University of Virginia Library

Search this document 


  

collapse section 
 1. 
 2. 
 3. 
 4. 
 5. 
  
collapse section 
  
  
collapse section 
 1. 
 2. 
 3. 
 4. 
 5. 
  
collapse section 
  
  
collapse section 
The Transcription of the "Clerk's Tale" in MS HM140: Interpreting Textual Effects by William McClellan
 1. 
 2. 
 3. 
 4. 
 5. 
 6. 
 7. 
 8. 
  
collapse section 
  
  
collapse section 
 1. 
 2. 
 3. 
 4. 
 5. 
  
collapse section 
 1. 
 2. 
 3. 
  
collapse section 
 1. 
 2. 
  
collapse section 
  
collapse section 
 1. 
 2. 
  
collapse section 
 1. 
 2. 
 3. 
 4. 
 5. 
 6. 
 7. 
 8. 
 9. 
 10. 
 11. 
 12. 
 13. 
 14. 
  
collapse section 
  
  
collapse section 
 1. 
 2. 
 3. 
 4. 
 5. 
 6. 
 7. 
  
collapse section 
  
  

collapse section 
  
  
  
  
  
  
  

89

Page 89

The Transcription of the "Clerk's Tale" in MS HM140: Interpreting Textual Effects
by
William McClellan [*]

HM140, a late fifteenth-century anthology manuscript housed at the Huntington Library, has been prominent in the renewed discussion of the early transmission of Chaucer's work, primarily because it contains an "independent" version of the Clerk's Tale. The tale, according to Daniel Silvia, was the most frequently excerpted of the Canterbury Tales. HM140 has received this attention, in part, because it belongs to a category of manuscripts that have been recently revalued by Derek Pearsall, who has argued that such manuscripts provide us with invaluable information for surmising the moral and aesthetic taste of their audiences.

HM140 has already been the subject of a groundbreaking article by Seth Lerer, who applies a codicological analysis of it to his interpretative reading of the tale. Lerer argues that the "excision" of the "Prologue" and two key transition stanzas just before the envoy strips the tale of its dramatic Canterbury context.[1] A major consequence of these changes is to narrow the aesthetic and moral scope of the story, modifying it to fit a conventionally pietistic, Lydgatean "homology" (311). This serious tone is reenforced with the addition of Chaucer's ballad "Truth" as a coda. Lerer thus gives us a useful characterization of the anthology's thematic logic, while presenting a plausible interpretation for the changes he finds in the text of the tale.

However, in his discussion regarding the "excisions" Lerer makes certain assumptions about the scribe's editing of the tale. For example, he assumes that the changes in the text of the tale were made by the scribe, and, further, that these changes reveal the scribe's "implicit reading of the Tale" (316). The suppositions that Lerer makes about the scribe's "act" bear further scrutiny because they refer to the crucial issue of "intentionality," in this case, editorial


90

Page 90
intentionality.[2] Using the approach he initiated as point of departure, I want to examine the question of how we determine the scribe's or editor's "intention" from the remains of the manuscript. I will pursue this through a close analysis of the transcription of the texts in the first booklet of the manuscript, focusing on the Clerk's Tale. Such an analysis will give us insight into the manuscript's production process and will provide us with information for evaluating how the tale was appropriated and read.

First, I will give an overview of the transcription of the texts in the first booklet of HM140 through an analysis of the scribal work stints and an examination of its ordinatio. Second, I will analyze the transcription of the Clerk's Tale, with the aim of evaluating possible reasons for the text's "excisions." Third, I will discuss the implications of this codicological analysis for an interpretative reading of the tale. Finally, I will suggest some possibilities for employing codicological analysis in the interpretation of texts from early in our tradition.

Transcription of Texts

There are two significant patterns in the transcription of the manuscript's text that can help us construct its production history: one is created by the different scribal hands and the other is generated by the different work stints of the scribes. The manuscript as it exists today is comprised of what were three different booklets: the first booklet consists of quires 1-6, the second, quires 7-9, and the third, quire 10.[3] At least eleven different scribes were involved in the writing of the manuscript. In the first booklet, which contains the Clerk's Tale, five different hands are evident in the copying of the text. All of the scribes wrote in secretary hands and their writing is relatively neat and fairly easy to read. The large number of different hands is unusual, and the pattern of alternation they form is even more so, especially in the sixth quire, as even a cursory analysis reveals.[4]


91

Page 91

The bulk of the writing in the first booklet of the MS was done by three scribes, with over half, 46 folios (the first two quires and quire 4, which contain about 70% of the Life of St. Albon) written by Scribe 1.[5] Scribe 2 transcribed about twenty folios: all sixteen folios of quire 3 (ff31-46) and the recto side of folio 63 in quire 5. Both quires contain the Life of St. Albon. In addition, Scribe 2 copied almost three folios in three different stints in the sixth quire. Scribe 3 is responsible for all except one folio of the fifth quire, where the first four folios contain the text of St. Albon and the rest, ten folios, hold the Clerk's Tale. He copied virtually all of the Clerk's Tale. In all he is responsible for twenty two folios—fourteen in quire 5 and eight in quire 6. The other three folios contain different shorter works. Scribes 4 and 5 are minimally involved in the copying of the manuscript, and only in the eccentric sixth quire. Scribe 4 has three different stints amounting to about four folios, and scribe 5's total contribution to the transcription is only two lines on f92v.

For the most part in the first five quires of the first booklet, the different scribal work stints coincide with the quire divisions. Scribe 1 inscribes the first three quires, and has no more involvement in the copying of the manuscript. Scribe 2 writes the fourth quire. This pattern is slightly modified in the 5th quire: Scribe 2 copies the recto side of f63, the first folio in the quire. Then scribe 3 takes over and continues his stint to f83, line 28, whence scribe 2 finishes up the Clerk's Tale and the ballad, "Truth." In addition, scribe 2 has two more short stints in quire 6.

