Description of the Manuscript

Date

S. xv1. George Kane and E. Talbot Donaldson assign this manuscript to the first quarter of the fifteenth century, a dating with which the late Dr. Jeremy Griffiths concurred. Dr. Griffiths, who was kind enough to supply us with his draft description of the manuscript in the summer of 1994, was of the opinion that it is "probably earlier rather than later even within this dating" (Private communication, 4 August 1994).

Physical Description

The manuscript is made up of ninety-three vellum leaves bound within four paper end leaves with a watermark of pot design. Cf. Briquet, No. 12691, dated 1580. The vellum is of uneven quality, and a number of folios (e.g., fols. 22 and 29 especially) are difficult to read because of bleed through from the opposite side.N Many holes in the manuscript were patched by Ms. Linda Lee in the 1989 restoration of the manuscript "using fine parchment and Goldbeater's skin." A number of small holes remain, but no text has been lost because they were present when the text was copied. At some point after the manuscript was given to Corpus Christi College, it was cropped but with no loss of the poetic text, though some marginalia are truncated.N

Collation

iii (end leaves)

i8+2 (+3, +8), ff. 1-10; F1.1-F4.177

ii8+2 (+2, +9), ff. 11-20; F4.178-F5.461

iii8+2 (+4, +7), ff. 21-30; F5.462-F6.89

iv8+2 (+3, +8), ff. 31-40; F6.90-F8.46

v8+2 (+4, +7), ff. 41-50; F8.47-F10.55

vi6+2 (+5, +7), ff. 51-58; F10.56-F10.675

vii8+2 (+3, +8), ff. 59-68; F10.676-F12.76

viii6+2 (+3, +6), ff. 69-76; F12.77-F14.86

ix8+2 (+4, +7), ff. 77-86; F14.87-F15.388

x10+1 (+7, -8, -9, -10, -11), ff. 87-93; F15.389-F16.383

i (end leaf).

Several facts suggest that the manuscript was copied in at least two different stints. The colored parasigns generally alternate between red and green in the first six quires. The opening with the last leaf of quire six and the first of seven has red and green parasigns on fol. 58v and red and blue on fol. 59r, and this combination of red and blue parasigns continues. Fol. 58v has just thirty-one lines instead of the more customary thirty-nine to forty-one. The scribe wrote this side in an unusually large and expansive style, even leaving a line empty, evidently padding to fill out the page, even though he was in the middle of a passus. The first leaf of the seventh quire has forty-three lines, as though the scribe needed to make up for space lost. It therefore seems likely that the scribe temporarily lacked an exemplar for completing passus 10. Moreover, a few changes in spelling conventions occur, suggesting that some time elapsed between the completion of quire six and the resumption of copying. Most notably, words beginning with /š/ are spelled <sh> from 1.1 through 10.675, and afterward tend to be spelled <sch>. There is also a pronounced shift in the scribe's preferred abbreviation for quod from using a <q> abbreviation over 75% of the time before folio 59 and a <qd> abbreviation 90% of the time after that point. At the same juncture, he also abruptly stops drawing exaggerated ascenders on the top line of each page. Before folio 59 they appear touched with red ink on over half of the pages. Finally, the Roman numerals in the incipits and explicits to passus are spelled out before folio 58v and expressed thereafter only as numerals.

There are no extant quire or leaf signatures and no catchwords. Foliation in modern pencil, 1-93, is followed in this edition.

Leaf Size and Arrangement of the Page

Size: 295 x 175 mm. One column, with pricking for column ruling with line and column ruling in silverpoint. The pricked frame area is c. 240-250 mm x 110-112 mm, with the text copied generally below the top line of ruling.

The average number of lines per page is about thirty-nine. Up to the end of the third quire as many as forty-six lines may appear. After the end of the fourth quire the average is closer to thirty-seven or thirty-eight lines per leaf, and in the last quires the hand is noticeably larger and the interlinear space greater. Seventy-eight sides appear with thirty-nine to forty-one lines. Another forty sides have over forty-one lines, while of the sixty-six sides containing fewer than thirty-nine lines, sixty appear after the fourth quire.

Script

The Middle English text is copied throughout by one hand in a current anglicana script in dark brown ink, with the Latin text in fere-textura, usually in red ink. The textura often shares features with the scribe's anglicana. Frequently, Latin passages open with an unflourished Lombard capital. Occasional marginal glosses, notae, and pointing hands from two or three different scribes appear, as well as some underlining added by a late fifteenth- or early sixteenth-century reader. The scribe wrote the Middle English text first, leaving a space for the Latin which he later added in red ink. He then highlighted important words and the beginning graph of each line, and the red paraph signs. Finally, he added the Lombard capitals and paraph signs in blue or green. On fols. 46v-47r the scribe neglected to enter the Lombard capitals, and there are several openings on which he failed to supply the colored paraph markers. Someone using a different ink and style (perhaps the original scribe's informal hand) is responsible for a handful of interlinear corrections; e.g., 4.388, 4.427, 9.325, etc. The same hand at 15.170 (fol. 84r) supplied the correct Latin tag pax vobis in the right margin. Ultra-violet light shows original þese wordis had been erased, but the scribe failed both there and at 14.15 (fol. 76r) to return with red ink to supply the missing text. The Latin tag seems simply to have been overlooked and never written on fol. 34v. At 14.115-116 (fol. 77r), the scribe has made the correction and left the corrector's note in the right margin.

Punctuation

The scribe uses four marks of punctuation—the paraph, the punctus, the punctus elevatus, and the virgule (or solidus). The caesura is steadily marked with a virgule or (less frequently) with a punctus elevatus. Occasionally the same marks appear within the half-line to indicate phrasal junctures. Beginning at the top of fol. 2r nearly every line closes with a punctus, and there is reason to think that the scribe intended so to end every line, though some dozens are omitted. Paraph markers appear in the left margin to mark both verse paragraphs and changes of speaker in dialogues. It is also possible, as David Benson and Lynne Blanchfield have argued, that they serve as the equivalent of marginal notae.N These paraphs are written in alternating red and green to the end of quire vi, after which red and blue tend to alternate except for the openings at fols. 67v-68r and 84v-85r. On a few openings, the scribe neglected to supply blue or green paraph markers, though marginal virgules mark the places where they would have appeared.N After fol. 1r-v the scribe steadily put a punctus at the end of virtually every line of verse, occasionally adding a missing one in red ink as he was supplying rubrication in an opening. Now and again a punctus appears at syntactic junctures within the half line, though rarely and without apparent pattern. In a few instances the scribe supplied both virgule and punctus to mark the caesura, but we see no pattern for that activity. He steadily supplied a virgule (less often a punctus elevatus) at the medial half-line boundary, but that he intended neither the virgule nor punctus elevatus exclusively to mark the metrical caesura is shown by their infrequent appearance at phrasal junctures.

Decoration

With the exception of the opening of folios 2v and 3r, the first letter of each line is regularly highlighted in red, and the first line of most pages before folio 59 is ornamented with exaggerated ascenders touched in red ink. Within the line, some proper nouns are written, and a few are underlined, in red ink. Important words are highlighted in red, frequently only on the first letter but from time to time with all or most letters highlighted. Passus divisions and some major intra-passus divisions are marked with ornamented capitals, pen-flourished in various combinations of red, blue, and green ink with sprays. Passus divisions exhibit more extensive marginal flourishes and sprays than those for intra-passus divisions. Ornamental capitals for intra-passus divisions appear at 4.158, 4.216, 4.345, 5.62, 5.188, and 5.308. At 5.135 and 5.394, the scribe left spaces for three-line ornamental capitals, but neglected to complete either. The ornamental capitals vary considerably in size as well as in the care and expertise with which they were produced.

Marie-Claire Uhart describes the decorational scheme in F as "characterized by enthusiasm rather than professionalism" and as "highly erratic, with variation in colour and extent of flourish at passus heads."N However, in comparison with its sister manuscript in the alpha family, MS Rawlinson Poetry 38 (R), which has a much more reliable text, F's success in continuing a complex scheme of decoration and textual emphasis from beginning to end is remarkable in the tradition of the poem. As Uhart notes, Piers Plowman manuscripts in general tend to display "unsteady, rather flagging decoration," but she remarks that such incomplete schemes "may signify no more than a poorly organised book trade."N

Illustrations

On fol. 1r inside a historiated initial <A>, a male figure, "yrobed in russet," seated "vpon a brood banke" with his head resting in his left hand and wearing a hat and a three-fingered (laborer's? gentleman's?) glove on his right hand, appears to represent the dreamer asleep.N There is disagreement as to whether the structure before his feet represents a stool (Scott), a tower (Weldon), or possibly even a walled city.N As Weldon notes, the green color of the "brood banke" links "the Prologue of Dream 1 to those of Dreams 2 and 3, and to F's restructured Dreams 4 and 5 and the 'grene launde' of those spurious prologues."N The <A> is inlaid with gold leaf, flourished in blue and maroon with white highlighting, and extends down the left side of the page and across the bottom with a bar border.N

At the top of fol. 11r, the beginning of the second quire, the ascenders are flourished more elaborately than usual and embellished with a crude line drawing of a male face on the left and a lion's head on the right.

A similar crude pen-and-ink drawing of two human heads and two grotesques elaborated from the ascenders appears at the top of fol. 76v.

Following the conclusion of the text of fol. 93r is a pen and ink drawing with colored wash, probably dating from the late seventeenth century, of a pelican, the emblem of Corpus Christi College, with a blank banderole around its neck and clutched in its raised right foot.

Binding

Dimensions: 305mm high x 185mm wide x 27mm deep. The manuscript is bound in a sixteenth-century limp vellum with four slip stations. It was rebound in 1989 by L. J. Lee, who noted that "There is evidence of a previous sewing with six sewing stations and endbands."N The number "10" appears in ink on the spine. "Peers the Ploughman" is written in an early modern hand on the upper cover. Traces of fantail ties remain on upper and lower covers by the fore-edge.

Provenance

The manuscript was given to the College by a fellow, William Fulman, M.A.N Nothing is known of its earlier history. Fulman donated at least one manuscript (CCC 174) to the college as early as 1676.N The college acquired his personal papers in 1707.N An inscription at the foot of fol. 1r reads "Liber C.C.C Oxon. Ex dono Gulielmi Fulman A.M. hujus Collegii quondam Socii." The current shelfmark appears in pencil inside the upper cover. The obsolete Bernard catalogue number and shelfmarks which appear on the third front flyleaf and in ink on fol. 1r, read "No. 1668. 201. D.3.10." The entry in Edward Bernard's Catalogi Librorum Manuscriptorum Angliæ et Hiberiæ in unum collecti, cum Indice Alphabetico (Oxford, 1697), p. 56, entry 1668, is the barest list, carrying next to no information about the manuscript except to show that it was in the possession of the college before 1697.

The Scribe

Dr. A. I. Doyle has recently discovered in the collection of Ushaw College, Durham, two binding fragments written by the F scribe. The fragments, Ushaw College MS 50, consist of upper and lower portions of the first leaf of the Southern Recension of the Prick of Conscience, with a new prologue of twelve lines and lines 1-55 (with two additional lines before l. 26 and two after l. 30). Dr. Doyle describes the hand as "clearly by the same hand as Corpus Oxford 201, and the initials possibly also the same."N It was discovered unattached and therefore cannot be connected with the binding of any book, though Dr. Doyle thinks that the rounding of the corners and the horizontal fold in the lower part of the leaf make it likely to "have been in some sort of wrapper of a small book." He describes the border ornament and cadel sketches as done in a provincial style, that is, "not the most fashionable metropolitan modes of c. 1400, about when it may belong." The historiated initial on the recto of the first fragment illustrates Christ of the Apocalypse.

Fol. 1r Upper portion of the Ushaw fragment recto

Fol. 1r Lower portion of the Ushaw fragment recto

Fol. 1v Upper portion of the Ushaw fragment versoN

Fol. 1v Lower portion of the Ushaw fragment verso

Sean Taylor has pointed to what may be a third instance of the same scribe's handwriting, a rubric entered into another B manuscript of Piers Plowman,N the Bodleian Library's MS Rawlinson Poetry 38, and Dr. Doyle agrees that the hand is "not unlike." Though this evidence may place the R manuscript in the hands of the F scribe, Taylor's claim that F is itself copied directly from R is consistent with neither the evidence of textual variation between the two witnesses nor the dialectal history of F. For a photocopy and discussion of the altered rubric in R, see Robert Adams, "Langland's Ordinatio: The Visio and the Vita Once More," The Yearbook of Langland Studies 8 (1994): 51-84, esp. Appendix I.

Text

Editorial Method

The Color Facsimile

We were fortunate in choosing F as our first text since it immediately challenged our initial assumptions about the kind of text we wanted to present by putting before us some difficulties that less complicated physical documents might have spared us. Initially, we had intended to make digital facsimiles from black and white microfilms of each Piers Plowman manuscript. Experience with digitizing the microfilm of this manuscript quickly changed our minds. This image of folio 9r demonstrates the decent quality of images obtainable by this method.N Nevertheless, the black and white images failed to convey textual information encoded in color in the original documents. Compare fol. 9r as seen above and the color image. The problem of bleed through in F proved so extreme on several leaves that they were rendered completely unreadable when digitized from black and white microfilm. Color images proved as easy to decipher as are the originals. Indeed it turned out that in at least one case in which the manuscript itself is difficult to read even with a powerful hand glass and in a good light (bottom of fol. 29v), the digital image proves to be much easier to read after fairly minimal manipulation. Since digital color images are not so expensive as reproductions for a printed text, we have decided, when libraries will permit us new color photographs, to provide color digital facsimiles of every page of every Piers Plowman manuscript with each facsimile leaf hypertextually linked to its transcription.

