Introduction

I. Description of the Manuscript: London, British Library, MS Additional 35287

I.1 Date:

S. xv in. The text is in one anglicana hand, though the writing varies considerably. George Kane and E. Talbot Donaldson report that "A. I. Doyle discounts the view that more than one scribe was concerned, pointing out that the formality and size vary within continuous passages, and the constituents of the writing are the same throughout. He fixes a point on fol. 42b where there is a change to a lower (medium) grade of script, but emphasizes the scribe's ability and finds him, on fols. 19-23, writing a script like that of the Ellesmere MS."NGeorge Kane and E. Talbot Donaldson, eds., Piers Plowman: The B Version, rev. ed. (London, 1988), p. 11, n. 73. Doyle repeats his view that the handwriting "sometimes [has] strong resemblances to that of Hengwrt/Ellesmere," ("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, Eng., 1986), p. 39).

I.2 Contents:

104 vellum leaves, containing only the B version of Piers Plowman, beginning:

In a somer seson  whanne softe was the sonne

And sith he gradde after grace . til I gan a-wake Explicit hic dialogus petri plowman

Other manuscripts of the B version with this explicit are C, C2, G, L, O, W and Y.NThere are two further explicits beneath, the first "Penna precor siste  quoniam liber explicit iste," probably in text hand, the second "Explicit iste liber . qui obsecro transeat liber," in a later hand. For comments on these, see the annotation there.

I.3 Collation:

The 104 leaves are foliated in modern pencil. There are 13 quires of 8, with catchwords at the end of every quire except the fourth,NPassus 7 begins at the top of the fifth quire. hence at the foot of fols. 8v, 16v, 24v, 40v, 48v, 56v, 64v, 72v, 80v, 88v, 96v. Quires 1-11 are numbered above or below the catchwords, with the seventh quire also numbered at the top of 49r. Some leaf signatures are visible, e.g. on fols. 1v-4r (aj-aiiij), 9r (bj), 39r-40r (Evij, Eviij), 49r-52r (gi-giiij), the last on 98r (Nij).

I.4 Physical Description:

Overall 305 x 195 mm. Cropping has not affected the text, and the pricking is visible on many folios, but has perhaps been cropped in quire 10 (fols. 81-88). Some leaf signatures have perhaps been lost by cropping. The leaves of the last quire (fols. 97-104) are cut irregularly at the foot. There is considerable worm-damage to the first folios; otherwise the manuscript is in reasonably good condition.

I.5 Arrangement of the Page:

Ruled in brown lead for 39-42 lines, the border consisting of a double frame on each side and below, with a double line forming a separated column for initial letters. This ruling, though elaborate, is often crudely done, extending beyond its bounds. Fols. 5rI and 7vI give good examples of the arrangement. The area ruled for writing is 225 x 140 mm. Through the first folio the scribe mistakenly begins the line to the right of the separated column; thereafter the initial letters are offset within the column, though the scribe is not entirely comfortable with the arrangement and becomes more haphazard as the copying proceeds. On 47r,I for example, the first line begins to the right of the box but subsequent initial letters are correctly placed. Among B manuscripts, L and R similarly have offset initial letters.

I.6 Handwriting: The Main Scribe:

The main body of English text is written in anglicana formata, varying in quality and detail over different copying stints. A good example of such variability is fol. 23v,I where the hand gets larger and less controlled until the middle of the page, at which point the scribe begins to write a much smaller and tidier script. At first glance upper and lower parts of the page look quite different, but the details of the script show it to be the same hand. There is a general degeneration in the script as the writing progresses through the manuscript, with less control and precision in letter-forms. There is certainly a general resemblance to the Hengwrt/Ellesmere scribe, and indeed to the scribe of W, as is to be expected of anglicana hands of similar training, but there are considerable differences of detail.

The scribe uses several methods to call attention to particular words and phrases. Red ink is used for passus headings and the ornamental capitals at the beginnings of the passus. Many Latin quotations and phrases are surrounded by a red ink box (not always completed), as are some individual words and phrases. Shorter stretches of Latin and some proper names are often simply underlined in red ink.NFor a list of "Words and Lines Emphasized" in manuscripts of the B version, see C. David Benson and Lynne S. Blanchfield, The Manuscripts of Piers Plowman: The B-version (Cambridge, Eng., 1997). Those in M are listed on pp. 170-1, with comparative tables on pp. 238-313. These include the references to longe wille (15.158) and Chichestre as Maire (13.274). Very infrequently, this underlining extends to material in the margins, such as the names of the deadly sins (5.65ff). In places where omitted lines have been added outside their expected place in the text, red lines and other signs often call attention to their proper placement (e.g. P.141-145). Less frequently, text has been underlined in brown ink. Many of these instances are additions by later readers (at P.132-138, for instance, the underlining is clearly attributable to the hand that wrote vox angeli in the margin in an italic script) but it also seems likely that many are by the main scribe (reddite at 5.474, for instance). From the middle of fol. 44v (10.274), the scribe began to write Latin text and many proper names in a noticeably larger and bolder script, though this is not entirely consistent (compare the enlarged script of pecuniosus in 11.58 with the Latin line that follows).I We have marked instances of this display script, including the one instance before 10.274 (5.173), with the attribute "display."

There is a carelessness about the scribe's work, evident in the frequent repetitions and omissions of words that then have to be corrected, in the eclectic underlining of the Latin lines in red or in text ink, and the lack of control over the vertical alignment of text (e.g. on fol. 88v).I Overall the manuscript has a rather unfinished appearance with a handful of paraphs indicated but none executed.

The main scribe uses a number of variant letter forms, which we here review selectively. There is little sense of development away from the more complex anglicana forms of some letters towards simpler secretary forms.

The scribe uses three forms of <A>, the most common an enlarged version of the double-chambered <a>, with a variant form where the top loop embraces the body of the letter.I The straight two-legged form is also frequent.I For the variety see the initial letters of 4.59-61. Another form is seen in And (7.28). The forms of <a> vary between anglicana and secretary versions, though not in any discernible pattern. Thus the ratio of anglicana double-chambered <a> to the single-chambered secretary version is 17:1 on 2r, but on 5v the secretary version predominates 3:1 (both in the first quire). In quire 11 on 85v there are twice as many double as single; on 101v in the final quire there are none of the secretary forms.NMS. W uses the double-chambered <a> almost exclusively. The single-chambered <a> is found in the Hengwrt-Ellesmere scribe's side-notes, but very rarely in the formal script of the text.

<B> has two distinct forms, one resembling the modern form (14.61),I the other considerably more complex (Bedleem [17.123]).I

Double <c> is not distinguishable from double <t>: see peccat (17.200).I

<D> may be an enlargement of the small letter, as in Deþ (18.30)I or a circular form as in Diden (14.89).I

<E> may be circular in shape, as in Englond (P.179),I or with a point on the left, as in Er (2.206,I 6.186I). The enclosed form of <e> is generally that used at the end of a word, the open form medially, though when the latter is used finally it often has an extra loop. See comaunde, conscience (4.6).I When carelessly written, the open form can resemble <o> e.g. catel (14.8).I

The single-chambered form of <g> is never used early in the manuscript but it is sometimes found later, e.g. vengeaunce vengeaunce (17.296)I.