The continuing presence of scribe 2 at key junctures in the writing indicates that he probably functioned in a supervisory role in the transcription of the first booklet, at least in the medial or later stages of it.[6] His initial stint, which includes quire 4, ends on the recto of the first folio of quire 5. Most likely he was marking out the page format for scribe 3, who takes over the copying and completes quire 5 and about one third of quire 6 in his first stint. That scribe 2 was acting as a supervisor gains plausibility because he begins his second stint on line 29 of f83, taking over from scribe 3 to finish up the Clerk's Tale. He also executes two more short stints in quire 6 including the last work in this booklet, Lydgate's Testamentum. So scribe 2 has continuing, if interrupted, involvement in the transcription of the booklet, copying at least some part of five of the eight different works that are in it.


92

Page 92
These include all three of Chaucer's works in the manuscript, as well as the major works by Lydgate, St. Albon and the Testamentum. And his stints come at key junctures: he ends the Clerk's Tale and begins "Anelida" and completely transcribes the ballad, "Truth." However, scribe 3 also plays a significant role in the copying of the MS. He ends St. Albon, with its elaborate colophon, and begins the Clerk's Tale. He also begins Lydgate's Testamentum, including its title, and the short poem, "Uppone a Cross," which he titles and copies completely.

The analysis thus far shows that the two scribes, 2 and 3, who were involved with the transcription of the Clerk's Tale played key roles in the overall transcription of the first booklet. It also shows that they worked in close proximity to each other, probably in the same shop, and that scribe 2 acted in a supervisory role.[7] My analysis also suggests that either the ordinatio was not fully worked out when the copying of the booklet was begun, or, more likely, that the original production plan was abandoned in the course of its transcription. A more detailed examination of the pattern formed by the work stints of the different scribes in the eccentric sixth quire strongly suggests that the latter was probably the case: the production of the booklet was increasingly interrupted and finally halted at the end of the sixth quire.

The Eccentric Sixth Quire

Four of the five scribes who transcribed the first booklet wrote in quire 6. This alone suggests that the production procedure for this part of the manuscript was extremely unusual. Doyle and Parkes state that the usual work stint for a scribe was governed by the quire structure; that is, the smallest customary unit of work was the quire. But what is even more peculiar is that the pattern of alternation revealed by change in hand shows there were no less than twelve changes in the quire:

                           
Scribe 3 ff78-83 line 28  Clerk's Tale (5½ ff) 
Scribe 2 ff83 line 29-84  Clerk's Tale (1 f) 
"Truth" (½ f [½ of folio is blank]) 
Scribe 3 ff84v-85 line 16  "Uppone a Cross" (¾ f) 
Scribe 2 ff85 line 17-85v line 9  "Anelida" (½ f) 
Scribe 3 ff85v line 10-86 line 20  "Anelida" (½ f) 
Scribe 4 ff86 line 21-88v  "Anelida (1 f [½ f87 blank]) 
"Midsomer Rose" (1½ ff) 
Scribe 3 ff89-90  "Midsomer Rose" (1 f) 
"Song of Virtue" (1 f) 
Scribe 4 ff90v-91 line 9  "Song of Virtue" (1¼ f) 
Scribe 3 f91 lines 10-13  Testamentum (4 lines, plus title) 
Scribe 2 ff91 line 14-91v line 20  Testamentum (1 f) 
Scribe 3 ff91v line 21-92v line 6  Testamentum (½ f) 

93

Page 93
   
Scribe 5 f92v lines 8-9  Testamentum (two lines) 
Scribe 4 f92v lines 10-35  Testamentum (¾ f) 

This breakdown shows that the stints of the scribes are short and the changes of hand are frequent. The twelve changes, marking thirteen scribal work stints, is a phenomenal number, given that the standard work portion was a quire. The stints range from two lines to five-and-a-half folios with most of the stints (eight) a folio or less. Even more significant is that both the shortness of the stints and frequency of change increase towards the end of the quire, where fourteen stanzas of Lydgate's Testamentum are placed, suggesting that the writing was much interrupted and finally discontinued before the transcription of the work was completed. Although it cannot be absolutely determined from the residue of the manuscript whether or not the transcription of the Testamentum ended at f92v, I suspect that it did. I arrive at this conclusion because the next folio, which was the last of the quire, has been cut, most likely because it was blank, indicating that the copying of the work had been discontinued.[8]

A plausible narrative of the booklet's production is one that describes it as moving away from a well organized effort where the ordinatio of the manuscript was known and where the work was being portioned out in the usual fashion. The regular pattern of the scribal writing stints in the first five quires, which follows the standard length, supports this thesis. The exception at the beginning of quire 5 where the second hand writes only the first page of f63 (one half folio) can easily be explained in terms of scribe 2's presumed supervisory role. The execution of the ordinatio in these first quires would also support such a thesis. In these, the writing, format, and decoration of the booklet's relatively plain ordinatio is well done and consistent. Although the ordinatio changes in the fifth quire, from the more elaborate formatting of St. Albon to the plainer layout of the Clerk's Tale, it is not until the sixth quire that the quality control of the booklet's production radically decreases. For example, rubrication ceases at f78v, the second page of the first folio in the sixth quire. At this point, the orderly production process was increasingly disrupted, and the transcription finally was halted in the initial stanzas of the Testamentum.

The pattern formed by the different scribal hands and stints thus points to a narrative in which the production of the first booklet moves from a well organized effort to one that is much less so, indeed one that leaves the booklet unfinished. Such a pattern of destabilized and incomplete production can been seen in the booklet's larger organizing structure, its ordinatio.

