Introduction: What is a Text? Author(s): Seth Lerer and Joseph A. Dane Source: Huntington Library Quarterly, Vol. 58, No. 1, Reading from the Margins: Textual Studies, Chaucer, and Medieval Literature (1995), pp. 1-10 Published by: University of California Press Stable URL: http://www.jstor.org/stable/3817894 Accessed: 11/12/2010 11:21

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=ucal.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

University of California Press is collaborating with JSTOR to digitize, preserve and extend access to Huntington Library Quarterly.

http://www.jstor.org Introduction

What Is a Text?

SETH LERER AND JOSEPH A. DANE

We are all now for "bibliographical"methods, keenly on the watch for every least indication of disturbancein the accurate transmissionof a text, sorting out by many subtle and ingenious methods the first, sec- ond, or third stage of the composition, the original draft, the first completed form, the revisionof this, that, and the other purpose, and so on. But there is much more in these modern methods of researchthan used to be understood by "bibliography,"and I am not sure that the recent extensions of the term have been altogether justifiable. The virtues of bibliographyas we used to call it were its definiteness, that it gave little scope for differencesof opinion, that two persons of reason- able intelligence following the same line of bibliographicalargument would inevitably arrive at the same conclusion, and that it therefore offered a very pleasantrelief from criticalinvestigations of the more "lit- erary"kind.1 nR, > B. McKerrow'sremarks of sixtyyears ago maywell standas an epi- graph for this special issue of HuntingtonLibrary Quarterly. We are o now, too, all for "bibliographical"methods: advances in computersci- ence and the attendant technologies of collation, ideological critiques of the presuppositionsof textual criticism itself, and the sheer volume of newly discov- ered manuscriptsand earlyprinted volumes have all made bibliographya central topic of academicdebate. Few might still claim that the provinceof bibliography lies with the unarticulatedintuitions of "personsof reasonableintelligence," but many would likely agreethat what we mean by "textualscholarship" necessitates a knowledge of the historicity of texts and readersand that such scholarshipcan only benefit from the communication among critics of many and variousfields.2

1. R. B. McKerrow, An Introduction to Bibliography(New York, 1927), 2. 2. See D. C. Greetham, TextualScholarship: An Introduction (New York, 1992), 2.

HUNTINGTON LIBRARY QUARTERLY - 58:1 2 -", SETH LERER AND JOSEPH A. DANE

Nonetheless, although "textuality"has become something of a term to conjure with-and disciplines as diverse as film studies, cultural anthropology,and the history of science imagine themselvesworking with the meaning and the mode of "texts"-the processesof transcribing,collating, describing, and editing books and manuscriptsof earlierhistorical periods remain, for the most part, relativelyfamil- iar.Editions from the age of McKerrowmay seem, at first glance, prettymuch the same as those from the atelierof JeromeMcGann. And, to a largedegree, debates framed by the Bedierists and Lachmanniansof nearly a century ago still find themselvesplayed out in the prefacesand articlesof professionaleditors.3 The question still evadesus: What is a text?Or, to ask it more precisely,what are the relationshipsbetween the methods of textual study and the privilegingof a class of objects called texts?How does an artifactbecome a text?Does it become one only when subjectedto the inquiriesof textual analysis?For that matter,can a text lose its status and revertto the merely artifactual? Such questions guide our inquiries into the elements of textuality itself. Are pictures-illustrations, marginalia,historiated initials-texts, and how can the editor incorporatesuch information in the making of the modern edition? For example, no modern scholarlyedition of The CanterburyTales (excluding, for the moment, facsimiles)prints the pilgrim portraitswith the General Prologue.Yet Frederic Furnivall'slate-nineteenth-century edition of Hoccleve's Regimentof Princesprints a facsimile of Hoccleve's portrait of Chaucer precisely where it appears in one of the earliest manuscripts of the poem (British Library,MS. Harley 4866).4 We may ask the question more radically:Can texts themselves function as pictures?Can, in other words, displays of writing function not as strings of words to be conveyed and edited but ratheras icons of something else? The very notion of "facsimile"implies a transformationof text back into picture. And, as current photographic technologies improve, the spate of new facsimiles may come to stand as virtual (in both senses of the word) replacementsof old texts. The facsimileof the famous Ellesmeremanuscript, recently copublished by the Huntington Libraryand Yushodo Co., Ltd., is no simple reproduction,how-