High quality color digital images enable access to manuscript texts in ways previously unimaginable. We have experimented with two different ways of making color digital images. For experimental purposes, Dr. Andrew Prescott in the spring of 1994 provided color images made on the British Library's then newly acquired Roche/Kontron ProgRes 3012 digital camera. Full leaves (or details) could at that time be scanned at resolutions of 2000 x 3000 pixels per inch in 24-bit color with truly astonishing fidelity to the originals, often making it possible to see clearly what cannot be seen with the naked eye. We also produced quite satisfactory digital images from 35mm slides. Using a Nikon LS-3510 Slide Scanner attached to a Macintosh Quadra 800 running Adobe PhotoShop, we experimented with scans at input resolutions varying from 635 to 3175 dpi and output resolutions between 72 and 1000 dpi. These JPEG and TIFF files varied in size from a low of 168 KB to just over 35 megabytes. It should go without saying that the largest of these files are expensive to store and slow to call to the screen and to manipulate. We found that a rational compromise between the highest and lowest grades, producing facsimiles more than adequate for most manuscript leaves, involved input resolutions of 1587 dpi, output resolutions of 500 dpi, and JPEG files of 750-1000KB. Such images may be doubled four times without pixillation, and they run very efficiently on most 486 and Pentium machines, on upper-end Macs, and on RISC stations.

The original digital images of CCC 201 were made at the Bodleian Library by Dr. David Cooper using a Dicomed 7520SE digital back on a Sinar process camera at an input resolution of 600 dpi in TIFF format. Files average between 75 and 95 megabytes per leaf. Those image files are archived in Oxford at the Bodleian Library. High quality JPEG reductions are available on the World Wide Web in the "Early Manuscripts at Oxford" archive at the following URL:

http://www.image.ox.ac.ukN

Presentation of Text: Levels of Inscription

Except in those rare instances in which the scribe is also the author, medieval manuscripts are like palimpsests. Each surviving copy represents the work of its immediate copyist. Each also reflects traces of the efforts of a usually indeterminant number of scribes whose work separates the immediate copy from the author's original text. The evidence suggesting the possible number of copyists is inferential, derived from collation of variant readings or, as is the case of texts surviving in single copies, from analysis of the language.N In this instance, we cannot know the number of hands and intelligences intervening between the poet and the immediate copyist of F, but inferential evidence permits us to distinguish five layers of inscription after the author's fair copy. Initially we intended to present texts of the last two such layers of scribal labor, the one a diplomatic transcription of the text of the final copyist (F-Scribe) and the other a lightly annotated critical text of the work of a scribal-editor (or set of scribal-editors) which we have designated as F-Redactor.N We are now convinced that F-Redactor's text cannot be distinguished from that of the immediate scribe with sufficient consistency to claim that our critical text is his. When, for example, a line or phrase is missing in F that is present in R and other B witnesses, it is usually impossible—except in clear instances of homeoteleuteon or other form of mechanical error—to determine with confidence that the omission reflects the decision of F-Redactor or an error by the immediate or an intermediate scribe in the tradition. On the other hand, since F-Redactor's revisions represent the intentions of an intelligent medieval reader/editor, it seems worthwhile to offer as close an approximation of his text as the technology permits while at the same time offering a close rendering of the immediate scribe's achievement. To that end, we now offer a lightly edited reading version of the text to accompany the literal transcription and facsimile edition of the immediate scribe's work. In spite of our inability always to distinguish in individual lections the level from which each derives, we nevertheless think that there were layers of scribal copies. We attempt to lay out immediately below what we take to have been the determinable layers of inscription in this manuscript.

1) The Authorial Text:

We know next to nothing of William Langland, the man who wrote these poems, for the life of fictional Long Will, the dreamer-narrator of the poem, cannot be assumed to correspond to the life of the poet. The linguistic evidence in the surviving witnesses tends to corroborate his identity as a Southwesterner, and many descriptive details in all three versions demonstrate that he must have lived for at least a while in London. We cannot claim more than that. Though The Piers Plowman Electronic Archive ultimately aims to determine and restore the authorial texts, we do not attempt at this level of the Archive systematically to distinguish the work of scribes from his.N

2) The B Archetype (Bx):

At the second level F reflects the work of a scribe who produced the already defective manuscript copy from which all extant B manuscripts descend. Detailed reconstruction of the Bx-scribe's work lies outside the primary concerns of this edition of F. It will then appear paradoxical that we nevertheless cite in our textual notes the readings of Bx. When we refer, as we will numerous times in this edition, to the readings of Bx, those lections represent a preliminary working hypothesis about that text, constructed from analysis of the variants provided in the Kane-Donaldson edition. Having already transcribed some thirteen B manuscripts, we have had many occasions to check the Athlone collations, and we have found them remarkably accurate. Moreover, preliminary collations by Adams and Hanna in the years before the Archive began have suggested the essential correctness of the now conventional view that there are two major manuscript families in the B tradition, alpha and beta.N Though a few cruxes remain in those lines where F's relation to Bx is at issue, in general it is a simple matter to determine the Bx reading, at least in the light of present evidence. We expect that in a number of details our working hypotheses about Bx will require correction, and it is one advantage of the electronic text that it so readily permits that kind of adjustment.

Though our reconstructions of the readings of Bx must appear logically circular, a moment's thought should suggest that some such provisional reconstruction is inevitable and need not be logically vicious. In one sense, we are beginning the editorial project ab ovo, editing each manuscript witness afresh with the goal of working inductively to construct from the corpus of variant lections the intermediate hyparchetypes, archetypes, and eventually the critical texts of the three canonic versions. A critical question at once poses itself: if we already know the readings of Bx sufficiently to cite them for understanding the relation of this manuscript to others, what is the point of laboring to transcribe documentary editions or of doing either elaborate collations or complex thinking about the relationships among the manuscripts? The fact of the matter, however, is that we are not starting at the beginning, that we come to this task in the middle of a long-standing scholarly project and after more than a century of editorial work on these poems and these manuscripts. A great deal of careful and reliable work by a variety of scholars has been done, and a number of theories about the work exist which, in the light of presently available evidence, we take to be valid. Though we must reconsider the most basic issues in the light of new evidence, we necessarily use those theories we have found persuasive until such time as we have reason to think them wrong. We begin our editorial project with fundamental hypotheses about the author, the number of versions, the relations among the manuscripts, the governing features of the poet's metrical rules, as well as a number of assumptions about what it is that editors ought to do. We are aware that all of these are contestable, some of them hotly. Like scholars in other fields, we can only attempt to revise our hypotheses in the light of the data, as fresh data becomes available. That is, the process of editing a textual archive such as this will consist of a series of provisional passes through the evidence, and we anticipate that at least some of the hypotheses we have formulated now will require revision at a later date when we have more precise and full data to bring to bear on the reconstruction of Bx. Therefore, our citations of Bx are offered as provisional, and we do not devote annotations in this edition to the still-to-be-constructed text of the archetype.N

3) The Alpha Family:

At the third level F contains the work of a scribe whom we, following Schmidt, will call "alpha."N As is the case with Bx, our ad hoc reconstructions in our notes of the text of alpha are also provisional. They are based upon our detailed comparison of F's lections with those of its sister manuscript R, a task made easier by the careful work of Kane and Donaldson in their edition of the B text.N We offer here neither systematic comparison of F with R nor a detailed reconstruction of alpha—both must come at a later stage in the construction of the Archive—but we have incorporated into our apparatus numerous textual notes calling attention to readings in the text of F that are owed to the efforts of alpha.

4) The F-Redactor:

At the fourth level F contains the work of a scribe whom we call "F-Redactor." The F-Redactor's work is conceivably the product of several scribes at different stages in the ancestry of CCC 201. However, we think it more likely that the bulk of the work was done by a single revising editor-scribe who rationalized the dream structures and converted Langland's twenty-one passus to sixteen. That is, it seems to us probable—though not certain—that the medieval "editor" who made the macro-changes was also responsible for most of the hundreds of apparently intentional, unique substantive micro-revisions in this manuscript. Many of these revisions bear no clear evidence of motive beyond the occasional hint of a need for semantic adjustment or the lure of metrical whimsy. But a significant number of whole lines, phrases, and even omissions in CCC 201 appear to reflect A-version influence and may result from the use of an A-version manuscript (probably related textually to M and the VH family) at some point to correct the scribe's primary exemplar. Such interversional correcting, if it occurred, is not unparalleled and might have been regarded by the scribe as mere prudence since he is unlikely to have shared our acute sense of the distinctive identities of Piers A, B, and C. At any rate, whether these various types of micro-revision are the work of one scribal editor or several, we attribute them here heuristically to the F-Redactor.

The most remarkable contribution of F-Redactor to the poem is his revision of the passus and dream-vision structure of the poem. Weldon is correct to claim for this scribal response to the B text that it

represents a sophisticated, intelligent, and consistent mediaeval reading of Piers Plowman B. Reader F seeks textually and visually to interpret the poem and to clarify its essential structure, for the "ordinating" passus arrangements together with the spurious lines and illumination do not add meandering digressions or innovative interpolations but heighten existing formal elements in a conventional way. While we may lament his textual distortions, we can also admire Reader F's interpretative guidance for a poem whose only verifiable authorial structure rests on passus division and dream vision sequence.[emphasis his]N

5) The F-Scribe:

Finally, the manuscript also contains the work of a person (or collection of persons) whom we call "F-Scribe." Ordinarily, such references can be taken to mean the immediate Essex scribe who copied the Corpus 201 manuscript. However, because an indeterminate number of copyists separates the Essex scribe from the editorial F-Redactor, and him, in turn, from the alpha scribe, we also use the "F-Scribe" label in a more comprehensive way to describe all the producers of unintentional textual variation that accumulated both before and after the exemplar left the hands of the F-Redactor. Unfortunately, the work of the F-Scribe (or group) is neither in logic nor in practice always separable from the F-Redactor: the latter, for all of his sophistication, presumably introduced at least a few inadvertent errors and inherited others from copies between his own and alpha, whereas the immediate scribe of our present manuscript, a fairly careful and modest copyist, may intentionally have introduced some semantically equivalent phrasal variants without appreciating their textual uniqueness. Likewise, he may have imported some conflated A-version readings under the mistaken impression that he was simply correcting. Though most commentators on the poem have projected onto the immediate scribe their sense of the eccentric high-handedness of F-Redactor—Elsie Blackman, for instance, described his work as "so bad that it is useless"—there is a good bit of evidence that he intended to copy his exemplar faithfully.N There is little evidence in the manuscript itself of on-going revision but rather its opposite.N Everything suggests a scribe careful to correct his own omissions and miswritings. His success in that effort can only be described as uneven—witness the openings in which he failed to supply rubricated Latin text or the colored paraph markers—but there is little reason to think of the immediate scribe as a textual innovator.N

At the margins of probability, then, our labels and claims for these two (or more) sources of textual input necessarily seem hypothetical, circular and non-historical: if the evidence for the F-Redactor consists of the apparently intentional editorial changes wrought in the text after the work of the alpha scribe, F-Scribe is our name for the source of apparently unintentional divergences from the work of alpha. To say this, however, and to make these two strata available in the form of separate style sheets, is not to imply that we think such a tidy distinction reflects the temporal reality of F's text in more than a broadly accurate pattern.

6) The Nachleben Scribes:

In addition to scribal copies of the text, the manuscript contains the work of a small number of marginal annotators. These readers, beginning in the fifteenth century and continuing in late fifteenth- or early sixteenth-century secretary hands, added some marginalia, mainly notae, crosses, and pointing hands. Though potentially interesting as indicators of what some readers found noteworthy, their significance is, in general, remarkably opaque. We have treated these additions to the margins of the text exclusively in our notes.

These six levels of inscription must have occurred in this sequence, but we do not consider them steadily distinguishable. A number of unique alpha lections, for instance, probably reflect authorial readings. F-Scribe may conceivably be F-Redactor working at a later date, though dialectal evidence suggests that is improbable. Certainly, it is almost inevitable that some of the erroneous lections (which we take to be unintended and thus the product of F-Scribe) will have been introduced by the F-Redactor. Kane and Donaldson argue that the scribal editor we have called F-Redactor had access to a manuscript superior to Bx, while Schmidt thinks that improbable.N Our primary access to the work of these scribes is MS CCC 201, which is the direct product of F-Scribe and the later marginal annotators, but we rely upon R for evidence about alpha and upon other B manuscripts for the B archetype.

Presentation of Text: Style Sheets

Starting from this understanding of "F," we conceive our editing project in the following ways. First, we attempt to present the scribal text of F in as unmediated a form as is practical. To that end we are providing high quality color digital images of the manuscript. At this least speculative level of our interpretation, we provide as well an electronically readable, searchable, and analyzable transcription of F, supplemented occasionally by notes which will communicate information about F-Scribe's work which is not fully captured in the images. Such an "objective" level of interpretation is likely to be useful to scholars working on a wide variety of questions. However, it is unlikely that we can anticipate all of the specific needs of these scholars. Paleographers, for instance, may well wish to add markup to our base text that would distinguish allographic variants, or linguists might wish to identify dialect forms, or metrists the patterns of stress. The electronic text permits future users to build upon textual work without fresh transcription.

Second, using both notes and SGML tagging, we attempt to interpret both these physical products and the mass of indirect evidence concerning the work that went on in the F textual tradition for which we have no direct physical evidence. In general, that consists of identifying various features of the manuscript and asking about them: "At what level of copying did this feature originate?" For the most part, this question is answered through our knowledge of the products of the other B scribes, who presumably had direct and indirect access to some of the products we would like to describe. Our analysis of the physical evidence and inferences from the textual tradition has led us to formulate the following policies:

• 1) Using SMGL markup and five different style sheets in the Multidoc Pro browser, we distinguish, where it is possible to do so, those lections which are likely to be the work of the F-Redactor from those of the F-Scribe as well as from those added by the Nachleben scribes.N The readings of the latter appear usually in the form of notes. Distinctions between unintentional scribal errors which we somewhat arbitrarily attribute to the immediate scribe and the readings we think must have been in the scribe's exemplar are made using SGML markup. Scribal errors are presented inside <SIC> tags and the correct readings inside <CORR> tags. The former are displayed in purple for the reader using the style sheets labeled "F-Scribe," "NoPalTags," and "AllTags." The "F-Critical" style sheet is designed to indicate what we believe F-Scribe intended to have written and thus to approximate the text of F-Redactor. Emendations displayed in the "F-Critical" style sheet appear in the conventional square brackets. Since the text displayed in the "F-Critical" style sheet is a reconstructed, putative text, it lacks the color features that appear in the more nearly diplomatic transcriptions of the manuscript. We have supplied line references to the Athlone B-text both for the convenience of readers and to provide a basis for later machine collation of documentary texts.