The Ordinatio of the First Booklet

The ordinatio of the manuscript, as a whole, moves from the relatively well defined and elaborate to the reduced and indefinite. But even though the ordinatio of the MS in the first booklet is inconsistent, virtually all the works are set off by clear division boundaries and discrete formatting. The


94

Page 94
exception to this pattern is the conflation of Chaucer's Clerk's Tale and the ballad, "Truth." None of Chaucer's works is attributed to him and none has a title. By contrast, some of Lydgate's works are given titles and attribution, and two of them, St. Albon and "Uppone a Cross," have both. St. Albon, given a short title at the beginning, has an extended colophon that indicates, in addition to the author, the name of the person for whom Lydgate translated the work from French and Latin.[9] The elaborateness and detail of title and attribution are unique in the MS. For other ascribed works, as the poem, "Uppone a Cross," the attribution is by initials only, i.e. "JL," identifying Lydgate as the author. Two other poems that have the monograph "JL" at the end, "Midsomer Rose" and "Song of Virtue," do not have titles.[10] The title of "Uppone a Cross," which follows the Clerk's Tale-Truth, consists of the first two lines of the poem in Latin (f84). The other work that has a title, but no attribution of authorship, is Lydgate's Testamentum. So if one judges by the ordinatio, the works of Lydgate are accorded more importance than Chaucer's, with the Life of St. Albon, the first and longest work in the manuscript, having first place in the hierarchy.

An examination of the mise-en-page of the folios reveals even more clearly the changes that occurs in the design and execution of the booklet's formatting. Although the format of HM140 is plain compared with many other manuscripts (such as the richly designed and executed Ellemere MS), the ordinatio of its first booklet changes from the fairly detailed format of St. Albon, where, for example, rubrication is used for marking text divisions and capitals, to a much simpler page format in the other works, as a comparison between St. Albon and the Clerk's Tale will demonstrate.

For St. Albon, the layout, rubrication and capitalization are clearly well planned and consistently executed throughout. The first folio has thirty-three lines of text, and there are between thirty-three and thirty-six lines per folio, with an average of thirty-five. The text is organized in a rhyme royal stanza, its form set off by rubricated initial capitals and by spacing between stanzas. Decoration of the text is simple, but the use of rubrication is extensive. Besides the initial stanza capitals, the initial line capitals are ornamented with red. In addition, rubrication is used to underline key words in the text, usually proper names or nouns. There are from twelve to twenty-two of these underlinings per folio, which I think act as a guide to reading the text.[11] Whole


95

Page 95
words are rubricated as well, usually the divisio markers, highlighting the ends and beginnings of sections within the narrative. For example, on f14, at the end of the first part, a division marker appears in red letters: "Here Endith the First Boke and Begynnyth the Prologg of the Second Lyne of the Conuersion of the Blessid Prothomartir of all Inglond Seint Albone."[12] Occasionally there are marginal glosses in red ink, such as "Verba Auctoris" (f13v). Finally, at the end, the colophon is in red ink.

The design and execution of the formatting of the text of St. Albon are done with obvious care. The work is titled and attributed. The stanza form is clearly marked with rhyme, rubrication, enlarged capitals, and spacing. Text divisions are marked in red, as are line capitals, and key words are underlined in red.

In contrast, the layout of the second work in the manuscript, Chaucer's Clerk's Tale, is both less elaborate and less clearly marked. The Clerk's Tale begins on f68 and ends on f83v. If "Truth" is considered a conclusion to the tale, as I believe it should be, the ending is on f84. Unlike St. Albon, the tale contains few signals or directions for reading in its formatting. It has no title, no headings, no division markers of its parts, and no attribution of authorship. There is no "Prologue," the text beginning at the opening of Chaucer's Tale, E57 (IV.57): "There is ryght atte the West Syde of Ytaile. . . ." "Truth" is appended directly after the end of the tale's envoy, without title or break in the continuous line format, with an "Explicit" following the ballad.

The number of lines per folio page is irregular, ranging from thirty-one lines on f68 to thirty-seven on f77, with the average being thirty-three lines. The initial letter of each line is capitalized and sometimes the initial capital of a page is slightly larger than the rest. The only large rubricated capital in the text is the tale's initial "T," which is three lines high. The other initial line capitals are only splashed with red, and this decoration ends at f78. There is no other rubrication. Except for the rhyme scheme of rhyme royal, which is disrupted, the stanza form is not marked or isolated; consequently, the text runs continuously, without spacings from beginning to end. None of the features which Parkes argues were commonly used to mark a verse format (i.e. spacing, illumination, capitals, rubrication, underlining, or paragraph markers) is present in the format of the tale ("Punctuation" 130). The tale in HM140 thus has a reduced, prosaisized format.

Comparison of the formatting of St. Albon and the Clerk's Tale clearly shows that a shift occurred in the book's production between these two works.


96

Page 96
The well designed and executed format of St. Albon gives way to the much less carefully formatted and decorated text of the Clerk's Tale and the plainer format of the shorter works that follow in the booklet.[13] The modification in the format plan and execution of the Clerk's Tale marks a significant shift in the booklet's ordinatio and reflects a change in the status of the manuscript's production. The most obvious indication of this change is in the decorative detail, the splashing of the initial capitals with red ink, which stops about two thirds of the way through the text of the tale.

This downscaling in the priority of the booklet's production might have resulted from a decommissioning of the manuscript's project. But without documentary evidence it is impossible to determine what the causes were for lowering the production priority, and for concomitant changes in the bibliographical text of the booklet. Consequently, the ordinatio or intentional structure of the manuscript remains ambiguous. The alteration in the booklet's ordinatio is also reflected in the transcription of its texts, where we see a similar reduction in the quality of the scribal copying near the end of the first booklet.