<h> has a descender tucked under its body, e.g hire (3.47),I or pointing right, particularly at the end of words, e.g. Flessh (3.55),I hadde (6.246).I Both may appear medially.II <H> is a larger and more elaborate version of <h> with a loop on the ascender as well as a looping descender often curved back to touch the body, e.g. Hendeliche (3.29).I

<I> has a hook, e. g. I (3.60),Ior sometimes a right-angled top, as in I[t] (5.210)I; the letter has sometimes been rewritten to widen the angle of the top.I Both forms appear in free variation.I In proximity to other minims <i> has a curved tick.I

The shapes of <n> and <u> are often clearly distinguished: compare lened and leued (13.387).I

<P> is often distinguished from <p> by means of a dot or a small line as in Piers (7.148).I On fewer occasions, the distinction is established by size alone, as in Piers (6.203).I

Long <r> is often joined to the next letter, as in apparailled (2.217).I The 2-shaped form of <r>I is used particularly after <o> and may have a tail, e.g. for 18.36.I

Long <s>I is used initially and medially, as in falsnesse (2.203),I sigma <s>I initially and finally, as in shewen (2.216),I and 8-shaped <s>I finally, as in churches (2.224),I though later in the text final <-s> is more commonly of the sigma form except for the more formal Latin lines. <S> is an enlarged form of sigma <s> e.g. Shal (3.32).I An unusually elaborate initial <S> appears in the last line of fol. 94v (19.230).I

Two forms of <v> are used, with the first stroke curving beneath, as in vp (5.624)I or, rarely, rising above, as in evam (5.626).I

The form of <w> is indistinguishable from <W>. The second element is double-chambered with a top loop often curving over to touch the body of the letter. The vertical strokes are sometimes straight, sometimes curvy, sometimes angled (for several examples see 3.61,I and compare where-of [3.56]).I

The forms of <y>I and <þ>I are quite distinct, since <y> has a sweeping descender, so that the dot occasionally written above it is unnecessary.

There are often flourishes on final <-c>,II <-d>,I <-g>,I <-k>,II <-p>,III <-r>,I <-t>,I and bars through <-h>II and <-ll>;II for further description of these and their interpretation see III.1 Transcription of the Manuscript.

For discussion of the paleography of the correcting hands see II. The Text and Its Correctors.

I.7 Decoration and Textual Presentation:

The largest textual division of Piers is the passus, clearly marked with a centered bastard anglicana Latin incipit in red. The first passus heading is Passus primus de visione (fol. 4r)I and, apart from Passus iij(us), they continue in that style up to Passus 11.NThe rubricator's heading Passus x(us) for Passus 11 has been corrected in brown ink. Passus 8 adds & incipit in(qui)sicio p(rima) de dowell, and Passus 10 adds et ij(us) de Dowel. From Passus 12 onwards the passus simply have the number. Other B manuscripts to have this scheme are R and S.NFor a study see Robert Adams, "The Reliability of the Rubrics in the B text of Piers Plowman," Medium Ævum 54 (1985), 208-31, and "Langland's Ordinatio: The Visio and the Vita Once More," Yearbook of Langland Studies 8 (1994), 51-84. For a contrary position, see Lawrence M. Clopper, "A Response to Robert Adams, 'Langland's Ordinatio,'" Yearbook of Langland Studies 9 (1995), 141-146. However, the guides to the rubricator include more extensive instructions that are often legible or partly so, especially for the last five passus (e.g. at the foot of 77r, pass(us) xvj(us) & p(ri)m(us) de dobet,I on 86r in large faint script ... xviij(us) & iije ...,I on 98v, pass(us) xx(us) et .... de dobeste).I It appears that the scheme first intended for M was very similar to that envisaged for L, and in L also the instructions to the rubricator are not always followed.

Each passus begins with a plain ornamental capital in red. On fol. 1r this capital <I> is 10 lines tall.I Other capitals are smaller, 3-5 lines tall. Those beginning Passus 4 and 5 have simple infilling of foliage (15v);I that to Passus 3 has the infill erased (10v).I On fol. 81r the initial for Passus 7 is an unusually small and clumsy <I> in a red ink which is different from that of the passus heading, and the opening lines are not indented to allow for anything larger, though a very large <I> is outlined in text ink beside it.I The passus number is recorded in a later hand at the top of almost every page from fol. 4r, "I(us) pass(us)," and onwards.

As do manuscripts WLRYNA. I. Doyle, "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, Eng., 1986), pp. 39-40, compares LMRWY, B text manuscripts that may come from a commercial London workshop but have West Midland dialect characteristics. M divides the poem into verse paragraphs, but no colored paraphs have been supplied in M as they have in WLRY. Double virgule guidemarks for paraphs occasionally appear on top lines because in this position there was no preceding blank space to act as a guide. There are also a few of the <cc> type paraphs in text ink that were perhaps intended for rubrication, though others are probably later, not always easy to distinguish from "n(ota)" (e.g. 84r top).I An interesting and rather puzzling example is the first leaf of the second quire, fol. 9r,I which begins with a double virgule in the box for the initial capital, and 14 lines down at the beginning of the third paragraph has <cc> in the box. On fol. 2r, P.83I has a <cc> where, perhaps by oversight, no space has been left between lines, since WLY all have a space and a paraph (R is lacking). We have recorded these marks as paraphs where we are reasonably confident that is intended.

I.8 Punctuation:

Each line of Piers Plowman begins with a littera notabilior, and we have so interpreted those that are not distinct from the small form of the letter. The caesura is regularly marked with a punctus, virgule, or punctus elevatus. The original scribe normally used a punctus, but the corrector often altered this to a punctus elevatus, and it is evident that the punctus elevatus was often an addition made in conjunction with other corrections. There are clear examples of such corrections in 3.29 where a punctus is converted to a punctus elevatus, in 3.31 where a punctus elevatus is added where there was nothing before, and in 3.33, where a punctus elevatus is written over an erased <-e>.I This pattern of correction occurs up until 59r, with a faint punctus elevatus at the top of 59v (13.104). After that the plain punctus is regular, with an occasional punctus elevatus, though there is a significant increase in the use of virgules in Passus 17 and 18.

A final punctus is sometimes used at the end of the line, but some of these are probably not deliberate, and an end-of-line punctus elevatus occurs rarely.

Other punctuation appears occasionally. Virgules sometimes emphasize phrasing, in both Latin (1.181, 7.65) and English (3.270, 11.49, 12.35, 12.204). A punctus is sometimes used for the same end (7.129, 9.134, 15.493) or to set off individual words for emphasis (8.125).

I.9 Marginalia:

More than with any other B manuscript, readers of M have filled its margins with notae, comments, finding aids, and other jottings. This process began with the main scribe himself and was continued by several others. Perhaps the obvious, and often sloppy, presence of thousands of "improvements" in the orthography of the text of the poem itself (see section II. The Text and Its Correctors) diminished the manuscript's finished appearance and encouraged later readers to feel free to appropriate the manuscript's temptingly generous margins for their own interests and needs.