• 2) We provide textual notes to call attention to or to explain various aspects of the scribal text in relation to earlier levels of inscription, though we have made no attempt to represent every difference between the lections of this highly eccentric manuscript and those of the B archetype. A number of textual notes are intended to supply the texts of lines omitted from Bx either by alpha or F, or to identify lines uniquely attested in F or alpha. We provide a few notes interpreting difficult words and passages. Other linguistic notes provide notice of ambiguous relict forms or of the appearance of nonce words as well as of letters corrected by the original scribe. A lemmatized glossarial concordance is under construction, and we expect with future documentary editions to be able to supply full glossaries. Except in a handful of lines where they were required to make a textual point, we have not supplied interpretive historical or literary notes but will supply them at other levels of the Archive. Many notes at this level of the Archive are concerned with discussion of physical features of the manuscript or the scribal hand, which many readers may well find footling. Those readers whose interest in seeing detailed comment on otiose curls, oddities of letter forms, malformed letters, marginal notae and pointing hands, etc. is limited might well choose to avoid using the "F-Scribe" style sheet, selecting instead the style sheet labelled "NoPalTags" which displays the transcription of the manuscript but with all codicological and paleographic notes made invisible.

• 3) In matters of ordinatio, abbreviations, and punctuation we identify what we believe is absent from the manuscript in relation to F-Scribe's intentions and supply what we believe he intended to have produced either in textual notes or with <CORR> tags. This is particularly the case with planned but unwritten capital letters and paraphs as well as with omitted rubricated half-lines of Latin text. From time to time the scribe omitted a line-terminal punctus, but we have not marked them for correction in the edited text. In the case of what may well have been otiose tildes on a handful of words, we have transcribed the doubled <n> or <m>, but readers should know that most instances of <(n)n> or <(m)m>, especially in unstressed syllables, are to be treated as suspect.

• 4) Where the manuscript is physically defective, we supply, to the best of our knowledge, what had been lost. However, in the case of this manuscript, all of the holes in the parchment were originally present, and there is no loss of text.

Most problems appear in connection with item 1 above, with determining the intentions of F-Scribe. For instance, the scribe at 14.135 wrote the following words:

Withoutyn wem in-to þis world / a kaue child she broghte.

Determining scribal intention is not always so straightforward. A typical problem arises in the first lines of the text. F uniquely begins the first line "Al in somer sesoun . . . ." We may speculate that F-Scribe's motive for revision lies in his (or his supervisor's) desire to begin the text with a historiated capital letter containing a portrait of the Dreamer. An <A> provides more scope for such illustrative work than an <I>, so we may assume that F-Scribe intended to begin the poem with Al in spite of the fact that Bx begins "In a somer sesoun."N That is, we take Al to reflect scribal intention, though we cannot be sure that intention is to be identified with the immediate scribe, the F-Redactor, or some intermediate scribal copyist. Was F's omission of the indefinite article in this first half-line also intentional? Or had the loss occurred by accident in the tradition of copying between alpha and the immediate scribe? It soon becomes obvious in the first dozen lines that this scribe's treatment of articles is distinctively problematic. For instance, his lection for the second line lacks Bx's article before sheep:

I shoop me in-to shrowdes / as y [a] sheep were.

& on [a] May morwe / on malverne hillis.

I slumbrede in-to [a] slepyng / it swyȝede so merye.

I beheld in-to [þe] Est / an heyȝ to þe sunne.

We make no effort to mark all of F's changes from Bx. For instance, the unique inversion in 1.5 of Bx's Me bifel to By-fel me goes without notice, since it represents either F-Scribe's or F-Redactor's intention. The same may be said in 1.7 of F's substitution of Vpon for Bx's Vnder or similar substitution in 1.8 of wawys for watres. Again, we have not sought utter consistency in this respect, and in some dozens of lines we have called attention with textual notes to some of F's more interesting or characteristic revisions.

Frequently F reflects the activity of alpha; those lections which are unlikely to represent Bx are flagged in textual notes. At 10.805 F reads "Synne scheweth vs good semblaunt / & sory gan he wexen." In this case, erroneous scheweth is owed to alpha. Beta reads "Synne seweþ vs euere quod he and sory gan wexe." F revises patently erroneous alpha to supply an appropriate direct object for the verb. In such a case where F has taken a defective reading from alpha and then attempted to make sense of it by adding grammatically appropriate words, we use a textual note to convey this information. The F-Scribe is not responsible for the initial error, and a <SIC> tag would be inappropriate since he has obviously intended the revision.

As we have noted, distinguishing intentional revisions of the text from unintentional failures of attention is a matter of judgment, and decisions frequently are required for which inadequate evidence exists. We have, for instance, usually taken F's unique omission of an alliterating word to have sprung from unconscious error except in those instances in which we could perceive in context a reason for revision. We claim no infallibility in marking with <SIC> tags what we take to be unintended scribal substitutions in such cases as 1.66, where lawhte appears for correctly alliterating rawhte. We think that the scribe in such cases would probably have corrected the error had it been called to his attention, but certainty is not possible.

Our edited text attempts in a modest way to open new territory in textual editing. An edition of a lost text at an indeterminate remove from the extant document, but not an edition of the poem that provides its raison d'être, our texts of F-Scribe's (and to some degree, F-Redactor's) poem attempt to represent an almost contemporary reader's response to Langland's text. Ideally, present-day readers of such a text would want to know where it differs from Langland's original text, from Bx, and from alpha. The kind of inversion noted above in 1.5, where we have chosen not to annotate, or the substitution of wawys for Bx's watres ought, one might reasonably argue, to be recorded in the apparatus or at least mentioned in a note. However, consistency in this regard involves begging too many questions, requiring us to edit Bx and alpha before we have accumulated the data on which such editions must be based. What we have attempted now is less ambitious, but more consistent with the long-range goals of the Archive.

To summarize, readers may choose, by selecting the appropriate style sheet, to view one of five versions of the text. The F-Scribe style sheet offers a diplomatic transcription of the manuscripts together with a set of descriptive and explanatory notes. It presents the scribal text with all of its corruptions and incompetences so that the reader will need to click on the P or C icons for explanations or suggested interpretations of codicological or paleographic matters or on the gray dog-eared document icon for notes on matters of textual or historical interest. The F-Critical style sheet offers a reading text purging the text of the errors of the F-Scribe and his immediate predecessors, presenting the text as the F-Redactor would have wished to see it. This text is, of course, a hypothetical construct based solely on the criterion of good sense and adherence to the formal features of the form, the working hypothesis behind it being that the F-Redactor and his successors would not knowingly have written nonsense. The text has more practical than theoretical justification in that it offers a corrected (though peculiar) reading text of Piers Plowman. The NoPalTag style sheet presents exactly the same text but spares the reader some of the distraction of the paleographic and codicological notes. A fourth style sheet, AllTags, shows the contents of all the SGML tags except for <DAMAGE>, <SPACE>, and <GAP> tags. Finally, a fifth style sheet we have called Diplomatic omits all note tags and does not display the color coding for <SIC>, <CORR>, <ORIG>, or <REG> tags.

Presentation of Text: Transcriptional Policy

We have two major goals in creating documentary editions. Since we intend ultimately to produce critical editions of the authorial texts, we look at each manuscript text and its documentary edition as a step toward restoration of an authorial text. From that perspective, much of what interests us in a manuscript lies in its relations to other witnesses and to the texts that lie between it and the archetype. Because each documentary text will be electronically collated with all the other texts, we have transcribed a few more aspects of the manuscript than has become the fashion in editing vernacular texts in the late twentieth century. We are also aware that these manuscript texts are often of considerable interest in their own right. As George Kane has argued, the recent trend in literary criticism to speak of such manuscript versions as medieval "readings" of the poem tends to sentimentalize the scribal role.N We have, however, found in the work of F-Redactor a highly interventionist scribe and editor whose interest in the text was steadily intense, if not as steadily intelligent; we have found a medieval editor who did not hesitate to revise his exemplar to satisfy his own sense of the poem's structure, its meter, and its meaning. In this case, it is not too much to claim that the scribal text constitutes a serious response to and rethinking of Langland's poem.

In the interest of reflecting as accurately as possible the features of the scribal document, we mark with SGML tags all changes in hand, style of script, or color of ink. We retain scribal punctuation, introducing none of our own. Except those entered by the later marginal annotators, which we mark with codicological notes (symbolized by the C icon in the browser), we tag all textual insertions in this manuscript and attempt, where possible, to identify the hand responsible for such additions. Similarly, we mark deletions, subpunctions, and significant erasures. In the case of erasures, the quality of the vellum is uneven, and we found that discoloration or roughness of texture could look like an erasure even in the original. We concluded that it was most reasonable to record erasures only when the text differs from the reading of Bx or where the erased text is still legible. We record our resolutions of all abbreviations, suspensions, and brevigraphs, sometimes including material of dubious significance. The F-Scribe tends, for example, to put what appears to us to be a meaningless curl over various letters throughout the manuscript. When we eventually do machine collation, such elements will almost certainly constitute only distracting informational noise. Nevertheless, those sporadic and otiose curls appear to represent a scribal intention, though what he intended is unclear, and we have preserved them in a series of paleographic notes. If they have significance, perhaps that will become clear to a reader who examines them with greater penetration or imagination. And possibly, though not, we think, probably, they may correlate with problems of textual interest in other manuscripts. We do not, on the other hand, mark what appear to be unintended ink trails and blots.N

The scribe's mise en page, his use of colored paraph markers, as well as changes in style of script and color of ink all provide textual information available to medieval readers and usually lost in modern printed editions. Using SGML tagging, we are able to render on screen a partial representation of that information. Viewed in the style sheets for the diplomatic transcription of the scribal text (labeled "F-Scribe," "NoPalTags," and "Diplomatic" in the Multidoc Pro command line under Styles), such changes are signaled by changes of font or color. The basic anglicana text hand is represented in roman type with resolved abbreviations and suspensions in roman italics. Texts written in fere textura are represented in bold roman italic, and when such text is rubricated, it appears in red roman italics and increased in size by five per cent. Resolved abbreviations and suspensions appear as roman characters inside the normal italics for textura scripts. Ornamental capitals appear in bold roman at 25 point size, without regard to their relative size in the manuscript, a twelve-line capital displayed the same size as one of only four lines.N Lombard capitals are represented by a change of font size to 18 point, making them intermediate in size between the capitals touched in red and the ornamental capitals that mark major structural divisions.

Our edited pages are not intended to reproduce literally the manuscript page, only to represent abstractly its salient features. For those who want to see the manuscript page, color images of each page are readily available with a click on the icon at folio breaks. For that reason we have taken advantage of color in our display to mark in the diplomatic transcription a small number of other features, most notably oddities in scribal word division (presented in lime), scribal insertions (presented in aqua), or errors in the scribal text (presented in purple). Needless to say, the manuscript is not lime, aqua, or purple at those points. In the edited text the corrected errors are represented more conventionally by displaying the emended words or phrases in square brackets, while the regularized word divisions are not marked at all. For example, at line 1.2b, the diplomatic text read through the F-Scribe style sheet renders the manuscript lection "as y sheep were" with the word sheep appearing in purple. A reader reading the edited text through the F-Critical style sheet will see the display: "as y [a] sheep were."

Transcription into an electronic medium is quite as interpretive an activity as that into printed texts, though the electronic edition offers greater flexibility. Extensive as our present markup of the text is, it would have been possible to provide an even more fine-grained transcription than we have chosen to do. The scribe, for instance, offers three distinctive letter forms for <r>.N Two distinctive forms of <s> appear. We could have distinguished sigma <s> from long <s> with entity references. However, since those distinctions represent allographic forms with a readily determinable rationale for their distribution, we, like Peter Robinson and Elizabeth Solopova in their transcriptions of the manuscripts of The Canterbury Tales, concluded that our text might most reasonably aim at graphemic rather than graphetic representation.N To that end, we have not distinguished allographic forms.N On the other hand, we have distinguished <z> from <ȝ>, though the letter forms are identical. The characters developed independently from quite distinct origins. Moreover, though SGML markup permits transcribers to specify expansion tags in some detail, distinguishing suspensions, contractions, brevigraphs, superscriptions, etc., the range of abbreviations in F is entirely conventional for both Middle English and Latin texts, and such detailed specification seemed to us an instance in which any potential gains in specificity are offset by added complications for transcription and proofreading. That game, we think, is not worth the candle. Indeed, so conventional are the abbreviations in this manuscript that users who prefer to insert more extended tags (or entity references) can easily supply them by a search/replace procedure on their word processors. Experienced readers of English late medieval hands, for example, will know instantly what range of abbreviations lies behind the display in italics of <es> or <er> or <pro>. Moreover, publishing the hypertextually linked color facsimile provides both experienced scholar and novice with a picture of the graphs themselves as the scribe produced them.

