A New Descriptive Catalogue of the Manuscripts of the CanterburyTales

75 A New Descriptive Catalogue of the Manuscripts of the Canterbury Tales Daniel Mosser In this paper I propose a new catalogue=and a new kind of catalogue=of the manuscripts of the Canterbury Tales, to be published alongside the materials generated by the Canterbury Tales Project. At present, the closest approximation we have to a comprehensive, descriptive catalogue of the manuscripts of the Canterbury Tales is volume 1 of Manly and Rickert, currently out of print and woefully out of date. There are numerous deficiencies in the Manly and Rickert descriptions: inadequate palaeographical terminology1; inaccurate and misleading collations; a paucity of facsimile illustrations; out-of-date and sketchy linguistic analyses; and perhaps most important of all, the lack of integration in terms of an overarching description of the manuscript relationships and the history of their production.2 These deficiencies result from the work’s being produced at time when the resources for making detailed descriptions were largely undeveloped, and from the need to produce an affordable publication. In the last two decades, however, many of these obstacles=both methodological and technological=have been surmounted. Methodological advances have occurred in the fields of paleography, codicology, and dialectology, to name the most obvious. Malcolm Parkes (1969) has developed a terminology for the description of Middle English handwriting that has been widely accepted and employed. The publication of the Linguistic Atlas of Late Mediaeval English marks the advent of a new era for the study of language variation and its distribution in the late Middle English period. The resources of the LALME will allow for far more accurate descriptions and localization of dialect features. The data that the Project’s transcripts and collation tools promise to provide will have the additional benefit of creating linguistic profiles for each scribe and each manuscript that will be far more detailed and thorough than those of the LALME. Users of this database will be able to map shifts of spelling preferences without enduring the tedium that the LALME methodology entails.3 The availability of the electronic database will enable the linguistic analyses in the Descriptive Catalogue to be far more thorough, and it will allow users of both tools=the Project’s materials and the Catalogue=to test the analyses for themselves against the raw data. The importance of conjoining the two undertakings is manifest on this point. The field of codicology=the ‘archaeology of the book’=has also come into being since Manly and Rickert’s time. Practitioners emphasize the need to consider the ‘whole book’ as a physical artifact and to analyze the Computer-Assisted Stemmatic Analysis THE CANTERBURY TALES PROJECT OCCASIONAL PAPERS I, 1993, 75-84 76 Canterbury Tales Project Occasional Papers I relationship of the physical features of the codex to the text-block it conveys. This perspective is aptly summarized by Siegfried Wenzel: During the past two generations, codicology and paleography have become immensely refined disciplines and have produced new bodies of information that would rightly be the envy of our elders. In contrast to them, we can no longer consider a codex as a mere receptacle that happens to have preserved the text under investigation; instead, a modern editor will have to look at the manuscript ‘holistically,’ as a total unit about whose physical makeup, composition, and history he or she will want to know as much as possible. (1990, 14; see also Delaissé 1976 and Gruijs 1972.) The investigation of the physical make-up of the codex, and the relationship of that structure to its text and prospective audience has borne fruit on several fronts. It is now widely accepted, following on the work of Pamela Robinson, Ralph Hanna, and others, that an important structural constituent in the book-production process was what has come to be known as the ‘booklet’: ‘a group of leaves forming at least one quire, but more likely several, and presenting a self-contained group of texts. Booklets thus form units intermediate in extent between the quire and the full codex’ (Hanna 1986, 100-1). Manly and Rickert’s imperfect understanding of this structure, and its relationship to textual content, led them to several unfortunate surmises.4 Ian Doyle and Malcolm Parkes’s analysis of the scribe’s ‘Progress of Copying’ in the Hengwrt Manuscript (in their ‘Paleographical Introduction’ to the facsimile volume) exemplifies the significance of the booklet structure for the production