There are several distinct strands among the general tangle of marginal annotations. But each strand is itself so variable in appearance, and the individual annotations so brief, that assigning any given annotation to a specific strand is a very tentative enterprise, and we offer our conclusions on this matter with some diffidence. The variability can be seen quite clearly even in those annotations which were made by the main scribe himself and were authorized in the manuscript's process of rubrication. These include the names of the sins on folios 19v-24r, and three brief markers: nota de fratribus (70r), quid est caritas (71r), and nota (103r). As a group, these annotations vary considerably in script and the rubrication is sometimes a box, sometimes simple underlining. Taken out of context, one might not decide to attribute them to the same agent, but in context one must. This realization has led us to be generous in attributing multiple annotations to the same agent.

In addition to the annotations noted above, we have also attributed others to the main scribe: the various catch words and two of the three explicits on fol. 104r.

As Hand2 we have classified the corrections within the line that the available evidence suggests are not by the main scribe, and we have, as a matter of convenient conjecture, assigned textual corrections in the margin to the same hand (see section II.1 Correcting hands).

We have sorted the rest of the annotations into several categories. The vast majority we have assigned to a hand we label Hand3, a hand we believe to date from the mid- to late-fifteenth century. These annotations, sometimes written in Latin, sometimes in English, and sometimes in a mixture of the two, appear in a variety of formats. Often, but not always, they are preceded by a flag-shaped longa, <#>, a punning form of nota from musical notation, as is indicated by the presence of the superscript <a> on a <#> on fol. 70v.I When an annotation is long enough to require two or more lines in its writing, the second line sometimes begins immediately under the <#> and sometimes immediately under the text following the <#>. This variety of formats and languages, as well as the variety of shades of ink and nib, suggests that the annotations were made on different occasions, perhaps spread over a significant period of time. The great majority of the annotations are markers to the action or to the argument, such as thassemble at medis mariage (fol. 8r). A whole series of notes draws attention to definitions of Dowell, beginning on 37r with Prima discripcio de dowell and ending on 64r with decima [recte undecima] discripcio de dowell. It would be tempting to connect this determined attempt to enumerate ten definitions of Dowell to the note on fol. 104v enumerating ten post-resurrection appearances of Jesus, but the hands, though somewhat similar, are not the same. Though this hand largely confines itself to annotation, we believe that on fol. 1v it altered the text of the poem itself at P.58, P.66, and P.67.

About 30 instances of annotation appear in a hand that is neater, smaller, and more compact than the annotations tagged as Hand3. These never appear with the <#> and are always written in Latin. We have tagged them as Hand4.I

One other series of annotations is widely represented in the manuscript, a hand that has written a series of running headers, generally in the upper right corner of the rectos of the manuscript, but sometimes centered on the page (fols. 11r, 22r, 34r, 35r, etc.) Most often these are simply the word passus and a roman numeral, but on occasion (fols. 78r , 79r, 81r, 86r, etc.) the unit is designated liber rather than passus. On fols. 87r and 91r this hand adds de passione to the formula. We believe that this hand, which sometimes uses a distinctive form of underlining, is responsible for eleven other annotations in the manuscript. All these are tagged as Hand5.

There remain several dozen other annotations and many solitary marks, notae, paraph marks, x's and other jottings which cannot be identified with any of these hands. These we have lumped together under the tag of HandX. Four of these (attached to P.132,I 6.96,I 6.147,I 15.537I) are written in distinctively italic hands, probably from the sixteenth- or seventeenth-century, but the hands are not all identical.

The annotations are themselves occasionally altered. The notation at the top of the first leaf has been revised, possibly to accord to Protestant sympathies. The note at 9.61 has been edited. At 12.284, an earlier annotation has been replaced by a later one. And in most cases, a correction to the text noted in the margin has been erased after the correction was entered into the text.

The manuscript's last leaf, fol. 104v,I contains many jottings in a variety of later hands, unrelated to the poem Piers Plowman but perhaps of interest. The earliest, at the top, is a Latin nota concerning the ten post-resurrection appearances of Christ:

Nota quod in euaungelio legitur Cristus decies apparuisse . primo Magdalene . Iohanne 20 . & Marci vltimo ./ secundo alijs mulieribus cum predicta Magdalena . Iohanne 20 ./ tertio petro lucae vltimo ./ quarto duobus discipulis euntibus in emaus lucae vltimo. quinto discipulis absque thoma . Iohanne 20 . sexto discipulis presente thoma . Iohanne 20 . septimo ad mare tiberiadis Iohanne 21 . octavo in galilea in monte vbi constituerat ille Iesus . Mathei vltimo ./ nono in ierusalem quando comedit cum eis partem pissis assi & fauum mellis . lucae vltimo . decimo in monte oliueti quando ascendit . lucae octavo & Marci vltimo .

Beneath this list, the leaf is dominated by a competent drawing of an elaborate mace standing on a base; the spike of the mace is not completed where it meets the Latin nota, which must therefore have preceded it. A scroll beneath the foot reads:

With this mace . be he smeteThat al the worlde . may it weteThat geuyth a-way . his owne thyngAnd goth hymsilff . a-Beggyng

To the left of the mace are two lines in Latin, not identified and only partially decipherable:

decepit multos fallax dilectio stultos Si modo plus dicam, faciam mihi tunc q........N We are grateful to Professor Gregory Hays of the Classics Department, the University of Virginia, for helping us transcribe these hexameters.

Beneath these lines, in a different hand and ink, an unidentified line of Latin:

Tristia post leta . post tristia sepe quieta

danet the.

To the right of the mace are a set of pen trials, beneath which is a brief memorandum of expenses for Bruges satin in an early sixteenth-century hand:

Item for a yerd and a quarter of saten ebregys ijs vj d

Below these two columns of jottings, running across the shaft of the mace, a mid sixteenth-century italic hand has written two lines from Ovid, Heroides V.7-8:

Leniter ex merito quicquid patiare ferendum estQue uenit indigno pena dolenda venit

I.10 Binding:

The binding is modern, in red leather.III There are two modern flyleaves frontII,II and back,II,II the first backfly with a note dated "Oct. 10th /99," recording that the manuscript was examined by E. S.

I.11 Provenance:

According to the note on the first fly leaf, the manuscript was bought by the British Museum at Sotheby's as lot 77 on 1 May 1899. Like Hm, M had formerly been in the Ashburnham collection, Append. no. 129.I Its earlier history is unknown. The common place-name or personal name Mordou(n) is written in the right margins of 65v and 100v, and Joh(a)nes scribbled in faint pencil on 72v. The monogram DEN dated 1545 on 104v has not been identified.I

I.12 Previous Descriptions:

C. David Benson and Lynne S. Blanchfield, The Manuscripts of Piers Plowman: the B-version (Cambridge, Eng., 1997), pp. 64-67 and 169-183.

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. Gregory Kratzmann and James Simpson (Cambridge, Eng., 1986), pp. 35-48.

George Kane and E. Talbot Donaldson, eds., Piers Plowman: The B Version, rev. ed. (London, 1988), p. 11.