The scribe highlighted some characters, words, and phrases throughout the manuscript. He wrote most of the Latin text in red ink, further emphasizing the Latin with a change of hand from bastard anglicana to textura, though not always maintaining consistency in letter forms. On most leaves the parasigns are done in red, green, or blue ink, though tell-tale ticks appear in the left margins on those leaves where the rubricator failed to achieve his intentions. Similarly, though he clearly meant to mark every initial character in every line with a touch of red ink, he infrequently failed to make the touch in one line or sometimes several in a row. On folios 2v-3r, he began the red touches on the first several lines, then neglected to finish them on that opening. We think these lapses to be of no significance. Also, throughout the manuscript he emphasized words and phrases by underlining them or touching them with red ink. A few words are treated to both red touches and underlining. In general, we have tried to represent as closely as possible what is on the page, but because this is a handwritten document, we have sometimes disagreed about whether the scribe intended to underline the whole word or to touch in red every graph. For instance, when the ligature <st> appears with a red touch between the characters, do we take the <s> or <t> alone to have been emphasized or the full ligature? We think in such cases that the rubrisher intended to highlight the ligature, and we have tagged both characters as "touched in red." Generally, we have taken a word partially underlined to have been underlined wholly, since we assume that the scribe intended to emphasize a word rather than a collection of graphs. When ascenders on the first line are extended and touched in red, we have marked the red touches, though we take them to be only decorative. We do not imagine that we have achieved utter consistency in rendering these points of emphasis and decoration, and indeed, had we done so, such consistency would better reflect our concerns than those of the scribe or his readers.N

Word division in handwritten documents is irregular. This text is no exception, and we have not attempted to represent in detail irregularity in word spacing, rendering both large and small spaces with single spaces. We mark with <gap> or <damage> or <del> tags only those instances in which a larger than usual space is created by an erasure or damage to the manuscript.N The difference in some instances between words and phrases is a matter of convention established within print culture. For instance, in modern British usage for ever is a phrase, while US publications will print the single word forever. We have in ambiguous cases taken the headword of the Oxford English Dictionary as our standard. When words appear in the manuscript with a clear space between morphemes which are treated in the OED as a single word, we represent that fact with hyphenation; e. g. a-begge, bonde-man, brest-boon, mys-dedys, i-callid, y-armed, etc.N Whether the scribe took these morphs to represent separate words is not determinable, for many words appear with and without spaces between prefixes and the stem. In a small number of words, usually those in which elision in the spoken language might easily have led to misanalysis, we have recorded in the diplomatic text the spacing of the manuscript and regularized in the edited text; e.g. a tese, a toones, etc. However, we have ignored erratic word divisions such as that appearing at 14.105 where the scribe has written "& ȝe lordeynes havey loost . . .," interpreting the graphs according to what we take to have been the scribe's intention as "have y-loost."N

In addition to marking abbreviations with superscripted letters, the scribe also tended simply to supply the vowel above the line. In such instances, we initially transcribed about one-third of the manuscript recording those superscriptions and then analyzed the data to see whether the difference between, say, þe and þe or was and was is graphemic. Since they appear in free variation and carry no distinction in meaning, we concluded that the difference is not graphemic and took no notice of that graphetic difference. Words most commonly so written include brouhte, nou, nout, þe, þoru, what, and was. Barred single and double <l> are ambiguous, and we have resolved them according to context either as <-le> and <-lle> or as <-les> and <-lles>/<-llis> depending upon the usual spellings of the words in which they appear. Barred <ll> is frequently a meaningless ornament in late medieval vernacular hands, but by searching for the scribe's treatment of <ll> in words where the inflection was spelled out, we could determine from both the words in which it appeared and the distribution of its appearances that it is a genuine abbreviation for this scribe. A suspension following word-terminal <d> is resolved by context either as <-e> or <-is/-es> except following quod. Other abbreviations and suspensions are typical of late medieval vernacular hands and require no special comment, though as we have noted above, readers interested in studying the scribe's forms may readily search for <EXPAN> tags in our text and then consult the facsimile. For thumbnail images of the most common abbreviations and suspensions in the manuscript, click here.

The scribe tends to write a curled down-stroke after word-terminal <-f, -g, -k, -s, -lle -t> and the <-st> ligature. On some occasions, as at 1.67, 2.29, 2.38, 2.137, 14.338, etc., where it appears at the medial caesura and the scribe has not written a solidus, we have taken it to reflect his intention to provide a caesural marker and have rendered it with </>, but in general it appears to be meaningless. Its appearance on words such as wille where the horizontal bar already represents final <-e>, as well as its appearance in the Latin text where no inflectional ending is wanted, suggests that it is usually a meaningless ornament, and we have not marked it. For some instances in which we have not taken it to mean anything, see its appearance in the following words: yf, departyng, Pekook, Inparfyght, lust, etc. In a few instances, as at 14.47, which reads "Boþe as long & as large / ...," where it appears on the <g> of long and at a phrasal juncture within the a-verse, the mark is thoroughly ambiguous. It also occurs in some dozens of instances at the end of an a-verse and where the solidus is still written, suggesting that the scribe sometimes used it to indicate a solidus and sometimes a flourish.

As we have noted, the rubrishing scribe failed occasionally in some openings to make the color parasigns. Since the places where parasigns should go were marked when the text was written with small virgules in the left margin, we have in the "F-Critical" style sheet provided the parasigns inside square brackets.

Line numbering in the printed editions is not consistent from edition to edition. In general, Sister Carmeline Sullivan's tripartite division of the Latin quotations and tags provided a basis for the Athlone and other modern editors for deciding when to provide serial line numbering and when to treat the Latin materials as supplementary to the English text.N Derek Pearsall's explicit statement of his practice in his edition of C is essentially that of modern editors: "The practice here is to number Latin lines which contain any word of English or which are integral to the syntax of the surrounding English lines. The remainder, mostly biblical quotations, are unnumbered and indented."N The problems we face in providing hypertextual linkages between every manuscript text and the editorially constructed hyparchetypes, archetypes, and critical texts have led us to provide for each documentary edition a set of absolute line numbers of both English and Latin texts. Even that policy is not entirely straightforward, since long prose passages often occupy more than a single line of manuscript text. We have let the scribal indications of structure determine when and where to assign a single line number. An indication of the problematic nature of the manuscript text may be seen at 11.129-131, where the scribe has written two lines separately and the remainder without a line break. His practice here appears to be eccentric, and Kane-Donaldson's decision to print the entire quotation as one single unit of prose is entirely reasonable. In any case, we have, in addition to our own line numbers, also supplied a reference to the Kane-Donaldson line numbering in every line in the poem. That is, we have assigned to every line its unique identifier in a line tag (<l>) and a reference number which will serve eventually as a basis for hypertextual linkages among the documentary texts. We are using the Kane-Donaldson numbers for reference in the early documentary texts both because they represent a rational modern standard and because they can serve as place-holders until such time as we have established our own text of B. The Kane-Donaldson numbers are displayed immediately following the F passus and line number in parentheses, but only under the F-Critical style sheet. F, of course, has a number of lines rejected in the Kane-Donaldson text, and these are designated by the number for their last line before the added material followed by a decimal and numbers for each additional line. For instance, F7.286 is labelled "KD10.272.1."

We have used <MILESTONE> tags to provide readers an indication of foliation and the differences between F's passus divisions and those of the B text.

Linguistic Description

A Linguistic Atlas of Late Mediaeval English (LALME) places the manuscript in Essex, a few miles north of Maldon near present day Witham.N We have found no evidence inconsistent with LALME's finding and a good bit to confirm it, certainly sufficient to be confident that the immediate scribe was from southern East Anglia. Since LALME's standards of accuracy have been questioned by T. L. Burton, especially the Southern sections, we have compiled the evidence afresh for each of the relevant questionnaire items.N The machine-searchable text has made it possible to survey quickly and accurately all forms, and it should surprise no one that we have both added to the recorded forms and in some instances discovered that LALME's relative frequencies were in error. The LALME project, though it was able to make sophisticated use of computer technology in compiling and analyzing the masses of data accumulated in field workers' notes, necessarily lacked significant access to reliable electronic texts of the manuscripts from which they collected that data. Hand compilation of such data is subject to all of the problems of fatigue and inattention that plague all scribal productions. It has, therefore, come as no surprise to find that machine searches of the electronic text have produced a number of spellings and forms not picked up in the original LALME survey. Nevertheless, we have been impressed by the general accuracy of LALME's account of the formal features of this manuscript. Nothing we have found is inconsistent with its placement of the dialect of the immediate scribe. (Click here to see the table.)

In the following account of the immediate scribe's language, we have given attention primarily to those features of the scribal language likely to be of assistance in localizing the language of the immediate scribe and in determining what can be recovered from relict forms of traces of anterior copyings. In the case of vowels, we have concentrated primarily on the development of native sounds and have relatively neglected words of romance origin. Forms occurring rarely are listed in parentheses. All the features we list and discuss below are consonant with an East Anglian provenance, while features listed in paragraphs 13, 14, 16, 30, 32, 38, 41, and 44 tend to identify the text as broadly East Anglian. Relicts more consonant with Norfolk than Essex appear as minority forms in paragraphs 6, 34, and 45. As M. L. Samuels has noted, F was "thoroughly translated" into the immediate scribe's dialect, showing "very few signs of relict forms."N However, occasional relict forms survive, and we have attempted to note the most significant of these below, from Langland's dialect, from the B archetype, from the alpha hyparchetype, and from the scribe we have called the F-Redactor. The process of determining these layers is to a degree speculative, but because of the recent work of M. L. Samuels and others on Langland's dialectN it is now possible to determine at least some features of the language of those earlier copyists. It might go without saying that the analysis we offer here is more than usually provisional and subject to change in the light of our analyses of other manuscripts in the B tradition. In this instance, it would have been especially useful to have completed analysis of the language of Rawlinson Poetry 38 (R), the other alpha family witness, before we attempted to disentangle the layers of copying in F. It should eventually be possible to draw some conclusions about at least some of the features of the dialect of the scribe who copied alpha, but definitive analysis must await our collation and analysis of R.

In the case of the F-Redactor, the scribe who restructured Langland's passus divisions and made hundreds of revisions at every level, our preliminary study in relation to LALME maps suggests a location in Central Norfolk.N The salient features are these: two appearances of kyrke spellings added by a scribe after alpha, the -s forms of third person singular indicative verbs, the appearance of the spelling er for the coordinating conjunction "or," the spellings feyȝr, fyȝr for "fire,"N the appearance of ȝeet for "yet,"N and ke(e)me for "came." The unique occurrence of þerknesse for "derknesse" is consistent with that placement.N All these elements point to the activity of an East Anglian scribe more northerly than the immediate scribe. Lacking, as we do, any means of determining the number of scribes intervening between Langland and the immediate scribe, we cannot prove that any particular change in the text is to be attributed to the medieval "editor" who revised the passus divisions, but it is reasonable to think that the interventionist scribe who changed his text in large matters also felt free to emend in small. Alternative explanations are, of course, possible, since it cannot be determined whether a single hand is responsible for the scribal forms and changes or whether each is a relict of a different copyist.

Phonology

Vowels:

Tonic Syllables

a-brood 3.178; aloone 4.495; cloþ(e) 5.210 ~ clooþ 15.290; a-roos 15.53; froo 5.284; goost (22x) 2.38 ~ gost (1x) 7.248 ~ gast (1x) 3.213; groone 5.1140 ~ grone 5.913; hoot 5.965 ~ hote 1.216; pope 1.102 ~ poppe-holy (1x) 10.293; soor(e) 10.586; stoon 11.502 ~ stones 3.15; wones 4.224 ~ woones 14.271; wore "were" (2x) 7.235, 13.174 <ON váru; woot 1.39; etc.

slowhȝ "slow" 10.419; sowle (65x) 2.39 ~ soule (25x) 2.128 ~ sowhle (1x) 5.1110; etc.

wan 5.464; can 1.105; fram (85x) 1.122 ~ from (2x) 13.8, 13.106; game 5.415; man 2.113; name 1.101; shame 4.374; etc.

a-mong 1.188; hond (15x) 5.590 ~ hand (7x) 13.221;N hange (12x) 1.163 ~ honge (5x) 1.170; longen 3.46; lo(o)mb 5.563, 16.36 ~ lambren 11.222; stonde (10x) 5.355 ~ stande (2x) 13.50; etc.

drank 5.177 ~ dronk 10.104; sank 14.69; þank- 6.104 ~ þonk- 3.150; etc.

barefoot 14.11; blood 2.152; book (40+x) 1.95 ~ boke (5x) 5.238; dome (8x) 3.207 ~ doom(e) 4.310; flood 6.267 ~ flodis (1x) 10.520; foode (15x) 1.38 ~ fode (2x) 5.673; looke (25x) 2.207 ~ loke (15x) 2.143; oþer(e) 1.47 ~ oyþer (3x) 2.173 ~ oiþer 7.283; weet-shood 10.638 ~ wetschod 14.1; etc.

apostl- (12x) 7.35 ~ apoostlis 12.169, 13.29; box 11.230 ~ boox 5.650; borwe 4.451; corps 2.135 ~ koors 10.9; cros (16x) 5.10 ~ croos (4x) 6.92;N folk 1.16; foox 16.44 ~ foxes 5.682; god (244x) 1.39 ~ goode 7.345;N holpe 4.511; look 2.200 ~ lokkys (n.) "lock(s)" 10.379; moos "moss" 11.303; northȝ 14.168; ofte 3.19; pekok 9.366 ~ pekook 9.348; poot- 5.839 ~ pot 10.261; spottys 10.284 ~ spoot 10.285;N stook "trunk" 12.7; top 4.129; etc.

