick, Ruth Perry, and me (my com-ments are largely reprinted in section 2 above). Otheruseful readings are Elizabeth Witherell’s presidentialaddress to the Association for Documentary Editing,D. F. McKenzie’s concluding remarks at the ElvethamHall conference on humanistic scholarship and tech-nology, and my “Reproductions and Scholarship”(which contains many references to related material, asdoes a forthcoming article of mine entitled “TheFuture of Primary Records”).

The committee hopes that the present addition tothe literature of this subject—in the form of the state-ment and the articles offered below—will arouse furtherinterest in the cause of preserving textual artifacts. Bythe end of 1995, various members of the committeewill have spoken on this subject at the New York Pub-lic Library, the University of Kansas, and meetings ofthe American Library Association, the American Insti-tute for Conservation, the College Language Associa-tion, the National Council of Teachers of English, andthe Rare Book School of the University of Virginia.Thecommittee has also scheduled two sessions at the 1995MLA convention in Chicago, one to be presided overby Philip Lewis, on decision making in libraries, andthe other to be chaired by Alice Schreyer, on shared

decision making. A sharing of ideas and discussionamong relevant professional organizations (many ofwhich have already given thought to these questions)is the heart of the recommendation made at the end ofthe statement, and the proposed task group would pro-vide a way for all interested parties to pursue the issuestogether. The MLA can usefully act as a catalyst in set-ting this joint activity in motion; it has taken the firststep in implementing the committee’s recommendationby accepting an invitation from the Preservation of Li-brary Materials Committee of the Association of Re-search Libraries to form a joint working group. In thisspirit of cooperative action, the committee encouragesmembers of the MLA to distribute the statement topersons who might not see it otherwise and to bring itto the attention of other organizations with which theyare connected. The MLA office welcomes letters frommembers reporting on such initiatives or commentingon related matters.

Works Cited

Franklin, Phyllis. “Scholars, Librarians, and the Future of PrimaryRecords.” College and Research Libraries 54 (1993): 397–406.

McKenzie, D. F. “Computers and the Humanities: A Personal Syn-thesis of Conference Issues.” Scholarship and Technology in theHumanities: Proceedings of a Conference at Elvetham Hall, Hamp-shire, UK, 9th–12th May 1990. Ed. May Katzen. London:Bowker-Saur, 1991. 157–69.

Tanselle, G. Thomas. “The Future of Primary Records.” Encyclope-dia of Library and Information Science. Vol. 58. Ed. Allen Kent.New York: Dekker, 1996. Forthcoming.

———. “Reproductions and Scholarship.” Studies in Bibliography42 (1989): 25–54.

Wendorf, Richard, ed. Rare Book and Manuscript Libraries inthe Twenty-First Century. Proc. of Houghton Library Fiftieth-Anniversary Symposium, Sept. 1992. Cambridge: Harvard ULib., 1993. Also printed as Harvard Library Bulletin ns 4.1 and4.2 (1993).

Witherell, Elizabeth Hall. “ADE Presidential Address.” Documen-tary Editing 16 (1994): 1–2, 20.

32 • Significance of Primary Records: Introduction

What Is the Future of thePrint Record?

J. Hillis Miller

It is a great honor and a great responsibility to be amember of the MLA Ad Hoc Committee on the Fu-ture of the Print Record. The future of the print recordis jeopardized in two quite different ways these days.The first threat: approximately one hundred millionbooks and other materials in United States librariesprinted on acid paper will become unusable duringthe next several decades. They are oxidizing, slowlyburning up, becoming brittle, crumbling away, andbecoming unreadable. Second threat to the print rec-ord: new electronic communication technologies arebringing about a revolution as great as was the shiftfrom manuscript culture to print culture. Books andother materials printed on paper will become, indeedhave already become, less and less important in the newelectronic culture we are rapidly entering. Computers,e-mail, faxes, the Internet, electronic books, and mul-timedia materials are already decisively transformingresearch and teaching in the humanities. They are do-ing this in ways we have hardly begun to understandfully, since we are in the midst of the revolution. Bookswill be with us for a long time, decisive in the lives ofmany for the foreseeable future, but already the sensi-bilities, the ethos, the politics, the sense of personalidentity of many of our citizens, including college stu-dents and faculty members, are determined more bytelevision, cinema, and video than by printed books.

Both these changes are, for better or worse, irrevers-ible. Those brittle books are going to fall apart. Theelectronic revolution has already, to a considerable de-gree, occurred. It joins the end of the Cold War andthe globalization of university research (which meansthat universities more and more serve transnationalcorporations rather than the nation-state) as one of thethree major factors that are rapidly transforming Amer-ican higher education.

In preparing these remarks, I have asked myself whatI really do think about the use of original materials.One thing is clear to me. The first obligation of theMLA is to support vigorously those efforts in textualpreservation, now funded to a large degree by theNEH, that will at best be able to save only twenty-fiveor thirty percent of the titles printed on acid paper.

The second obligation: the MLA needs to makeevery effort to study the effects of the electronic revo-lution, along with those of the globalization that goeswith it, and to make sure that it happens in ways thatwill be beneficial to our interests. To study this revolu-tion means supporting the radically new graduatetraining that will make our young scholars and teach-ers appropriately educated for the study of many cul-tures (as in, for example, global literature in English orUnited States literature in languages other than En-glish) as well as for the study of those media that mixlanguage with other visual and auditory materials,media such as cinema, television, and video, whichhave such influence on our lives today. To make surethe electronic revolution proceeds in ways beneficial toour interests means resisting the rapid commercializa-tion of the Internet that is at this moment occurring. Itmeans also doing our best to make sure that electronicstorage of printed materials carries as much as possibleof the history that is embodied in the physical artifacts:for example, all the illustrations in Victorian novelsand all the information in the dust jackets, title pages,


The author is Distinguished Professor of English and ComparativeLiterature at the University of California, Irvine. A version of thispaper was presented at the 1994 MLA convention in San Diego.

PROFESSION 95

end pages, and so on of physical books in general thatelectronic books now characteristically leave out butcould easily include. We need to ensure the preserva-tion of full bibliographical information about the orig-inals when books are electronically or photographicallystored. We need to urge care in the choice of exemplarsto be copied. We need to urge those who prepare elec-tronic transcriptions to follow the guidelines of theText Encoding Initiative.

The third obligation is to attend closely to the usesof original materials and to save as many of those arti-facts as possible. I strongly urge that the MLA appointa joint working group with the Association of ResearchLibraries to make decisions about preserving originalmaterials where the artifactual value is questionable. Inurging that this joint group be formed, I join Betty G.Bengtson, chair of the ARL Preservation of LibraryMaterials Committee. In her recent letter to PhyllisFranklin she said: “The issue of cost is critical, giventhe magnitude of the preservation problem and thevast number of endangered research materials. In acontext where choices will be made, it is vital to distin-guish between materials that have significant artifactualvalue and those for which surrogates can be createdthrough electronic, photographic, or other means.”

The difficulties will come in trying to make suchdistinctions. Let me give a little example. Last summeron Deer Isle, Maine, where I spend my summers, I wasrereading a novel by Anthony Trollope, Ayala’s Angel(1881). I brought with me my old copy of the OxfordWorld’s Classics reprint of this novel. Originally pub-lished in 1929, this reprint was reissued several timesthereafter (my copy is dated 1960) as part of a more orless comprehensive edition of Trollope’s novels. Theywere included over fifty years ago in the World’s Clas-sics series (in which Ayala’s Angel was number 342),long before personal computers were invented. A noteat the end of the book tells me it was “set in GreatBritain at the University Press, Oxford, and Printed byJ. W. Arrowsmith Ltd, Bristol.” It cost “10s 6d. net inU.K. only,” and I bought it in London in the late1960s. I first read the novel in this edition. It is a quasi-sacred object for me, one with which I have a long per-sonal association. I have carried it from place to placeas part of my library. My relation to this object is anexample of the way so many readers of my generationand many generations before mine have participated ina reasonably benign fetishism of the book.

Ayala’s Angel, however, was also available to me onDeer Isle in another way: as an electronic book, part ofthe Oxford Text Archive collection of such books. Ihad access to that by way of my laptop computer and

the modem that connected me by courtesy to theInternet server at Colby College. What is the differencebetween reading Ayala’s Angel in book form and read-ing it in electronic text form? I have stressed the physi-cal embodiment of Ayala’s Angel in the World’s Classicsedition. Not only is the text of the novel caught in themateriality of the book, it is also tied by way of thebook’s paper, cardboard, ink, and glue to the historicaland economic conditions of its production and distri-bution. The edition was part of a moment in Englishpublishing history when one of the great academic-commercial English publishers made classic books ofWestern literature available in inexpensive form. Thismoment was preceded by earlier moments, first by theinitial publication of the book in 1881, then by subse-quent cheap editions. Many of Trollope’s novels werereprinted as yellow-bound paperbacks sold in railwaystations in the late nineteenth century. The twentieth-century World’s Classics version was thus a later stagein English publishing history. It depended on the exis-tence of a large literate middle-class reading public inBritain. It also depended to some degree on the factthat television was not yet available.