3. Witness, for example,the exchangesin the essayscollected in StephenG. Nichols, ed., TheNew Philology, a specialissue of Speculum60 (1990); and see Tim William Machan, TextualCriticism and MiddleEnglish Literature(Charlottesville, Va., 1994). 4. F.J. Furnivall,ed., TheRegiment ofPrinces A.D. 1411-12, from the HarleianMS 4866, EarlyEnglish Text Society,extra series, 72 (London, 1897). Of the forty-threemanuscripts of this work, only a few contain,or everdid contain, a Chaucerportrait where the text seemsto demandone. See M. C. Seymour,"The Manuscriptsof Hoccleve'sRegiment of Princes,"Edinburgh Bibliographical Society Transactions 4, pt. 7 (1974): 253-97. WHAT Is A TEXT? s 3

ever;and a numberof textualscholars have argued directly that the entireenter- priseof facsimileproduction is editorialfrom beginningto end.5In the "Notes and Documents"section of this issue,Anthony G. Cainsdescribes the rebinding and conservationof the Ellesmeremanuscript that was undertakenas a compo- nent of the recentfacsimile publication. He revealsthe waysin which binding affectsthe productionof a facsimile-what one can see of the page,both gener- ally and in detail. More broadly,his analysissuggests the difficultboundary betweenthe textualand the artifactual;not only naturaldegradation but alsothe variousbindings and repairshave interfered with the pigmentsand inks,the dec- orationof the manuscriptin particular,which, as other essaysin the volume show,is increasinglythe subjectof textualinquiry. In whatways does the availabilityof facsimileeditions alter the characterof textualquestions? For the EllesmereChaucer, the answerlies in the future.But for,say, Beowulf the evidencehas been around for centuries.One needonly recall the readingwundini gold-a phraselong thought to evidencethe linguistic archaismof the poem-to reflecton how editorsworking from a transcription(in this case the Thorkelintranscripts of the Beowulfmanuscript) could discerna pieceof text not to be foundin the actualmanuscript and then incorporatethat text in publishedfacsimile and criticaleditions.6 What, then, are the elementsof textuality?At the local level,where do we drawthe line betweena "variant"and an "error"?In the vicissitudesof scribal copying,how can we distinguishbetween a lapseand an intrusion-and, in the caseof what appearto be intelligentand meaningfulintrusions, how do we dis- tinguish between the variantand wholesalerewriting?7 The transmissionof

5, See, for example,T. H. Howard-Hill,review of MichaelWarren, The CompleteKing Lear (1608-1623) (Berkeleyand Los Angeles, 1989), Reviewof EnglishStudies, n.s., 43 (1992): 420-22. 6. The phrasewundini gold in Beowulfhad been understoodto be a survivalof an earlyOld Englishinstru mental form, and thus was valuedas evidencefor an earlydate (seventhor eighth century)of compositionof the poem. Kevin Kiernandemonstrated that this readingappears as an editorialconjecture in the transcript of the poem made by G. J. Thorkelinin the late eighteenthcentury. It was then reproducedin Julius Zupitza'stransliteration of the text in his publishedfacsimile (Beowulf, Reproduced in Facsimile[London, 1882]) and was maintainedby twentieth-centuryeditors (notably C. L.Wrenn),who had not consultedthe originalmanuscript. Kiernan's examination of that manuscriptshows that the readingshould, in fact, be wundmigold -scribal nonsense,but one that may be more conservativelyedited as wundumgold, not an archaicphrase at all and one consistentwith the date c. 1000 for the manuscript.See Kiernan,Beowulf and the BeowulfManuscript(New Brunswick,N.J., 1980), 30-37. 7. This questionhas been exploredfor a varietyof vernacularliteratures in such recentstudies as SylviaHout, FromSong to Book:The Poetics of Writingin Old FrenchLyric and LyricalNarrative Poetry (Ithaca, N.Y., 1987); Seth Lerer,Chaucer and His Readers:Imagining the Author in LateMedieval England (Princeton, N.J., 1993); and John Dagenais,The Ethics of Readingin a ManuscriptCulture: Glossing the "Librode beunamor" (Princeton,N.J., 1994). ' 4 SETH LERER AND JOSEPH A. DANE