The doubled <oo> spellings do not appear necessarily to reflect lengthening of OE /o/. The spellings apoostlis, boox, foox, look, and pekook are not otherwise recorded, and MED, s.v. stok, marks its single instance of stook from Lydgate's Troy Book as "?error." The appearance of moos, poot, and spoot in a variety of East Anglian texts shows that the spellings are to a degree conventional in that area. Cf. paragraphs 13, 18, and 21 below for parallel development of other OE short vowels. Note as well the apparent indication of length in spellings of unstressed syllables recorded in paragraph 28 below.

bold (7x) 1.179 ~ boold (3x) 13.13; gold (20x) 1.69 ~ goolde (2x) 4.25; moolde (13x) 1.58 ~ molde (4x) 1.202; word- (65+x) 2.144 ~ woord- (29x) 2.13; etc.

a-bowte (30+x) 1.28 ~ abowhte (16x) 3.11; clowdes 14.418 ~ clowhde 4.181; clowtys 10.61; doun 4.434; how (70x) 1.96 ~ hou (8x) 7.92; mous 1.193 ~ mows 1.173; prowd(e) (7x) 4.167 ~ prowhd(e) (4x) 10.56 ~ prouhde (1x) 7.75 ~ proud- (1x) 8.244; etc.

botere 5.446; curse 4.131; flyx 5.179 ~ fluxes 16.81; ful (107x) 1.16 ~ fulle (6x) 4.53; love 4.282; pulle 9.379; etc.

found 10.543; hound 5.365 ~ hownd 5.261; moorne 4.17; tunge (40+x) 4.120 ~ tonge (1x) 16.162; etc.

beelis 16.84; fyȝr 5.1039 ~ feer 7.433; her(e) 5.560 ~ hyȝre (n.) 5.566; heren 5.767 ~ heere (v.) 5.848 <OE hýrian; keen 5.794 <OE cýna; wyȝshede 10.81 ~ wischede 16.194 ~ wysshed 5.354; etc.

ken (5x) 2.190 ~ kyn (5x) 10.389 ~ keen "kin" (2x) 4.192, 6.259 (Bx=kynde) <OE cynn; dyde (21x) 2.28 ~ dide (14x) 4.188 ~ dede (8x) 5.508; gylt- (10x) 4.97, 6.273 ~ gelt- (3x) 7.273; gilt- (9x) 1.74, 4.443; hedde 7.452; heddyn 9.30; hellys 1.206 ~ hillis 1.4; kyll- (11x) 12.145 ~ kulled 15.144; lyft 3.5 ~ left 5.590; mychil (32x) 1.173 ~ myche (28x) 4.134 ~ mychel (7x) 2.37 ~ meche (2x) 9.251 ~ michil (3x) 4.518 ~ miche 10.742;N synne (60+x) 2.141 ~ sennys (2x) 10.663; þenne "thin" 15.408 <OE þynne [cf. OFris þenne (OED)]; w(h)ich(e) (35x) 5.786, 7.28, 9.236 ~ wheche (3x) 1.68; etc.

The unrounding of OE /y/ and /y:/ to /i/ and /i:/ and subsequent lowering is characteristic of spellings in East Anglian manuscript spellings in LME.N However, the <ee> spelling for keen "kin" 4.192, is puzzling in suggesting an unaccountable lengthening of the tonic vowel. The same spelling in deene 14.64, is more readily explicable by the change of /y/ to /e/ and its subsequent lengthening in an open syllable.

beurde 14.120N ~ beerde 4.15; keende (26x) 3.78 ~ kynde (16x) 1.112; meende (5x) 8.264 ~ mynde (4x) 11.311; etc.N

blythe 3.159; chyde 4.166 ~ chyȝde (1x) 7.321 ~ chide (1x) 5.973; knyf 5.81; lyf (111x) 1.116 ~ lif- (4x) 2.27 ~ lyȝf (4x) 4.415; pyȝk- 5.756, 16.261 ~ pyk 5.480; ryde (7x) 4.351 ~ ryȝde (1x) 4.202; syȝd(e) (8x) 5.81, 5.197 ~ syde (4x) 1.7 ~ side (1x) 5.378; styf 11.513 ~ stef 6.28;N tyme (84x) 1.78 ~ tyȝme (15x) 4.212; wyȝd(e) (9x) 5.543 ~ wyde (4x) 1.3; wyȝn (10x) 2.31 ~ wyn (1x) 1.219; wise (9x) 7.8 ~ wyȝs (7x) 1.199; wyȝsdom (7x) 4.370; etc.

bitterly (2x) 14.408 ~ betterly (1x) 5.111; (by)twixe 5.98 ~ by-twexe 4.119; lyȝnde 2.153; lyve (36+x) 4.533 ~ lybbe(n) (12x) 4.215 ~ leve (1x) 15.214; neme 12.72; wyȝnd (4x) 4.324 ~ wynd (7x) 6.26; wyght (13x) 2.64 ~ whyȝt (6x) 5.117 ~ wyȝht- (2x) 4.215 ~ wyȝght (1x) 6.65 ~ wyht (1x) 9.145 ~ wyȝt (1x) 7.177 ~ whyht (1x) 6.145; wyke 5.911; y-wrete (1x) 2.198 ~ y-wryte(n) (2x) 5.1104; wrete 9.378; etc.

This feature is consonant with East Anglian, though it is common in other northerly dialects. See Norman Davis, "A Paston Hand," RES n.s. 3 (1952): 216, and Wilhelm Dibelius, "John Capgrave und die englische Schriftsprache," Anglia 23 (1901): 189-91.

be (400+x) 1.47 ~ by "be" (3x) 5.1080; bedeman 4.47 ~ beedemen 11.221; beech 5.592; brede 9.347; crede 5.743; deme 4.520; feede (8x) 1.84 ~ fede (1x) 8.193; feet 4.428; grene 5.936 ~ greene 9.1; heede 7.206 ~ hede 5.667;N kene 1.216 ~ keene 15.309; keep(e) 3.47 ~ kepe 1.69 ~ kep (1x) 10.281; meede (80+x) 3.19 ~ mede (1x) 4.234; meet(e) 8.26 ~ mete 5.100; seeke 5.1162 ~ seke 4.341; spede 4.263 ~ speede 4.159; sweete 13.211 ~ swete 1.80; etc.

Some spellings of "fed" (pret. and ppl.) indicate the expected shortening before a long consonant (Jordan, §§ 23, 33.1); e.g., fed 12.132 (pret. sg.), fedde 11.301 (pret. pl.), but two spellings with <ee> suggest retention of length: e.g., feede (pret. pl.) 11.431, y-feed (ppl.) 11.318.

best(e) 7.388 ~ beest(e) 5.22 (adj.); dowel (64x) 5.1108 ~ doweelis 10.803; fedre 15.420 ~ feedre 4.380; fellyn 2.120 ~ fyllyn 2.117; feestrid 13.190; goweel 6.146; ieestys 3.86 <OF geste ~ iestys 7.20; rek(e)ne (5x) 10.585 ~ ryk(e)ne (3x) 8.129; web(be) 5.199 ~ weeb 5.112 <OE web(b); weed 10.371 <OE wed(d); wel (100+x) 1.63 ~ weel (15x) 2.198; etc.

beende 5.577; beendiþ 5.175; beerne "barn" 4.403;N bleenden "blind" 14.141 ~ ablendiþ 7.282; elde (16x) 5.193 ~ eelde (2x) 10.365 ~ heelde 8.53; feld(e) (3x) 2.2 ~ feelde (2x) 4.489; seelde(n) (11x) 5.413, 5.128 ~ selde(n) (10x) 7.419; send- (11x) 5.980 ~ seende (4x) 14.395; etc.

deele 13.308 ~ dele 2.199 ~ deyle 5.729; drede (20+x) 1.145 ~ a-dred 15.22 ~ adrad 15.310 ~ dreede (3x) 3.210; heele 10.353 ~ hele 9.350; hees "hest" 5.66; lasse (18x) 3.46 ~ lesse (4x) 5.224; laste (v.) 4.29 ~ leste 10.343; leene "lend" 1.84 ~ lene 5.669; see 4.471; swete 10.269; weet- 5.543, 10.648 ~ wet- 14.1; whete (5x) 4.403 ~ wheete (1x) 4.41; etc.

bak (7x) 4.184 ~ baak (1x) 5.1087; bathed 13.192; craft 2.135; dale 1.14; fadir 2.65; smale 12.84; wasshe 5.580; etc.

beem 7.282; betyn 9.110; breed 5.173; chepe 5.326; deed(e) 10.803, 14.64 ~ ded(e) 2.184, 14.66; dreme 8.4 ~ dreem 12.22; ere 4.357 ~ eere 11.420; gret(e) 1.67 ~ greet(e) 5.388; hed 5.195 ~ heed 3.34; heep 5.329 ~ hep 5.233; leef 2.153; leep(e) 15.127 ~ lepe 1.191; reed(e) 3.14 ~ rede 3.11; etc.

heyȝ(e) (10x) 1.12 ~ hey(e) 1.134, 9.158 ~ hy 11.87 ~ hyȝ 2.170 ~ hyȝe 2.74; ne(e)yhe- (5x) 2.178 ~ neyȝh(e)- (4x) 13.156 ~ neyȝ (3x) 5.94 ~ neeyh 4.134 ~ neyghe- 5.262; þeyhȝ (22x) 4.149 ~ þeyȝ (6x) 4.343 ~ þey 10.716, 11.405 ~ þeih 10.506 ~ þeyh 5.559 ~ they 1.176;N etc.

al 1.1 ~ alle 1.49; barn 3.3; calf 5.942; falle 1.56; flex 5.665; salte 10.39; shafte 9.76; vnfelde 13.264N; weex 13.287 (n.); wex(e) (27x) 4.324 ~ wax(e) (3x) 4.293; etc.

beden 15.145 ~ beedyn 4.28; brest (4x) 5.399 ~ breest (1x) 11.205; cleve 8.228; creepe 16.44 ~ crepe 2.194; depe (6x) 1.15 ~ deep (2x) 1.14; feend- (13x) 6.38 ~ fend- (9x) 2.41; flee (6x) 3.211 ~ fleyȝ (5x) 3.212 ~ fley (1x) 13.186; frend- 11.191 ~ freend- 10.151; leem 14.128; syke 5.414; sykness (7x) 5.782 ~ seknesse (4x) 5.422; spewen 7.42; theef 9.309 ~ thevys 10.163; etc.

cherl 5.363 ~ chirlys 15.56; derk(e) 3.104 ~ dirk 2.60; erl 6.85; fele (44x) 1.22 ~ feele "many" (2x) 8.328; feer "far" 6.75; herte 2.139; hevene 1.97; leerne 5.45 ~ lerne 2.134; -selue (180+x) 5.496; sterre 14.234; sterve 9.111; werk 1.155 ~ wirkis 5.1196; worchepe 2.8; world 1.3; etc.

blewe 15.343 ~ blewh 5.352; brewh 5.219 ~ brew- 15.404; rewe 12.150; sew (7x) 10.385 ~ seew (1x) 15.348; trewe 1.114; etc.

Unstressed Syllables

clergyȝe (6x) 7.178; doweelis 10.803; drapeer "draper" 5.255; drunkelewhȝ 6.80; fysyȝk 16.376 <OF fisique; grasees "grasses" 10.522; inparfyȝt 11.106; ioyneen "to join" 14.257; lovyȝe 8.182; lusyfeer 14.407 ~ lucifeer 14.310, 14.356; mateer 11.81; meryght 2.180; morteer 15.329; pereyl (8x) 5.771 ~ pereil 5.746; pylgryȝm 5.755; profyghte 10.241 ~ profyȝht 4.465; sodoom 10.555; vn-lowkeeþ 14.263; etc.

Consonants:

qwhyle 4.324; wan "when" 5.767, 16.196; where "were" 1.59, 2.135, 3.65, etc.; where "wear" 1.158; wyt "white" 11.425; wyche "which" 5.302; whythyes wyse "like a withy" 5.531; whont "accustomed" 16.368; whonen "dwell" 3.76; whot "wot" 3.79, 3.123, 5.784; etc.

Since Langland himself alliterated words that historically began with /hw/ with words beginning /w/, this feature does not distinguish the scribal dialect from the poet's. However, it is very clear that /hw/ has become /w/ in this scribe's dialect.N The single instance of qwhyle, "while," is probably a relict form from the Norfolk redactor.

The evidence of inverse spellings suggests that initial aspirate /h/ has been lost in the dialect of the immediate scribe. The phenomenon is common in East Anglia in late Middle English.N

er-of 5.986 ~ here-of 7.255; hagews "agues" 16.84; hax "ax" 9.233; hetyn "eat" (pres. pl.) 7.10; herly "early" 5.308 (a reading not in Bx), 5.329; hers "arse" 7.328; hews "use" 1.93; evy 16.2; evynesse 14.249; huttyrly 6.232; etc.

Neutralization or loss of /h/ may well have been nearly complete in all contexts in the scribe's dialect. At least, his spellings of reflexes of OE /u:/ or OF <ou> with <owh> and OF <au> with <awh> or OF <eu> before dentals suggest near complete neutralization of <h> in the scribal dialect:

agewh 10.347; baw wawh 15.404; bewh "beau" 5.1161; blewh 5.352; brewh 5.219; corlewh 10.521; drowh "drew" (5x) 5.359; Iewh 7.137; knewh 10.190, 15.421; lowh (3x) 5.140 ~ lowhȝ (3x) 5.602; lowhly (2x) 5.563; Matthew (3x) 5.892 ~ Mattheu (3x) 4.244 ~ Mathewh (1x) 15.266 ~ Mattheuh (1x) 8.253; nowh (2x) 5.863 ~ nowthe (4x) 5.858; ouer-threwh 5.360; rewhliþ 6.139; etc.

The series he ~ hey ~ hyȝ ~ hyȝe for "high" supports the hypothesis that the velar has disappeared in virtually all contexts in the immediate scribe's dialect, as does weyȝ for "wye." If further evidence of the loss of the velar in the scribe's dialect were needed, these inverse spellings would prove the case: lyth "light" 6.153 and lythlokere 5.581.

