University of Virginia Library

SIR DEGREVANT IN THE "FINDERN ANTHOLOGY"
(CAMBRIDGE, UNIVERSITY LIBRARY MS FF.1.6)

by Michael Johnston *

Cambridge, University Library MS Ff.1.6, known as the Findern Anthology (which I will hereafter call it), has long been recognized as one of the most interesting and noteworthy surviving manuscripts of Middle English literature.1 It contains extended excerpts from John Gower's Confessio amantis, many of the shorter poems of Chaucer, and numerous texts by Hoccleve, Lydgate, Roos, and Clanvowe—in short, a veritable anthology of England's greatest late medieval writers. But this manuscript is also worthy of note for its large number of uniquely surviving lyrics, many of which may have been composed by aristocratic women of Derbyshire and Staffordshire. Some scholars even suggest that women acted as scribes in the manuscript—hence, my gender-neutral he/she pronoun choice when referring to the scribes in what follows.2 This codex also presents us with

quite reliable evidence about its late medieval readership, for a number of identifiable individuals from southwestern Derbyshire and southeastern Staffordshire signed the volume, indicating that it was being rather widely handled—and, one assumes, read—by the gentry of this region in the late fifteenth and early sixteenth century.3 Such reliable provenance evidence is uncommon for compilations containing Middle English literature.

Yet in spite of this manuscript's manifestly interesting and valuable content, and the great deal of attention already paid to its codicological make-up, we still know remarkably little about how it was produced and what relationship the texts bear to one another. Was it produced in the household of the Findern family, or of another local family, who subsequently gifted it to the Finderns? Was it copied out by these gentry landowners and their associates, or did the original owners pay scribes already in their employ to copy out these texts? Alternatively, might they have sent off to London to have these texts copied, piecemeal, over a period of time, or hired out a local scriptor to accomplish this? And what of the literary texts? What does the manuscript evidence tell us about what role Chaucer, central to our conception of the Middle English canon, played in the cultural imaginary of these gentry readers? And what role did the only romance in this compilation, Sir Degrevant, play? Most modern critics, after all, lump romances in this period into the broad category of "popular literature,"4 setting them against more courtly productions, like Chaucer's and Gower's, so to see such texts cheek-to-jowl in this manuscript raises questions about these gentry readers and their relation to the cultural pulse of their time.

In the mid-fifteenth century, when this manuscript was created, it most likely belonged to the Findern family of Findern, several of whose members signed it. We cannot, however, connect it to them with absolute certainty until the mid-sixteenth century, when two separate notes from this period were inscribed on

blank leaves, the first recording a "A rekenyng be-twne John Wylsun & mester fynderne" (fol. 59v) and the second a list of "the parcellys off clothys at fyndyrn" (fol. 70r). Although in this essay I apply Ockham's razor to the production of this manuscript by calling it the Findern Anthology, were evidence to emerge that the Finderns only acquired it later and that its production should be attributed to a different neighboring family, the analysis to follow would not change substantially. As Kate Harris remarks, referring to the other families whose names appear in this manuscript, it may be just as accurate "to call the manuscript the 'Cotton,' 'Frauncis,' or 'Shirley Anthology,' as it is to refer to the volume as the 'Findern Anthology.' However, such detailed consideration places in dispute neither the social 'milieu' nor the geographical area in which the manuscript was produced: the opposite is the case—it apparently confirms the origins of Ff.1.6 in a country house just to the south of Derby."5