II. The Text and Its Correctors:

II.1 Correcting Hands:

The work of the main scribe has undergone extensive correction by contemporary hands, usually over erasure, and this attention has been recorded by the note cor in red and sometimes in black as well at the end of quires 2 (16v),I 4 (32v),I 7 (64v),I 10 (80v),I and 11 (88v),I in red only on 3 (24v)I and 8 (72v),I and in black only at the end of the first quire (8v) (together with examinatur underlined in red),I on fols. 12r,I 16r,I 23r,I 33r,I 37r,I 47vI (head of Passus 11), 50v,I 53vI (head of Passus 12), 57v,I 58r,I 61r,I 69r,I 77r,I and 83r.I It is difficult or impossible to identify with confidence the correctors responsible for erasures and small alterations, and even in cases of more substantial rewriting identification is complicated by the variability in the main scribe's hand. Corrections are made by insertion or erasure; the erasure is often curiously half-hearted and it is then possible to read what has been erased, often without special lighting. On other occasions it is necessary to indulge in a certain amount of guesswork to establish the erased word, but since most corrections conform to a set of formulas, this is not as hazardous as it may seem. There are still a good many instances where we have been obliged to indicate that the erased text is illegible, and also, caveat lector, there are undoubtedly cases where we have not noticed that a word is written over an erasure or a letter inserted later.

The identity of correcting hands is often difficult to determine, so it is as well to begin with instances where there can be little doubt. Hand1, that of the main scribe, is certainly responsible for some of the corrections, those entered in the process of transcription. Every now and again he anticipates a word, a phrase or a half-line, erases his error and adds what he has missed in his recopying. Clear examples of such self-correction may be seen in 17.43,I where the scribe anticipates Abraham and corrects it to any man, and 17.277, where the scribe erases doþ and writes the longer word fordoþ over it before completing the line.I Other near-certain cases of Hand1 are where a missing line or passage has been inserted into a space between lines or in the margin, and subsequently corrected for spelling by Hand2. On fol. 90r,I for example, the scribe misses three lines (18.328-30). One is inserted in the blank space between paragraphs, and two are added at the foot of the page, with two carets to mark the place for insertion. That the ink is darker and the nib sharper suggests that the addition was made later; it was subsequently revised by erasure. The insertion of 18.395-7 (again subsequently revised) was perhaps a self-correction at the time of copying, since the nib and ink match the text.I All those alterations that are themselves corrected for spelling are probably in the main hand. 10.416 offers another example of what is probably the main hand supplying a missed line, in a darker ink and corrected for spelling.I In this case there is evidence that the exemplar also omitted the line, which might suggest that the main scribe had access to a second exemplar for correction. It is likely that Hand1 went back over his work correcting it against the exemplar(s) he had been following. In 8.45 it looks like the main scribe who has corrected to keputh with the Western spelling of the third person singular ending; yet the erased marginal correction reads kepeth. Possibly this indicates that someone else was involved in checking the scribe's accuracy, but there is no clear evidence of such co-operation. However, brief insertions are often very difficult to assign. In 15.382 the inserted newe is in the same ink and same style as the surrounding text, and has two other characteristics of probable Hand1 insertions: the apparent absence of an erased guide word in the margin, and two carets, one beside the insertion and one at the appropriate point within the line.I

The majority of the corrections are in hands that resemble that of the main scribe but are sufficiently different to incline us to the view that they were executed by correctors trained in the same "school" as the main scribe. These revisions introduced at some point after the copying of the text often present problems of identification. They seem to have been made piecemeal: the color of the ink and the thickness of the nib change frequently, often after only a few leaves. We can be sure there were at least two other revisers who have signified their corrections with two styles of corrector's marks throughout the manuscript. We have found no way of regularly distinguishing them, so our designation Hand2 necessarily refers to two or more contemporary correcting scribes. In the absence of distinctive scribal habits, a number of factors may be adduced for ascribing corrections to this group Hand2 rather than to the main scribe. Most such corrections were done after the manuscript had been rubricated, so that the rubricated boxes of surrounding lines are partly erased, as in 5.507I and 7.65.I The pervasive revision of Hand1's spelling system, such as the erasure or addition of final <e>, may be credited to Hand2, and presumably to a single scribe. Hand2 corrections and additions adopt the revised spelling system which is often distinguishable from that of Hand1, such as yow (6.121 and 14.161) and heed (1.166, with heued in the margin), forms the main scribe never uses. At least one corrector revised the text by comparison with one or more manuscripts of the poem related to WCr, and we have found no clear evidence that Hand1 corrections are of this type. Corrections made by Hand2 are very rarely themselves corrected, and then never for spelling. 15.77 is an exception: a Hand2 correction later revised in the direction of the W reading. We conclude that the great majority of corrections were made by Hand2, and so we assign a correction to Hand2 unless there is reasonably secure evidence to show that it is attributable to the main scribe.

The most extensive example of a correcting hand that is similar to but probably not that of the main scribe is that of the inserted lines 17.185-6 on fol. 83v.I In corrections the <y> that so often replaces <ȝ> usually has a dot above it, whereas original <y> usually does not. The open form of <e> is preferred, even at the end of words, and there is a preference for the single-chambered form of <a>. The trouble is that these simpler letter forms are also used by Hand1 in corrections, e.g. in 10.416.I The problems with identifying distinctive palaeographic features of the correctors may be further illustrated by tracking examples of corrected <v>: in hovues (P.211),I <v> has an exaggerated forward loop not characteristic of the main scribe, and the smaller letter in vnfolden (2.74)I and verrey (15.274)I is pretty certainly the same hand. Quite different, though even more dramatic, is the sweeping lead-in stroke of the <v> in prives (2.180),I and the disproportionate stroke in Davy (5.325),I but are these two by the same corrector? We judge vnrostud (5.625),I with a small loop beneath the letter, to be a correction by Hand1, which may be compared with Hand1's similar form in virginem immediately below. Yet vncomely (9.173),I with the loop above the letter similar to Hand1's via & veritas immediately above, seems likely to be a reviser's hand and ink. The alteration of <u> to <v> in avowe (11.88)I can be compared with the same alteration in crovne (20.183)I; the latter spelling is quite uncharacteristic of the spelling-systems of either Hand1 or Hand2.

Corrections that are in hands different from the main scribe or the Hand2 "group" include the line from Psalms inserted in a paragraph space as 5.426,I the only line in the text written in red ink, which has no parallel in other manuscripts (though it also appears more appropriately as 4.36). At the foot of 40r is the erasure of partly legible þei tyne and the substitution in a different hand of get they neuyr (9.183),I which again is without parallel. On fol. 2r (P.74) a considerably later hand has crossed through bonchid and written blessid above.I These later corrections tend to be sloppily executed and to introduce readings not paralleled by any other texts of the poem. We have grouped these as HandX. If M was indeed used as an exemplar for further copies of the poem (see II.5 below), these readings may be the work of the later scribes working on new copies. On fol. 1v a single hand has made changes in P.58, P.66 and P.67,I and we have assigned these alterations to the much later Hand3 responsible for the majority of the marginal annotations.

II.2 The Nature of the Corrections: Spelling:

The great majority of the corrections are motivated by concerns over spelling rather than differences in the textual tradition. A corrector using a paler ink (who again resembles the main scribe) is particularly active from 21r. On that page, for example, the pale ink has rewritten the sub- of subpriour (5.173, W Suppriour), inserted a above the line before Child (5.176), rewritten knowest (5.185), changed what might have been be or bi to by (5.187), respelled heruy (5.191) and inserted they above the line (5.195). None of these alterations occurs at a point where there is variation in the textual tradition. However, in 5.188-189 what is apparently the same correcting hand has altered the readings to hym, hym and his, readings shared only with Cr, for me, me, and my in other texts. A point the significance of which will become clear is that the mid-line punctus elevatus is supplied in pale ink over the deleted final <-e> of mouthe (5.181). The writing is often large and clumsy, and no attempt is made to match it to the surrounding script.