Many spellings using <ȝt> are conventional, predictable on the basis of etymology—fyȝt(e), hyȝt(e), knyȝte, etc. Others are explicable on grounds that the <ȝ> appears after front vowels as a lengthening graph: apetyȝt (stressed /xx) 5.919; byȝte 16.359; coueyȝteþ 15.309; delyȝte 7.383; disseyȝt 14.333; fayȝten 5.1088; feyȝth 14.15; petyȝt (/x) 5.1044; and whyȝt "white" 5.791. The <y> following <e> is itself already a lengthening graph, and in apetyȝt and petyȝt the stress is not on that syllable. At 12.106 we have treated the single instance of fytȝ, "fight," as scribal error, but as East Anglian scribes sometimes wrote <-tȝ> in such contexts, the spelling may have been intended.

These spellings appear in unstressed syllables in a number of words in which they never appeared historically, suggesting that the velar spirant /x/ had been completely lost before /t/: en-habyȝtid 10.294; hermyȝtis 5.799, 5.840; meryght 2.180 ~ meryȝt 5.387, 8.182; spyryȝt 13.354; tyȝhtly 8.53. In a dialect in which the velar spirant before syllable-terminal /t/ had been spelled <-ht>, its loss would lead to spellings such as awhter (3x) 4.52, bawhde 5.725; cawhdel 5.364; defawhte (10x) 3.141; defrawhdiþ 5.1056; sawhter (9x) 4.226; bewhte 9.169; lewhte 1.120; sewht 5.496; abowhte(n) (20x) 3.11; clowhde 4.181; cowhde (18x) 1.173; dowhte (3x) 9.111; lowhde (5x) 1.123; lowhtede 4.105; prowhd (3x) 10.287; rowhte (9x) 1.166; etc.N

dunk "dung" 4.302;N dunkele "dung hill" 5.76; and þynk "thing" 10.405. The form dryng "drink" 10.796 is an inverse spelling. A West Midlands phenomenon, it is perhaps surprising to find F as the only witness among B manuscripts to such spellings in each case. There is no reason to think these can be relict forms, but they are not usually taken to be features of East Anglian dialects.

by-qwache 14.250; chafte "shaft" 6.155; felachepe 3.195; fychȝ 16.45 ~ fys(s)hȝ 5.391 ~ fysch- 6.50 ~ fyssh 5.177; florcheþ "flourisheth" 10.773; lordchepe (10x) 3.46 ~ lordschepe (1x) 10.738; marchal 4.189; punche(n) "punish" 10.766, 15.198 ~ punshe (4x) 3.49, 7.286; werkmanchippe 13.262; worchep- (18x) 2.8; etc.

The evidence suggests that this form is inherited by the scribe and is not his usual form, since at 10.561 he corrected cheltroum to sheltroum. At 4.67 he corrected punschyn to punsshyn and at 4.195 lordchepe to lordshepe. Perhaps contradictory evidence appears at 5.1054 where some scribe in the tradition has written shapman, intending chapman (itself an error). The form schapmen appears in the N-town cycle, also written in East Anglia.N It is here very possibly the work of the Norfolk revisor responsible for a number of other non-authorial forms. Similar spellings appear in Capgrave's work.N

englyȝs (3x) 5.39 ~ englysch (2x) 11.27 ~ englyshȝ (3x) 5.1100 ~ englyschȝ (2x) 11.394; manshid(e) 3.40, 9.210 ~ manschid 16.221 ~ manshede "mansed" 7.299; sordych "Shoredich" 10.351; wysse 5.565 ~ wisshe "wisse" 5.148, 5.546, 6.108, 7.163, 7.406; etc.

byȝte "bite" 16.359; lyȝk "like" 15.8; nyghne 13.156 ~ nyhȝne "nine" 5.379; paradyȝs "paradise" 7.485; wyȝnys "wines" 7.383; wyȝse "wise" 13.340; etc.

Before a voiced stop, yogh steadily appears as a lengthening graph after front vowels: abyȝde 3.214; by-tyȝde 3.119; hyȝd(e) 15.463; leyȝd(e) 7.51, 5.244; priȝd(e) 6.114; seyȝd(e) 6.162; syȝd(e) 5.81; etc.

a-cursethȝ 14.110; dethȝ 14.145; feythȝ (8x) 2.183; forfeetythȝ 16.25; forthȝ 13.288; frythȝ (2x) 9.42; furthȝ (2x) 1.213; kepithȝ 11.366; lyȝthȝ 14.388; morderythȝ 13.337; mowthȝ 15.291; nazarethȝ 15.139; northȝ 14.168; schryuethȝ 16.288; wernethȝ 16.12; etc.

There is little reason to think the spelling indicates any unusual pronunciation of /θ/, though it is possibly related to other East Anglian spellings in which a word-terminal <t> is followed by <ȝ>.N See section 43 below as well.

englys(c)hȝ 11.394; fys(s)hȝ 10.93 ~ fychȝ 16.45; fles(c)hȝ 10.775; Frenchȝ 5.239; freshȝ 5.965; pars(c)hȝ (3x) 5.44; and weschȝ 13.62.

borghȝ 3.89; borwhȝ 5.1073; bowhȝ "bough" 5.30; drunkelewhȝ 6.80; y-nowhȝ 10.510 ~ y-nowh 5.1078; lowhȝ 5.602 ~ lowh 5.140; plowhȝ 5.755 ~ plowh 1.19; þeyhȝ "though" 4.149 (but also "they" 7.313); þor(g)hȝ 1.67; þowhȝ "thou" 6.68 and "though" 11.199.

Though most of these are explicable historically, the scribe's writing "thou" as þowhȝ and "they" as þeyhȝ suggests strongly, especially in light of the evidence in other contexts above, that the velar spirant had been lost in the scribe's dialect, that he wrote yogh in various places where it had never indicated the velar (e.g. after <th> because <th> came to be written for <ht> in dialects where /xt/ > /t/).

Cf. agewh "ague" 10.347; bewh "beau" 5.1161; blewh "blew" 5.352; brewh "brew" 5.219; corlewh "curlew" 10.521; Iewh 7.137; knewh "knew" 10.190, 15.421; and Mathewh 15.266 for corroboration of spelling neutralization.

slewte "sloth" (1x) 5.458 ~ slewhte 16.158 ~ slewhthe (2x) 5.443 ~ slewthe (7x) 1.41 ~ slewþe (5x) 4.302; mowht "mouth" 1.207, 10.57; etc.

These spellings, though possibly simple scribal error, are consonant with a dialect in which the velar spirant had been lost before a syllable-terminal /t/.

Two instances appear in F, the first in 7.459 in an OF word quantȝ, "when," and another at 12.106 where "fight" is written fytȝ. The latter possibly reflects simple scribal error, and we have marked it with a "sic" tag. See the discussion in A Linguistic Atlas of Late Mediaeval English, ed. Angus McIntosh, M. L. Samuels, and Michael Benskin (Aberdeen: Aberdeen University Press, 1986), 2.xvii, of the forms of <ȝ> and <z>. The editors suggest that in some areas a word-terminal form that looks like <yogh> and <z> is derived from neither but from the French and Latin abbreviations for <-et>. Their examples are not precisely parallel to these, but it seems likely that the scribe attached no phonological value other than /t/ to the graph.

The scribe writes spalm(es) (4x) 5.904 for "psalm(es)" and speudo for "pseudo" 11.522.

lych "like" (4x) 5.119; machamet(h) (7x) 4.322 ~ makameþ 11.424; patriarch(is) (3x) 9.242 ~ patriark(is) (10x) 10.170; etc.

At 16.237 cheve is written for Bx's chewe. The a-verse is repeated almost verbatim at 14.203. Since cheve makes at least a modicum of sense in context, it is possible the scribe intended to write that word. However, at 11.489 scheven is written for shewen. At 4.286 and 16.172 howe is written for OE hūfe, spelled howfes at 1.202. Instances of <u> for <w> appear in luscheburue 11.368 and stroued 10.48. <w> is written for customary <v> or <u> in lowe "love" at 10.145, while lowlyest appears at 10.304 where Bx reads louelokest. Note also chewsaunce at 16.16. It seems very likely that these are relict forms from the revising Norfolk scribe. See Jacob Bennett, "The Language and Home of the 'Ludus Coventriae'," Orbis 22 (1973): 51, for this feature of that Norfolk text.

Morphology

Parentheses indicate optional elements. When Bx appears after a reading, it attests not necessarily the spelling, but the feature: i.e. <-ith> (Bx) means that the <-th> form for the plural verb inflection is in the archetype, but the vowel might be <e, i, y>.

46. The Status of Final <-e>:

Since we are dealing with a scribal copy, the metrical evidence that might provide a clue to whether <-e> was sounded is not relevant, and we are, therefore, dependent upon the distribution of spellings. Such evidence suggests that the immediate scribe has preserved at least some systemic remains of the grammatical system in which final <-e> was sounded, but it is curiously mixed. A good bit of evidence suggests that <-e> was systematically written on the stressed stems of weak and plural adjectives, but not steadily on the unstressed syllables of dissyllabic adjectives: his grete grace 9.392, his owene grace 13.285, hise manye talys 3.219, hise leve seyntys 4.385, hise lawe-ful doomes 8.144, hise wikkide werkis 16.368, for siluerene plates 12.151;N etc. On the other hand, though most stressed stems of weak and plural adjectives are marked with <-e>, some adjectives are exceptions to rule, either appearing without the expected <-e> in weak or plural adjectives (as in his cheef lyflode 5.89, hise chef chyldryn 5.631, and fele fals be-hestis 16.118) or with unhistoric <-e> (as in & false doome 14.29, grete wisdom 7.416, of grete god 6.186.N The evidence for <-e> in verb inflexions is similarly mixed.

The analogical formation in which hise/hyse appears before plural nouns and his before singular is carried through with remarkable consistency. However, the same pattern is not manifested in hir(e), "her," myn, nor our(e). Though the four instances of myne all modify plural nouns, myn modifies both singular and plural nouns in free variation, while in the case of our(e) the form with <-e> modifies both singular and plural nouns.

NOUNS

Abraham-is 13.11; adam-is 14.224; Cayn-is 6.259; Iesu-is 14.103; Iohan-is 14.324; mannys (22x) 7.496 ~ mannis (1x) 5.497 ~ manne (1x error?) 15.440; meme[n]to-is 5.474; Peers 14.21 ~ perses 14.26 ~ Peersis (5x) 14.23 ~ Peersys (2x) 15.268; Penitencia-ys 5.480; and pharao-is 5.1163.

The B archetype at 5.480 reads "penitencia his pik." Such a "his" genitive, which tends to be a feature of Southwest Midlands texts, may well have meant little to the Essex scribe, though it is perhaps remarkable that he has in so many cases retained the space between stem and inflection. We have in each case treated the <is>/<ys> as an inflection and hyphenated. The "his" genitive was not completely unknown in East Anglia, since the reading "to haue mynde of cryste es Passion" appears in Julian of Norwich's Revelations of Divine Love in British Library, MS Additional 37790, fol. 97r. See Tauno F. Mustanoja, A Middle English Syntax, Part I: Parts of Speech, Mémoires de la Société Néophilologique de Helsinki, no. 23 (Helsinki: Société Néophilologique, 1960), 159-62.

Abraham 13.216; addir 5.88; broþir 7.281; chirche 7.267;N Constancius 7.344; hevene (Bx) 10.636; Prioresse 5.158; sowle 5.544, 7.266 ~ soule 14.370;N trewþe 11.102; etc.

blamegeris 10.92; bodijs 2.193 ~ bodies 12.131; childryn 1.33 ~ children 7.345 ~ childur 13.28; clerkys 10.11 ~ clerkis 10.136; dettys 10.10; dysshes 10.82; dubbler(e)s 10.82; eyȝen 5.359 ~ eyes 10.806; erys 4.488 ~ eryn 1.71;LG ferlijs 1.56; flappis 10.64; foon 5.96; frydaes 10.462; grasees 10.522; infynytis 10.129; Iewis 10.212; ladijs (7x) 1.90 ~ ladies (2x) 11.338 ~ ladyes (2x) 8.17; londys 10.211; men 10.36; menstralis 10.232; morterelis 10.108; pesys 5.848;N pylgrimes 10.22; preestys 10.11; prentyȝs "prentices" 5.321; Sarsynes 10.212; schoos 16.219 ~ schoon 10.812; scyenses 10.126; sones 10.123; thevys 10.163; wafres 10.272; weyȝes 5.1011 ~ wyȝen 9.390; werkis 10.148 ~ werkys 10.320; wordis 10.148; ȝeer(e) 5.196, 8.16 ~ ȝeeris 1.56; etc.

bisschopis 11.464; kyng(g)ene 2.104 (< Bx.) 15.81; mennes 4.139 ~ mennys 15.385; wyfene (< Bx) 5.27; etc.

Adjectives and Adverbs:

bygger(e) 7.326; blessedere 8.258; clennere 10.305; cursedere 15.425; deppere 7.193; feblere 11.367; frenlokere 7.237; geltyer(e) 9.205; hastlyere 15.475; leuere 5.155; lythlokere 5.581 ~ lyghtlokere 9.277; lyghtere 10.725; needfullere 2.21; plentevousere 7.85; etc. Though <re> is the most common comparative for bettre, it does not otherwise appear. F uniquely has double comparatives at 2.189 averouserere, and at 10.203 and 14.419 leuerere.

be(e)st(e) 1.156, 5.22; brunneste 5.961; derrest 3.12; dowhtyest(e) 15.136; lowlyest 10.304; mervylokest 6.64; neest 13.345. The multiple superlative at 10.521 most klennest is archetypal.N

Verbs:

Non-finite forms:

asken 10.706 ~ aske 4.445; bere 16.91; faren 7.100 ~ fare 5.1092; fecchen 10.383 ~ fecche 3.182; helpen 10.682 ~ helpe 2.17; kennyn 10.445; lystne 6.62; pleesen 10.698 ~ plese 1.29; seen 5.542 ~ see 4.55; techen 11.424 ~ teche 2.142; etc.

a-bydynge 13.103; beggyng(e) 4.207, 15.233; beryenge 8.72; doenge 11.520; dysyryng 10.367; etynge 9.17; iang(e)lyng(e) 3.86, 6.298; lawhyng 12.163; lyenge 10.333; lookenge 8.324; slepyng 1.9; tayl-ende 6.78 (< Bx); wenyng 16.33; etc.

abydynge 15.297; dryvende 16.100;N etynge 7.109; hippynge 13.157; lawhenge 4.495; lurkende 3.218; pleyende 13.90; slepynge 14.299; syttynge 4.338 ~ syttende 12.149N; waggynge 6.26; wenynge 10.289; etc.