Most questions about the manuscript's production and early provenance remain unanswered, previous attempts to provide answers being markedly speculative. Rossell Hope Robbins, for example, suggests that the manuscript arose "through the cooperative efforts of itinerant professional scribes and educated women living in the neighborhood" of the Finderns.6 The Finderns, if they were indeed the original owners, had numerous connections with London, as Maureen Jurkowski's exhaustive studies of the family demonstrate, lending credence to Robbins's suggestion that they could have hired itinerant scribes from the capital.7 Ralph Hanna says of the Findern Anthology, "Often considered an example of the spread of London poetic taste into the hinterlands, it was copied in south Derby, almost certainly in a gentry household during the third quarter of the century."8 In my analysis of the manuscript, I will confirm Hanna's suggestion—that this manuscript was almost certainly produced in Derbyshire in and for a gentry household. The main piece of evidence I will draw upon is my conclusion that the manuscript comprises mutually dependent booklets; that is, none of the codicological units—with a few noteworthy exceptions, as I discuss below—was produced independently of the others. Analyzing the scribal stints and the division of paper stocks across the manuscript reveals that nearly every booklet was made in conjunction with every other booklet, suggesting that most all of it was produced in one place. Given the numerous signatures of these gentry readers, and given the linguistic evidence I have gathered, that place was most certainly a local household. There is one significant exception to the mutually dependent production of the booklets, however: the copy of the romance Sir Degrevant. As I will argue, this text was likely produced separately from, and earlier than, the

rest of the volume, which was built up around it. Thus, Sir Degrevant stands as the core of this Middle English miscellany, with the more "courtly" literary texts and the uniquely surviving lyrics being added around this core.

The Booklets Of The Findern Anthology

In any analysis of the composition of this manuscript, the question of booklets is of the first order.9 Kate Harris, whose study of this manuscript's make-up is impeccable, argued that it does not break down into individual booklets but rather was planned as a single codicological unit. However, Ralph Hanna subsequently demonstrated, using Harris's evidence, that indeed the Findern Anthology does comprise a series of independent fascicles, produced as separate units and only later brought together into a single volume.10 The appearance of the same scribe and same paper stock across the various sections, Hanna shows, is not evidence against the case for fascicularity, as Harris contended, but rather in favor of it. Hanna goes on to identify where individual texts correspond with the end of a quire, a likely sign of the completion of a booklet.

The booklets of the Findern Anthology divide up as follows. In this division, I follow Hanna's earlier assessment, with the exception of his Booklet V, containing Sir Degrevant, whose concluding quires were produced separately, as I argue below. Hanna offers the following division into booklets:11

• Booklet I: quires A-B-C, fols. 3–20, 29–44
• Booklet II:12 quire bb, fols. 21–28
• Booklet III:13 quire E-D, fols. 45–63, 68–76
• Booklet IV:14 quire e, fols. 64–67
• Booklet V: quires F-G-H, fols. 81-115
• Booklet VI: quires I-K-L-M, fols. 116–142
• Booklet VII: quire N, fols. 143 -164

• Booklet VIII: quire O, fols. 166–179
• Booklet IX: quire P, fols. 181–188

Since the ending of Sir Degrevant in Booklet V was, I will argue, produced separately, I hereafter modify Hanna's division by subdividing this booklet into Va (the part that was produced along the same lines as the rest of the manuscript) and Vb (produced separately).

My own analysis reveals that each of these booklets, save II, IV, Vb and VII, is mutually dependent, with paper stocks, scribes, or exemplars connecting them to the other booklets. Such a state of mutually dependent booklets strongly suggests that all these parts of the manuscript—that is, almost the entirety of the Findern Anthology—was made by a team working together. By the term team, I do not mean to suggest that the production of the manuscript was an organized affair, as all the evidence points to quite ad hoc methods. Rather, the appearance of the same scribes and materials in so many of these booklets indicates that they were produced near one another within a relatively small window of time. The logical conclusion to follow from this is that this team was local, somehow connected to the region of these gentry readers, either as scribes already in their employ or available for hire from a nearby town (perhaps Derby?), a conclusion affirmed by my linguistic analysis, discussed below.

In assessing the relationship between the component parts of this manuscript, we must start with the watermarks and scribal stints. If a watermark or stint appears in multiple booklets, then we can reasonably assume that the production of those two booklets is connected—likely made in the same locale in a time not too far from one another (excluding a scribe who added "filler" to the end of a booklet, which could always be done after the completion of the main scribal stints). As I will show, all the booklets—save four—are mutually dependent. Two of these four outliers (viz. Booklets II and IV) may have been produced commercially, purchased and placed within the compilation, as I discuss below. But in the case of Sir Degrevant, as I will show, its booklet should be considered the centerpiece of this codex.