The Transcription of the Clerk's Tale

Two different scribes transcribed the text of the Clerk's Tale: scribe 3 copied most of it, from f68 to f83 line 28; scribe 2 completed the tale, including the appended ballad, "Truth," f84 line 11. My own examination confirms Manly's original assessment that the transcription was ". . . a bad copy of a MS of good tradition," and that the writing was much interrupted, with little or no supervision and few corrections (435, 434). The frequent interruptions in the transcription, I think, may account for the many missing lines in the text of the tale. Besides the "Prologue," the two other sections of the Clerk's Tale that others have noted as missing are the seventy-two lines (E189-260) near the beginning of the tale, probably as the result of a lost folio following f69,[14] and, as I have already mentioned, the "Lordynges" and "Wife of Bath" stanzas (E1163-1176). I have found seven other major lacunae, totalling thirteen lines, including a complete stanza, as well as several instances of line transpositions, and one repetition.[15]


97

Page 97

In considering the pattern of scribal "error" I am discounting the absent "Prologue" because none of the "independent" versions of the Clerk's Tale has it, and it is external to the tale.[16] I think that it is highly unlikely that folio 68, between the end of St. Albon on f67v and the beginning of the Clerk's Tale on f68v, was left blank for copying the "Prologue" at a later time, because there is insufficient space here to fit its fifty-eight lines. Except for the line transpositions on f83v in scribe 2's stint, all of the other errors were made by scribe 3. The configuration of missing lines and transpositions forms an alternating pattern between folios where there are no mistakes and those where there are, with the alternations increasing in frequency towards the end of the tale.[17]

Comparing the errors in the transcription of quire 5 with those in quire 6 shows even more clearly the decreasing quality control of the scribe's writing.[18] Not only is the frequency of "error" higher in quire 6, but the total


98

Page 98
number of missing lines is greater.[19] In addition, the kinds of "errors" made in quire 5, such as the three line transpositions, suggest that the scribe was more attentive to his work here than in quire 6, which has only missing lines.[20]


99

Page 99

This points to a conclusion that the quality of scribe 3's work decreased substantially towards the end of his writing stint of the Clerk's Tale.

Implications for Interpretation of the "Clerk's Tale"

Exactly what kind of guidance does a reconstruction of the production process for HM140 provide for evaluating the absence of major sections, the "Prologue" and the "Lordynges" and "Wife of Bath" stanzas, from the text of the Clerk's Tale? These omissions are crucial for constructing any interpretation of the tale in its HM140 context because these sections of the tale provide the reader with a guide for reading the work. I have argued elsewhere that, in The Canterbury Tales, the "Prologue" introduces two of the narrative voices which constitute the tale's dialogic discourse (Multi-Voiced). The two transitional stanzas towards the tale's end also promote the dialogic voicing of the tale because they allude specifically to the performative frame of the Canterbury context. The absence of these passages in the HM140 site has the effect of reducing the complex and contradictory voicing of the tale to a more monologic narrative voicing, which fits a more conventional moral viewpoint. So it is important for any interpretation of the tale in the HM140 anthology to evaluate whether their absence is the result of deliberate editing, omission in the exemplar, or scribal mis-transcription.

This analysis shows that it is very probable that the scribe simply missed these two stanzas in his copying. Their omission comes right at the end of his stint, just before the final five lines on f84, and he had just dropped a complete stanza several folios earlier (f81; E1009-1015). This, combined with the fact that the frequency of the line omissions increased towards the end of his stint, makes it plausible that he simply missed the lines, possibly because he was interrupted. The fact that his stint ended just three lines before the end of the folio page strongly suggests that it was not planned but the result of some distracting interruption, which may very well have been the reason for the termination. But my conclusions that the missing stanzas were the result of mis-transcription are also based on an analysis of the tale in other manuscripts.

Such a conclusion, however, is not immediately apparent. When I examined the twenty-two manuscripts in the group Manly identified as having a close textual relationship with HM140 I found the following.[21] Virtually all of the manuscripts in this group (except Np and Ra1) have the "Prologue," which is not surprising because all, except Np, contain the tale in its Canterbury context. Most have the "Wife of Bath" stanza (except Ld2, Lc, Mg, Np),


100

Page 100
and all have the "Lordynges" stanza. One conclusion that can be drawn from this analysis is that the version of the tale in HM140 is unusual for the group with which it shares a close textual affiliation because it is missing important contextualizing sections of the tale that virtually all others have.

In a detailed analysis comparing all the missing lines in HM140 with those of the four other manuscripts that Manly identified as having the closest textual affiliation, I found that none has any of the omissions that HM140 has, and only one has a single set of similar line transpositions.[22] The fact that there are so many missing lines unique to HM140 confirms my conclusion that the omissions in the text, including the "Lordynges" and "Wife of Bath" stanzas, could have resulted from disruptions in the copying process itself, and not from editorial changes or omissions in the exemplar. An examination of 6 other manuscripts randomly chosen from the group sharing a close textual affiliation with HM140 (Ry1, S12, Ld2, Np, Cp, La) further confirms this evaluation. Except for La, which transposes lines 456/457, none of these MSS has any of the line transpositions of HM140. And except for Np, none of these manuscripts is missing any of the lines lacking in HM140.[23]

But it is possible that some changes in the text of the tale are the result of the tale's new non-Canterbury context. Unlike virtually all of the manuscripts with which it shares a close textual affiliation, HM140 contains an "independent" version of the tale. When we look at this class of manuscripts, a slightly different pattern emerges: none of these manuscripts has the "Prologue" or "Wife of Bath" stanza. Thus we can deduce that the usual ordinatio of the tale in non-Canterbury sites is without the contextualizing frame that the Prologue provided. And the omission of the "Wife of Bath" stanza is not unusual even in the Canterbury context, where it is absent in about twenty manuscripts. So it is possible that the exemplar for HM140 had no "Prologue" or "Wife of Bath" stanza. Both these patterns, which suggest different histories for the tale's transcription, raise serious doubts about the deliberate editing out of the two transition stanzas in the HM140 context.

