Compiling the Canterbury Tales in Fifteenth-Century Manuscripts

Abstract

In recent studies of the manuscript tradition of the Canterbury Tales, scholars have considered the possibility that differences in the content and arrangement of the earliest manuscripts might be indicative of discrete stages of authorial revision. Such a view is not new; Walter W. Skeat was one of the earliest editors to raise such a possibility.1 Skeat viewed the order of the Hengwrt manuscript as representing Chaucer's first attempt at arranging his work.2 This early stage was followed by four subsequent revisions, three of which Skeat viewed as authorial, the final as work of a later editor. The first of these rearrangements is represented by Petworth House MS 7, the next by Oxford, Corpus Christi College MS 198 and its textual twin London, British Library MS Lansdowne 851. Skeat attributed the final Chaucerian stage of revision to London, British Library MS Harley 7334, while the ordering of the Ellesmere manuscript, the basis of most modern editions of the poem, he considered to be a later scribal reworking.3 More recently, N. F. Blake has raised the possibility that some of the earliest manuscripts of the Canterbury Tales may have been written under Chaucer's supervision, and that differences in content and tale order would therefore represent separate stages of authorial revision.4 Such evidence, Blake suggests, could be witness to different versions of the work: “a Canterbury Tales A text, a Canterbury Tales B text, and so on.” If accepted, the proposals of Skeat and Blake would transform our view of the text, its contents and arrangement, and how it is edited and studied.A related, yet distinct, view was proposed by Charles A. Owen Jr., who suggested a more radical understanding of Chaucer's ordering of the Canterbury Tales, and of its reflection in the arrangements preserved in the extant manuscripts. Owen discounted any concept of a single authorized order, claiming that evidence of completeness in the early manuscripts is purely the result of scribal intervention. Having accepted this situation, Owen argued, it becomes possible to trace earlier stages in Chaucer's plan for the work and the process by which it evolved. Rather than seeing the Canterbury Tales as a coherent and organized work, Owen viewed the text as a “collection of fragments reflecting different stages of [Chaucer's] plan for the work as a whole.”5 The first stage that Owen identified within these fragments is a considerably simpler scheme, comprising just the Man of Law's Prologue followed by the Tale of Melibee, the Wife of Bath's Prologue followed by the Shipman's Tale, and the Parson's Prologue followed by the final tale. This scheme was subsequently expanded after two discrete periods of literary activity. The first period was characterized by “religious concern,” during which Chaucer wrote the Parson's Tale and Retraction, and translated De contemptu mundi, an important source for the Man of Law's Tale. The second period was more diverse but included the writing of the tales of the so-called Marriage Group, and with that the expansion of the role of the Wife of Bath. In proposing these distinct stages of authorial revision, Owen focused on references within the links and tales themselves, rather than on the evidence provided by the manuscripts. He argued that the Man of Law's reference to a prose tale in his prologue indicated an earlier scheme in which he was to tell the prose Tale of Melibee, subsequently allocated to the pilgrim Chaucer. Similarly, the female pronouns used by the speaker of the Shipman's Prologue indicate that it was originally intended to be followed by the Wife of Bath's Tale. John Fisher has made similar claims for the identification of earlier stages in the composition process.6 Fisher argues that the Man of Law's Epilogue was intended to function as a link between the Man of Law's Tale of Melibee and the Shipman's Prologue and Tale, told by the Wife of Bath. According to this theory, Chaucer's original intention was that the Wife of Bath should interrupt the Host and the Parson, insisting that she tell her tale next. For Fisher, this evidence is indicative of larger developments in the overall conception of the work, which he divides into two separate stages. The first stage, represented by the tales of Groups C-I, reflects Chaucer's earliest efforts on the collection, while Groups A-F emerged during a second stage in which the concept of “dramatic interplay” had fully formed in Chaucer's mind as an integral part of the work as a whole. While the proposals of Owen and Fisher may well uncover vestigial traces of earlier plans in Chaucer's composition process, they do not have the same status as the distinct authorial versions proposed by Skeat and Blake. Where Skeat and Blake envisage a scenario in which Chaucer authorized the release of copies of different versions of the poem, now witnessed by our surviving manuscripts, there are no surviving manuscripts