The effect of the multitude of non-substantive corrections is to alter the language in the direction of the London English written by the most professional scribes, that is to say the standardized form known as Type III, classically represented by the Hengwrt and Ellesmere manuscripts of the Canterbury Tales.NM. L. Samuels, "Some Applications of Middle English Dialectology," English Studies 44 (1963), 81-94; Jeremy J. Smith, "The Language of the Ellesmere Manuscript," in The Ellesmere Chaucer: Essays in Interpretation, ed. Martin Stevens and Daniel Woodward (San Marino, CA, 1995), pp. 73-74; and Simon Horobin, The Language of the Chaucer Tradition (Cambridge, 2003). The best and most consistent example of this language among Piers manuscripts is W, although it shows differences in detail from Hengwrt/Ellesmere.NThese differences are analyzed in Thorlac Turville-Petre and Hoyt N. Duggan, eds., The Piers Plowman Electronic Archive, vol. 2: Cambridge, Trinity College, MS B.15.17 (W), SEENET, Series A.2 (Ann Arbor, 2000), "Linguistic Description," correcting M. L. Samuels, "Chaucer's Spelling," in Middle English Studies Presented to Norman Davis, ed. Douglas Gray and E. G. Stanley (Oxford, 1983), pp. 22-3. LALME's Linguistic Profile for the Hengwrt/Ellesmere scribe is LP 6400. In almost every case it is appropriate and instructive to compare the forms in W with those adopted by correction in M. Since there is no extant rule-book of London English, such corrections offer an insight into what a professional scribe considered elements essential to the conventions.NFor fuller discussion, see Thorlac Turville-Petre, "Putting it Right: The Corrections of Huntington Library MS. Hm 128 and BL Additional MS. 35287," Yearbook of Langland Studies 16 (2002), 41-65.

The simplest of all changes, and the most pervasive, is to remove or alter initial yogh. Just one instance remains of the spelling ȝif in the text (5.261), though it is not changed in the catchword on fol. 24v. Every other instance has been altered by deleting yogh (see below, IV.1.2.7). The other words beginning with yogh are respelled by erasing yogh and replacing it with <y>. A complete list of those words that escaped such attention is: ȝaf (19.123), ȝate (20.376), ȝe (7 cases), ȝede (20.184), ȝow (3 cases) ȝoure (2 cases), ȝowre (10.152). None of these words is ever spelled with yogh in W, where the practice is to use initial <ȝ> only before <i>/<y>: so ȝit but yet (the solitary exception to this rule is Yiftes W.3.101). In M there is also a tendency to alter yogh to <y> within the word.

Some other examples of spelling changes introduced simply to conform to the London norm are:

Other spelling changes alter forms characteristic of the scribe's (and Langland's) South-West Midland dialect to forms more familiar in London. In the scribal dialect the vowel derived from Old English short /y/ in open syllables is often spelled with <u> and usually corrected to <i> (IV.1.1.2.13). A striking case is which(e), the only form in the text, occurring 45 times, five of which are apparently original. What is likely to underlie the corrected forms is the reading wuche that remains in the catchwords on fol. 88v (IV.1.1.2.12). The form, mapped by LALME in their Item Maps as item 11 (2.46, and recorded in their County Dictionary (4.19-23), has a distinctively South-West Midland distribution, centering on Gloucestershire. The London spelling of W is which(e). Other such South-Western spellings that have regularly been corrected are dud, left once only at 5.37 (but probably sometimes underlying the regular correction to did(e)), and þuder, corrected to þyder in 5.533 and 581.

OE /y/ (or shortened /y:/) in closed syllables is often <u> in M, occasionally corrected but usually not (IV.1.1.2.12). So churche is corrected to chyrche at 1.183 and churchus to chyrchus at 6.12 (W has chirche(s) and once cherche). The distribution of the form with <u> is again predominantly South-West Midland, but it is not restricted to that area and indeed there is another cluster in London, according to LALME dot map 386 (1.401) and County Dictionary item 98 (4.144-146). Similarly, the scribe's spelling furste is generally left to stand (always first(e) in W), but there is the corrected form fyrste at 2.201. The forms fust, gult- and hull- (dot map 995) are not corrected where W has a mixture of forms: fust and gilt- always, but both hull- and hill-.

In words whose stem vowel descends from OE /y:/ in open syllables the scribal spelling varies between <u>, <uy>, <y> and <i> (IV.1.1.2.14). The corrector often alters the South-Western rounded form <u> to <y>. Thus prude is corrected to pryde at P.23, 5.634, 8.117 and 20.382, though the form pruyde is preserved at 3.66. In thirty other instances the scribe himself appears to have written pride, with three instances of pryde (all three in the last two passus). So, too, the scribe's fuyr is not corrected at 13.170 and 14.48, whereas fire at 7.55 is a correction of fure. The scribe's usual form of the word is fyr. The noun "hire" is written huyre at 6.142 and 6.199, while the verb hured is corrected to hyred at 6.117. The noun hyre is over an illegible erasure at 5.576 and 6.202. Spellings in W are generally with <i> or <y>, with fuyr and huyre once each, perhaps relict forms from the exemplar or archetype.NThis notion is further supported by appearances of the noun form huyre on seven occasions in R (KD.2.123, KD.5.550, 552, 556; 6.139, 195, 198), four times in L (KD.6.139, 195, 198; 14.143). Verbal forms with huyr- appear twice in R and in L, both in the same lines (KD.6.114, 312). L and R are the two manuscripts closest to Bx.

Rounding of OE /eo/ is sometimes retained in scribal spellings (IV.1.1.2.23). As in W, three words of predominantly Western distribution, burn(e)/buirn, heo and leode (beside le(e)de) retain their Western forms and are not corrected. The <u> spelling of "learn" is not found in W; it is used by the M scribe and is uncorrected six times,Nlurne 5.211, 8.109, 10.240; lurned 7.51, 7.145; lurnynge 11.154. but on at least five other occasions lurne is corrected to lerne. OF <ue> sometimes appears as <oe> in moeue (also spelled moue and meue). For "people" the scribe has the spellings poeple, peple, and puple; on some 75 occasions the corrector alters puple to peple except for 14.209. In W peple is the dominant form, though the scribe spells it poeple once.

It appears that the correctors have no specific desire to remove dialect forms, but a more general wish to adopt the standard London spellings of common words. The point is confirmed by tracking the widespread use throughout M of <u> for the vowel in final unstressed syllables. This is characteristic of Western dialects and never occurs in the spelling of W, and yet the corrector is not particularly assiduous in altering it. On the verso of the first leaf, for example, it is apparently corrected to <y> (saintys P.47), but not corrected a few lines further down (hokud P.53). The use of this unstressed <u> for the ending of the past tense and past participle of weak verbs is common and rarely corrected. It is often used for genitives and plurals of nouns, where the corrector sometimes alters <-us> to <-es>, but as often lets it stand (IV.2.2.2, IV.2.2.4). So, too, the vowel in verbal endings, though usually <e>, is occasionally spelled <u>: it is left uncorrected in infinitive helpun (9.121) (IV.2.5.1.1); <-uþ> or <-uth> are the endings of the present third person singular nine times and the plural twice, though these <u> spellings are only used, or at least are only preserved uncorrected, between the end of Passus 5 (helpuþ 5.645) and the beginning of Passus 9 (kepuþ 9.17); the present plural is four times <-un> (IV.2.5.1.9); in over 100 instances the weak past tense and participle forms are <-ud> or <-ude> (IV.2.5.2); and the strong past participle form at P.71 is y-brokun (IV.2.5.3.6). Correction of other instances to <e> is likely but is often difficult to determine. Final <-ur> is common in M, and corrected only sporadically, leaving aftur (82x, LALME item 29), hungur (15x), oþur and othur (45x), vndur (22x), watur (12x), etc. Petur is the dominant spelling of the personal name.