One of the watermarks, by itself, unites much of this manuscript: Mark 4, that is, can be found in Booklets III, VI, VIII and IX.15 To this we can add the evidence of common scribal stints: in particular Scribe 6 appears in Booklets III and VI, which both contain the common watermark 4, as well as Booklet I, thereby bringing Booklet I into the orbit of sections of this manuscript that must have been produced together.16 Scribe 6 writes in a very erect Secretary script, dotting his/her –i with a long stroke that runs upward at a 45-degree angle, putting a cross through his/her double –l, and employing a large otiose stroke on final –n. Moreover, he/she alternates between an Anglicana and Secretary –w, with the Anglicana form of this graph particularly noteworthy because of the scribe's unusual practice of separating the two loops of its body. In sum, then, my palaeo-

graphical analysis confirms the identity of Harris's scribe 6 across three distinct sections of this manuscript. The use of a common exemplar for the selections from Gower's Confessio amantis further supports the connections between these booklets, and also draws Booklet Va into this grouping of mutually dependent parts: as Harris elsewhere shows, all of the selections from Gower's Confessio in this manuscript derive from a common exemplar.17 In this case, the Gower bits appear in Booklets I and III (already connected via watermarks) and Va.

The appearance of Mark 4, Scribe 6, and a common Gower exemplar in much of this manuscript leaves only Booklets II, IV, Vb and VII unaccounted for. Booklet VII was certainly produced locally and should thus be thought of as similar to the mutually dependent booklets discussed above. It contains a single composite quire, one Hanna aptly labels a "monster."18 This quire is definitely an anomaly within this compilation, for its opening and closing leaves are left blank and it is filled entirely with lyrics, most of which survive uniquely in this volume. This quire provided the main place for these gentry readers to copy out their own lyrics, placing them in a self-contained quire that could easily be passed around to allow for various individuals to try their hand at poetasting. Such a condition would explain the unusual circumstance of the opening and closing leaves being blank. The composite nature of this quire, with individual bifolia being added accretively, would underwrite the informal and ad-hoc nature of amateur lyric composition, for surely the addition of these lyrics did not require any great planning but was rather done on the spur of the moment. The Finderns, then, reserved this quire for their less formal writing. This leaves us only with Booklets II, IV and Vb as potentially produced separately from the rest of the manuscript.

In fact, I think it quite likely that all three of these sections were produced separately from the remainder of this codex. Booklets II and IV are, as Harris shows and as Hanna confirms, later additions nested inside of quires B and E, respectively. Booklet II, was, I would argue, likely produced commercially, purchased by the Finderns and added to already-constructed quires within their growing compilation of Middle English verse. One clue to its commercial production comes in the quality of script and the attention paid to its mise-en-page, something noticeably lacking in the rest of this manuscript, one which Ethel Seaton rightly describes as a "very unprofessional manuscript."19 This quire contains Clanvowe's Book of Cupid, copied in a quite competent Secretary script, with even spacing between stanzas, which lends the page a clean and pleasing appearance, as if the scribe put a good deal of planning into the execution of the text.20 The second independent booklet is IV, which contains the story of Piramus

and Thisbe from Chaucer's Legend of Good Women. Like the scribe of Booklet II, this scribe has left a space for a two-line initial, something one would expect of a booklet produced by a commercial scribe, leaving the owner to have the initial executed—should he/she desire. The most important piece of evidence pointing to separate production for these two booklets is that neither their paper stock nor their scribes appear elsewhere in the Findern Anthology. Given all this evidence, it is quite likely that these were produced by scriptores, but whether in London or some other book-producing city I have not been able to determine.21 There is precedence for the commercial production of independent booklets containing the works of important Middle English authors (as these two booklets contain): Julia Boffey and John J. Thompson survey numerous such "mini-anthologies" of Middle English authors circulating in the fifteenth century.22 Such an origin also seems quite likely for these "two extraneous units,"23 leaving only Booklet Vb.