of at least some Chaucer manuscripts.5 Manly and Rickert’s descriptions of the manuscripts and their relationships were also coloured by their conception of the processes of fifteenth-century book production. Their presumption of large, commercial scriptoria finds almost no purchase in present-day discussions of the manuscript and textual history of the Canterbury Tales.6 The belief that Chaucer’s texts were copied by organized, drone-like groups of craftsmen, labouring together under one roof, in an ‘assembly-line’ fashion has largely been supplanted by a different kind of picture: that of a ‘bespoke’ book trade, consisting of stationers, who took orders for books, and then subcontracted different parts of the process=copying, decorating, binding=to independent craftsmen, whose engagement with vernacular literary works was probably only a part-time occupation. This revision in our understanding of the purposes and contexts of fifteenth-century book production has significantly influenced studies of Chaucer manuscripts and texts in the fifteen years since Doyle and Parkes’s landmark essay in the Ker festschrift, and it will certainly be in evidence in the proposed Catalogue.7 Technological advances include the application of fibre-optic light and the development of the Video Spectral Comparator (in use at the British Library, for example; Barker 1990). Both tools have great value for the study of palimpsests, faded inks and crayon, and watermarks. In the latter area, the rich body of work by Stephenson, Spector, Needham, Lyall, Thompson, and others provides bibliographers with the tools and terminology necessary to carry out the collations of manuscripts previously thought to be incapable of 77 collation. The resulting descriptions can now be made detailed enough to enable the comparison of books and manuscripts that, for one reason or another, may not be available for examination to a given investigator in situ.8 Another new technology, digital photography, in conjunction with improved data-compression tools, will make it possible to provide highresolution colour facsimiles. The frontispiece to this volume shows the first page of the Wife of Bath’s Prologue from Corpus Christi College, Oxford, MS 198 photographed by a Kontron ProgRes 3012 digital camera. It is hoped that in the later stages of publication, the Project will be able to include facsimiles of entire manuscripts on CD-ROM. With all of these advances (and one can only begin to imagine the rapidity with which new technologies and techniques will become available), it is now reasonable to propose the creation of an electronic, descriptive catalogue of the 84 manuscripts of the Canterbury Tales and the early printed editions=through which a user will be able to navigate by means of ‘hypertextual’ links=that would include, for each entry, the following information: •COLOUR FACSIMILES, illustrating a range of features for each scribal hand and script, illumination, binding, and format. The catalogue itself will likely limit the number of facsimiles to those necessary to illustrate particular points. The Canterbury Tales Project, however, does envision providing facsimiles of entire manuscripts. This feature represents one distinct advantage of electronic publication: the primary costs are those of the digital camera and royalty payments; the equivalent hard-copy would cost the consumer thousands of dollars (or pounds). •A LINGUISTIC PROFILE for each scribe and analysis of scribal language to determine as closely as possible the linguistic provenance of the scribes and the layers of language accrued in a given manuscript resulting from the processes of textual transmission. The profiles will be compiled using the data supplied by transcripts and collations of the Canterbury Tales Project. Since the Project’s data will, in effect, allow for the production of a complete linguistic concordance, it will be a relatively simple matter to track shifts in spelling preferences, which can then be analyzed for evidence of scribal layering and shifts of exemplar. •A CODICOLOGICAL DESCRIPTION, including paleographical analysis, information on materials, binding, collation, illumination, relationship of physical structure to text, and format. The electronic format will allow for side-by-side comparison of scribal hands, decoration, and page layout. For each hand, it will be possible to create a sample scribal script: an alphabetized repertoire of letter-forms for each type of script employed. It will also be desirable to include digitized beta-radiographs of watermarks (for those manuscripts copied in part or entirely on paper), which