medievalliterary documents has been an obvioussite for such reflections.To some, variationis the markof unreliability,the distinctivemanifestations of what Chaucerwould lament as the miswritingand mismetering"for defaute of tonge" thathe imaginesas the scribalafterlife of Troilusand Criseyde.To others,variation offers testimony to the creativelyfluid natureof the medievaland the early mod- ern text. BarryWindeatt incorporatessuch miswritingsinto his edition of Troilus and Criseyde.He presentsChaucer's poem "in the contextof the corpusof vari- ants drawn from the extant manuscripts,not only because those variantscan be of editorialvalue, but alsobecause they are held to be of a positiveliterary value, to embodyin themselvesa formof commentary,recording the responsesof near- contemporaryreaders of the poetry."8But variationdid not end with the advent of print.In a curioustroping on Chaucer'senvoy, Stephen Hawes concluded the PastimeofPleasure (first published by Wynkynde Wordein 1509) with his fears not of scribalbut of typographicalerror: Go lytellboke I praygod the saue Frommysse metrynge / by wrongeImpressyon.9 And it went on. Phenomena recordedas variantsby modern editors may all too oftenbe the meremistakes of the compositor.Sir Thomas Wyatt's "Satire to John Poynz"condemns the courtiers, "Of them that list all vice for to retain,"a line mangledin Tottel'sMiscellany to "Ofthem thatlist all nice for to retain"(a read- ing, by the way, correctedback to "vice"in the first three reprintingsof the volume,yet againtransformed to "vile"in laterones).10 And moresignificantly, many Shakespeareanvariants have recently been revealedto be not witnesses to competingversions of the playsbut similarmistakes at the printshop. Some dif-

8. B. A. Windeatt,ed., Chaucer:Troilus and Criseyde(London, 1984), 25. 9. W. E. Mead, ed., ThePastime of Pleasureby StephenHawes, Early English Text Society,original series, 173 (London, 1928), lines 5803-4. For a discussionof this passagein the contextof a largerargument on Hawes'sthematic attentions to print cultureand literaryfame, see Lerer,Chaucer and His Readers,176-93. 10. Tottel'sreading is recordedas a variantin the edition by KennethMuir and PatriciaThompson, Collected Poemsof Sir ThomasWyatt (Liverpool, 1969), 88; and in the study by RichardHarrier, The Canonof Sir ThomasWyatts Poetry (Cambridge, 1975), 171; though not in the edition by RonaldA. Rebholz,Sir ThomasWyatt, The Complete Poems (Harmondsworth, England, 1978). It is likelythat his readingis not a true variantbut rathera compositor'smistake: perhaps not the printingof an n but of a turnedu (wherethe compositormisread vice and uice).Such turnedletters are commonplace in the printingsof the Miscellany; see the list of variantreadings and misprintsin HyderE. Rollins, TottelsMiscellany (1557-1587), 2 vols. (1928-29; reprint,Cambridge, 1965), 1:263-326. The specificsof the line in the "Satireto Poynz"are recordedon 283 (keyedto 1:85, line 41). WHAT IS A TEXT? 5

ferences between the Quarto and Folio editions of King Lear, for example, may well be missettings of a markedup copy-text.11 The instabilityof medieval and early modern literarytexts is thus a problem not only for the editor but also for the author and the reader.Chaucer's attentions to miswriting in the Troilus,together with his well-known lament to Adam Scriveyn, suggest a thematic attention to the vagariesof textual transmission.12 The author'swords are alwayssubject to the garblingsof others. As Chaucerput it to his Adam: "Andall is through thy negligence and rape."But often, the mis- takes of writing hands are not due to the ministrationsof the scribe but to the anxieties of authorshipitself. And now my penne, alias, with which I write, Quaketh for drede of that I most endite.13 This couplet from the proem to Book IV of Troilusand Criseydestands not just as a mark of Chaucer'snarratorial self-consciousness. In the century after his death, it also came to articulatethe position of any author writing in the post- Chauceriantradition. Lydgate mimes these lines half a dozen times, as do a string of poetasters stretching into the sixteenth century.14Even Sir Thomas Wyatt complains,"Myhand doth shake, my pen scant doth his part,"in a ballade that queries the very possibility of authorialcontrol of the literarytext. I do remain scant wotting what I write, Pardonme, then, rudely though I indite.15 If the instabilitiesof texts give voice to the anxietiesof authors and their readers, they have become the nodes of celebrationfor more modern, theoreticallymind- ed critics. Late-twentieth-centuryattentions to the reader'saffective responses to