A plausible narrative regarding the omissions is that both the "Prologue" and "Wife of Bath" stanza were absent in the exemplar and that the scribe


101

Page 101
miscopied the "Lordynges' stanza, as he did the earlier stanza (f81 seven lines [E1009-15] missing between lines 2 and 3). The possibility of a miscopying of the stanza has an even greater plausibility because it occurs right at the end of scribe 3's stint. But an even more important point is that we cannot determine from the manuscript "evidence" whether these omissions of text were intentional "excisions," as Lerer claims (and if they were, at what point in the process of manufacture they occurred) or whether they were a consequence of the problematics of production. These conclusions regarding the poor quality of the transcription and the standard ordinatio of the "independent" versions raise serious doubts that the absent sections of the text were deliberately edited out.

What I think this detailed codicological analysis of HM140 makes abundantly clear is that the quality of the manuscript's production was not high, or rather that it suffered a radical decrease in the sixth quire, where most of the Clerk's Tale was transcribed. Consequently, the inferences that can be drawn about the missing parts of text have to be carefully qualified. The analysis shows that the manuscript "evidence" is, at best, ambiguous, throwing real doubt on the definitive status of the "intentional structure" of the tale's ordinatio. The lack of a clear conception of this structure makes it impossible to ground any interpretation of the tale. The destabilization created by the low quality control of the tale's transcription is also advanced by other ambiguous features in the residue of the manuscript: the unfinished state of the first two booklets and the heterogeneous collocation of booklets in the manuscript's compilatio.

Coda: Brief Thoughts on Future Directions for Codicological Analysis in the Interpretation of Texts

As the analysis of MS HM140 shows, it is impossible to ascertain the scribe's or editor's "intention," especially when external documentary evidence is lacking as is often the case with texts that have been transmitted to us from early in our tradition. But while the editor's specific "intentions" are not recuperable, I think a codicological analysis of a manuscript can provide us with information which we can use in interpreting its texts. I would like to briefly suggest the direction such an approach might be developed.

In addition to analyzing the transcription of texts, this approach would consider the presentational features, such as page layout and decoration, as well as the larger organizing concepts, compilatio and ordinatio, that A. I. Doyle and M. B. Parkes have shown are important for the interpretation of texts (186). It would also take into account Jerome McGann's argument that the bibliographic text is crucial for interpreting the linguistic text (57). Most importantly, it would have recourse to Bakhtin's general theory of dialogic discourse, which ties together the analysis of presentational features with the social and collaborative aspects of textual transmission. For Bakhtin, who understood discourse as fundamentally social, the formal features of a verbal work of art are a constitutive part of its meaning (259).[24]


102

Page 102

Such an approach will give us a way of construing an interpretative model that has the virtue of utilizing the precise information generated from an analysis of the manuscript, the material residue of the text's construction site, to help construct an "intentional structure" for reading the texts it contains. Yet it also takes into account the overdetermined, social aspects of textual transmission. While this may make us more aware of the problematic nature of that transmission, it also gives us a more nuanced way of imagining the historical world(s) presented in those texts dispatched to us.

Finally, I think the most important issue that HM140's problematic archaeology helps us to confront is that manuscripts are the leftovers of construction sites whose traces and marks have to be interpreted. The manufacture of information about their compositional process constitutes an interpretation, and hence does not stand outside the hermeneutic circle to be used as "evidence." They are not self-explanatory compositions with autonomous set(s) of meaning(s), but residues of communicative structures that we must reconstruct so that they can speak to us. This reconstruction is an on-going, event-ual constituting of their "truth."[25]

Works Cited

  • Bakhtin, Mikhail. The Dialogic Imagination. Trans. Caryl Emerson and Michael Holquist. Austin: U of Texas P, 1981.
  • Chaucer, Geoffrey. The Riverside Chaucer. Ed. Larry Benson. 3rd ed. Boston: Houghton Mifflin, 1987.
  • Doyle, A. I. and M. B. Parkes. "The Production of Copies of the Canterbury Tales and the Confessio Amantis in the Early Fifteenth Century." Medieval Scribes, Manuscripts & Libraries. Ed. M. B. Parkes and A. G. Watson. London: Scolar P, 1978. 162-210.
  • Dutschke, C. W. Guide to Medieval and Renaissance Manuscripts in the Huntington Library. 2 vols. San Marino: Huntington Library, 1989.
  • Lerer, Seth. "Rewriting Chaucer: Two Fifteenth-Century Readings of the Canterbury Tales." Viator 19 (1988): 311-326.
  • Manly, John and Edith Rickert. The Text of the Canterbury Tales. Vol I. Chicago: U of Chicago P, 1940.
  • McClellan, William. "The Consequences of 'Treuth': Reading Two Versions of the Clerk's Tale." Genre (forthcoming).
  • ___. "A Codicological Analysis of the Quire Structure of MS HM140 and Its Implications for a Revised Ordinatio." TEXT (forthcoming).
  • ___. "The Dialogic Other: Bakhtin's Theory of Rhetoric." Bakhtin and Otherness. Discours social/Social Discourse 3 (1990): 233-250.
  • ___. "Bakhtin's Theory of Dialogic Discourse, Medieval Rhetorical Theory, and the Multi-Voiced Structure of the Clerk's Tale." Exemplaria 1.2 (1989): 461-488.