The conjunction "than" and the adverb "then" are both generally spelled þanne/thanne by the scribe. They are, of course, the same word etymologically. A corrector systematically (over 60 times) alters the spelling to þan/than in the case of the conjunction, and this is the spelling distinction maintained in W. The adverbs nouȝt and out are among many other words regularly corrected by deleting final <-e>.

II.3 The Nature of the Corrections: The Grammar of Final -e:

All these alterations are a matter of routine, but to remove or add final <-e> when it is an inflectional ending, and to do it consistently and correctly, requires specialist knowledge. Again, a comparison or M's practice with W's is instructive. The W scribe has a good grasp of the practice and carries it out with reasonable consistency.NThe evidence is reviewed in Thorlac Turville-Petre and Hoyt N. Duggan, eds. The Piers Plowman Electronic Archive, vol. 2: Cambridge, Trinity College, MS B.15.17 (W), SEENET, Series A.2 (Ann Arbor, 2000), "Morphology." In his text adjectives of one syllable ending in a consonant (whether of OE, OF, or ON origin) will add <e> in the plural, and in the singular when following a defining word such as þe, þis, oure, or used with a proper noun or in a term of address. Final <-e> may be used to mark a dative singular of a monosyllabic noun, particularly at line-end. It is optional after the ending <-yng>, in the imperative singular, after <-ed> in the preterite, but not used after past participle <-ed>, nor after the preterite first and third persons singular of strong verbs.

The original scribe of M did not have the W scribe's concern for, nor his knowledge of, the correct use of final <-e>, and later correctors have made thousands of alterations to revise the text so that it conforms to these rules. These alterations were made after the red boxing of the Latin lines, and after the scribe added missing lines on fol. 90r (18.328-330), since these lines themselves have erasures. Yet none of the corrections in pale ink has been altered and the punctus elevatus written in this ink over an erasure in 5.181 (and many places elsewhere) shows that the erasing preceded this. The sequence of correction is neatly demonstrated in 18.112, crist com{e} \of/ hire kyngedom{e}; the rubrication of the following line has been affected by the erasure of the <e> of come, and of has been inserted over the erasure.I For examples and analysis of such corrections affecting nouns, see IV.2.2.3; for adjectives and adverbs see IV.2.4.

Since the scribal tendency is to add unetymological <e>, cases where etymological <e> has to be supplied are less common, but nevertheless the correctors are on the lookout for error here too. On fol. 1r there are at least two nouns which have been supplied with <e>: the dative at wille (P.37), and lykame (OE līc-hama)(P.30). Other cases are the nouns wrathe (OE wræððu), heuene (always so spelled in W), at inne (OE æt inne) (8.4), synne (8.50) (as always in W) and wele (OE wela)(18.211); the plural adjective alle (9.154), whiche (6.135); the adverb manlyche (OE manlīce) (10.97), the infinitive wrathe (6.155), and so on.

II.4 Classification of the Text:

Until Passus 17, the uncorrected text of M is a good and independent witness to the B tradition.NSee the recent study by Robert Adams, "Evidence for the Stemma of the Piers Plowman B Manuscripts," Studies in Bibliography 53 (2000), 173-94. Adams lists agreements of M with LRF, arguing that "their shared readings represent those of the common B archetype" (p. 175). Apart from repeated words and some omissions of words or lines, many repaired by the original scribe, there are very few unique readings. It has no strong affiliations with any one manuscript or group of manuscripts, though it commonly agrees with YOC2 and the larger group YOC2CB, sometimes joined by L. It shares a number of exclusive variants with F (particularly in Passus 5),NKane and Donaldson list agreements with F on p. 44, n. 65, but their list is incomplete, partly because they do not always record M's unrevised readings. Agreements are: 1.50; 4.66, 67; 5.19, 46, 71, 117, 150, 184, 198, 300, 306, 411, 556, 657; 6.59, 68, 198, 269, 270, 317; 7.16, 92, 139, 160; 8.8, 45, 67; 9.187; 10.49, 108, 287; 11.155, 229, 264; 12.156, 210; 13.165, 240, 342; 14.76, 129, 241, 293; 15.121, 148, 203, 254, 266, 311; 16.159; 18.436; 19.193, 458. with H,NExclusive variants shared with H's B text (Prologue - 5.128) are P.86; 2.83, 137, 186; 3.183; 4.112; 5.7. with Hm,NM shares variants with Hm at P.116, 178; 4.121; 5.240; 6.176, 229; 10.183; 11.314; 12.286; 13.251, 287; 19.330. and sporadically with B.NM shares variants with B at: 5.98, 114, 219; 10.74; 11.223; 13.320. Of some significance are the agreements with readings of A manuscripts not noted by Kane and Donaldson.NSome variants shared with A manuscripts are: 1.106, 138, 184; 2.169; 3.183, 212, 289; 4.19, 195; 5.202; 6.69, 90, 112, 269, 332; 7.147; 8.70, 72; 9.25, 56. From Passus 17 to the end of the poem, there is evidence of a switch of exemplar in the uncorrected M tradition, since the original text now shows affiliations with the WHmCrS group, and more specifically with Cr.NAgreements are listed by Kane and Donaldson, pp. 42-3, and p. 44 n. 65

In addition, Kane and Donaldson point to the evidence that a number of corrections align M with the larger group WHmCrS, and they conclude that "indications are that the connexion of WHmCrS was not only genetic but also somehow local."NGeorge Kane and E. Talbot Donaldson, eds., Piers Plowman: The B Version, rev. ed. (London, 1988), p. 50. So for the last four passus the scribe or his exemplar had access to a text of the WHmCrS group, and this connection was then reinforced by corrections throughout the manuscript. The close relationship with the sixteenth-century texts CrS in particular is quite convincing, and indeed the rewritten line 17.162 at the foot of 83r is by itself enough to demonstrate that this corrector of M had access to their common ancestor. The order of the two halves of the line With-Inne hem thre . the wyde world holden is shared with CrS alone. So, also, are the readings merueille which a corrector substitutes for truþe in 18.126, picked up from the line above, and bad inserted above the line in M (20.272) to make sense of the omission of the verb heet. There are quite a number of such corrections. NSome are listed by Kane and Donaldson p. 51.

It is, however, altogether likely that the correctors used more than one manuscript of the poem. Other revised readings are, for example, shared with H (P.58, 2.189), with B (5.400), with G (6.208), with F (5.540, 583; 6.43, 10.111 (and see note there), 10.178, 213, 235, 241; 15.233), with the GYOC2CLR group (5.31), with A manuscripts (1.52; 2.189; 4.101; 5.97; 6.43; 10.63, 213), and with A and C manuscripts (6.122, 220), etc.