Sir Degrevant in the Findern Anthology

Booklets Va and Vb contain the following texts:

Booklet Va (quire F, fols. 81 –88; quire G, fols. 89–99):

• Gower, Confessio amantis, Book IV, ll. 2746–2926; fols. 81r–84r
• Gower, Confessio amantis, Book VIII, ll. 271–846; fols. 84v–95r
  → fol. 95v blank
• Sir Degrevant, ll. 1 –564; fols. 96r–99v
  → one leaf cancelled after fol. 99

Booklet Vb (quire H, fols. 100 -115):

• Sir Degrevant, ll. 565–end; fols. 100r–109v
• Chronicle of English kings and saints; fols. 110r–113r
• Heraldic notes; fol. 113r–v
  → fols. 114—115 missing, likely cancelled

Booklet Vb, consisting only of Quire H, was, I will argue here, produced separately from the rest of the Findern Anthology and was likely acquired early in the process of building up the compilation. Hanna believes that quires F, G, and H form a single booklet, which he labels Booklet V: "F–G have in common stock 7 (F also includes a unique sheet of stock 6). H contains the separate

unique stock 8, but given the shift of scribal stints at the boundary of quires F and G noted above, this variation simply reflects the paper available to a scribe who may not even have copied in physical proximity to other producers of the manuscript. But the common and unique stocks suggest a separate fascicle."24 There is good reason for Hanna's conclusion since the manuscript's copy of Sir Degrevant straddles the divide between quires G and H, usually a sure sign that two quires are part of a common codicological unit. Hanna here recognizes that F–G and H may not have been produced together, though he still denominates them a single booklet: the cancelled leaf at the end of G, he writes, "reflects, not a fascicle boundary, but the effort at regularizing the mid-text border between two scribal stints."25 In this case, however, quires F–G, containing the beginning of Sir Degrevant, and quire H, containing its conclusion, should be considered separate booklets—with respect to their production. That is, H was likely produced earlier and F–G later. Of course, since H was bound up with F–G, the three quires form a single booklet, as refers to their usage by the Finderns and their associates. In this case, establishing facts about the production of these quires is essential to dating the manuscript as a whole, for quire H contains the only securely internal evidence for its copying date. It would appear that this romance was originally copied as a single text spanning two quires but that the opening quire was subsequently lost and a later scribe added the beginning, possibly from a different exemplar, to what is now quire G. In short, to borrow the terminology developed by Erik Kwakkel, F–G and H represent distinct production units but the same usage unit, and thus I label F-G as booklet Va and H as booklet Vb.26

Without question, in its present state in the manuscript, there are two scribes responsible for Sir Degrevant, each contributing to independent quires. Scribe 21 copied the opening 564 lines of the poem, with the end of his/her stint corresponding with the end of quire G, and what I believe to be the end of Booklet Va, followed by a cancelled leaf. Scribe 22 copied the concluding lines of the poem.27 There are two cancelled leaves at the conclusion of Quire H, which also coincides with the end of a text—a sure sign that this marks the end of a booklet. But there is plenty of evidence that the two Degrevant sections were not produced together but were only subsequently joined, thus working against the idea that they were initially produced as a single booklet. To begin with, quire H is on a paper stock that appears nowhere else in the manuscript.28 The scribes in quire H are also not attested elsewhere in the manuscript, further evidence for the unique production circumstances of this particular quire.29 The second Degrevant-scribe is clearly a more experienced textual copyist than the first scribe: he/she writes

in a relatively neat Anglicana media hand. It is noteworthy that he/she is the only scribe in the whole manuscript to employ an Anglicana script, which again points to separate production circumstances from the rest of the volume—separate also from the opening lines of Sir Degrevant. His/her hand is marked by a notable effort to include "feet" at the base of each minim, though it does grow progressively sloppier towards the end of the stint. Further compounding this sense of different circumstances of production for Sir Degrevant is the fact that the quire containing the second stint is one of the few in the manuscript that has evidence of pricking and ruling. The scribe here uses a regular number of lines/page, and he/she has added decorative strokes to the letters on the top line of many leaves, all features lacking in almost all the other texts in this manuscript.