The forms for both gerund and present participle are remarkably consistent. The dominant form by far, with over 350 occurrences, is <-yng(e)> for both. The <-ende> in five of its six occurrences marks a present participle, and tayl-ende as a gerund is a pun retained from Bx. The form <-eng(e)> appears on gerunds in all but six of its twenty-five occurrences. There are no instances of -ing(e), -ind(e), or -ynd(e).

y-armed 15.146; a-basshid 7.305; y-blessed 15.392 ~ y-blessid 11.47; a-cumbred 2.201 ~ a-cumbryd 2.34 ~ a-cu[m]brid 2.194; bowht 1.167; browht 4.141; i-callid 11.245; cast 10.251; clepid 7.432 ~ clepd 4.103; for-beete 16.198; y-huslyd 15.403; lent 5.247; mawngid 5.913; peyed 10.399 ~ y-payed 4.417; y-qwit 13.364; sent 1.72; triȝed 2.205; went (Bx) 4.273; etc.

A few instances of participial adjectives appear with a final <-de>, but usually on strong or plural adjectives or when the adjectives are used as nouns. Otherwise only finite forms have final <-e>, except for callide 16.280, where the suspension is possibly an error. For weak and plural participial adjectives, see crabbide 9.276; crownide 8.323; cursede 5.814; hokede 16.226; (y-)le(e)rnede 4.39, 6.10, 15.88; lettride 7.418; etc.

bakyn "baked" 5.846; bownde 10.724 ~ bownden 10.726; broke 5.1012 ~ brokene 1.62; chose 8.113 ~ chosene 8.114 (serving as a nominal); dryve 5.623; faren 5.3; y-feed 11.318; for-getyn 5.406; gyve 3.122; holpe 4.511, 11.147; knowe 9.93; shryve "shriven" 5.313, 5.422; seyȝ 9.113; wasshe 10.475 ~ waschen 14.394; wrete 8.229 ~ wretyn 7.475 ~ wryten 13.118; wrooke 3.196, 14.392; etc.

Both weak and strong past participles are frequently marked with initial <y-> or (less often [5x]) <i->:

y-bakyn 5.834; i-blessid 8.163; y-born 14.236; i-boterased 5.600; y-broke 15.349; y-fownden 11.246 (varies with the weak form y-fownded 11.341); i-paraylid 5.529; y-seyn 13.359; etc.

Finite forms:

Indicative:

Present

com(e) 7.232; deme 5.115; for-gyve 4.443; heyȝle 5.101; leve 1.32; sey 1.193; stumble 4.498; swere 5.228; walke 5.148; warne 1.199; wysshe 2.43; etc.

dotyst 2.136; dryest 2.25; gredist 15.433; leernyst 4.355; lyvyst 3.126; lyxt 5.164;N myghtyst 1.206 ~ myght 5.875; seest 2.5; wyrchist 4.63; woost 4.168; etc.

a-faytethȝ 10.775; akeþ 5.911; beryþ (10x) 4.403; bit (Bx) 5.1055; bryngþ (3x) 6.281 ~ bryngeþ(1x) 11.84; coueryth 15.299; fareþ (9x) 6.33 ~ faret 10.46; fynt (3x, all Bx) 4.473 ~ fyndes 5.1126 ~ fyndis 11.193, 15.450;N folweth 3.187; for-fret 12.30; gyfþ 4.127 ~ gyveþ 4.327 ~ gyviþ 10.800; halt (Bx) 13.203, 13.243; lykþ 14.32; lyþ 9.299; longeþ 7.143; makþ (9x) 4.112; seyþ 4.224; sekþ 10.574; sleeþ 10.570; smyt (Bx) 9.114; spekþ 7.469 ~ spekeþ 5.1032 ~ spekiþ 11.62; stant (Bx) 14.45; stynkþ 7.317; takþ (13x) 3.34; tellyþ (15x) 1.93 ~ telliþ (9x) 1.95 ~ telleþ (6x) 2.88 ~ tellyt 4.244 (1x) ~ tellethȝ (1x) 9.294; þynkþ 5.423; wytnessyþ 3.76; etc.

a-bide 12.28 ~ a-byden 16.79 (Bx); aske 4.211 ~ asken 5.1066; by-tokneþ 5.1153; burgoneþ 11.84; crave 4.213 ~ craven 4.210; cropyþ (Bx) 5.685; dwellyþ 6.121; fare 8.65 ~ faren 3.185; fecche 5.811 ~ fecchen 4.397; folweth 4.341; helpe 4.232 ~ helpen 5.632; lystneþ 10.788; makþ (Bx) (1x) 6.27; nemeþ 5.667; schryuethȝ 16.288; see 15.249 ~ seen 2.52; seweth 1.41; smerteþ 13.384; teche 7.289 ~ techen 11.104; tristne 4.113; wernethȝ 16.12; etc.

The distribution of these dental suffixes across the text is of some interest. The most common form is <-eþ> with almost 600 instances, most appearing in the last seventy per cent of the text. The others are <-yþ> (162x, mostly in deciles 3-7 of the poem), <-ith> (57x, none in first decile), <-eth> (8x, seven in the first two deciles), <ethȝ> (5x), <-yth> (5x), and <-ythȝ> (3x, all in the last two deciles).

Preterite:

awakid 6.1; a-wakned 15.488; batered 4.187; be-hyghte 14.332; borwede 5.752; clepide 15.9; covrbet 3.1; dyȝede 5.751; drempte 7.503; fawht 14.370; grette 7.231; herde 1.180; kyllyde 4.175; lenede 1.8; lefte 4.185; restid 14.7; wayted 10.354; wente 1.6; etc.

conseyledist 4.194; dyedist 5.496; eggiddist 14.288; feddyst 5.503; madist 5.232 ~ madyst 5.493; reddist 4.250; seydist 9.279; þoledyst 15.176; wroughtist 4.96; etc.

a-cordid 8.41 ~ a-cordyd 4.500;N armed 16.123; askyde 5.77; be-sowhte 2.166; bowht(e) 3.3; brouht 8.3 ~ broghte 11.518; callede 2.4; dyȝede 6.55; enbawmed 13.168; gronede 16.309; hente 5.826; kawht 5.361; lakkede 6.166; parled 5.41; sweltride 16.105; wepte 10.806; etc.

a-mortysyd 11.337; a-noyed 3.168; assentid 1.166 ~ assentide 3.69; belevid 1.63 ~ belevedyn 14.312; carolden 14.433; carpid 7.110 ~ carpedyn 7.112 ~ carpeden 10.223; casten 12.145; helyde 5.845; kneliden 15.84; openeden 14.322; pulliden 5.764; sente 16.307; etc.

be-come 8.202 ~ by-come 15.39; be-knewen 15.151; chose 16.237 ~ chosen 16.236; come 5.1086 ~ comen 15.153 ~ keme 4.389 ~ kemen 1.64; felle 5.386 ~ fellen 12.112 ~ fellyn 2.120; hongyn 2.170; knew 8.241 ~ knewe 6.12 ~ knewen 9.273; kestyn 3.152; lope 4.495 ~ lo(o)pen 2.114, 14.312; lurn 9.247; mette 3.214 ~ mettyn 14.125; stoden 3.73; sungen 14.324; tooke 4.74 ~ token 4.419; woxen 12.60; wooke 15.158; wopyn 5.1021; etc.

Subjunctive

carpe 13.220; fayle 5.669; folwe 6.45; gyve 5.107; lyke 6.47; rede 4.349; werke 4.488; etc.

aske 5.432; carpyn 7.53; coueyte 16.251; don 5.878; grucche 5.869; haven 10.649; lykyn 10.189; lyve 5.45; loven 5.574; mettyn 5.823; reuerence 14.258; sewen 2.184; weryn 8.286; etc.

Imperative

be 16.48; com 5.584; coueyte 5.585; hav 10.528; lakke 3.48; sitte 5.918; etc.

be 3.139; beggiþ 5.1076; byddeþ 5.611; led 11.94; lediþ 3.136; let 5.53; seke 5.59; stynte 5.588; strike 5.589; wadiþ 5.580; weteþ 3.76; wyrcheþ 3.135; wytnessyþ 3.76; etc.

Pronouns:

The first person singular ich does not appear in the manuscript, but the unique reading But Cheeste with Charyte be / y chylle on ȝow pleyne (10.110) shows that some scribe in the tradition used that form at least on occasion. Since R at that point agrees with beta family witnesses, the scribal revision occurred at some point between alpha and the present manuscript. Though such revision is consonant with the other changes we have attributed to the F-Redactor, the palatal is not a feature of his Norfolk dialect. The other spelling suggesting a palatal ich at 5.228, so þeech, is probably to be attributed to alpha, since it is shared by R as so theich. Curiously, perhaps, the archetypal and probably original form there is ik in the phrase "so thee ik." Langland's dialect joke—the same as Chaucer's in The Reeve's Tale—is at the expense of Sire Hervey, the Norfolk manifestation of covetousness.

The feminine singular form he is a relict from Langland's dialect as shown by alliteration in 14.156.N He appears for "she" at 3.134, 4.281, 5.199, 7.499, 8.104, 8.107, 14.156, 14.175, 14.178-80 and 15.161-62, usually reflecting an alpha lection.

The scribe perhaps preserves a for "he" at 5.808, though since most B witnesses read a-bosted the reading is insecure. This perhaps reflects another occasion when the difficilior lectio may not in fact be more probable than the easier reading.

The appearance of he for "they" at 3.149 and 6.243 is owed to alpha, and the scribe at 1.198 appears to have taken it to be singular. See also the note to 10.317.N F substitutes þei for Bx's he at 7.140, 10.615, and 15.33.

The scribe wrote out myn in only four of its 151 occurrences. He appears to use it in free variation with my since it appears immediately before consonants as well as before vowels, both conjunctively and disjunctively.LG

þi and þy appear only conjunctively. Only one of 100 instances appears before a vowel (5.1062). þyn appears both disjunctively and (far more commonly) conjunctively. In some twenty-eight instances, it appears before a vowel, though as with the first person myn the two forms are essentially in free variation.

Of 157 occurrences of hise and 7 of hyse, only one (1.165) appears before a singular noun. Four appear before the formally singular but notionally plural collective nouns meyghne (2.106, 11.152), conseyl (5.971), and felachepe (3.195). All others precede plural nouns or are used disjunctively with a plural reference (10.258, 13.326, 16.61). The singular form his appears with plural nouns only twelve times in over 590 occurrences, and the nineteen occurrences of hys all modify singular nouns. Taken together, only thirteen exceptions in over 775 occurrences of his/hys/ys ~ hise/hyse strongly suggest that the scribe intended to distinguish singular from plural forms.

The neuter forms his and hise appear at F9.375, F13.303, and F15.213 and are in each case archetypal.

IV. List of Manuscript Sigils

The following list of sigils of the manuscripts of Piers Plowman differs in some respects from the traditional sigils used since Skeat's edition. To a degree the inconsistencies in the sigils reflect the sequence of discovery of the relationships among them. If we were to use the traditional sigils, we would court ambiguity in an electronic text with identical sigils representing different manuscripts and different sigils identifying single manuscripts. British Library Additional 10574, for instance, has no sigil for A, is B's Bm, and C's L. We have, therefore, chosen to represent each manuscript with a unique sigil.

For descriptions of the B manuscripts see George Kane and E. Talbot Donaldson, eds., Piers Plowman: The B Version, Will's Visions of Piers Plowman, Do-Well, Do-Better and Do-Best: An Edition in the Form of Trinity College Cambridge MS B.15.17, Corrected and Restored from the Known Evidence, with Variant Readings., rev. ed. (London, Berkeley and Los Angeles, 1988), 1-15; A. I. Doyle, "Remarks on Surviving Manuscripts of Piers Plowman," in Medieval English Religious and Ethical Literature: Essays in Honour of G. H. Russell, ed. G. Kratzmann and James Simpson (Cambridge, 1986), 35-48; and C. David Benson and Lynne Blanchfield, The Manuscripts of Piers Plowman: The B-Version (Cambridge, 1997).

1. B Manuscripts

2. A Manuscripts

3. C Manuscripts

4. AB Splices

5. AC Splices

6. ABC Splices

Bibliography

Editions

Brewer, Charlotte, and A. G. Rigg, eds. Piers Plowman: A Facsimile of the Z-Text of Bodleian Library, Oxford, MS Bodley 851. Cambridge: D. S. Brewer, 1994.

Kane, George, ed. Piers Plowman: The A Version: Will's Visions of Piers Plowman and Do-Wel, An Edition in the Form of Trinity College Cambridge MS R.3.14 Corrected from Other Manuscripts, with Variant Readings. London: Athlone Press, 1960, rev. ed., 1988.

Kane, George, and E. Talbot Donaldson, eds. Piers Plowman: The B Version, 2d ed. London: Athlone Press; Berkeley and Los Angeles: University of California Press, 1988.

Kane, George, and George Russell, eds. Piers Plowman: The C Version. London: Athlone Press; Berkeley and Los Angeles: University of California Press, 1997.

Morris, Richard, ed. The Pricke of Conscience (Stimulus Conscientiae): A Northern Poem by Richard Rolle de Hampole. Berlin, 1863.

Pearsall, Derek, ed. William Langland: Piers Plowman. The C-Text. 2d ed., Exeter Medieval English Texts and Studies. Exeter: University of Exeter Press, 1994.