11. See JosephA. Dane, "'Whichis the Iustice,which is the theefe':Variants of Transpositionin the Text(s)of KingLear," Notes and Queries42 (1995): 322-27. For some of the broaderimplications of mechanical variationin Shakespeareantexts, see PeterW. M. Blayney,The TextsofKing Learand TheirOrigins, vol. 1, NicholasOkes and the FirstQuarto (Cambridge, 1982); and, more generally,the discussionin Greetham, TextualScholarship, 285-90. 12. "Chaucer'sWordes Unto Adam, His Owne Scriveyn,"in LarryD. Benson et al., eds., TheRiverside Chaucer, 3d ed. (Boston, 1987), 650. For a discussionof the place of this poem in Chaucer'sthematics of correction, togetherwith a broaderargument about the "openness"of medievalmanuscript culture, see GeraldL. Bruns, "The Originalityof Textsin ManuscriptCulture," Comparatve Literature 32 (1980): 113-29. 13. Windeatt, Troilusand Criseyde,IV. 13-14. 14. For Lydgate'suses, see Hans Kurathet al., eds., MiddleEnglish Dictionary (Ann Arbor, Mich., 1954-), s.v., quake. 15. Rebholz,Complete Poems, CLXXVII.19, 23-24. 6 m SETH LERER AND JOSEPH A. DANE

a work, together with poststructuralistpredilections for the jeu of signs and sig- nifying, may lead editors as well as critical interpretersto celebratethe variation that inheres in textual transmission.The ideal of the editor as an intentionalist and of the critical edition as recoveringthe meaning of a work as posed by its originaryauthor may well be past. As David Greethamputs it, "it seems that all we have are competing texts and competing readers."16What we are "allfor," to return to McKerrow, is just about everything-be it cultural studies, history, materialism,politics, the problematicsof representation.It might be hard now to distinguish between editing as a professionalpractice and "criticalinvestigations of the more 'literary'kind." The interfacesof the history of authorship,the practicesof scribesand print- ers, and the institutions and the ideologies of the academylead us to ask not only "What is a text?"but also "What is a textual critic?"The discipline of bibliogra- phy may have originally seemed a handmaiden to the literary: editions were produced to make texts availableto readers,critics, and appreciators.Recently, however,textual work has taken on an increasingindependence. To some degree, JeromeMcGann's plea for a new "criticalediting" offers a manifesto of the bibli- ographer'sautonomy. Both linguistic and bibliographicaltexts are symbolic and signifying mechanisms.Each generatesmeaning, and while the bibliographicaltext typicallyfunctions in a subordinaterelation to the linguistic text, "mean- ing" in literaryworks results from the interactive agency of these two semiotic mechanismsoperating together.17 From this position, it may seem a short step to Greetham's"Textual Forensics," in which textual scholarshipis not merely on a par with literarycriticism but also in some way profoundly apart from it, yet a powerful influence over it. Textualscholarship is an antidisciplinebecause it does not occupy a per- manent or consistent epistemological position and because it has no definable Fach, or subject matter.And textual scholarshipis a postmod- ernist antidiscipline because it consists of coopted and deformed quotations from other fields.18

16. Greetham,Textual Scholarship, 342. Remarksin this paragraphare indebtedto Greetham'sgeneral discussionof the relationsbetween new developmentsin textualcriticism and literarytheory (pp. 341-46). 17. JeromeJ. McGann,"What Is CriticalEditing," TEXT 5 (1991): 27. 18. D. C. Greetham,"Textual Forensics," PMLA 111 (1996): 32. WHAT IS A TEXT?