The stint of the first Degrevant-scribe is remarkably less well executed, and looks more like the amateur efforts of the other scribes in the Findern Anthology. He/she writes in a very sloppy Secretary hand, correcting the text in a haphazard fashion and exhibiting great variation in the number of lines/column.30 He/she also frequently pushes too hard on the pen, resulting in ink bleed throughout the copying stint. Moreover, this scribe—unlike the second Degrevant-scribe—has not pricked or ruled his/her page, resulting in quite uneven margins. Both Degrevant scribes copied their texts in double-columns, a standard format for Middle English verse romances but one found nowhere else in this manuscript. However, since the second scribe ruled the page, his/her columns appear neat and tidy, as I noted. The first scribe, by contrast, did not rule the page, yet still wrote in double columns. Since, as I argue below, I believe the second scribe actually copied his/her section earlier than the first, it is a possibility that the first scribe (working later) was attempting to reproduce the mise-en-page of the second scribe, though without taking the time to rule the page, and thus the final product is much less pleasing. Moreover, it appears that, as both Hanna and John Thompson note, the scribe of the first Degrevant stint begins to cram more material onto the page as he/she gets closer to the completion of copying, indicating that he/she was trying to make the work fit into the end of a quire so that it would meet up with an already-completed section of text.31 The conclusion I draw from this fact is that the second scribal portion of Sir Degrevant was already completed and the first scribe was copying out the opening lines and thus had to make the ending of his/her section match up with the beginning of the second scribe's part. Moreover, the fact that a leaf was cancelled at the end of G further supports my suggestion that the first Degrevant-scribe was trying to make his/her portion of text match up with an already-completed section.

The textual evidence from quires F and G also contributes to the impression that these two quires were produced after quire H and were made to fit with what was already in that quire. In particular, the textual selection from Gower's Confessio that immediately precedes the opening of Sir Degrevant does not con-

tain a coherent narrative, but rather stops abruptly, unlike other selections from Gower in this manuscript. Every other Gower text in the Findern Anthology exhibits some coherence, standing alone as a text with clear beginning, middle and end.32 But the selection in quires F and G ends quite abruptly, with no clear resolution to the tale. This particular selection, spanning fols. 84v–95r, covers lines 271–846 of Book VIII—"The Tale of Apollonius of Tyre"—commencing where the tale proper begins, omitting Genius's prolix disquisition on incest in Old Testament times and going straight into the story of Apollonius. The beginning of the selection, that is, was well chosen and makes narrative sense. The selection concludes, however, in a rather odd spot in the narrative, right where the Princess of Pentapolim has fallen in love with Apollonius: "Touchinge þis man of Tire / Hir hirt is hot as eny fire" (fol. 95r). There is no obvious reason to wrap up the selection here, for the love between these two characters has only just begun. All the other selections leave off at an obvious stopping point, leading to the conclusion that the scribes who copied "The Tale of Apollonius" did not adequately plan for how much space they would need. Once they neared the middle of quire G, these scribes likely stopped copying in anticipation of the space needed for the opening 564 lines of Sir Degrevant, which is copied into the final four folios of this quire. Or, quite possibly, the first Degrevant-scribe had already begun copying this romance into the end of quire G, and the scribes of Gower's text realized they needed to truncate their text. Whatever their motivation, these scribes then left fol. 95v—the folio between the end of "Apollonius" and the beginning of Sir Degrevant—blank, likely when they realized that there was no way to reach a suitable stopping point in the narrative, giving up and leaving off copying with no real conclusion to the text.33

The dialectal evidence likewise contributes to the impression of difference between these two scribal stints.34 Scribe 2 can be localized rather confidently

to the area of southern Derbyshire, southeastern Staffordshire, northwestern Leicestershire, or north-central Warwickshire (i.e., an area within twenty miles or so of the Finderns and their associates). The following forms help us to localize him/her such:

• They: thei (they) (Dot Map 30)35
• Each: ylke (Dot Map 84)
• Much: mychel (michel, muchel) (Dot Map 109)
• Might: my3th (mote, mete) (Dot Map 330)
• Did (sg/pl): dude (Dot Map 400)
• Fire: fuyre (Dot Map 412)36

Scribe 1, by contrast, has seemingly concocted a thoroughgoing Mischsprache. He/she attests a wide diversity of linguistic forms from several mutually exclusive linguistic regions. The most prominent of these isoglosses include:

• Her: hur. This form is attested throughout the West, Southwest, and Southeast (Dot Map 23).
• Them: hom (hem, tham): Hom, his predominant form, is attested most strongly in the Northwest Midlands, not far from Derbyshire, and is only attested in scattered form elsewhere (Dot Map 48). His secondary forms present somewhat contradictory evidence: tham, for example, is a northern feature (Dot Map 41), whereas the – h forms of this pronoun, as used by this scribe in hem, are not found in the North (Dot Map 40).
• Such: swych (suche, seche): Swych is attested everywhere but the Central Midlands and the Northwest; i.e., not in Derbyshire (Dot Map 74).
• Each: ych (eche, ilk): Ych is a Midlands and East Anglian form (Dot Map 87); eche is found in East Anglia, the Southeast and Southwest, but rarely in the

  ---

  82

  
  
  
  area of Derbyshire (Dot Map 86); ilk is a Northeast and Northern form (Dot Map 84).
• From: fram. This form is only attested in the South and does not extend as far north as Derbyshire (Dot Map 174).
• If: yeff. This form is only attested in the South, and does not extend as far north as Derbyshire (Dot Map 209).
• Hundred: hondred, houndred. This latter form, houndred, is only attested in the South (Dot Map 450).
• –ly suffix: –ly (–lech, –lich): –lech is attested in the Midlands, Southeast and Southwest (Dot Map 604).

Given that this scribe employs a series of linguistic forms from dialects that are mutually exclusive of one another, and given that he/she uses such a wide variety of forms, it is impossible to determine this scribe's linguistic origin.37 He/she may well have come from the region of the Finderns and their associates, as I found with Scribe 2, but until one sorts out which forms are scribal and which are from his/her exemplar—something I have not been able to do—any further determination seems impossible.

Whether or not these two scribes were from the same place, it is clear that they produced very different texts. So how did the manuscript come to have a bifurcated copy of Sir Degrevant? The main clue to answering this question comes in the water damage found throughout Quire H, which contains the second Degrevant stint. There is heavy damage throughout the center of each leaf, damage seen nowhere else in this manuscript.38 This state suggests the possibility that a now-lost quire originally contained the opening of Sir Degrevant and that damage subsequently struck both this and quire H, destroying the now-lost quire and leaving H with stains. The current opening 564 lines of Sir Degrevant were then "retro-fitted" to the conclusion of the poem—made by a different scribe at a different point in time. Given that the second Degrevant-scribe copies about 35 lines/column, a quire of four would have perfectly accommodated the opening 564 lines. Such a scenario, in which what is now the second Degrevant section was actually copied before what is now the first section and originally contained the romance in its entirety, can account for why this second section is remarkably more polished in appearance.

It is important to note that the paper stock and scribes of quire H are unique to this quire, appearing nowhere else in the manuscript. This further supports the notion that this bit of Sir Degrevant was produced under circumstances separate from the rest of the manuscript. By contrast, the opening stint of Sir Degrevant is in a quire whose production can be tied to the rest of the Findern Anthology, as I outlined above. Although the scribes of Booklet Va (i.e., quires F–G) do not

appear elsewhere, the appearance of selections from Gower's Confessio appearing here tie this quire to the other quires with Gower texts, as they were likely taken from the same exemplar.39