The Oxford University Press in the twentieth cen-tury has been, moreover, an international operation. Itsbooks have been marketed all over the world, but espe-cially in cities in what were once British colonies orparts of the British Empire. The globalization of theEnglish language did not occur by accident or becauseof some intrinsic superiority of that language. The listof cities—printed on the page facing the title page—where the Oxford University Press in 1960 asserteditself as located reads like a litany of sites associatedwith British colonialism and imperialism: Glasgow,New York, Toronto, Melbourne, Wellington, Bombay,Calcutta, Madras, Karachi, Kuala Lumpur, Cape Town,Ibadan, Nairobi, Accra. The sun never sets on theOxford University Press. In all these ways, and in oth-ers space does not allow me to specify, the little bookthat I hold in my hand is embedded in history, embod-ies that history in material form, and gives me access tothat history.

The electronic text version of Ayala’s Angel is cut offfrom all signs of historical context. Or, rather, it isgiven a strange new historical placement in the cyber-space of today. A date of original publication is indi-cated, and that is about all. The novel exists not asembodied in material form, or at least not material inthe fixed way of a printed book. It exists as a largenumber of bits of information, zeroes and ones in-scribed as magnetic differences on a hard disk or onmagnetic tape or as minute scratches on an optical disk

34 • What Is the Future of the Print Record?

or as electrical pulses on the wired and wireless trans-missions of the Internet. Ayala’s Angel as an electronicbook takes on a new meaning when it is placed in thisnew context, when it floats in cyberspace. It is de-tached from its local historical context and becomes atext in the context of an enormous and incoherentabundance of works of all kinds—verbal, pictorial, andauditory—on the Internet. As such it might nowbecome the object of a globalized “cultural studies” byscholars who are themselves more and more trans-formed—in part by their use of the computer and bytheir inhabitation of cyberspace—in their relation tothe culture of the book. This transformation is occur-ring even though it is still a primary goal of literary his-tory and literary criticism in the modern languages tounderstand and interpret that culture of the book.

To show the difficulties involved in deciding whichoriginal materials to preserve and which not, I havedeliberately chosen an example, Ayala’s Angel in theOxford World’s Classics edition, that is not original inthe ordinary sense. A well-known essay by R.W. Chap-man long ago demonstrated how unreliable as texts theOxford World’s Classics editions of Trollope’s novelsare. Probably this edition of Ayala’s Angel would notqualify as a book worth saving in its original form forits artifactual value, whereas a first edition of Ayala’sAngel might conceivably do so. My example is meant

to show, however, that much can be learned about his-tory, even with such secondary or tertiary editions, fromclose attention to the materiality of the book, its bind-ing, dust jacket, title page, and so on. My example isalso meant to show how difficult it is, in practice, todistinguish between the book as artifact and the bookas the bearer of pure verbal information, data thatmight be transcribed unchanged and without loss intoany form, including electronic, just as it might be trans-lated, without loss, into another language. This doesnot weaken my allegiance to the three obligations Ibegan by identifying, or the hierarchy in which I placedthem, but it does indicate the extreme difficulty ofdeciding which books to save in their original form fortheir artifactual value, as they cannot all be saved. Nev-ertheless, we must decide. I hope the MLA will play animportant role in that process.

Works Cited

Bengtson, Betty G. Letter to Phyllis Frankin. 7 Dec. 1994.Chapman, R. W. “The Text of Trollope’s Novels.” Review of English

Studies 17 (1941) 322–31.Trollope, Anthony. Ayala’s Angel. 1881. World’s Classics 342.

Oxford: UP, 1929. Online. Oxford Text Archive. World WideWeb. August 1994. Available FTP: black.ox.ac.uk./ota/english/Trollope/ayala.1873.

J. Hillis Miller • 35

Discussion

Traces of a Lost WomanSusan Staves, Brandeis University

Elizabeth Griffith was the author of six plays, threenovels, and a variety of other works published between1757 and 1782. With her husband, Richard Griffith,she also published six volumes of A Series of GenuineLetters between Henry and Frances. Although we do notknow to what extent the letters have been edited sincethe manuscript letters have not survived, these never-theless do seem to be genuine letters between Richard,writing as Henry, and Elizabeth, writing as Frances.The letters chronicle first a seduction attempt resisted,then a marriage, and finally many happy years inwhich the two share love and common literary inter-ests (see Tompkins 1–40).

Until recently, scholars were inclined to dismiss Grif-fith as unworthy of serious inquiry. They contentedthemselves with fleeting mention of her works in com-prehensive literary histories of the drama or the novelor produced perfunctory notes explaining who she wasin cases where she or her works were mentioned in thetexts of canonical writers. Not surprisingly, the lowlevel of scrutiny she was thought to deserve accountsfor a significant amount of disinformation in referenceworks. For instance, in Bibliotheca Britannica the titleof Griffith’s play A Wife in the Right is transformed—whether through the accident of a printing error or bya Freudian slip on the part of the nonfeminist com-piler—into A Wife in the Night.1

Now, of course, critics and scholars are very interestedin refinding, researching, and reevaluating the works ofwomen writers of many historical periods and manycountries. Some early women writers have been pro-foundly lost, with no trace of their works apparently re-maining; others only relatively lost, as in the case wherewe have copies of at least some works and reason to sus-pect the existence of more.2 Three facts pose particulardifficulties for those compiling lists of the works of earlywomen writers we are now trying to rediscover. First,

many women circulated their works in the form of un-signed manuscripts or published them anonymously orwith formulas like “By a Lady” instead of the author’sname. Second, some women wrote kinds of textsscorned by collectors and libraries, like children’s books.And, third, most women changed their names on mar-riage, and some women were married more than once.

The truism we do not know what we do not know isespecially relevant to the study of neglected womenwriters. Let us suppose that we replace the survivingprinted copies of Griffith’s works with reformatted ver-sions, anything from microfilm to electronic text.What sorts of information present in the existing paperdocuments might disappear in the reformattings?

Much, of course, depends upon the principles of se-lection used. Suppose that our reformatting programdecides to select “good copies” of first editions of origi-nal works—a frequent and unsurprising choice.We thenlose evidence of subsequent authorial revision, and, inGriffith’s case, sometimes more important, the evidencecontained in preliminary matter to later editions. Forinstance, in the second edition of volume 1 of GenuineLetters Griffith added a new dedication, “To my Sex.”(For comments on the significance of this and othervariations between different editions, see Bernstein.)

No one, after all, can really tell what a good copy iswithout collating it with others. The value of the tex-tual critic’s maxim that every copy is unique untilproven otherwise is amply demonstrated by the workon Griffith that Brandeis graduate students and I havedone (see Staves, “Revising the Pedagogy”). HoraceWalpole’s copy of her first performed play, The PlatonicWife, has pasted into it a printed announcement withthe heading “To the PUBLIC from the Author of thePLATONIC WIFE,” an attempted defense of herself


PROFESSION 95

against charges of “Indelicacy” in the play. This lookslike it was clipped from a contemporary newspaper,but we have not been able to find this text in any sur-viving newspaper.

One copy of the Dublin second edition of volumes 1and 2 of Genuine Letters (currently at the BeineckeLibrary of Yale University) is quite imperfect, missingmany pages, and, as the librarians say, “mutilated” invarious ways by marginalia and crossings out of wordsand passages. The mutilation, however, seems to havebeen done by Griffith herself, who used this particularprinted copy as a base text for making revisions in-tended for a future edition (although these revisionswere not incorporated in later editions). Many of thedeletions are of low phrases, descriptions of Elizabeth’spoor health, references to money, or references toRichard’s passions—all revisions designed to make thetext more genteel and more belletristic. Griffith alsodecides to excise a playful early threat that she will pub-lish Richard’s letters, presumably because it makes herappear too aggressive, too poor, and too mercenary:“[I]f I am reduced, I vow, I will print your Letters—I think they will keep me in Tea, clean Linen, andPlays. . . .” The printed letter is signed “Your affection-ate Pauper,” but “Pauper” is crossed out and replacedwith the more decorous “Frances” (55). Richard’s wishfor the “Enjoyment” of her “Person” is transmuted intoa chaster hope for the “certainty” of her “Love” (122).His willingness to marry her is made less casual andmore eager. Instead of forming “a Sort of vague Deter-mination in his Mind, to marry her,” he forms “a De-termination” and the caveat that “he had not resolvedwith himself on the Time” is stricken (163). When shewrites with the direct question of whether he intendsever to make her his wife, a deleted passage unromanti-cally explains that he deferred a reply until he saw her,“for he did not chuse to give any thing under his Hand,which might be construed into a Contract” (268).