In the annotations we have recorded agreements which may throw light on the genetic affiliations of M's text, including unique readings, agreements of the original text with one other manuscript, and agreements involving a substantive correction. We have not noted dialect variants or variation between final <-e>/<-en> and the like.

II.5 The Function of the Manuscript:

We have observed that the corrective activities in M represent a determined attempt to make many thousands of corrections so as to alter the language of M in the direction of London English on the model of W, while allowing certain features characteristic of the South-West Midlands to stand. The nature of some of the substantive corrections shows that at least one of the correctors used an ancestor of the WHmCrS group, and implies that he had access to the products of the London network that produced manuscripts such as W. In making the alterations, the correctors made a considerable mess of the manuscript, clearly without regard for its final appearance and so presumably with no concern about its financial value. A. I. Doyle writes that M was corrected "very possibly to serve as an exemplar for further copying,"NA. I. Doyle, "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, Eng., 1986), p. 40. and this offers a convincing explanation. Can we envisage the situation?

A Gloucestershire scribe with London training copied a text of Piers, introducing the paragraphing and other features of the layout of metropolitan texts such as W and R or inheriting these from his exemplar. The scribe had a tendency to omit words or copy them twice; his writing was somewhat uncontrolled, and he grew less careful of presentation as he proceeded. He or a rubricator underlined or boxed the Latin text in red rather carelessly, and the rubricator added passus headings and simple red capitals at the start of each passus. M has the appearance of a manuscript that was never finished to the standard originally planned.

The manuscript, perhaps once intended for a provincial owner, was then adapted to function as an exemplar for texts for the London market, and at this stage the corrections to "Londonize" the language were made. The systematic spelling corrections were made after the red boxing, which has sometimes been damaged by the erasures, and after the first round of textual corrections, as shown by the form of the marginal guide word ȝe for 1.22, revised as ye in the text, and the revision of the marginal addition 17.185-6. And yet it is probable that alteration of ȝet to yet took place before the red passus initials were painted, since the only example of uncorrected Yet is the first word of Passus 2, where the tiny guide letter <y>, still visible, may have replaced the scribe's <ȝ>. There must have been a considerable need for exemplars that could be copied to meet the demand for texts of Piers, and professional scribes must have practiced a system of loaning or exchanging texts among themselves. A manuscript to be hired out to other London scribes did not need to be attractively finished so long as it was legible and "accurate." What had been erased could be left partly legible as long as the intended reading was clear. Corrections could be overwritten or inserted without any attempt made to blend in the alterations by matching the ink or size of script to the surrounding text; in fact the presence of obvious corrections would reassure the borrower that attention had been paid to the text. Some such situation seems a satisfactory explanation for the correctors' activities in M.

III. Editorial Method:

III.1 Transcription of the Manuscript:

The scribe uses the standard abbreviations and suspensions, which we have expanded. In style sheets other than "Critical," resolved abbreviations appear in italics. In English words the scribe uses a superscript <t> in w(i)t(h)I but only occasionally in þ(a)t (19.295),INThe abbreviated form is used in corrections, e.g. 19.22. and indicates with a superscript vowel the omission of <r> before the vowel, as in c(ri)stendom (12.280)I or t(ra)uaill (14.166).I At the end and within a word, <er> is indicated by a loop, as in þ(er)eI, oþ(er)I (5.32), or man(er)eI (19.185),NSee note to prison(er) 15.355. and <ur> and <ra> are indicated by superscripts, as in curato(ur)s (20.281)I and g(ra)ce (19.268).INThe superscript for <ra> represents <a> alone in walsingh(a)m (5.232),I Bokyngh(a)m (2.112),I and Abrah(a)m (16.84).I A bar through the descender of <p> represents either <per> or <par>, as in p(ar)fitlich[e] (19.194)I and p(er)sone (19.195);I a loop through the descender indicates <pro>, as in p(ro)ue (20.274),I while <re> is represented by a backward loop after the <p>, as in p(re)lates (20.126).I A slanting line through long <s> represents <er> as in s(er)uauntz (19.442).I Nasalization is indicated by a line over the preceding letter or group of letters, and we have expanded as appropriate. We have not recorded instances where it is used without meaning: for example, frequently after the last three letters of pardoun (e.g. 7.113, 116, 118). The corrector sometimes went to the trouble of erasing such otiose tildes (e.g. 7.187, 188). The abbreviation of l(ett)re is indicated by a bar through <l> (20.325);I the abbreviation of sp(irit)ualte is marked by a tilde over <u> (5.151).I

It is not always easy to distinguish between meaningful abbreviations and meaningless ornamentation. Loops and curls on final letters are often difficult to interpret. In particular, there are bars through <-h>I and <-ll>,I and the flourishes on final <-c>,I <-d>,I <-g>,I <-k>,I <-p>,I <-r>,I <-t>.I

The bar generally through final <-ll> is never essential and we have ignored it. There is often a bar through final <-h> in the combinations <ch>, <gh>, <ssh>, <th>. It occurs occasionally even on words such as with (7.12)I where it cannot possibly be meaningful. We have not expanded barred final <-h>, but medially we have interpreted it as an abbreviation in reh(er)ced 11.408I and h(er)te (13.48)I.

We can find no significance in the loops and tails on final <-c>, <-d>, <-g>, <-k> and <-t>. Since a corrector is concerned to erase final <-e> when it is not appropriate, he might be expected to erase such curls and loops if he considered them significant, but although he regularly erases <-e> from clerke, e.g. in 4.170,I he never removes the curl (e.g. Clerk [3.9]I). It is also relevant to note that the corrector, having erased <-e>, sometimes adds what must be mere ornamentation to the consonants now in final position. Thus at 4.166 cokewoldI a tail is added to what has become final <-d> when <-e> has been erased. Similarly, at 5.324 the final <-e> was erased from ClerkeI and the flourish supplied on <-k>.

<p> often has a tilde above it, especially in final position, but its pattern of use (for example, it is often used to end bisshopI and bisshoppI) suggests the tilde is not significant. Having deleted final <-e> from warpe, the corrector adds a tilde above the <p> (5.89).I

The hooked backward loop on final <-r> (e.g. aftur, 5.600)I is not so easy to interpret, and the scribe seems inconsistent in its use. Where his practice elsewhere shows a strong preference for a spelling with final <-e>, we have so interpreted the loop. For example, sir(e) in 9.12I is the sole instance of the word where the scribe did not write out final <-e>. So, too, her(e) infin. "hear," (15.112),I needs its <-e>. But in many other cases his own usage, and indeed the usage of the scribe of W, argues against expansion. Aftur is never spelled with <-e>; it twice ends with a loop (5.319, 5.600), and 90x without. The same pattern of distribution is found in words like rather, wastour, watur and others, and so in such cases we have not expanded. The usage of the corrector is more consistent and straightforward in the matter of final <-r>. He has a version of the curl that resembles a ball above the letter. It is used for the feminine singular pronoun hir(e) at P.165, 3.149, 5.47 and 12.89,NAt 2.15 the corrector writes hir without <-e> for the feminine pronoun. Nine times he writes hir for the plural, which the main scribe never does. The usage in W is both with and without <-e> for both feminine and plural pronoun. In 14.146 the corrector erases the <-e> from the pronoun but not from the noun in hir{e} hire, in line with the spelling in W. for lier(e) (1.40 and 2.205),I and wher(e) (8.4). The rewritten ending of Cheier(e) 5.630I has the corrector's form of curl over <r>, and a final <-e> is etymologically justified. The curl is added by the corrector to an original <-r> in porer(e) (20.49)I and flaterer(e) (20.323).I An original curl is erased at the end of power (3.287)I and wintur (5.562).I In all these cases the curl is clearly meaningful and in most cases it is paralleled by the usage of W, as is so much of this corrector's spelling practice.