For RandallMcLeod, who has calledour attentionto, in Greetham'swords, "a technicaland epistemologicaldisjunction between the local evidencederivable from analyticalbibliography and the generalizedevidence resulting from read- ing,"19the veryact of writingand publishingtextual scholarship becomes a tour de forceof typographicalinvention. One need only scan the concatenationof visualcodes in McLeod'swork-facsimiles of FirstFolio type, photographs of sig- natures,skewings of lineationand justification,orthographic representations of verbalgarblings, right down to the presentationof the critic'sname as, variously, "RandomClod" and "RandomClovd"-to see the buildingblocks of textual scholarshipreassembled into somethingof an abstractexpressionism of the word.20McLeod's text becomesan objectof textualstudy itself. One can imagine the nightmareof proofreadingsuch a document;more to the point, one might note that the critic'sown work enactshis dicta of textualscholarship: that we must avoidreading texts in orderto seethem.21 This volumeof essaysseeks to engagethese directions in textualscholarship from a varietyof criticalpositions, and it does so stimulatedby the occasionof the New ChaucerSociety's biannual meeting, held in LosAngeles in 1996. That such an occasionshould provoke a volumethat may seem to querythe veryidea of authorialautonomy may be somethingof a paradox.Ralph Hanna III and A. S. G. Edwards,in the first essayin the volume, treatthe annotationsand ephemerasurrounding the text of TheCanterbury Tales in the Ellesmeremanu- script-includingRotheley's poem on the De Verefamily, written on the opening flyleavesof the manuscript-not merelyas curiositiesof historybut alsoas testi- mony to the historicityof Chaucer.They explorethe provenanceof Ellesmereto discernhow Chaucer'swork was read and understoodby late-medievalaudiences and also to recognizethat such an understandingwas a profoundlysocial and politicalact. Their new editionof Rotheley'spoem, presentedas an appendixto the essay,illustrates the waysin whichthe methodsof traditionaltextual scholar- ship maybe pressedinto the serviceof decenteringthe canonicaltext. For JuliaBoffey also, the recoveryof a new text is centralto recalibrating Chaucerianauthority-here, not as the authorof a singlecanonical work but as the name attachedto a canon of shorterpoems. The poet as proverbialistis importantto much fifteenth-and sixteenth-centuryreception of Chaucer,and

19. Ibid., 37. 20. RandomClod, "InformationUpon Information,"TEXT 5 (1991): 241-81. 21. See the pictureof McLeodreading upside down and the discussionin Greetham,"Textual Forensics," 39. - 8 SETH LERER AND JOSEPH A. DANE

Boffey explores the ways in which a little stanza of proverbialverse by Walton came to be misattributedto Chaucer.Bringing to bear the contextsof late- medievalpatronage on manuscriptproduction, Boffey's essay helps us understand the waysin whichChaucer's poems were brought together to definelate-medieval literarywriting itself. In the contextof Boffey'songoing project of revisingthe IndexofMiddle English Verse, the essayalso provokes insight into the criticalpre- suppositionsunderlying the constructionof such referenceworks. The original Indexorganized autonomous poems by firstlines; it createdhierarchical relation- shipsbetween "texts" and "fragments."Many Middle English lyrics are cut out or cobbled togetherfrom longer narrativepoems. Bits and pieces of Chaucer's Troilusand Criseyde,Lydgate's Temple of Glass,Stephen Hawes'sPastime of Pleasure,and otherworks were often excised,reassembled, and recastinto short lyricsof desireor verseletters of complaint.Where do we drawthe line between quotationand a new text?When does an extractbecome a new poem?How do the organizationalprinciples of a projectsuch as the Index ofMiddle EnglishVerse shapethe answersto thesequestions, and in the process,how does sucha schol- arly resourceredefine the canon-and the very materialsto which it providesthe index? Suchquestions also motivate Ardis Butterfield's contribution to this volume. Her closeexamination of the manuscriptsof Troilusand Criseydenot only reveals the modesof late-medievalreading in termsof the ordinatioof layout,rubrica- tion, heading,gloss, and commentary;it also helpsus to appreciatethe waysin whichthe Troilusembedded a varietyof genresin its narrative.Lyrics, songs, and epistlesare, to somedegree, the buildingblocks of Chaucer'samorous and philo- sophicalstory, and Butterfieldshows in greatdetail how the scribesof the poem indicatedthese variousgeneric distinctions in their manuscripts.Such distinc- tions, too, may drawon the habitsof Frenchmanuscript organization, and this essaypoints toward further research in the historicalrecovery of medievalnotions of the text. Troilusand Criseydemay not be best consideredas a singletext but rather,as the scribesperhaps indicate, an arrangementand a synthesisof differ- ent forms of discourse,each signaledvisually by distinctivecodes of textual representation. Textualrepresentation is at the heartof David Boyd'saccount of Chaucer's Cook'sTale in BodleianLibrary, MS. 686. Like the precedingcontributions, Boyd'sessay locates textuality in the materialand social conditions of makingand reception.He readsthis versionof the fragmentary,difficult, and both poetically andpolitically challenging Cooks Tale in termsof the socialpositions of its likely audience.Situating a singletale in a singlemanuscript, he nonethelessenlarges the WHAT IS A TEXT?