a user will then be able to compare to other paperstock samples in the catalogue. The electronic format will allow regular updates for these entries as circumstances change; in Corpus Christi 78 Canterbury Tales Project Occasional Papers I College, Oxford, MS 198, for example, several leaves that were disordered in the previous binding (the state described by Manly and Rickert 1: 92-3) were repositioned when the manuscript was conserved and rebound in 1987.9 •A BIBLIOGRAPHY. An advantage of electronic publication will be the ability to update a feature such as this on a regular basis. Particularly significant references (e.g., Doyle and Parkes’s essay on the Trinity Gower manuscript) could be included on the CD-ROM, providing issues of copyright can be resolved satisfactorily. •INFORMATION ON PROVENANCE AND HISTORY. With each entry serving as part of an electronic database, it will be possible to search the entire Catalogue for repetitions of the names of owners and locations. Had this been a feature of Manly and Rickert, a number of connections might not have been overlooked. For example, the following information is presented as a series of discrete facts in Manly and Rickert: Delamere and Trinity College, Oxford MS Arch. 49 enjoy a textual relationship with the Cardigan subgroup only in Knight’s Tale; the earliest traceable provenance for both manuscripts is in Nantwich, Chesire (Manly and Rickert 1: 113-16, 540-41); both were likely copied later than Cardigan, and certainly later than the Cardigan exemplar (of the two, Delamere is the older=1450-60=while the Trinity manuscript is assigned the date of ‘1461-83’, 1940, 1: 110, 535). What can we conclude from this? It is improbable that, if the scribe of either manuscript had access to the entire Cardigan exemplar, it would be resorted to only as a secondary source for Knight’s Tale. It therefore seems reasonable to suppose that a copy of Knight’s Tale was produced from the Cardigan exemplar and taken to Nantwich. This is a relatively straightforward connection that is easily overlooked in the Manly and Rickert presentation. A further advantage of electronic publication will be the ability to check references to physical evidence of ownership against colour facsimile images of that evidence. For example, in Dd.4.24, the scribe’s name (presumably) appears at the ends of several tales (in the formula ‘Quod Wytton’ or ‘Amen quod Wytton’), and on fol. 8r is the name ‘hungerford’: probably not that of Walter Lord Hungerford, but that signature, as well as surviving examples of Hungerford signatures from other sources, could be included for ready comparison. Again, as new discoveries are made, information can be updated easily in an electronic format. •TEXTUAL AFFILIATIONS. This is the area in which the collaboration between the Catalogue and the Canterbury Tales Project finds its most ideal fit. With access to the collations and cladograms promised by the Project, the makers of the Catalogue will be able to test our own hypotheses, as well as those of Manly and Rickert, Blake, Owen, and others, against that 79 considerable body of data; of course others, in their turn, will be able to test our analyses against their own interpretations of that same data. •DATE. There have been numerous revisions of Manly and Rickert’s proposed dates for the manuscripts. For example, their proposed dating of Gg.4.27 (‘1420-40’, 1940, 1: 173) has been challenged by Parkes and Beadle (1979 3: 7), who observe that ‘the closest script analogues … were copied before 1420, and in our opinion 1430-40 would be too late for the handwriting.’ •INCIPITS AND EXPLICITS. McCormick and Heseltine’s The Manuscripts of Chaucer’s Canterbury Tales provides a useful catalogue of the incipits for those manuscripts; including something along those lines would be useful, as well as relatively easy to produce from the Project’s collations. In addition, a user would have available hypertext links, enabling comparison of related manuscripts (on the basis of format, scribal identity, paleography, language, format, decoration, provenance, paperstock and other materials). Finally, the compilation of the Catalogue, along with the Project’s transcripts and collations, has an overarching aim: a narrative account of the manuscript relationships and textual transmission of the Canterbury Tales. Those categories that overlap with the Canterbury Tales Project are areas on which we plan to work conjointly. For the most part, however, the two projects should be regarded as complementary, with each supplying gaps in the other. Together, they represent the ambitious prospect of ‘redoing’ Manly and Rickert vols. 1 (‘Descriptions of the Manuscripts’), 2 (‘Classification of the Manuscripts’), and 5-8 (‘Corpus of Variants’), thereby supplying the tools for the eventual, new ‘Text of the Canterbury Tales: Studied on the Basis of All Known Manuscripts.’ This time, however, users will have the advantage of built-in search tools, a collation program, and hypertextually-linked databases. The collaboration between the Project and the Catalogue makes electronic publication particularly appropriate. Much in Manly and Rickert is, as I have illustrated above, badly outdated: this alone would compel publication of a preliminary ‘First Edition’ of the Catalogue early in the life of the Canterbury Tales Project. This early version of the Catalogue will provide a necessary guide to the manuscripts for the users of the Project’s materials. However, the avalanche of information the Project will produce about the individual manuscripts and their relations means that this ‘First Edition’ will itself need constant revision as the Project makes available new tranches of data and publishes new analyses. Through the medium of electronic publication, the Catalogue may become a dynamic and growing entity, reflecting new discoveries about the text through the life of the Project. In May 1993, I demonstrated a ‘shell’ of the Catalogue at the International Medieval Congress in Kalamazoo; in July, I made a similar presentation to the staff of the Project in Oxford. Later that month, at the Third Biennial Conference of the Early Book Society, in Sheffield, Peter Robinson, Elizabeth Solopova, Norman Blake, and I discussed some of the obvious ways in which our two projects fit together. We decided then to pursue the obvious A New Catalogue of the Manuscripts 80 Canterbury Tales Project Occasional Papers I advantages of a collaboration in order to allow both projects to be published in the same format and by the same press, with the hope thereby of avoiding some obvious redundancies and of allowing users of both databases access to compatible resources. The prospect is at once heady and daunting. However, the advantages of collaboration between the Project and the Catalogue, along with the manifest benefits of electronic publication (both in terms of cost and range of possible uses), make that prospect less daunting and the reasons for accomplishing it more compelling. I have considered ‘redoing’ volume one of Manly and Rickert since the days when I was at work on my dissertation on the Cardigan Chaucer Manuscript.10 Since that time, I have examined all but a handful of the manuscripts, and my conviction about the need for such a project has time and again been reinforced. Several manuscripts (Cardigan, Devonshire, Sion College, Delamere, Phillipps 8136, Phillipps 6570) have changed owners since Manly and Rickert was published. A new manuscript has been discovered (Harris 1983, 31). Significant revisions of the Manly and Rickert descriptions have been published by a number of scholars. One only need look at the ‘Bibliography’ section in the latest volume of Studies in the Age of Chaucer (vol. 15, 1993) to note the contributions being made. In addition to Owen’s work on the manuscript tradition, there is an essay by Julia Boffey and Carol Meale on Rawlinson C.86, and a dissertation by Barbara Rae Kline on British Library MS Harley 7333. We look forward expectantly to Kathleen Scott’s book on the illumination of later Gothic manuscripts.11 These are representative of an expanding corpus of work on the manuscripts and text of the Canterbury Tales, and each contribution to that corpus has, in its own way, anticipated the project described in the preceding pages.12 Notes 1The editors were well aware of this shortcoming: ‘Unfortunately there has been little systematic study of literary handwritings in the 14th and 15th centuries … We have tried to describe the hands in intelligible terms, but again it is true that there is no received system of nomenclature’ (1: 23). 2Charles Owen (1991) and Norman Blake (1985) have both undertaken this task. Both works fall short of the mark, however: Blake’s focus is entirely on the first generation of manuscripts and his analysis of their relationships is based primarily on the presence or absence of the most notable of variants (including a reliance on the most obvious kind of variant: tale-order), a premise explicitly rejected by Manly and Rickert: ‘A striking variant is too often the product of assignable causes which may operate independently upon unconnected scribes to afford a sound basis for classification’ (2: 22). The collations being prepared by the Canterbury Tales Project, which extend to the level of spelling variation, will make the soundness of both the Blake and Manly and Rickert methodologies assessable. Owen’s analyses are ultimately wanting because of what Hanna (1993) has referred to as an ‘underreliance on modern codicology’ (this is also the case with Blake’s study; Hanna 1993, 247), and to this must be 81 added his obvious over-reliance on Manly and Rickert’s analysis of the manuscript relationships. In the same place, Hanna articulates the need for ‘a full narrative history of the transmission of Chaucer’s Canterbury Tales.’ 