Among versions of texts “of no authority” likely to beignored in reproduction programs designed to preserve“intellectual content” are translations.3 Translations arealso likely to get short shrift in reproduction programssupported by government funds and designed to pre-serve particular national heritages. Yet translation haslong been an important literary medium for womenwriters, and earlier norms of translation practice oftenmade less of a distinction between translation andadaptation than we do now. Griffith did a number ofbook-length translations from the French, which oninspection add considerably to our knowledge of herideas. She believed that translators had a right to com-

ment on the texts they translated, so her translationscharacteristically contain interpolated, even on occasionfeminist, commentary. Despite her English literary per-sona as a champion of sentimental virtue, she translatedsome French libertine texts, notably The Memoirs ofNinon de L’Enclos and Claude Joseph Dorat’s The FatalEffects of Inconstancy.4 When one of Dorat’s libertineprotagonists discusses women’s incapacity for disinter-ested resistance to seducers, Griffith retorts in a note:

Such is the artful and insidious manner of arguing, with alllibertine wits; but ’tis certainly most unphilosophic. Theyseem to speak of Women, not only as of a different gender,but of a different species, too, from Men. There is no distinc-tion of Sexes in virtue or vice; and whatever has been oncedetermined to be the point of honour, in man or woman, willbe equally defended, by each. (9–10)

Who would have guessed a hundred years ago thatin 1995 roughly half the new professors of literaturewould be women or that so many of them would wantto study women’s writing? Not only do we now notknow what we do not know, we cannot predict whatscholars or what society will want to know a hundredyears from now. But we certainly can proceed to letdecay or to destroy the printed texts that could supportthe inquiries of 2095.

Notes1A less amusing example in a more recent work attributes the

novel The Gordian Knot to Elizabeth Griffith, though it was in factwritten by her husband Richard (Martin, Mylne, and Frautschi).That this error should arise is not startling, since Elizabeth andRichard together published Two Novels: In Letters. By the Authors ofHenry and Frances, the set containing one novel by her, The DelicateDistress, and another by him, The Gordian Knot.

2For example, we know that Ann Masterman, Griffith’s contem-porary, was the author of one novel that survives, The Old Maid,but contemporary sources say she wrote more than this. We havenot yet found another title that can be attributed to her. See Staves,“Matrimonial Discord.”

3For the phrase “intellectual content,” see United States 6.4For a discussion of the significance of Griffith’s involvement

with libertine texts and details about her interventions in Memoirsof Ninon, see Staves, “French Fire.”

Works CitedBernstein, Susan David. “Ambivalence and Writing: Elizabeth and

Richard Griffith’s A Series of Genuine Letters between Henry andFrances.” Eighteenth-Century Women and the Arts. Ed. Frederick M.Keener and Susan E.Lorsch. NewYork: Greenwood, 1988. 269–76.

Bibliotheca Britannica; or, A General Index to British and Foreign Lit-erature. 4 vols. 1824. New York: Franklin, n.d. S.v. “Griffith,Elizabeth” and “wife.”

Discussion • 37

The Diaries of Queen Lili‘uokalaniMiriam Fuchs, University of Hawaii, Manoa

Griffith, Elizabeth, trans. The Fatal Effects of Inconstancy; or, Lettersof the Marchioness de Syrcé, the Count de Mirbelle, and Others. ByClaude Joseph Dorat. Vol. 1. London, 1774.

Griffith, Elizabeth, and Richard Griffith. A Series of Genuine Lettersbetween Henry and Frances. The Second Edition, Revised, Corrected,Enlarged, and Improved. By the Authors. Vol. 1. Dublin, 1760.

Martin, Angus, Vivienne G. Mylne, and Richard Frautschi. Bibli-ographie du genre romanesque français, 1751–1800. London andParis: Mansell and France Expansion, 1977.

Staves, Susan. “French Fire, English Asbestos: Ninon de Lenclosand Elizabeth Griffith.” Studies on Voltaire and the EighteenthCentury 314 (1993): 193–205.

———.“Matrimonial Discord in Fiction and in Court: The Case ofAnn Masterman.” Fettered or Free? Collected Essays on Eighteenth-Century Women Novelists in England, 1670–1815. Ed. Cecilia Ma-cheski and Mary Anne Schofield. Athens: Ohio UP, 1986. 169–87.

———. “Revising the Pedagogy of the Traditional Scholarly MethodsCourse: The Brandeis Elizabeth Griffith Collective.” Eighteenth-Century Women and the Arts. Ed. Frederick M. Keener and Susan E.Lorsch. New York: Greenwood, 1988. 255–62.

Tompkins, J. M. S. The Polite Marriage. Cambridge: UP, 1938.United States. Commission on Preservation and Access. Preserving

the Intellectual Heritage: A Report on the Bellagio Conference,June 7–10, 1993. Washington: Comm. on Preservation and Ac-cess, 1993.

38 • Significance of Primary Records

Many factors in Hawaii work against the preservationof print media. They include Hawaii’s semitropical cli-mate—high humidity and frequent rain—and longtradition of the open-window system of air cooling,which takes advantage of the trade winds. State anduniversity buildings that house important documentsare usually air-conditioned but not always in all rooms,and the conversion to air conditioning has been grad-ual and slow. For example, only one floor of the under-graduate library at the University of Hawaii at Manoa iscurrently air-conditioned; my office still waits. Roomsused by the English department were only recentlyconverted, putting an end to such dramas as classroomwindstorms and visits by local birds. Sunlight, evenwhen filtered through windows on the opposite sideof a room, has a quick, devastating effect on printmedia. Books lose their color, and the writing fades.Hardbound covers attract a type of mildew that leavesthem soft and with a slightly furry surface. Paper clipsoxidize and leave documents with brown imprints.Paper stays soggy, and as the ocean salt works its wayinto expensive machinery, printers jam and computerinnards begin to corrode. Cockroaches eat the glue ofbindings until books come apart. What the cockroachesneglect, the bookworms undertake. They burrow theirway through the text and leave behind them pin-sizetunnels, sometimes from cover to cover.

Manuscripts and records are therefore guarded withvigilance, and they are not always easy to gain access to.In particular, eighteenth- and nineteenth-century doc-uments from the time when Hawaii was a monarchyand, as such, a Pacific island nation tend to be kept outof sight in restricted sections of library collections, statearchives, and museums, so cool that one might think

they were refrigerated. But as sovereignty for Hawaiianshas become a pressing ethnic and political issue, in-terest in Hawaiian language, history, and culture hasgained momentum, and increasing numbers of peopleare requesting access to print materials that once wereof interest primarily to scholars and historians. Amongthese documents are memoirs, official reports, and news-paper accounts containing information about the 1898annexation of Hawaii to the United States as well asabout Hawaii’s last reigning monarch, Lili‘uokalani. Theyear 1993 marked the hundredth anniversary of heroverthrow and 1995 the hundredth anniversary ofher formal statement of abdication. These landmarkdates and numerous commemorative activities con-tinue to generate interest in Lili‘uokalani, who couldreadily serve as a rallying symbol for the Hawaiian sov-ereignty movement.

Interested in Lili‘uokalani myself, I did research onthe book Hawaii’s Story by Hawaii’s Queen, whichLili‘uokalani wrote from late 1896 through 1897 andwhich is widely known and read in Hawaii. Publishedin the United States only months before Congress wasto vote for or against the annexation of Hawaii, thebook was Lili‘uokalani’s final effort to intervene in thepolitical process, from which she had been removed byher overthrow, forced abdication, trial, and imprison-ment in her own palace (which now stands restored inthe middle of Honolulu). In the book, Lili‘uokalaniargues and pleads with the American people not to takeover her country. She declares “absolute authority” insaying that “the native people of Hawaii are entirelyfaithful to their own chiefs, and are deeply attached totheir own customs and mode of government; that theyeither do not understand, or bitterly oppose, the scheme

of annexation” (370). Hawaii’s Story by Hawaii’s Queenis rich in Hawaiian history and will undoubtedly becited in future debates on the contested history andstatus of indigenous Hawaiians and of their land andpolitical rights.