  • 103

    Page 103
  • MacCracken, H. N. Ed. The Minor Poems of John Lydgate. Pt 1. London: EETS es 107, 1911.
  • McGann, Jerome. The Textual Condition. Princeton: Princeton UP, 1991.
  • Parkes, M. B. "Punctuation, or Pause and Effect." Medieval Eloquence: Studies in the Theory and Practice of Medieval Rhetoric. Ed. J. Murphy. Berkeley: U of California P, 1978. 127-142.
  • ___. "The Influence of the Concepts of ordinatio and compilatio on the Development of the Book." Medieval Learning and Literature. Ed. M. Gibson and J. Alexander. Oxford: Clarendon, 1976. 116-141.
  • Patterson, Lee. "On the Margin: Postmodernism, Ironic History, and Medieval Studies." Speculum 65 (1990): 87-108.
  • ___. Introduction. Literary Practice and Social Change in Britain, 1380-1530. Ed. L. Patterson. Berkeley: U of California P, 1990. 1-14.
  • Pearsall, Derek. "Editing Medieval Texts: Some Developments and Some Problems." Textual Criticism and Literary Interpretation. Ed. J. McGann. Chicago: U of Chicago P, 1985. 92-107.
  • Robinson, P. R. "The 'Booklet' A Self-Centained Unit in Composite Manuscripts." Codicologica 3 (1980): 46-69.
  • Silvia, Daniel. "Some Fifteenth Century Manuscripts of the Canterbury Tales." Chaucer and Middle English Studies. Ed. B. Rowland. Kent, Ohio: Kent State UP, 1974. 153-163.
  • Strohm, Paul. "Chaucer's Fifteenth-Century Audience and the Narrowing of the 'Chaucer Tradition.' Studies in the Age of Chaucer 4 (1982): 3-32.
  • Vattimo, Gianni. The End of Modernity: Nihilism and Hermeneutics in Postmodern Culture. Trans. Jon Snyder. Baltimore: Johns Hopkins UP, 1991.

Notes

 
[*]

I want to thank the Huntington Library, where HM140 is housed, for making the MS available to me and for providing me with a congenial working environment. I want to thank the NEH for supporting the initial stages of this study with a summer grant, and Baruch College and the CUNY Research Foundation for their continuing support of my work. I also want to thank Martin Stevens for his supportive criticism. Finally, I thank Steven Kruger and Peter Hitchcock who read and offered valuable suggestions in the final stages of this essay. The mistakes, however, are mine.

[1]

One stanza is addressed to the "lordynges" and the other refers to the "Wife of Bath" (E1163-1176). All references to the modern edition of the "Clerk's Tale" are to the Riverside Chaucer.

[2]

Lerer himself seems to be aware of this difficulty, for early in his essay he suggests that he is using the term "scribes" in the widest construal of those who "read, wrote and rewrote Chaucerian manuscripts for their own purposes" (311). His employment of "scribe" as a metonymy for the whole process of textual transmission covering a period of more than half a century is a helpful and convenient way of specifying agency in the transcription of a text. Yet the way he uses the term in his discussion can be confusing. Sometimes he seems to use it as a marker of "intentionality" to explain textual changes. Consequently, the sense that these changes are overdetermined, which is included in his original formulation, tends to be neglected. I think we always have to be careful to regard these changes as the product of a multiple set of scribal/editorial practices carried out over an extended period of time. Hence, lacking external "evidence," we should always be aware of the difficulty, if not impossibility, of specifying exactly when and why and by whom they were made.

[3]

Unlike Manly and Dutschke, I consider quires 7 to 9 to be a separate, unfinished booklet that was combined with the other two booklets in the early sixteenth century when the manuscript was bound. I discuss the issue of the manuscript's composite structure in an article, "A Codicological Analysis of the Quire Structure of MS HM140 and Its Implications for a Revised Ordinatio."

[4]

I am following Manly and Dutschke in the identification of the hands. I also consulted with Ralph Hanna III about several key junctures, such as the change of scribal hands at the end of the Clerk's Tale. The following is a breakdown of the hands in the first booklet: Scribe 1: ff1-30v (quires 1&2), ff47-62v (quire 4—i.e. if textual ordination is followed, this should be quire 3, but probably was transposed with quire 4 when the MS was bound). Scribe 2: ff31-46v (quire 3, but as already noted should be quire 4), f63 (quire 5), ff83 line 29-84, ff85 line 17-85v line 9, ff91 line 14-91v line 20 (quire 6). Scribe 3: ff63v-83 line 28 (virtually all of quire 5 which ends at f77, and quire 6, ff78-83, ff84v-85 line 16, ff85v line 10-86 line 20, ff 89-90, f91 lines 10-13, ff91v line 21-92v line 7 (all in quire 6). Scribe 4: ff 86 line 21-88v, ff90v-91 line 9, f92v lines 10-35 (all in quire 6); Scribe 5: f92v lines 8-9 (quire 6).

[5]

As I noted above, the order of the third and fourth quires was transposed when the manuscript was bound. As it stands now Scribe 1 inscribed quires 1, 2, and 4. Scribe 2 copied quire 3 which was originally quire 4.

[6]

Although Manly notes "no supervision; few corrections . . ." (434) in the copying of the Clerk's Tale, his evaluation was probably based on explicit marks of supervision, such as marginal notation, of which there are none.

[7]

Lerer concludes that the manuscript was probably produced in the household of the "northern gentry" (312). This accords with Ralph Hanna's suggestion to me that production features of the manuscript indicate that it might have been produced in a private household, not a commercial shop. Such a provenance might help to explain the transcription problems in quire 6. But as Parkes and Doyle have argued, scribes often worked in close collaboration on commercially produced manuscripts. From the external evidence we have it is not possible to determine with any certainty where the manuscript was produced.

[8]

The text of the Testamentum runs to the end of the verso of f92. My examination shows that this folio has been cut away leaving a one inch strip.