3The LALME, while invaluable, is not without its shortcomings. See the reviews by Burton (1991; and Benskin’s reply 1991), Seymour 1991, Wilson 1988, and Shippey 1986 for some helpful caveats. Norman Davis’s comments (1983) in ‘The Language of Two Brothers’ express additional reservations. Essays by the editors of LALME have been collected in Laing 1989 and Smith 1988. 4For example, I have suggested (1985) that Manly and Rickert’s analysis of HM 144 overlooks some rather obvious physical evidence in order to preserve their theory of the textual organization of the manuscript as a series of ‘books’. In other words, they fail to reconcile the physical evidence with the textual evidence, instead spinning elaborate hypotheses to explain what, according to their analysis, seems anomalous. 5Doyle and Parkes 1979, xxvi-xxxiii. They do not employ the term ‘booklet’, but rather ‘structural section.’ These sections are, however, explicitly labeled ‘booklets’ by Hanna (1989), an essay in which he meditates productively on the implications of this evidence. Doyle and Parkes’s interpretation of the evidence has gained widespread=though not universal=acceptance; cf. Blake 1985. 6Germaine Dempster (1946, 403-6) summarizes Manly’s opinions on ‘shop manuscripts’ or ‘manuscripts of the commercial class.’ 7This picture of the book trade emerges from Graham Pollard’s work (1937). Doyle and Parkes (1978) extend it to the production of manuscripts of Chaucer and Gower in compelling fashion, and it has since become a virtual commonplace. The consensus that has emerged from the Doyle and Parkes essay is in clear evidence in the essays collected by Griffiths and Pearsall (1989). 8For example, Manly and Rickert give up on Cambridge, University Library MS Ee.2.15, declaring that the collation is ‘[i]mpossible, because the folios are not mounted and interleaved’ (1940, 1: 126). See the Bibliography for representative works by Lyall, Needham, Spector, Stephenson, and Thompson. Gerhard Piccard’s catalogue of watermarks, whose publication is ongoing, is also, of course, invaluable. 9Other manuscripts that have been conserved and/or rebound in recent years include Hengwrt, Lincoln Cathedral MS 110, and Cardigan; there are plans to pull the text-block of Ellesmere as part of the process of producing a state-of-the-art facscimile, which will provide scholars with a rare opportunity to study the manuscript in an unbound state. 10 Mosser 1985. That material has largely been reworked and will appear as part of my as-yet-unpublished book, The Alpha Manuscripts and Text of the Canterbury Tales. 11 A Survey of Manuscripts Illuminated in the British Isles, ed. J. J. G. Alexander, VI: Later Gothic Manuscripts (as cited in ‘Design, decoration and illustration,’ p. 54, n. 9.) 12 I would like to thank the National Endowment for the Humanities, and the Department of English and College of Arts & Sciences at VPI&SU for support that made some of the research for this project possible. A New Catalogue of the Manuscripts 82 Canterbury Tales Project Occasional Papers I Bibliography Barker, N. 1990. ‘Recent Work with the British Library’s Video Spectral Comparator.’ In Déchiffrer les Écritures Effacées, ed. L. Fossier and J. Irigoin, 79-82. Paris: Éditions du Centre National de la Recherche Scientifique. Benskin, M. 1991. ‘In Reply to Dr Burton.’ Leeds Studies in English 22: 209-62. Blake, N. 1985. The Textual Tradition of the Canterbury Tales. London and Baltimore: Edward Arnold. Burton, T. L. 1991. ‘On the Current State of Middle English Dialectology.’ Leeds Studies in English 22: 167-208. Davis, N. 1983. ‘The Language of Two Brothers in the Fifteenth Century.’ In Five Hundred Years of Words and Sounds: A Festschrift for Eric Dobson, ed. E. G. Stanley and D. Gray, 22-8. Cambridge: D. S. Brewer. Delaissé, L. M. J. 1976. ‘Towards a History of the Medieval Book.’ Codicologica 1: 75-83. Dempster, G. 1946. ‘Manly’s Conception of the Early History of the Canterbury Tales.’ PMLA 61: 379-415. Doyle, A. I. 1983. ‘Retrospect and Prospect.’ In Manuscripts and Readers in Fifteenth-Century England: The Literary Implications of Manuscript Study, ed. Derek Pearsall, 142-6. Cambridge: D. S. Brewer. Doyle, A. I, and M. B. Parkes. 1978. ‘The Production of Copies of the Canterbury Tales.’ In Medieval Scribes, Manuscripts and Libraries: Essays Presented to N.R. Ker, ed. M.B. Parkes and A. G. Watson, 163-210. London: Scolar Press. =. 1979. ‘Palaeographical Introduction.’ In The Canterbury Tales. Geoffrey Chaucer. A Facsimile and Transcription of the Hengwrt Manuscript, with Variants from the Ellesmere Manuscript, ed. P. G. Ruggiers, xix-xlix. Norman, Ok.: University of Oklahoma Press. Griffiths, J., and Pearsall, D. (Eds.) 1989. Book Production and Publishing in Britain, 1375-1475. Cambridge: Cambridge University Press. Gruijs, Albert. 1972. ‘Codicology or the Archaeology of the Book? A False Dilemma.’ Querendo 2:87-108. Hanna, R. III. 1986. ‘Booklets in Medieval Manuscripts: Further Considerations.’ Studies in Bibliography 39: 100-11. =. 1989. ‘The Hengwrt Manuscript and the Canon of The Canterbury Tales.’ English Manuscript Studies 1: 64-84. =. 1993. Review of Owen 1991. Studies in the Age of Chaucer 15: 247-51. Harris, K. 1983. ‘John Gower’s Confessio Amantis: The Virtues of Bad Texts.’ In Manuscripts and Readers in Fifteenth-Century England: The Literary Implications of Manuscript Study, ed. Derek Pearsall, 27-40. Cambridge: D. S. Brewer. Laing, M. (Ed.) 1989. Middle English Dialectology: Essays on Some Principles and Problems. Aberdeen: University of Aberdeen. Lyall, R. J. 1989. ‘Materials: The Paper Revolution.’ In Book Production and Publishing in Britain, 1375-1475, ed. J. Griffiths and D. Pearsall, 11-29. Cambridge: Cambridge University Press. Manly, J. M., and Rickert, E. (Eds.) 1940. The Text of the Canterbury Tales. 8 vols. Chicago: University of Chicago Press. 83 McCormick, Sir William, and Heseltine, J. E. 1933. The Manuscripts of Chaucer’s Canterbury Tales: A Critical Description of Their Contents. Oxford: Clarendon Press. McIntosh, A., Samuels, M. L., and Benskin, M. (Eds.) 1986. A Linguistic Atlas of Late Mediaeval English. 4 vols. Aberdeen: University of Aberdeen. [LALME] Mosser, D. W. 1985. ‘Manly and Rickert’s Collation of Huntington Library Chaucer Manuscript HM 144 (Hn).’ Papers of the Bibliographical Society of America 79: 235-40. =. 1985. ‘The Cardigan Chaucer Manuscript and the Process of FifteenthCentury Book Production.’ Unpublished Dissertation. Austin: University of Texas at Austin. Needham, P. 1982. ‘Johannes Gutenberg and the Catholicon Press.’ Papers of the Bibliographical Society of America 76: 367-456. =. 1985. ‘The Paper Supply of the Gutenberg Bible.’ Papers of the Bibliographical Society of America 79: 303-374. Owen, C. A., Jr. 1991. The Manuscripts of the Canterbury Tales. Cambridge: D. S. Brewer. Parkes, M. B. 1969. English Cursive Bookhands. London: Scolar Press. Parkes, M. B., and Beadle, R. 1979. The Poetical Works of Geoffrey Chaucer: A Facsimile of Cambridge University Library Ms Gg.4.27. Vol. 3. Norman, Ok.: Pilgrim Books (in association with D. S. Brewer, Cambridge). Pollard, G. 1937. ‘The Company of Stationers before 1557.’ The Library 4th series, 18: 1-37. Robinson, P R. 1980. ‘The “Booklet”: A Self-Contained Unit in Composite Manuscripts.’ Codicologica 3: 46-69. Scott, K. 1989. ‘Design, decoration, and illustration.’ In Book Production and Publishing in Britain, 1375-1475, ed. J. Griffiths and D. Pearsall, 31-64. Cambridge: Cambridge University Press. Seymour, M. 1991. Review of LALME (McIntosh, A., Samuels, M. L., and Benskin, M. 1986). English Studies 72:73-80. Shippey, T. A. 1986. ‘Here be dialects.’ Review of LALME (McIntosh, A., Samuels, M. L., and Benskin, M. 1986). Times Literary Supplement (October 30, 1987) 1200. Smith, J. J. (Ed.) 1988. The English of Chaucer and His Contemporaries. Aberdeen: Aberdeen University Press. Spector, S. 1978. ‘Symmetry in Watermark Sequences.’ Studies in Bibliography 31: 162-78. Stevenson, A. H. 1951. ‘Watermarks are Twins.’ Studies in Bibliography 4: 57-91. =. 1961. Observations on Paper as Evidence. University of Kansas Publications, Library Series, 11. Lawrence, Kansas: University of Kansas. =. 1967. The Problem of the Missale Speciale. London: Bibliographical Society. Thompson, J. J. 1987. Robert Thornton and the London Thornton Manuscript: British Library MS Additional 31402. Manuscript Studies 2. Cambridge: D. S. Brewer. Wenzel, S. 1990. ‘Reflections on (New) Philology.’ Speculum 65: 11-18. Wilson, E. 1988. Review of LALME (McIntosh, A., Samuels, M. L., and Benskin, M. 1986). Review of English Studies n.s. 39: 94-6. A New Catalogue of the Manuscripts