The question that arose in my research on Queen Li-li‘uokalani resulted from my willingness to rely on pho-tocopies of her transcribed diaries from 1878 through1906. The original volumes are not readily available tothe public, but I anticipated no particular problems inusing the photocopied versions, which are convenientlyshelved in the Hawaiian and Pacific Collection at theUniversity of Hawaii. I offer the following paragraphsas a cautionary tale to illustrate the dangers of not usingoriginal documents, dangers that unfortunately becomeapparent only when researchers decide for some reasonto examine the original documents. Reproducing pri-mary materials by any method—even simple ones suchas transcription and photocopying—may have the ef-fect of distorting the original text.

Using typed copies of the diaries seemed altogetherreasonable to me—that is, until I began to think, verysurprisingly, that Lili‘uokalani may not have writtenHawaii’s Story by Hawaii’s Queen. It was not my inten-tion to discredit her authorship of a book that is sowidely known. Still, rumors of a ghostwriter or of avery liberal collaboration with the man who served for atime as Lili‘uokalani’s secretary have existed at leastsince the 1930s, when Lorrin A. Thurston, a grandsonof missionaries and a leader of the pro-annexationistparty, declared Lili‘uokalani’s authorship of Hawaii’sStory by Hawaii’s Queen a sham. In his own memoir ofthe overthrow of the monarchy, Thurston uses Lili‘uo-kalani’s diaries to support his embittered accusationsagainst the queen, who had worked so hard to thwarthis plans for a Hawaii that would be tied politicallyand permanently to the United States. After his politi-cal party confiscated the diaries and other personalpapers from the queen, Thurston, who had a lawdegree from Columbia University, examined them.Certain passages turned up in the case against thequeen when a military tribunal of the Provisional Gov-ernment put her on trial for treason. She was foundguilty of misprision of treason and sentenced to fiveyears of hard labor and a $5,000 fine. The sentencewas commuted to imprisonment, and when it wasover, she had spent eight months confined to one roomof her palace, five months under house arrest in herprivate residence, and for another eight months shewas forbidden to leave the island of Oahu (Allen 341).In Memoirs of the Hawaiian Revolution (1936), Thur-ston discredits the queen’s authorship even though,

while imprisoned in 1895, she worked on the first En-glish translation of The Kumulipo, a Hawaiian poemand chant of the Creation, which she published in 1897with her own introduction. (She was also one of Ha-waii’s most talented and prolific composers.) Thurstoninsists, however, that there is too large a disparity be-tween the style of the diaries and that of Hawaii’s Storyby Hawaii’s Queen. “I have checked the record in vari-ous sources,” he writes, and “[t]he English of the book,as compared with that of the diary, is evidence of mystatement. Lili‘uokalani personally was incapable of us-ing such clear-cut English as that published” (175, 180).

To my regret, the typed copies seemed at first toconfirm Thurston’s charge: the entries were often shortand oddly fragmented or elliptical, with a pattern ofabbreviations and errors. Lili‘uokalani seemed also towrite on standard-size paper, but strangely she ignoredits horizontal dimension; her sentences were foreshort-ened and rarely came close to the right-hand margin.There were many oddities in the diaries that I read forthe years 1878, 1885, and 1898 and many inconsisten-cies from diary to diary. Deciding it was necessary toview the original volumes, I learned that some weresequestered in the State of Hawaii Archives Buildingand others in the Bishop Museum, both in Honolulu.I went to the Bishop Museum, which stores thousandsof documents and artifacts from pre- and post-ContactHawaii, and consulted with an archivist there. Heagreed to remove the diaries from a room that is dehu-midified and air-conditioned twenty-four hours a day,which I was not allowed to enter. Returning to anotherroom in which I waited, the archivist handled thediaries with spotless white gloves and carefully placedone volume, then another, and then another on a tablebefore me, and he patiently turned down each fragilepage for me to examine but not to touch. I then dis-covered something that surprised me even more thanhad my initial skepticism of Lili‘uokalani’s authorship,something I would never have discovered had I notseen the original entries.

Utterly absent from the photocopied transcripts andutterly obvious in the originals was the way in whichthe physical dimensions of each diary determined theodd style of its entries. Only by viewing the originalsdid I realize that the diaries, in contrast to the photo-copied versions, come in a variety of shapes and sizes.Furthermore, nearly all the diaries are very small. Forexample, the 1886 diary is 3 by 41⁄2 inches, the 1898diary 3 by 5 inches, and the 1906 diary 23⁄4 by 51⁄2inches. Some of the diaries are so small they are moreaccurately described as appointment books with spaceonly for quick, hasty entries. The materials that I had

Discussion • 39

examined earlier were physically identical, all the stan-dard 81⁄2 by 11 inches, all in bound notebooks the sizeof academic theses. Also, and again in contrast to thephotocopied texts, the original diaries do not generallyhave conspicuous gaps between the text and the edgesof the paper. In fact, the original diaries show what thetyped copies camouflage, that Lili‘uokalani’s handwrit-ing often goes to the very edge of the page, leaving nospace whatever. In the smaller diaries the queen’s hand-writing is very cramped with occasional additions evenwritten upside down, providing evidence that sheworked hard to utilize all the space available to her andwas thus expressive rather than reticent.

The conclusion I drew from even a cursory study ofthe original documents was the opposite of the impres-sion I first received from the photocopied version:Queen Lili‘uokalani was more than capable of writingHawaii’s Story by Hawaii’s Queen. Many of the supposederrors and idiosyncrasies of her prose are a natural andlogical result of writing in what she considered a privategenre and in the physical books she chose for her dia-ries. The typed copies that are available throughoutHawaii give the impression that Lili‘uokalani used stan-dard American notebook paper, which she collectedand bound as a manuscript. They also suggest obviousimprobabilities—that, for example, she typed some ofthe entries or that when she wrote in Hawaiian, she alsotranslated those sentences. In retrospect, I see that theinsufficiencies of the copies are embarrassingly evident,and yet I was slow to recognize them. They corrupt theoriginals in ways that led me, at least for a few days, todivest Lili‘uokalani of authorship of her importantHawaii’s Story by Hawaii’s Queen.

Although without systematic study it is difficult togeneralize about the textual corruptions from one year’sdiary to another, I think it fair to say that the duplicat-ing process and the preparation of fragile materials forundergoing that process create significant problems. Infact, they produce “copies” that are dangerously un-faithful to the original. The decision, for instance, tophotocopy typed versions of handwritten diary entriesproduces odd differences and dislocations. The deci-sion concerning the relation between text and pagetends to magnify minor errors so that they appear glar-ing. Without sufficient editorial apparatus, copies thataren’t really copies emphasize qualities of Lili‘uokalani’sprose that seem idiosyncratic on a standard page but areabsolutely appropriate to the cramped format of theactual diaries. Lili‘uokalani’s prose, removed from itsoriginal context, is thus stripped of its history.

Thurston, who read the original diaries, surely un-derstood that the fragments and elliptical constructions

were the result of the dimensions of the diaries. GivenHawaii’s benign climate, which is anything but benignfor the preservation of print, and given Hawaii’s volatilepolitics, perhaps he believed that the diaries would notsurvive but that his accusations would. The crucial diaryof 1897, which Lili‘uokalani must have written in whileshe worked on Hawaii’s Story by Hawaii’s Queen andwhich would probably corroborate her authorship andoffer details of the drafting process, has never beenfound. In view of the history of Lili‘uokalani’s confis-cated possessions, I suspect that it never will be found.

Thurston proclaims Lili‘uokalani incapable of writingher own book. I conclude that Hawaii’s last indigenousruler not only wrote her own book but also was some-one for whom writing was imperative. Using thosepocket-size diaries was very practical. Lili‘uokalani wasable to keep one or two of them hidden in the folds ofher late Victorian dresses and write the instant she feltthe urge to write. Symbolically, if not in fact, she wasprotecting the record of her life from those who shefeared might gain political ascendancy and who eventu-ally did. They were the same people who imprisonedher, ransacked her private rooms, and confiscated everydiary and personal and official paper they could find, al-lowing her to keep only one document, her last will andtestament. We are uncertain why some of the diariesstill have not surfaced and why others show evidence oftampering and erasures, but we know this much: Lili‘u-okalani could not be stopped from writing. Not onlyare the diaries written in English and Hawaiian, but inlater years, after Hawaii became a Territory of the UnitedStates, Lili‘uokalani began to use a private numericalcode that was not broken until 1971 and occasionally apersonal shorthand. It seems certain that Hawaii’s Storyby Hawaii’s Queen was indeed written by Hawaii’s queen,and it is ironic that the original diaries, but not theirtyped copies, offer strong support for this conclusion.