[9]

"Here endith the glorious lyf and passyoun of the blessid martir Seint Albone and seint Amphiball which glorious livis were tranlatid oute of ffrenssh and latyn by dane John Lydgate monke of Bury at the request and prayer of Master John Whetamstede the yere of oure Lorde M CCCC xxxix [1439] and of the said Master John Whetamstede of his Abisse xix."

[10]

MacCracken thinks the monogram at the end of the poem "Uppone A Crosse" "looks like J L" (Minor Poems, Part I.254). "Midsomer Rose" ends with an Explicit followed by "JL" and "Song of Virtu" ends with an "Amen" followed by "JL." (See Minor Poems, Part II.785, 838). The point is arguable, but my own examination of the MS confirms this opinion. More to my point, however, is that if these are attributions of authorship they are far less explicit than the full one given to St. Albon.

[11]

For example, f13v has the following words underlined: Albon, martire, Albot, clerke, Anapt, lybrarie, coffers, poyettis, prudent, phelozyfires, ethiologie, whete, Whete, glenys, phelozifers, memory, Reportery, knyghthod, passion. Towards the end of the work, beginning with f65 the refrain 'O prothomartir of Brutis Albion" is underlined in red. It occurs twelve times in the closing stanzas.

[12]

Some other examples follow: On f15 the end of the prologue of the second book has a rubricated division marker, "Her Endith the Prollogg. And begynnth the Seconde Boke of the Glorious Prothomartir Seint Albon Howe he was made Gourernour of the Cite of Verolamye." F59 has a rubricated, inter-stanzaic line in Latin repeated four times, two recto and two verso: "Albanus Ver Egregius Martir Extat Gloriosus." On f60 the end of the second book is marked by the following rubric: "Explicit passio sancti Albani Incipit passion Sancti Amphibali sociorumque suorum cum Translatione Sancti Albani."

[13]

I think that there are a number of reasons for the change in the verse format of the Clerk's Tale to what I call a prose formatting, and that this change reflects more the textual and ideological status of the tale rather than production values per se. I discuss this reformatting in an article interpreting the tale in its HM140 context, "The Consequences of 'Treuth': Reading Two Versions of the Clerk's Tale." But the changes in decoration do reflect the downscaling of its production priority.

[14]

The 70 lines that are missing would very comfortably fit into the number of lines per page, which varied from 31 to 37. According to the pattern of the quire structure, this folio would have been of vellum.

[15]

In my examination of the transcription of the Clerk's Tale I focused on the major "errors" and lacunae, and for the most part noted those variations at the unit level of a line or greater. I did not record variations within the line, though there are many. I did note and record one scribal "correction" where a dropped line was inserted between two already written lines. The purpose of my examination is not to provide an exhaustive description of the copying of the tale, but to generate sufficient information to evaluate the quality of the scribes' work. In the listing of the variations in the transcription of the tale I also give reference to the Riverside Edition of Chaucer's Works, indicating either the missing line or the section where line transpositions occurred.

    Quire 5 (ff68v to 77; 9 1/2 folios):

  • 1 f68v lacuna: "Prologue" [E1-56]
  • 2 f71 lacuna: line missing between lines 24 & 25 [E353]
  • 3 f71v transposition: lines 19 & 20 between lines 23 & 24 [E379-385]
  • 4 f72v correction: line 6 squeezed between lines 5 & 7 [E399]
  • 5 f72v transposition: lines 31 & 32 transposed [E456-462]
  • 6 f73 repetition: line 1 repeats penultimate line (31) on f72v [E458]
  • 7 f73 transposition: lines 2 & 3 transposed [E459-460]
  • 8 f74v lacuna: line missing between lines 2 & 3 [E558]
  • 9 f74v lacuna: line missing between lines 8 & 9 [E565]

    Quire 6 (ff78 to 84, line 11; 6 1/6 folios):

  • 10 f78 lacuna: line missing between lines 4 & 5 [E798]
  • 11 f78 lacuna: line missing between lines 22 & 23 [E817]
  • 12 f79 lacuna: line missing between lines 21 & 22 [E886]
  • 13 f81 lacuna: 7 lines (stanza) missing between lines 2 & 3 [E1009-15]
  • 14 f83 lacuna: 14 lines (2 stanzas) missing between lines 21 & 22 [Lordynges and Wife of Bath stanzas, E1163-76]
  • 15 f83v transpositions: multiple line transpositions in Envoy, lines 18 to 24 [18, 20, 22, 23, 19, 21, 24]

[16]

The six extant manuscripts with "independent" versions of the tale are: Harley 1239 (Ha1), Harley 5908 (Hl4), Longleat 257 (Ll1), Naples XIII.B.29 (Np), Rawlinson C86 (Ra4), and HM140.

[17]

In the first two-and-one-half folios, ff68v to 70v, there are no major lacunae or line transpositions. In the next four, ff71 to 74v, there are eight major mistakes, at least one per folio (except f73v), and three folio pages that have two each, ff72v, 73, and 74v. In the next three folios, ff75 to 77v, there are none. In the first one-and-one-half folios of quire 6, ff78 to 79R, there are three major errors, two on one page. On the next one-and-one-half folios, 79v to 81v, there are none. On f81 there is a major lacuna of seven lines, followed by one-and-one-half folios, ff81v to 82v, where there are no errors. Finally, on f83, there is the major lacuna of fourteen lines, the "Lordynges" and "Wife of Bath" stanzas, and on f83v, Scribe 2, who has taken over the writing at the end of f83, has several line transpositions in the final lines of the envoy. The lacunae increase in frequency and size as the transcription of the tale progresses.

[18]

I am not counting the error I listed as #4 because it is, as I note, an interlinear insertion of a line, that is, a correction of a missed line.

[19]

In quire 5 there are three separate lacunae, all consisting of one line each. In quire 6 there are five instances of omission, three of one line each, one of seven lines, and one of fourteen lines, totalling twenty-four lines.