The Date of Sir Degrevant's Copying

As it happens, H is the only securely datable quire in the whole of the Findern Anthology. At the end of this quire, after the completion of Sir Degrevant, a different scribe added a chronicle of the kings of England. On fol. 112v the chronicle lists the length of each English king's reign, but the space beside "herry vj" was left blank, suggesting, as L. F. Casson notes, that "he was therefore still reigning" when this was copied out.40 On fol. 113r, however, the chronicler provides a more precisely datable entry, something scholars of this manuscript hitherto have missed: "And fro þe incarnacion of Ihesu crist til þe xx [blank] of kyng herry vj ml iiijc xlvj 3er." The reference to "ml iiijc xlvj" years from the Incarnation clearly indicates that this text was copied out in 1446, assuming that this scribe was not slavishly copying an earlier chronicle entry at a later date. The first part of the entry, reading "þe xx [blank] of kyng herry," indicates that the scribe was unsure of the regnal year (which is, after all, more difficult to determine than the calendrical year), knowing only that Henry VI had been reigning for twenty-some-odd years. The most plausible explanation for these facts is that the blank space was left for the final Roman numerals of the regnal year (which, in 1446, would have been 24 or 25 Hen VI), but was never actually filled in. The scribe copied this chronicle at the end of the quire containing part two of Sir Degrevant; hence, the date of 1446 provides a terminus ad quem for this manuscript's copy of part two of this romance.

As I argued above, this quire likely was produced along with another quire containing the opening, now lost. Thus, 1446 would mark the terminus for the original production of the entirety of Sir Degrevant, with the replacement opening quire being added at some indeterminate later point in time. Most analyses of either the script or the watermarks employed in the manuscript as a whole suggest that it was produced in the second half, or even last quarter, of the fifteenth century.41 My own analysis of the watermarks confirms Harris's suggestions, which are the most thorough of previous studies of the manuscript, on almost all points. Of course, dating a manuscript via watermark identification is a tricky endeavor, but there are enough different marks in this manuscript to control for the pitfalls of relying on Briquet too heavily. In sum, almost all of the paper dates roughly

from 1450–70, which accords with the earliest end of most attempts to date the manuscript on palaeographical grounds.42 Given the evidence of dating found in quire H, and given the fact that most of the rest of the manuscript seems to date from later in the fifteenth century, the Finderns' acquisition of Sir Degrevant most likely marked the opening of their compilation, with additional quires subsequently added around this romance. Richard Beadle and A. E. B. Owen hint at this when they note that "if [Sir Degrevant] is not the oldest text in the collection it is certainly the most archaic in diction and metre."43

So if the original copy of the entirety of Sir Degrevant arose differently from the majority of the Findern Anthology, where might it have been produced? I suggested above that the mutually dependent booklets in this manuscript were most likely produced by a team working within the region of the Finderns and their associates. My linguistic analysis of the second Degrevant-scribe reveals that he/she hailed from quite near the Finderns' home, which suggests that this romance was similarly a local production. Such an origin for these booklets accords well with Hanna's sense that the entire manuscript "seems typical of that informal creation of literary artifacts which one associates with the aristocratic menage, not the stationer."44 The "aristocratic menage," in fact, seems responsible for many of the surviving copies of Middle English romance from this period. It was, after all, a period in which anyone with access to paper, ink, and exemplars, and who was able to read and write, could produce his/her own books. Or, if one were disinclined to undertake the copying oneself, trained documentary producers and unbeneficed clerics were always at hand, often probably underemployed. The skills of reading and writing, of course, were limited, but the English mercantile classes and the landowning gentry—families like the Finderns—were the very sorts of people who would possess such skills. In the century before Caxton set up his press in Westminster, numerous such aristocratic menages gave rise to manuscripts of romance, as well as other sorts of vernacular literary texts. The Finderns and their associates were just a part of what Curt F. Bühler calls the "every man his own scribe" movement of the fifteenth century.45 The manuscript they left us is remarkable, for it shows us the wide range of gentry readers who handled this manuscript. But it also shows us, as I hope I have now shown, the central place that Middle English romance played in their literary and cultural world.