NoteMy thanks to DeSoto Brown, an archivist at Bishop Museum,

Honolulu, Hawaii, for his helpful reading of this paper.

Works CitedAllen, Helena G.The Betrayal of Liliuokalani: Last Queen of Hawaii,

1838–1917. Honolulu: Mutual, 1982. Lili‘uokalani. An Account of the Creation of the World according to

Hawaiian Tradition. Boston: Lee and Shepard, 1897.———. Hawaii’s Story by Hawaii’s Queen. 1898. Honolulu: Mu-

tual, 1990.Thurston, Lorrin A. Memoirs of the Hawaiian Revolution. Ed. Andrew

Farrell. Honolulu: Advertiser, 1936.


Discussion • 41

I have three anecdotes, three for the price of one: to-gether they indicate that there is more than one aspectto the business of manuscript versus microfilm. One ofthe anecdotes even suggests that we should be happythat some manuscripts have been preserved on film.

For the last ten years or so I have been working onthe manuscripts of Marcel Proust, trying to establishthe chronological sequence of everything that he wrotein preparation for his great novel A la recherche dutemps perdu, from 1909 (when it first took shape, andthe pillars were set down) through 1911 (when Proustcompleted a first draft) to 1913 (when a typescript ofhalf the novel was ready for a publisher; half this type-script was set in proof and half the proof published).My examples come from the first book of the novel,Du côté de chez Swann, the part entitled “Combray”;from the central portion, Le côté de Guermantes; andfrom the last part, now La fugitive or Albertine disparue.

The “manuscripts” are essentially exercise books, withthe addition of a few episodes written on loose sheetsand of a seven-hundred-page typescript that coversabout half the novel as it was envisaged in 1911. Proustwrote on the recto pages of his exercise books, fre-quently passing from one book to another, and maybeanother, and back again to the first; he used his versopages for subsequent additions. Occasionally he savedhimself the trouble of recopying by giving a whimsicalcross-reference (a sketch of a boat, or a butterfly, forinstance), but generally, as he refined his prose andreorganized his episodes, he rewrote the whole text.

This material was all in private hands until the earlysixties, when the Bibliothèque Nationale in Paris ac-quired the lion’s share (other exercise books have comeinto the library since then). It was not in good order,with exercise books falling apart, and so the books wereall “restored,” which means essentially that they wererebound, with the loose pages firmly fixed in place. Apractical method of classification was devised, assigningeach rebound exercise book to the part of the novel towhich it seemed most closely related; thus the booksthat work on the overture were numbered 1–7, thebooks on “Combray” 8–14, the books on “Un amourde Swann” 15–19, and so forth.

For twenty years scholars had a field day with thismaterial, gradually analyzing the contents, identifying

problems, proposing solutions. But the day came, inev-itably, when it was decided that too many fingers in theProustian pie were having a destructive effect on theprecious pages and that henceforth consultation wouldbe done on microfilm, in order to preserve the manu-script. It is not impossible to obtain permission to seethe originals, but all routine work has to be done firstunder glass. In the process, things can get overlooked.

One disadvantage of microfilms, as everyone whohas used them knows, is that all sense of the physicalreality of the book disappears. A microfilm is like ablack-and-white reproduction of an oil painting. Thewords are there, but everything has become intellectu-alized, distant. For the scholar in search not only of evi-dence but also of suggestions, some clues are no longerdiscernible—the color of the ink, for example. Tenyears ago I realized that the first part of the “1911”typescript, including nearly all of “Combray,” musthave been made toward the end of 1909, and I asked ayoung Japanese doctoral student, Akio Wada, who waswriting a thesis on the evolution of “Combray,” if hehad considered that possibility. “Yes,” he replied, “and,what is more, I can prove it.” His proof, which hewould not disclose at the time but which became pub-lic knowledge once his thesis had been defended, wasthat a note in Proust’s personal carnet, which we cancertainly date December 1909 or January 1910, indi-cates in red ink that there were passages added on cer-tain pages. On the typescript all these pages havemarginal additions, and they are all in red ink. Red inkis very rare in Proust. Luckily Wada noticed the red inkin time, before the manuscripts were taken from him.

With microfilms you proceed one dreary page at atime, without the sense of the whole, which you getphysically from the exercise books themselves. My sec-ond anecdote suggests that the mechanical process ofmicrofilm reading encourages laziness. It was always saidthat there was a gap in the manuscript version of thecentral portion of Proust’s novel, Le côté de Guermantes.The manuscript version is found in exercise books 39–43 and 49 of the Bibliothèque Nationale collection.Thelast page of 43 ends in mid-sentence, and the first pageof 49 begins in mid-sentence, and they are not the samesentence. The assumption, never questioned, was thatwe were in the presence of two different versions, the

Manuscripts on Microfilm: The Disturbing Case of ProustAnthony R. Pugh, University of New Brunswick

first version missing only its last exercise book, the sec-ond version known only through its last exercise book.Or maybe it was the other way around; there was somedispute about which was the first version and which thesecond. But nobody doubted that there were two dis-tinct versions. We were, however, in the presence of amodern-day “golden tooth.” The gap between 43 and49 is of only a few lines, as we see if we compare thedraft with the published text, and it seemed to me un-likely that 43 and 49 did not belong together. That theremaining portions of two different versions should fitalmost exactly was too much of a coincidence. Evidentlya few lines, perhaps one manuscript page, was missing.Could the missing text be found? More easily than Ihad imagined; for when I looked at the microfilm ofbook 43, I realized that when Proust had got to theend of his last page (writing, as usual, on the recto),he continued on the facing verso. But once again hereached the end of the page without finishing his sen-tence, and he picked up a new book (our 49) andcompleted it there.The scholars who had studied Guer-mantes had evidently not thought to wind back the reelto the previous image. It can get quite hot and sopo-rific in the microfilm room, but even so. . . . Can youreally afford not to bother to read the verso pages of acrucial manuscript?

My third example comes from later in the novel, thepart that never made it to the 1911–12 typescript.Here we find that the microfilm, far from obscuringthe evidence, actually leads us to the truth aboutthe original version. This part of the novel, reworkedby Proust during the war, is one of the rare instanceswhere he simply cut out pages and included them,without recopying them, in his new manuscript.Another Japanese scholar, Jo Yoshida, had spotted andidentified five such pages. When I was trying to recon-stitute the original 1911 version, using the microfilm, Inoticed a sixth page, of which only the margin (which

was empty) remained in the original exercise book; therest of the page had been pasted into the wartime exer-cise book. But in 1911 Proust had written somethingon the back as well, going of course all the way to theright-hand edge of the page, so when he cut along themargin line of the recto, which was the text he wishedto reuse, the writing on the verso was divided aboutone inch from the end of each line. The text could bereconstituted by putting together the two portions ofthe divided page, but this is quite unpractical onmicrofilm, even if one could for a few minutes use twoadjacent readers. So I preferred to add this question tothe list of questions I needed access to the originalsto answer. Imagine my surprise when I had the original1911 exercise book in my hand and could find no tracewhatever of the page I wanted to see. I can only assumethat the one-inch margin was “tidied up” when theexercise books were rebound and that fortunately thistidying up occurred after the microfilm was made. Ihave no proof of this, however. It shows why theJapanese scholar counted only five such pages: in theearly eighties, he was working straight from the origi-nals. It also presents a horrendous prospect to the con-scientious scholar, who would have to compare everypage of the originals with every page of the microfilmsto be really sure that nothing was missed. This isimpossible, practically speaking, but it shows just howcareful one has to be.

This final anecdote leads to another conclusion.Scholars are responsible for their evaluation of the evi-dence put before them. The onus is on libraries to givevery complete bibliographical descriptions of theirmanuscripts and to suppress no evidence, howeveruntidy it may be. Scholars need to contemplate themess and work it out for themselves. The assumptionsof librarians and archivists may not be beyond re-proach, and it is dangerous to present assumptions asfacts and tidy up the evidence.


Discussion • 43

Undergraduates come to my course on French civiliza-tion with a variety of backgrounds and with differentideas of what studying civilization is all about. Theshape the course takes depends to a great degree on theinterests and needs of the students. However, a basicgoal shared by all students is the discovery of connec-tions among such areas as history, art, music, literature,and religion. Students are introduced to major eventsin French history and to works of art produced inFrance from the Middle Ages to the present. Since wehave only fourteen weeks to cover such a large amountof material, we select a few works of art to illustrate eachperiod we study. Students participate in the choice bydeciding together what event or work of art they willdiscuss more in depth in class or in their written work.Conducted entirely in French, the course is primarily adiscussion class where students are expected to learn howto express their opinions about French civilization. Bythe end of the course they should be able to describeseveral French institutions, historical events. Theyshould also be able to recognize and define basic con-cepts of French society in terms of its social and politi-cal structures and literary and artistic movements.