that testify to any of Owen's or Fisher's earlier schemes. No manuscripts attest to the use of the Man of Law's Prologue as a link to the Tale of Melibee, nor is there any witness that follows the Wife of Bath's Prologue with the Shipman's Tale. While there is disagreement about the identity of the pilgrim who interrupts the Man of Law's Endlink, in no single manuscript is it the Wife of Bath.Recent studies of the codicology of some of the early witnesses have, however, identified physical evidence that points to uncertainty and adjustment of the placement and arrangement of certain tales; the evidence has then been linked with these theories of authorial revision. In a detailed codicological analysis of Corpus 198, Estelle Stubbs has argued that the insertion of vellum sheets of a thicker quality at certain key points in the manuscript is indicative of revision of, or additions to, the text: The implication of this codicological evidence is intriguing. If Scribe D [the scribe of the Corpus manuscript] was in the process of making structural alterations to his manuscript, some of which were in line with Hg [Hengwrt] additions and the Hg order, then copies of some tales must have been in hand, perhaps in a different form and in a different order. Could this have been the mini-Canterbury Tales of Owen and Fisher?7 Stubbs notes how deficiencies in the text of Corpus coincide with gaps in the Hengwrt manuscript, and she argues that these are indicative of the manuscript's access to early materials, perhaps even earlier than those acquired by Hengwrt, rather than evidence for the loss of such text at a later date: “Linking passages which Hg originally lacked or never acquired are also missing in Cp, a feature which surely suggests an earlier stage of assembly rather than the misplacement or loss of vital passages which has been assumed for manuscripts of the ‘c’ type.”8There is, however, an alternative way of interpreting such evidence. If we imagine scribes of the earliest manuscripts having access to similar exemplars, or parts of exemplars, then it would not be surprising to find them encountering difficulties at the same points in the text. Although it is apparent from textual that the Corpus and Hengwrt manuscripts were not from an it is possible that difficulties by one scribe in a might be by into the and of the scribes of early Chaucer manuscripts has how they those scribes the of and in at the of the this to the of one of the key for a scribe in a of the Canterbury Tales, and an important for theories of authorial revision. This is the arrangement of the four tales that Groups and the tales of the and The of these two of tales a of disagreement between the two earliest and most copies of the work: the Hengwrt and the Ellesmere manuscripts. Hengwrt these tales in the order with the Tale then between the and in Ellesmere we find the ordering The basis for the between these four tales that is by the modern has been by who has argued that there is no Chaucerian between Groups and the two are linked in surviving The the of traces of authorial revision be in the codicological and textual evidence of the manuscripts the placement and arrangement of these we have for authorial revision in the manuscript tradition have focused on the earliest manuscripts, the of witnesses written later in the important to this is the of the of the of the Tale by which with the manuscript in its how the earliest manuscripts evidence of later scribes similar in copies from This evidence, with a of to the of copies of the Tales, on the difficulties by and the those early scribes The Chaucer not apparent that there is any in the and there are passages that the order in which they should these links are not in surviving manuscripts, and in some they are different The most of these the earliest surviving manuscript, Hengwrt, in which the link is used to the Tale and the Tale. Blake argued that the Hengwrt scribe or an the links in an attempt to between these tales he that Chaucer had not provided This view that the Hengwrt scribe that was in Chaucer's and that no parts of the work had While Blake is in that there is no surviving evidence of of parts of the work, this not that it not Blake had we would to that tales links to the in this is not a it would more to imagine tales that these links were to of we do not to the of to the of passages in that these links are could be on a single would have been to and from an The is that the Hengwrt scribe not these links they had from the This is a more scenario than that a scribe who in the manuscript to links to these scholars that the codicological evidence that the Hengwrt scribe the and links after he had the tales The evidence for this is the way that the link to the has been a single with no between the of the tale and the and with the written well the text the scribe had a single to a but this to be for the link that he If the link was during there is no he should have been to the text a single in this Similarly, the