Rigg, A. G., and Charlotte Brewer, eds. William Langland: Piers Plowman: The Z Version. Toronto: Pontifical Institute of Mediaeval Studies, 1983.

Schmidt, A. V. C., ed. William Langland, The Vision of Piers Plowman. A Critical Edition of the B-Text Based on Trinity College Cambridge MS B.15.17 with selected variant readings, an Introduction, glosses, and a Textual and Literary Commentary. London, Melbourne, and Toronto: J. M. Dent & Sons Ltd.; New York: E. P. Dutton & Co., 1978, 2d ed., London: J. M. Dent & Sons, Ltd.; Rutland Vermont: Charles E. Tuttle Co., 1995.

—, ed. William Langland, Piers Plowman: A Parallel-Text Edition of the A, B, C and Z Versions: Vol. 1. Text. London and New York: Longman, 1995.

Skeat, W. W., ed. The Vision concerning Piers the Plowman together with Vita de Dowel, Dobet, et Dobest secundum Wit and Resoun by William Langland: Part 2. The "Crowley" Text: or Text B. EETS OS 38. London: Oxford University Press, 1869.

—, ed. The Vision of William Concerning Piers the Plowman, in Three Parallel Texts. 2 vols. Oxford: Clarendon Press, 1886.

Studies

Adams, Robert. "Editing Piers Plowman B: The Imperative of an Intermittently Critical Edition." Studies in Bibliography 45 (1992): 31-68.

—. "Langland's Ordinatio: The Visio and the Vita Once More." The Yearbook of Langland Studies 8 (1994): 51-84.

Alford, John A., ed. A Companion to Piers Plowman. Berkeley, Los Angeles, and London: University of California Press, 1988.

—. Piers Plowman: A Guide to the Quotations. Medieval and Renaissance Texts and Studies 77. Binghamton: MRTS, 1992.

Beadle, Richard. "The Medieval Drama of East Anglia: Studies in Dialect, Documentary Records and Stagecraft." D.Phil. thesis, University of York, 1977.

Bennett, Jacob. "The Language and Home of the 'Ludus Coventriae'." Orbis 22 (1973): 43-63.

Benskin, Michael. "In reply to Dr. Burton." Leeds Studies in English 22 (1991): 209-62.

Benson, C. David, and Lynne Blanchfield. The Manuscripts of Piers Plowman: The B Version. Cambridge: D. S. Brewer, 1997.

Bernard, Edward. Catalogi Librorum Manuscriptorum Angliæ et Hiberiæ in unum collecti, cum Indice Alphabetico. Oxford, 1697.

Björkmann, E. Scandinavian Loan Words in Middle English. Studien zur englischen Philologie, 7, 11 (1900, 1902).

Blackman, Elsie. "Notes on the B-Text MSS of Piers Plowman." Journal of English and Germanic Philology 17 (1918): 489-545.

Bowers, John M. "Hoccleve's Two Copies of Lerne to Dye: Implications for Textual Critics." The Papers of the Bibliographical Society of America 83 (1989): 437-72.

—. "Piers Plowman's William Langland: Editing the Text, Writing the Author's Life." Yearbook of Langland Studies 9 (1995): 65-90.

Brewer, Charlotte. "The Textual Principles of Kane's A-Text." Yearbook of Langland Studies 3 (1989): 67-90.

—. "Authorial vs. Scribal Writing in Piers Plowman." In Medieval Literature, Texts and Interpretation, ed. Tim William Machan. Binghamton, NY: MRTS, 1991.

—. "George Kane's Processes of Revision." In Crux and Controversy in Middle English Textual Criticism, ed. A. J. Minnis and Charlotte Brewer. Cambridge: D. S. Brewer, 1992.

—. Editing Piers Plowman: The Evolution of the Text. Cambridge Studies in Medieval Literature, no. 28. Cambridge: Cambridge University Press, 1996.

Bright, Allan H. New Light on Piers Plowman. Oxford: Oxford University Press, 1950.

Briquet, Charles M. Les filigranes, dictionnaire historique des marques du papier jusqu'en 1660: A Facsimile of the 1907 Edition, ed. Allan Stevenson. Amsterdam: The Paper Publications Society, 1968. (Original edition Paris, 1907).

Brown, Carlton, and R. H. Robbins, eds. Index of Middle English Verse. New York: Printed for the Index Society by Columbia University Press, 1943.

Burton, T. L. "On the Current State of Middle English Dialectology." Leeds Studies in English 22 (1991): 167-208.

Chambers, R. W. "The Authorship of 'Piers Plowman'." Modern Language Review 5 (1910): 1-32.

—, and J. H. G. Grattan. "The Text of 'Piers Plowman'." Modern Language Review 11 (1916): 257-75; 26 (1931): 1-51.

Colledge, Edmund, O.S.A. and Cyril Smetana, O.S.A., "Capgrave's Life of St. Norbert: Diction, Dialect, and Spelling." Mediaeval Studies 34 (1972): 422-34.

Coxe, Henry O. Catalogus codicum MSS. Qui in Collegiis Aulisque Oxoniensibus hodie adservantur. Oxford: E Typographeo Academico, 1852.

Dahl, Eric. "Diverse Copies Have It Diverslye: An Unorthodox Survey of Piers Plowman Textual Scholarship from Crowley to Skeat." In Suche Werkis to Werche: Essays on Piers Plowman In Honor of David C. Fowler, ed. M. F. Vaughan. East Lansing, MI: Colleagues Press, 1993.

Davis, Norman. "A Paston Hand." Review of English Studies n.s. 3 (1952): 209-23.

Dibelius, Wilhelm. "John Capgrave und die englische Schriftsprache." Anglia 23 (1901): 153-94, 323-75, 429-72; 24 (1901): 111-63, 269-308.

Donaldson, E. Talbot. "MSS R and F in the B-Tradition of Piers Plowman." Transactions of the Connecticut Academy of Arts and Sciences 39 (1955): 177-212.

Doyle, A. I. "Remarks on Surviving Manuscripts of Piers Plowman." In Medieval English Religious and Ethical Literature: Essays in Honour of G. H. Russell, ed. Gregory Kratzmann and James Simpson. Cambridge: D. S. Brewer, 1986.

Duffy, Eamon. The Stripping of the Altars: Traditional Religion in England, 1400-1580. New Haven: Yale University Press, 1992.

Duggan, Hoyt N. "Langland's Dialect and Final -e." Studies in the Age of Chaucer 12 (1990): 157-91.

Duggan, Hoyt N. "On Constructing Documentary Texts for The Piers Plowman Electronic Archive." In Rationality and the Liberal Spirit: A Festschift Honoring Ira Lee Morgan, Shreveport: Centenary College of Louisiana, 1997.

Ekwall, Eilert. The Concise Oxford Dictionary of English Place-Names. 4th ed. Oxford: Clarendon Press, 1960.

Forsström, G. The Verb "To Be" in Middle English: A Survey of the Forms. Lund Studies in English, no. 15. Lund: C. W. K. Glerrup, 1948.

Gelling, Margaret. "English Place-Names Derived from the Compound wīchām." In Place-Name Evidence for the Anglo-Saxon Invasion and Scandinavian Settlements, ed. Kenneth Cameron. Nottingham: English Place-Name Society, 1987.

—. Signposts to the Past: Place-Names and the History of England. 2d ed. Chester: Phillimore, 1988.

Hanna, Ralph, III. William Langland. Authors of the Middle Ages, no. 3. Aldershot and Brookfield, VT: Variorum, 1993.

—. "On the Versions of Piers Plowman." In Pursuing History: Middle English Manuscripts and Their Texts. Stanford: Stanford University Press, 1996.

Hecht, Hans, and Levin L. Schücking. Die Englische Literatur im Mittelalter. Wildpark-Potsdam: Akademische Verlagsgesellschaft Athenaion, 1927.

Jordan, Richard. Handbook of Middle English Grammar: Phonology. Trans. and rev. Eugene J. Crook. Janua Linguarum, Series Practica, no. 218. The Hague and Paris: Mouton, 1974.

Justice, Stephen., and Katheryn Kerby-Fulton, eds. Written Work: Langland, Labor, and Authorship. Philadelphia: University of Pennsylvania Press, 1997.

Kane, George. "The Text." In A Companion to Piers Plowman, ed. John Alford. Berkeley and Los Angeles: University of California Press, 1988.

Ker, Neil R. Catalogue of Manuscripts Containing Anglo-Saxon. Oxford: Clarendon Press, 1957.

Ker, Neil, and A. J. Piper. Medieval Manuscripts in British Libraries. Oxford: Clarendon Press, 1992.

Kihlbom, Asta. A Contribution to the Study of Fifteenth-Century English. Uppsala Universitets Årsskrift. Uppsala: A. -B. Lundequistska Bokhandeln, 1926.

Lucas, Peter J. "Consistency and Correctness in the Orthographic Usage of John Capgrave's Chronicle." Studia Neophilologica 45 (1973): 323-55.

—. "An author as copyist of his own work: John Capgrave OSA (1393-1464)." In New Science Out of Old Books: Studies in Manuscripts and Early Printed Books in Honour of A. I. Doyle, ed. Richard Beadle and A. J. Piper. Aldershot, Hants.: Scholar Press, 1995.

McIntosh, Angus. "Word Geography in the Lexicography of Medieval English." Annals of the New York Academy of Sciences 211 (1973): 55-66.

McIntosh, Angus, and M. L. Samuels. "Prolegomena to a Study of Medieval Anglo-Irish." Medium Ævum 37 (1968): 1-11.

McIntosh, Angus, M. L. Samuels, and Michael Benskin, eds. A Linguistic Atlas of Late Mediaeval English. 4 vols. Aberdeen: Aberdeen University Press, 1986.

Metzger, Bruce. The Text of the New Testament: Its Transmission, Corruption, and Restoration. 3d ed. New York and Oxford: Oxford University Press, 1964.

Mills, A. D. A Dictionary of English Place Names. Oxford: Oxford University Press, 1991.

Mustanoja, Tauno F. A Middle English Syntax: Part I: Parts of Speech. Mémoires de la Société Néophilologique de Helsinki, no. 23. Helsinki: Société Néophilologique, 1960.

Parkes, M. B., and Richard Beadle, eds. The Poetical Works of Geoffrey Chaucer: A Facsimile of Cambridge University Library MS GG.4.27. 3 vols. Norman, OK: Pilgrim Press, 1979.

Patterson, Lee. "The Logic of Textual Criticism and the Way of Genius." In Textual Criticism and Literary Interpretation, ed. Jerome J. McGann. Chicago: University of Chicago Press, 1985.

Pearsall, Derek. "Editing Middle English Texts." In Textual Criticism and Literary Interpretation, ed. Jerome J. McGann. Chicago: University of Chicago Press, 1985.

—. "Texts, Textual Criticism, and Fifteenth-Century Manuscript Production." In Fifteenth-Century Studies: Recent Essays, ed. Robert F. Yeager. Hamden, CT: Archon Books, 1984.

Reaney, P. H. The Place-Names of Essex. English Place Name Society 12. Cambridge: Cambridge University Press, 1935.

Robinson, Peter M. W. The Transcription of Primary Resources Using SGML. Office for Humanities Communication Publications, No. 6. Oxford: Oxford University Computing Services, 1994.

—, and Elizabeth Solopova. "Guidelines for the Transcription of the Manuscripts of the Wife of Bath's Prologue." In Canterbury Tales Project Occasional Papers 1, eds. Norman Blake and Peter M. W. Robinson. Oxford: Office for Humanities Communication, 1993.

Samuels, M. L. "Langland's Dialect." Medium Ævum 54 (1985): 232-47.

—. "Dialect and Grammar." In A Companion to Piers Plowman, ed. John A. Alford. Berkeley and Los Angeles: University of California Press, 1988.

—. "Scribes and Manuscript Traditions." In Regionalism in Late Medieval Manuscripts and Texts: Essays Celebrating the Publication of A Linguistic Atlas of Late Mediaeval English, ed. Felicity Riddy. Cambridge: D. S. Brewer, 1991.

Scase, Wendy. Piers Plowman and the New Anticlericalism. Cambridge: Cambridge University Press, 1989.

Scott, Kathleen L. "The Illustrations of Piers Plowman in Bodleian Library MS Douce 104." Yearbook of Langland Studies 4 (1990): 1-86.

—. Later Gothic Manuscripts 1390-1490. London: H. Miller, 1996.

Seymour, M. C. "A Fifteenth-Century East Anglian Scribe." Medium Ævum 37 (1968): 166-73.

Stephen, Leslie, and Sidney Lee. The Dictionary of National Biography. Vol. 7. Oxford: Oxford University Press, 1917.

Spector, Stephen, ed. The N-Town Play: Cotton MS Vespasian D.8.1. Early English Text Society, Special Series 11-12. Oxford: Oxford University Press, 1991.

Sullivan, Sister Carmeline. The Latin Insertions and Macaronic Verses in Piers Plowman. Washington, DC: Catholic University of America Press, 1932.

Taylor, Sean. "The F Scribe and the R Manuscript of Piers Plowman B." English Studies 77 (1996): 530-48.

Uhart, Marie-Claire. "The Early Reception of Piers Plowman." Ph.D. diss., University of Leicester, 1986.

Weldon, James. "Ordinatio and Genre in MS CCC 201: A Mediaeval Reading of the B-Text of Piers Plowman." Florilegium 12 (1995, for 1993): 159-75.

Westcott, Brooke Foss, and Fenton John Anthony Hort, eds. The New Testament in the Original Greek. London and New York: Macmillan, 1896.

Windeatt, Barry A. "The Scribes as Chaucer's Early Critics." Studies in the Age of Chaucer 1 (1979): 119-41.

Wyld, Henry C. "South-Eastern and South-Eastern Midland Dialects in Middle English." Essays and Studies 5 (1920): 112-45.