[20]

There are several possible reasons for transposing a line. The most obvious is that the scribe was following the transposition order of his exemplar, which may be the case in the transpositions in lines 31 and 32 on f72v and lines 2 and 3 on f73. Another is that the scribe just mixed up the lines. But a third may be that the scribe missed a line, realized he had done so, and then added it after the line he had just transcribed. Looking at the one correction I listed in quire 5 and the cluster of errors at the end of f72v and the beginning of f73 will give a good illustration of scribe 3's working procedure and make my conclusion about his practice more concrete. In the first example, the missed line correction on f72, line 6 is squeezed between lines 5 and 7:

  • 5 As in a cote or in a ox stall
  • 6 |But noryshid in an Emprous hall|
  • 7 To euery wyght waxen she is so dere
  • 8 |But| And worshipfull that folke there she was
The "But" in line 8 is superlinear and partially erased. This correction is very interesting because it gives us a rare opportunity of "seeing" scribe 3 correct a mistake. I surmise that scribe 3 missed line 6 beginning with, "But noryshid" and later caught his omission. The partially erased superlinear "But" on line 8 indicates that he probably was going to place the omitted line there, but corrected this and squeezed it between 5 and 7, which is the proper order. This potentially missed line shows the scribe correcting himself and suggests that the quality control of the writing was still probably quite good. A second series of textual variations to consider occurs at the end of one folio and the beginning of another, which may account for some confusion. Near the end of f72v, scribe 3 transposed the last lines of the folio, 31 and 32. He then repeated the penultimate line (31) of 72v as the first line of f73 and transcribed lines 2 and 3 on f73:
  • f72v 30 He had assayed here inowe before
  • 31 Her for to tempt and alway more and more
  • 32 And founde here euer goode what nedith it
  • f73 1 Here for to tempe and all way more and more
  • 2 But as for me I say euyll it sitte
  • 3 Thought som men preyse it for asotill witte
Here is the passage from the Riverside Chaucer for comparison:
He hadde assayed hire ynoght bifore,
And foond hire evere good; what neded it
Here for to tempt, and alwey mooore and moore,
Though som men preise it for a subtil wit?
But as for me, I sey that yvele it sit (456-460)
Here we have a cluster of three mistakes; the first two, at least, are closely connected. The earlier transposition of lines 31 and 32 probably resulted in the second mistake of line repetition. When scribe 3 started f73 he probably referred to line 32, the last line on F72v, beginning "And founde . . ." (one of the transposed lines). He then repeated the line beginning, "Here for to tempe. . . ." This suggests that his exemplar had the proper line order. The scribe then wrote the next two lines in transposed order, which suggests that the work had been much interrupted. Both these line transpositions were made in the texts of manuscripts that Manly identifies as having a close textual affiliation to that in HM140: lines 456 and 457 in Rawlinson Poetry 223 (Ra3) and lines 459 and 460 in Harley 7335 (Ha5) and Trinity College Cambridge R.3.3. So it is plausible that these transpositions, especially that of lines 459 and 460, were in the scribe's exemplar.

[21]

Those manuscripts in the group where Manly placed HM140 are: Additional 35286 (Ad3), Harley 7335 (Ha5), Rawlinson Poetry 223 (Ra3), Trinity College Cambridge R.3.3. (Tc1), Cambridge Dd.4.2 (Dd), Egerton 2864 (En3), Additional 5140 (Ad1), Harley 7334 (Ha4), Sion College (Si), Naples XIII.B.29 (Np), Royal 17 D.XV (Ry1), Laud 739 (Ld2), Additional 24178 (Ad2), Hatton Donat (Ht), McCormick (Mc), Rawlinson Poetry 141 (Ra1), Lichfield 2 (LC), Morgan 249 (Mg), Corpus Christi 198 (Cp), Landsdowne 851 (La), and Sloan 1686 (Sl2).

[22]

I made a detailed examination of the microfilm of these four MSS: Additional 35286 (Ad3), Harley 7335 (Ha5), Rawlinson Poetry 223 (Ra3), and Trinity College Cambridge R.3.3. (Tc1). Except for one set of two line transpositions in the passage E456-460, none of these manuscripts has any of the other omissions or transpositions that HM140 has. All have the Prologue, and three, Ad3, Ra3 and Tc1, have the "Lordynges" and "Wife of Bath" stanzas, as well as the envoy. It is not possible to ascertain whether or not Ha5 originally had these stanzas because the text breaks off at E1105, probably because of lost folios.

[23]

Naples XIII.B.29 is a special case, because its missing lines may very well be the result of folios lost after the transcription of the tale. The text of the tale starts at E92 and is missing lines E1114 to 1194, the section where the "Lordynges" and "Wife of Bath" stanzas are located. Since Np contains an "independent" version of the Clerk's Tale, the prologue may have been eliminated to begin with. Either the "Wife of Bath" or the "Lordynges" stanzas may not have been included in the original transcription of the MS since the 80 lines that are missing add up to about 10 more than the average number of 35 lines per folio in the MS. There are, however, folios of 41 lines in Np so it is entirely possible that they were originally there.

[24]

To quote Bakhtin: "Form and content in discourse are one, once we understand that verbal discourse is a social phenomenon—social throughout its entire range and in each and every of its factors, from the sound image to the furthest reaches of abstract meaning" (259). For a synoptic overview of his theory of dialogic discourse see my article, "The Dialogic Other: Bakhtin's Theory of Rhetoric."

[25]

See Gianni Vattimo for discussions of the relation of a work of art and "truth" in the chapter, "Ornament/Monument" (79-89), and of our relation to the classic texts of the Western tradition in the chapter entitled "Hermeneutics and Anthropology" (145-163, especially 161).