link was for the Hengwrt manuscript, not it was the the to the Tale, the scribe the he was to an single to the this provided with more than was for the link a the scribe an for the and considerably larger gaps and after the most scholars that the links that the scribe subsequently were those now in which the scribe was to to the order of tales that he had The evidence for this is the Hengwrt versions of these links to the that the For the Ellesmere link has the Host on the to tell a tale of it is a he the Hengwrt on the to tell a for that is a of which he it of for as as any it a for as as any The Hengwrt is a which no of the which has no in the of the with the scheme are apparent the link has been for the Ellesmere the becomes the to the with the codicological evidence, with the and of the Hengwrt of the that these links are scribal rather than scribal Hengwrt of these two links an important in how a scribe this one who had access to early and exemplars, and may have been to Chaucer could Chaucer's text to an of his also to how a that might to indicate stages of authorial revision is as the result of in a of the identification of the scribe of Hengwrt and Ellesmere as and her that he was the by Chaucer in the of that there has been a to that this scribe had access to Chaucer's the evidence of these links that the difficulties by the Hengwrt scribe the placement of these and had access to the similar the of these to the evidence of two manuscripts by the same Harley and Corpus manuscripts were written early in the by a of literary manuscripts who with the scribe of the Hengwrt and Ellesmere manuscripts on a of In to his two Canterbury Tales manuscripts, has been identified as the of manuscripts of and single copies of De and the is made of a of of with two the and just of these codicological coincide with important of uncertainty in the text The first was used for the of the Tale of which was later to the by the Tale. The Tale of was on the of the Tale at the of and the of was then to of the tale. this was a later adjustment to the original and that the Tale of was not part of the original is from the insertion of the and by the use of the in the at the of The text of on and is followed by there is no at the of the nor is there a to link this to the The second on which the final line of the the link The link has been with and after with between the text and the The of the link on this with its at the of an that the scribe had some in the link and perhaps also in it within the that the which with the of the Tale, was written the link on the scribe of the Hengwrt manuscript, the Harley scribe to have the Tale and the Tale a link to the Harley scribe the Tale on a which would to the tales the link The link that the Harley scribe follows the Ellesmere in the and the however, there are that indicate with the in The Harley of as a tale for as as While this with Ellesmere in the to tell his tale the use of the and the a the in that the link by the Harley scribe was the for the Hengwrt and that he the to his The that the of was by the Harley scribe suggests that he not have access to the Ellesmere link and the link by the to The evidence of Harley is similar to that of he Scribe D not have the he acquired the they not his as a he the of the to them codicological of Corpus 198, it has been argued, also evidence of authorial revision. This manuscript's of the link might to to our that the was never The of this link is, however, to the same difficulties by the scribes of Hengwrt and Harley in the same of text. Where the scribe of the Hengwrt manuscript the and after he had the and never the the scribe of the Corpus manuscript never any of these three the of the link in Corpus was to the of by the Hengwrt scribe is apparent from the Corpus of the of the Tale, which at line In this would not be it would be witness to difficulties by the scribe in a of the text. this of the Tale with a in in the Hengwrt The distinct in at to a in the Hengwrt of the Tale. could more than the of a work or a the of this in with the loss of the of the tale in the Corpus must indicate that there was a physical in the used by scribes at this and that the of the text was not yet The Hengwrt scribe was to the text at that later to while the Corpus scribe never the second of the text. that the of the link in Corpus is to the same difficulties by the Hengwrt who the links for these tales after he had the tales the Corpus scribe not the he had no as to how to order these the scribe not the he also had no as to the Tale should be to this was to it after the Man of Law's Tale, with the Man of Law's as an This placement of the Tale after the Man of Law's Endlink, an arrangement that in manuscripts, and which to from the at B has been as an alternative authorial by the Ellesmere placement the link had been considered within the larger of access to the there no to this placement as than a scribal attempt to a of the Tale with no link it should The Man of Law's is a link that in