This year Diderot’s Encyclopédie was the documentselected to start a discussion on the French Enlight-enment. The short extract in the textbook could ofcourse not represent fairly the huge compilation thattook twenty-nine years to publish in full and eventu-ally ran to twenty-one volumes of text, twelve plates,and two indexes. The complexity of the document andits place in French eighteenth-century society are bestinvestigated using the document’s original form. Aftera class during which we discussed the background ofthis formidable work and its effect on French culture,the students were given specific assignments to be car-ried out in the university library. They were to meet insmall groups in the rare books collection and thenreport back to the class about a variety of issues. Wehad agreed on the entries they would check, such as“autorité politique,” “guerre,” “peuple,” or “droits féo-daux.” Two students chose to focus on the famousillustrations of the Encyclopédie. All the students re-ported their findings with enthusiasm and had many

interesting observations. They were struck by the num-ber of volumes (thirty-five) and their size and by thenames of the numerous intellectuals who participatedin Diderot’s project. The look and feel of the docu-ment were also discussed in class—by students whohad never seen or touched an eighteenth-century edi-tion. The calligraphy and spelling interested them. Therare books collection itself was an important discoveryfor them. Most had never been in that part of thelibrary and found the experience fascinating, thoughthe rare-books librarian did not seem to have enjoyedtheir visit much. Some students were reprimandedfor writing on pieces of paper placed directly on theeighteenth-century edition. With humor, they told dif-ferent anecdotes of their first contact with the rarebooks and with the librarian there.

The discussion of the different entries they looked upwas very informative.They had discovered, on their own,through the difficulty of consulting the original work,some of the subtlety of the Encyclopédie. We spoke, forexample, about censorship and self-censorship. I amconvinced that, in many ways, students have gainedmore from working with an eighteenth-century editionof the Encyclopédie than they ever did from workingwith a textbook reproduction of an Encyclopédie articleor two. For example, the students raised interestingquestions about the cost of such an edition and aboutthe number of readers who were able to acquire thesehuge volumes at the time.

Looking at, touching, and reading the Encyclopédie isa wonderful way to put oneself into the mind-set ofeighteenth-century France, to grasp what was consid-ered so sensational, shocking, and revolutionary then.The hundreds of engraved illustrations that show peo-ple at work making all manner of industrial productsdefinitely helped my students understand why theEncyclopédie embodies the Enlightenment and is one ofthe most important documents on eighteenth-centuryFrance in general.

I asked the students to evaluate our pedagogical ex-periment. Most were very positive even when they hadencountered difficulties. Many students told me thatusing an eighteenth-century edition had changed their

Twentieth-Century Undergraduatesand an Eighteenth-Century Edition of Diderot’s EncyclopédieManon Anne Ress, Temple University

conception of what a research project could be. Onestudent I met later told me that she had just got a goodgrade on a research project about Frederick Douglass (ina journalism course) because her use of Douglass’s arti-cles instead of the course’s compilation of secondarysources had helped her shape a personal and creativeapproach. She laughed as she was telling me about the

difficulties of finding his articles in various libraries andabout her perseverance. She struggled with the libraries’catalogs and the texts themselves but eventually learneda lot. At the end of our brief conversation the student,young but a scholar indeed, mentioned that she is stillthinking of trying to find original copies of The NorthStar and maybe even some of Douglass’s manuscripts.


On the Importance of Judging Books byTheir CoversGregg Camfield, University of Pennsylvania

Any proposal to make an exclusively electronic libraryis predicated ultimately on an extreme idealism, an ide-alism crudely but accurately expressed by the axiom“You can’t judge a book by its cover.” This axiom im-plies that the words alone count and that if we couldsimply pour them unmediated into our brains, wewould best be able to judge books. According to thisprinciple—by which we teach our classes out of cheappaperback reprints of authors’ words—the mediumdoesn’t count at all. Get the words right and the stu-dent can read them off microfilm, a VDT, or even lis-ten to them on audiotape.

But as the amount of money publishers spend oncovers suggests, we can—indeed we usually do—judgebooks by their covers. The Harlequin romances at thecheckout counter in the supermarket have covers thattell us not only what but also how we will read.Leather-bound, gilt-edged volumes also tell us how toread: reverentially in the face of transcendent genius,which we have the good taste to purchase and displayostentatiously. My point is simple and obvious: thephysical presentation of a piece of literature gives usessential clues about how we are intended to read itand gives us further clues about the means of its pro-duction and the social role it plays. In my classes onnineteenth-century literature, I insist that the studentstake the physical book into account as a part of theirreading experience. For that I need libraries to haveearly editions of the books I assign.

Let me give an example. In my class on Mark Twain,I want my students to understand what risks Twain ranand what benefits he sought in publishing his books bysubscription. I refer to his own words to describe hissense of his business, but these words themselves referto the physical artifacts quite concretely:

There is one discomfort which I fear a man must put up withwhen he publishes by subscription, and that is wretched paper& vile engravings. I fancy the publisher don’t make a verylarge pile when he pays his author 10 p.c. You notice that theGilded Age is rather a rubbishy looking book; well, the salehas now reached about 50,000 copies—so the royalty nowdue the authorship is about $18,000. (81)

Twain traded aesthetic pleasure for economic power,but what that tradeoff meant is not fully obvious tolate-twentieth-century readers.

To show what the tradeoff meant, I have my stu-dents study the physical copies of books that spanmuch of Twain’s career, from the early days when hedid not control the presentation of his books to thedays when he took control by publishing his bookshimself. Using copies of The Innocents Abroad, Rough-ing It, and The Gilded Age as examples of Twain’s ear-lier works, I have my students take note of the gaudycovers; the cheap woodcuts; the brittle, thin, yellowingpaper; the thin cardboard covers; the crude typesetting;and the simple heft of the books. These details con-cretely show that the books were relatively inexpensive.They suggest further that such books were directed toan audience not used to buying books, an audiencebuying books not for status but rather for entertain-ment, an audience wanting as much as possible for theprice and willing to trade quality for bulk. When I addto this an example or two of subscription book pro-spectuses, my students see all the more clearly the eco-nomic conditions of subscription publishing and theaudience expectations that to a large extent determinedthe range of content of the books.

Thus, reading the material artifacts helps my stu-dents read the content, especially when they turn tothose books that Twain published himself. Their new

insight into the entertainment value of Twain’s earlysubscription books allows them to see how having con-trol over the production of Adventures of HuckleberryFinn put Twain in something of a bind. He wanted thebook to sell as a typical subscription book, but he hadwritten a biting satire. Rather than alienate an audiencethat expected simple entertainment, he hired an artistto draw conventional comic illustrations, essentiallymuzzling the biting prose with silly pictures.

When my students examine the beautiful book thatis Connecticut Yankee, they can see how Twain tried toextend the range of meaning available to a subscriptionbook by changing its material circumstances. Thecover is sturdier, and its ornamentation is more taste-ful. Inside, the paper is relatively slick, allowing forcrisp printing of both text and illustrations. The illus-trations are beautiful and expensive, even thoughTwain hired a previously unknown illustrator in orderto hold costs down. The careful layout of each page tointegrate words and illustration into one visual wholehints at the effects soon to be exploited by popularmagazines of the twentieth century after typesettingcosts diminished with the development both of theMorganthaler typesetter and of several less expensivetechniques for reproducing graphics. But in Connecti-cut Yankee, the typesetting was done by hand, and thecost was high. Not surprisingly, Twain never mademuch money on the book, in part because his desire toexpand the range of the subscription book outran thetechnology available to him.

As the book’s presentation suggests in yet anotherimportant way, Twain may not have cared about profit-ability. Unlike in Adventures of Huckleberry Finn, inConnecticut Yankee he made no attempt to hide his po-litical intentions behind benign illustrations of harmless

clowns. On the contrary, he gave his illustrator, DanBeard, free rein to highlight the book’s satire. Beardpenned biting political cartoons, most of which point-edly attacked abuses of political or economic power. Hebrought to the fore the novel’s allegorical significance bydepicting contemporary political and literary celebritiesin his illustrations. To make illustration explicitly servepurposes other than ornamentation is a notable depar-ture from the formula for success in subscription pub-lishing. This and the other merely aesthetic departuresfrom the conventions of subscription publishing sug-gest that Twain meant his work to have enduring artisticand political value as opposed to ephemeral entertain-ment value. All of this is suggested not so much by thewords as by the book itself. Thus, the material artifacthelps students understand authorial intention as well asthe book’s social and economic circumstances.