issueto querythe verynature of an authorized,or authorial,text. Any historicist approachto literature,and by implicationany historicistconception of textual criticism,must recognize, he argues,that "an authorial text ... existssocially and historicallyas the author's(s') Work ... to a particulargroup of readersand own- ers of particularversions." This revisionist claim leads to a particularly postmodernconception of both the medievaltext and the professionof medieval studies."We cannot separate," Boyd concludes, "the processes of medievalbook production,construction, and transmissionfrom our studyof medievaltexts or fromissues of ideology,politics, and power, all of whichare naturalized in the spe- cificmanuscript matrices within which and through which they operate." For David Greetham,the possibilitiesof postmodernmedieval or textual studieslie not so much in exposingthe politicsof literatureor the literarypro- fession as in exploringthe controllingmetaphors that have shaped modern editorialpractice. The imagesof filiation,lineal descent,and genealogy;the notion of the text as a memberof a familyor a branchof a tree;the socialpara- digms of linearityand patriarchythat governtraining in literaryscholarship generally-all find themselveslaid barefor the culturalfigurations that they are. Throughanalogies to Indo-Europeanphilology, comparative biology, and physi- calpaleoanthropology, Greetham shows how all formsof historicalinquiry define the objectsof theirstudy in familial,linear, or generativeterms. What Greetham offers,in the end, is not a rejectionof thesemodels as such but a personalrecal- ibration.He posits"a model of embeddingrather than descent," a notion of the text as findingits placenot on the treeof textuallife but in the rhizomeof tex- tual diversity.And he suggests,in closing, that "the hypertextualmodel of free-floatinglinks," for all its seemingpostmodernity, "is a bettersimulacrum of medievaltextuality than the fixedcritical text of the codexever was; or at leastof some typesof medievaltextuality, the scriptiblerather than the lisible." Readindividually, each of theseessays exemplifies a tone, a technique,and a traditionof textualcriticism in medievalliterary study. Read in sequence,how- ever,they tracean arc that leadsus from the historicaldocument to our own historicalpositions as scholarsand readers.Along that arc we may be able to locate,if not a definitiveanswer to this introduction'stitular question, then at leastsome points of referencefor posingdefinitions. For what each of thesecon- tributorssuggests-sometimes explicitly, sometimes implicitly-is that a text is whateveris reproducibleas an objectof concern.The Rotheleypoem to the De Veres,a proverbialstanza of Chaucerianattribution, annotated manuscripts of Troilusand Criseyde,manipulated forms of the CooksTale, even the documents of professionaleditorial practice: all are "texts"worthy of study.Or, perhapsto ' 10 SETH LERER AND JOSEPH A. DANE

put it differently,what we witnesshere are the ways in which bibliographical study transformsphysical artifacts into objectsof study,and furthermore,the waysin which the practicesof editing,in turn,transform these objects of study into "texts."As scholars,we areall awareof theseprocesses (this is whatintellec- tual history is). Yet what this collection may enhance is our appreciationof the manyand competingforces that disturb, mediate, or occludesuch an awareness. As textualscholars, we all standat the nexusof objectand artifact:of physical thing and constitutivesubject of institutionalstudy. It may seem, in the end, merelyreductive to definea textas whatever textual scholars study. But by explor- ing the vicissitudesof textualstudy itself-its methods,histories, controlling metaphors,and ideologies-we might well, in the end, see more clearlywhat McKerrowcalled the "virtuesof bibliography."

StanfordUniversity Universityof SouthernCalifornia