manuscripts, not Hengwrt or Although a of these witnesses have the as the pilgrim who interrupts the Host and to the to tell his tale next have a tale that has a prologue that it must the one manuscript, Oxford, Library MS has and follows this link with the Shipman's Tale. is to this as evidence for a single authorial that that now in While this textual and codicological evidence may to the Man of Law's originally as a link to the Tale in an early of the Canterbury Tales, the disagreement in the identity of the pilgrim at B such a possibility into The content of the Man of Law's to such a possibility yet The of the link is for the the speaker in the Man of Law's with a that is in with the nor his nor the of as and in the Where the the Host with and in the the speaker is and the and the by he he no in in our a tale from Corpus 198, The of the of and its on the Man of and his of of with the of the as A more of this scenario is just as in the Hengwrt and Harley manuscripts, the Corpus scribe not a for the Tale and had no as to it should Hengwrt and Harley subsequently the the Corpus scribe and he was to his The the scribe was to the Man of Law's by the of the pilgrim to and this tale with the of the of the link in these early copies of the Canterbury Tales has suggested how evidence for authorial revision be as reflecting the of exemplars, the of the text into discrete physical and the of the which on single In the second part of this to the evidence of later manuscripts. in of evidence of authorial revision and with Chaucer these manuscripts, but these evidence of how scribes to with similar and they we to the of the link in Hengwrt, we find that this link was subsequently to later of the is now in extant Oxford, Library MS Library MS Petworth House MS Library MS and Library MS This of a in which the Hengwrt was by the in the to these scribes the a between and with the same The of how to the passages was in the of Oxford, College MS a scribe access to the Hengwrt and Ellesmere the Prologue as it in but it is by a of of a tale for as any he but as for be a tale is and tale is this This prologue is then followed by the Tale, which at the of with the the The is and then at the of have been these is the which is then followed by the Ellesmere form of the link this evidence as the work of a and that the scribe the in this way of some for for the is to the of the link as it in one of two manuscripts that the a that argues has no While is in such a his of the evidence and links do not the of a and with a the scribe might have the simpler and of rather than that the Prologue is missing from manuscripts, it more that the scribe not have the prologue he to the and so he the to the Tale to the Having this the scribe then of a of the link and it after the the is and in it the of scribes had to gaps and to an of as well as to and Chaucer's text to it as to the difficulties scribes in access to passages the evidence suggests that of the Canterbury Tales to in an just as they at the of the and that single were to and The of this manuscript, with the of its of the text, a to those who to Chaucer's in the in the early manuscript that similar the of these tales is MS original scribe not have access to the In this manuscript a second has the in the at the of the Tale, with the for this is not the final for a has to this is for it is for The same also notes that the Tale with the Tale is also the in line the same has written and then the of the link it the line of the Tale. The of these links in a about it at two separate stages of of the manuscript with two distinct The scribe the text as he a prologue to the Tale. he subsequently of a link to that the he to the in the text and it in the a in the text to indicate the of A second stage of by a second this a manuscript with the Ellesmere that the Tale should be followed by the Tale, and that the as a prologue to the Tale should in the Tale. This of of and of the text a to similar to that in While the manuscript and it represent a more and ordering of the The manuscript's to represent the text as as even at the of the overall of the manuscript suggests that the manuscript was intended to function as an for According to this theory, the notes by the second scribe would be to a later as to the ordering of the well as to of these links is a witness to the scribes in access to these passages in the form of the is in manuscripts, Oxford, Library MS which the Hengwrt of the with a of the in which the Host to the tale of This was his and by were a at had is a tale for to not be it be a she she an of of a is to she no in be it to it a tale as a be as as a tale for not a tale This link the which after the in those manuscripts that followed by two The first of these is the Epilogue and the second is a of the The to part of the Epilogue and the in is a intended to a from the form of the tale also written in and a surprising of literary This scribal link by of the which would

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.