I do not know how I could make such points real tomy students without the physical copies of the books. Iunderstand the difficulties of making nineteenth-centurysubscription books available. They are by nature fragile,and their fragility makes them increasingly rare. Rebind-ing robs them of some of their usefulness, but restoringold bindings is difficult and expensive. Yet in spite ofthese difficulties, the benefits of maintaining a multi-faceted understanding of our cultural history are worththe trouble and expense. If we wish to give our studentsa truly usable past, we must give them a realistic ratherthan a purely idealistic sense of the past.

Work Cited

Clemens, Samuel. Mark Twain’s Letters to His Publishers. Ed. Ham-lin Hill. Berkeley: U of California P, 1967.

Discussion • 45

Rekindling the Reading Experience of theVictorian AgeCatherine Golden, Skidmore College

I confess that part of the mystique of visiting the rarebook room at Skidmore College lies in its location; evengraduating seniors are as surprised of its existence asMary Lennox is surprised, in The Secret Garden, whenshe discovers her cousin Colin in a forbidden wing ofMisselthwaite Manor. Like Mary’s visits to the forbiddenwing, visits to the rare book room unlock a world: the

world of Victorian literature and culture. Thus I regu-larly bring students in my nineteenth-century literaturecourses to the rare book room in the Lucy Scribner Li-brary of Skidmore College.Together we examine worksin the Hannah M. Adler Collection, which features illus-trated nineteenth-century periodicals, part issues, books,and multivolume works by leading and popular writers.

Undergraduates reading nineteenth-century novelsoften marvel at the length of works by Dickens,Thack-eray, and Eliot. Seeing long novels in their original in-stallments, however, shows how these texts appeared tothe Victorians who read them as multivolume works,as parts of works published independently, or as serialsin the leading weekly and monthly journals of the day.Examining pivotal texts and periodicals—Eliot’s Mid-dlemarch (1871), Master Humphrey’s Clock, and Bentley’sMiscellany 1—rekindles the reading experience of theVictorian age. A first edition of Middlemarch in eightslim volumes shows how reading a long novel, a chal-lenge for many of my students, was in 1871 a far moremanageable task than that of absorbing the fat paper-backs we assign today. A more dramatic point is madewhen we examine serial publication in the collected vol-umes of Bentley’s Miscellany and in the unbound partissues of Master Humphrey’s Clock, which have theiroriginal blue-green paper wrappers, an artifact of Vic-torian publishing. As Linda Hughes and Michael Lundpoint out in The Victorian Serial, it is well known thatserialization brought forth some of the best literature ofthe age, but the experience of reading a novel over aperiod of two years (or longer), as the Victorians typ-ically did, is foreign to our students, who are givenapproximately two weeks to read the same work. Theinstallments of Dickens’s The Old Curiosity Shop andBarnaby Rudge in the part issues of Master Humphrey’sClock encourage students to ask themselves how readinglengthy novels one segment at a time, with breaks be-tween reading periods, affected the ways Victorian au-diences responded to and created meaning from theseworks as well as the ways in which Victorian authors andartists created fiction. Students today can thumb aheadto discover a mystery’s conclusion; the serial readercould not. Learning the story’s outcome thus was post-poned, gratification delayed. Moreover, when studentsactually see from the publishing format that illustratedserial novels unfolded over months and years, they morereadily understand how the illustrations provided con-tinuity between installments and an aid to memory.Illustrations, a vital part of the reading experience evenof sophisticated Victorians, were studied, as author-illustrator George Du Maurier has put it, “with pas-sionate interest before reading the story, and after, andbetween” (350).

The monthly installments of Oliver Twist from Feb-ruary 1837 to April 1839 (omitting June 1837, October1837, and September 1838) in Bentley’s Miscellany dra-matize the benefit of using primary materials to teachabout illustrated Victorian fiction. Dickens’s novels un-

folded through and with pictures, though modern edi-tions typically eliminate all or most of the original il-lustrations. The monthly numbers of Oliver Twist inBentley’s Miscellany show students all twenty-four illus-trations, allowing them to understand why many Vic-torian readers considered the memorable characters inTwist more Cruikshank’s than Dickens’s. Cruikshankwas known for his caricature art. In his illustrations, thebody of Fagin, who lures Oliver into his den of crime,shrivels in emaciation, while the overweight forms ofMr. Bumble and Mrs. Corney, parish beadle and work-house matron, call attention to the fact that, unlike thinOliver and the backdrop of famished children, thosewho oversee the workhouse have no need to ask formore. Pictures integral to plot and character develop-ment, such as “Monks and the Jew,” have been droppedfrom modern editions,2 and even the best reproductionsavailable today do not capture the nuance of detail andshading so vital to illustrations like “Fagin in the Con-demned Cell,” where the stippled effect on the cell wallcalls attention to Fagin’s circular, glazed eyes and crazedmanner. Seeing all twenty-four images reveals that thestory of Oliver Twist is told through Cruikshank’s pencilas well as Dickens’s pen and adds credibility to Cruik-shank’s grand claims, in his book The Artist and theAuthor, for authorship in his collaborations with Dick-ens and William Harrison Ainsworth.3

The novels we typically read in my Victorian litera-ture classes are all now part of the canon of British lit-erature. However, viewing these texts in their originallypublished format in the rare book room demythologizesthe aura of classic that now surrounds them. OliverTwist, collected in volumes 1–5 of Bentley’s Miscellany,demonstrates how a new serial was often started along-side a successful one nearing completion to entice es-tablished readers to keep purchasing the periodical. Involumes 1 and 2 (each volume contains six monthlynumbers), the installments of the new serial OliverTwist follow Songs of the Month, by Ainsworth, SamuelLover, and other authors now forgotten but then moreestablished than Dickens.The placement of Twist indi-cates the publisher’s lack of certainty about the successof this new serial, even though Cruikshank was quitepopular in the 1830s and Dickens, then better knownas a journalist than as an author, still had enjoyed suc-cess with The Posthumous Papers of the Pickwick Club(1837), which popularized serial fiction in the firstplace. Not surprisingly, Oliver Twist rises to the leadposition in the table of contents of volumes 3 and 4 ofBentley’s. But despite the enormous popularity of Twist,the final four installments drop to the sixteenth position


in the table of contents for volume 5.4 Instead, Ains-worth’s Jack Sheppard has first billing, which reveals thepublisher’s desire to excite his readership with the esca-pades of another Newgate novel.

Extending the boundaries of the classroom into therare book room of a library is more than a field trip toview dusty artifacts. Seeing novels in the very publish-ing format the Victorians once saw them in makes Vic-torian cultural constructs accessible to students. Therare book room thus becomes a window into Victorianculture as well as into the best literature of the age.

Notes1Master Humphrey’s Clock was a weekly periodical written wholly

by Dickens to carry installments of his full-length novels The OldCuriosity Shop (1840–41) and Barnaby Rudge (1841). Bentley’s Mis-cellany was a distinguished monthly periodical including importantcontributions by William Harrison Ainsworth and by Dickens, whowas also the first editor of Bentley’s.

2Many paperback editions that retain illustrations do so only er-ratically. For example, following Dickens’s own later selection of il-lustrations, the Oxford edition of Oliver Twist retains only eight ofCruikshank’s original illustrations, which editor Kathleen Tillotsonnumbers as twenty-five to include an illustration Dickens rejected.“Fagin in the Condemned Cell” and “Oliver Asking for More” areretained; but other, less memorable illustrations, such as “Sikes and

His Dog,” are included rather than, for instance, “The Last Chance”or “Mr. Bumble and Mrs. Corney Taking Tea.” Similarly, the Signetedition of Thackeray’s Vanity Fair retains some of the inset capitaldesigns and an occasional woodcut. Recently, there has been a trendto include illustrations in modern editions, and the complete illus-trations of Vanity Fair, for example, are printed in the 1994 NortonCritical Edition as well as in the Oxford University Press edition ofthat novel. Furthermore, the Norton Critical Edition of David Cop-perfield also includes all of Phiz’s illustrations. These are but a fewexamples of the revaluation of Victorian illustrations in recent paper-back editions.

3Cruikshank’s career ended in controversy over his exaggeratedclaims that his illustrations had altered Dickens’s original concept ofOliver Twist and had inspired Ainsworth’s The Miser’s Daughter andThe Tower of London. His struggles over authority ended his collab-oration with Dickens in 1841 and with Ainsworth in 1844.

4Other factors may have influenced this decision. The final mag-azine installments of Twist appeared serially after the three-volumepublication of Twist and after Dickens had resigned the editorshipof Bentley’s. Also, Ainsworth, the subsequent editor, may have beeneager to showcase his own fiction, Jack Sheppard (1839).

Works Cited

Cruikshank, George. The Artist and the Author. London: GeorgeBell, 1870.

Du Maurier, George. “The Illustrating of Books from the SeriousArtist’s Point of View.” Magazine of Art. Aug.–Sept. 1890: 349+.

Hughes, Linda K. and Michael Lund. The Victorian Serial. Char-lottesville: U of Virginia P, 1991.

Discussion • 47

Postscript about thePublic Libraries

Ruth Perry

The MLA “Statement on the Significance of PrimaryRecords” is part of a wider professional response fromthe library and scholarly communities to problems offunding and space that they are encountering in theirefforts to preserve the print record. The conservation ofdamaged or deteriorating books and the creation anddistribution of adequate reproductions of manuscriptsand printed texts for scholars and students without ac-cess to originals are two issues that concern those re-sponsible for transmitting to the next generation ourcultural and intellectual heritage. Less well understoodand publicized but also crucial to the future of our cul-ture is the threat posed to local public libraries by cut-backs in public spending. The closing of even smallneighborhood branch libraries, with the loss of theircollections, means that the public suffers an irreplace-able loss of a wide variety of books, many dating backto earlier eras and containing material as well as intel-lectual evidence of cultural history. Furthermore, whilethese closings may not directly affect the lives of aca-demic professionals with their university library cards,specialized private collections, and online computertexts, they accelerate our society’s alienation from booksand reading culture and thus in the long run underminethe work we do and the textual history we value. Mostimportant, the defunding of public libraries stymies thetaste for books and reading in those who are poor andless mobile and cuts off their access to a quiet, non-commercial place of respite and imaginative renewal.

Every state in this country has seen branch librariesclose or cut back their hours in the last decade. Oper-ating with reduced staffs, many small branch librariesstay open only a few hours a week, often when peopleworking nine-to-five cannot get to them. Cutbacks infederal funding in combination with the diminishing

tax base for many state budgets (a function of disap-pearing industry and declining property taxes) aremaking it impossible to keep libraries open at timesand in neighborhoods most convenient for the peoplewho most need libraries. Meanwhile Buck-a-Bookstores are springing up in many shopping malls, andlarge chain bookstores flourish, testifying to a newlystructured mass market for books and a new level ofcommodification of literacy and reading.

In California this year there is a statewide library cri-sis because of the reallocation of local property taxesfrom library budgets to school districts. It is an expres-sive double bind—to promote either public libraries orschools, either public ownership of books or institu-tional selection of reading materials, either autonomousbrowsing or regulated learning, either book culture ortextbook culture, but not both. Although the Californiacase has a special poignancy, libraries everywhere havebeen in trouble since at least 1990, given their shrinkingresources to maintain collections and pay expert staff.

Some larger public libraries, forced to raise moneyand save space, have been selling their older, out-of-print books to private dealers. Used book dealers willtell you that the market in rare books is picking up againas libraries quietly deaccession to make ends meet. Butthe cheapest way to dispose of books when they are be-yond repair or when bulging shelves need relief is simplyto throw them away. Susan Koppelman, known for heranthologies of nineteenth-century women’s stories, was


The author is Professor of Literature at the Massachusetts Instituteof Technology.

PROFESSION 95

dismayed to learn, on repeatedly submitting call slipsfor nineteenth-century American women’s novels andcollections of short stories in the Saint Louis Public Li-brary, that many rare and important volumes were per-manently unavailable because they had ended up in alandfill. Indeed, according to Koppelman, many of thesources from which she gleaned her first short story col-lection, Old Maids (1984), have been fatally deacces-sioned. It has been reported that the works of LangstonHughes and Edmund Wilson were removed from someNew England municipal libraries, probably because ofthe poor condition of the books, but the libraries appar-ently had no funding to replace them.

Built long ago as a public resource funded by tax-payers’ money, the network of public libraries in thiscountry belongs to everyone and to no one in particu-lar. These public buildings with their irreplaceable col-lections of books are our common property, like parksand beaches and highways. They are not a phenome-non of the market; not much money changes hands forthe service they provide. The current Congress is lessinclined to support public enterprises than to privatizethem, but while some private service industries, likethe mail businesses, are turning a profit, so far no onehas figured out how to turn a profit from the publiclibraries. Still, there is clearly more money to be madefrom selling books than from lending them, andlibrary collections, of no obvious fiscal advantage toanyone, are being allowed quietly to run down, despitethe dedication of many library workers and managersand the needs of the public.

Yet public libraries are the very cornerstone of a truedemocracy. They provide access to information andideas about everything a thinking citizen might wantto know to make a reasoned judgment, cast a vote, orregister an opinion. Democratic access to knowledge isessential to the free play of ideas, a concept originatingin the Enlightenment and very much conditioned bythe rise and spread of print culture. Books provide lit-erary art and entertainment. They support productiveaims as well and are the resource of inventors anddreamers. They supply information on everythingfrom building machines and filing patents to fertilizinggardens and raising children. A free society needs freepublic libraries.

In our century, public libraries have also been animportant wellspring of serious literary production.Many of our best writers educated themselves in publiclibraries, browsing the open shelves, absorbing influ-ences, coming upon unknown authors long out ofprint, following the trail of sudden interest and inspira-

tion. Libraries thus represent our literary future as wellas our literary heritage. Where will the poets and citi-zens of the future make their unexpected discoveries?Where will they roam uncalculatingly among writersout of fashion and not to be found in any undergradu-ate syllabus? Aspiring writers cannot buy everythingthey need to read. And how many poets can affordcomputer access to online texts—even if screen readingwere the same as hand-held book reading? Democraticaccess to books in our thousands of public libraries willnot soon be replaced by any electronic or marketmechanisms. Professors of literature in particularshould be aware of the cultural cost of the erosion ofthis public resource.

A letter written by the author Helena Maria Vira-montes in 1993 as part of a campaign to save a locallibrary makes the case eloquently:

Several weeks ago I was informed that a branch library in thecity of Orange [California], appropriately called “The FriendlyStop / La Parada De Amistad,” is being shut down.This deeplyconcerns me. I have been involved with the library which is atrailer situated in the barrio of West-Central Orange. The oneroom library is constantly visited by Latina/os primarily, mostlyteens, who have found the library a comfortable reprieve fromthe streets.They read, receive homework assistance, or becomeinvolved in the many bilingual activities the library has tooffer. . . .

How many of us Chicana/o-Latina/o writers grew up inbookless homes? How many of us found solace and rapture inbeing able to attend the library, sit in a quiet place and read orhave the right to exercise our imaginations? I, for one, madean office of a library chair and a piece of table where I wouldsit for hours and read, conduct meetings, write in my journal,dream, even nap. In a house with eleven people, this libraryspace was my private heaven. It was a space filled with floatinganswers, infinite questions, and the quiet time for meditation.It was a space for me like no other and we simply can’t sit byand let this experience be ripped away from our youth who sovery much need it AND want it.

Viramontes speaks for thousands of intellectuals andwriters who rely on such oases in our speedy and mate-rialistic world. More, she speaks for the poor and onbehalf of the young at risk.

Free public space is in increasingly short supply inthis country; there are few places to go any more and fewthings to do that do not cost money. Public libraries areamong the last places left where all people are welcome,qualified for admittance merely by their humanity,curiosity, and literacy. Public libraries symbolize thecommitment of our society to something other thancommercial exchange. They provide democratic accessto books and knowledge for a broad cross section of ourpopulation including the elderly, the self-educated, im-migrants, children, the poor. It is extremely shortsighted

Ruth Perry • 49

for academics to ignore the current defunding of publiclibraries and the real and symbolic threats it poses to thereading public and to extracurricular book culture. Weneed to defend our public libraries for the sake of aninformed citizenry and for our children’s children, thereaders and writers of the future. Our art and our poli-tics depend on the fullest possible access to the culturalrecord. If we do not make the effort to keep our librar-

ies intact during these lean years, we will jeopardize forall time what is best in our society.

Works Cited

Koppelman, Susan. Telephone conversations with the author. 25and 27 Apr. 1995.

Viramontes, Helena Maria. Fax to the author. 3 May 1995.

50 • Significance of Primary Records: Postscript about the Public Libraries

Documents

Significance of Primary RecordsRekindling the Reading Experience of the Victorian Age Catherine Golden Postscript about the Public Libraries Ruth Perry The texts that make up “Signiﬁcance