Common Latin words are often radically abbreviated. Thus Cristus,I Dominus,I fratre,I forms of habeo,I omnes,I quod,I similisI and others are all abbreviated in the standard forms.

We have not distinguished allographic forms, such as the three forms of <s>, or the single-lobed <a> from the double-lobed form. The letters <ȝ> and <z> are not distinguished by the scribe, but we have represented the velar spirant with <ȝ> and the sibilant with <z>, e. g., artz, baptize and dozeyne, etc.

Punctuation at the ends of lines, in particular at the foot of a leaf, is often impossible to distinguish from a flourish on a final letter or other decoration. We cannot claim to have been consistent, but we have marked what appears to be an intentional punctus, punctus elevatus, swung dash, and virgule.

Our capitalization follows the scribal use of litterae notabiliores, although there are some letters, in particular <w> and sigma <s>, where there is no clear distinction. We have interpreted such letters according to their context: thus they are capitalized at the beginning of the line, but printed as lower case within the line unless their context and enlarged size suggests otherwise.

The word-division of the manuscript is followed as far as practicable, though no attempt is made to represent the variety of spacing between words and letters. The interpretation of the scribe's word-division, though it is generally unambiguous, is occasionally a matter of fine judgment. There is sometimes no obvious space between the indefinite article and the noun, e.g. afreman (20.145), but there seems no good purpose in recording these as a single word. A hyphen in the transcription indicates a space in the manuscript within a word, or a compound or phrase conventionally hyphenated today; we have followed OED in doubtful cases. Conversely, some phrases, in particular or elles, at ese and at ones, are written as one word, orelles, atese, atones. The scribal form appears in the Scribal style sheet in lime and in the Diplomatic style sheet without any color difference from the surrounding text, the regularized form in the Critical style sheet, and both forms in AllTags in lime and olive respectively.

Proper nouns have been treated as English unless they have a Latin inflection: thus genitive Cesares is not given special treatment by the scribe in 1.53, and is not tagged, but in 1.54 Reddite Cesari is boxed in red and written in slightly enlarged script, though Cesari with a Latin dative ending in the same line is not treated specially by the scribe. Both are tagged as Latin. There are other cases where the decision is more arbitrary: transgressores in 1.98 is not marked out with underlining by the scribe, but yet we have inserted a <foreign lang="lat"> tag interpreting it as a Latin plural, whereas Mnames (6.248) is understood as English, with the backing of MED. In 5.430 it seems obvious that Beatus vir and Beati omnes, boxed in red, should have <foreign lang="lat"> tags, but the consequence is that the two instances of beatus vir with an English genitive inflection, beatus virres (10.328, 13.55), have also been tagged as Latin despite the English ending.

Scribal mispellings have been recorded with a <sic> tag and corrected with a <corr> tag; the former shows up in violet in the Scribal and AllTags style sheets, the latter is displayed in purple in the AllTags style sheet and inside square brackets in the Critical style sheet. The scribe has an odd tendency to miscount minims, so writing brennneþ (17.216),I synnne (5.20),I (18.181, 19.463), etc., and such instances have been tagged.NOther examples are spynnnen (5.218),I þannne (5.111),I þere-innne (9.144),I wynnnen (6.165).I These are sometimes corrected: inne (8.4),I þere-inne (1.12,I 6.64, 9.5, 9.54), with-Inne (5.564).I On the other hand withnne (13.361)I is short one minim.

III.2 Transcription of Corrections and Erasures:

Wherever possible we have attempted to distinguish between the text as originally written and as subsequently corrected by the main scribe or another hand. Where we are reasonably confident that we can read the erased letters, they are recorded within deletion tags which display in the Diplomatic and Alltags style sheets as deleted letters enclosed in wavy brackets. When erased text is illegible, we have indicated this with one punctus per deleted character up to six characters. When longer stretches of text are involved, we indicate deletions with "...?..." and deletions longer than a half line with "...?...?...".

Added text, whether or not written over a deletion, is displayed in dark gray in Scribal and Alltags style sheets.

We have not marked as a correction the corrected punctuation in the middle of the line unless it is written over an erasure. It is clear that in many cases the punctus elevatus is a conversion of an original punctus, but only occasionally can we be certain that this is so in any individual case.

III.3 Presentation of the Text: Style Sheets

Using XML markup, we offer four different views of the text accessible through four different style sheets: Scribal, Diplomatic, Critical, and AllTags.

The Scribal style sheet's presentation of the text represents as closely as possible both the readings and features of the manuscript text as well as the most information about editorial interventions. Changes of script and style are reflected by changes in the font style. The Middle English text's anglicana formata is represented in roman letters. Resolved abbreviations and suspensions appear in italics. Color in this style sheet serves two functions: red indicates the color of ink used by the scribe, while any other colors — aqua, dark gray, lime, olive, pink, purple and violet — mark editorial functions. For a detailed key to the conventions we have adopted for identifying editorial functions by means of color shifts, see the Instructions for First Time Users.

The Diplomatic style sheet suppresses all notes, marginalia not in the text hand, and indications of error or eccentric word division. Its text is otherwise identical to that presented in the Scribal style sheet.

The Critical style sheet is designed to indicate the text as it was intended to appear after correction. Since the text displayed is a reconstructed, putative text, it lacks the color features that appear in the more nearly diplomatic transcriptions of the manuscript. We conventionally use italics for Latin and French words and phrases in this style sheet. We have supplied line references to the Athlone B text for the convenience of readers. Eccentric word divisions are silently, at least in the surface display, corrected in this style sheet. That is, atones appears as at ones. A reader who wishes to find all such divisions can still search for them in the views provided by the Scribal and AllTags style sheets as well as in the underlying XML text.

The AllTags style sheet, as its name implies, is intended to display the full content of markup in XML tags.

An example of the effects of the four style sheets may be offered by the "shadow-hyphen," which we have used to join the elements of compound words that the scribe had left separate. See III.1.10. In the Scribal style sheet the elements of the compound are joined by a pink hyphen to indicate editorial intervention: so for-wandred, M.P.7. In the Diplomatic style sheet the two words appear as the scribe wrote them: for wandred. In the Critical style sheet the elements of the compound are joined without a space: forwandred. In the Alltags style sheet the pink hyphen again joins the parts of the compound for-wandred.

III.4 Presentation of the Text: The Annotations

Four sets of annotations are provided—codicological, lexical, paleographic and textual.

     (a) Codicological: These notes draw attention to physical features of the manuscript and to later additions in the margins such as brackets, names, pointing hands and other drawings. Codicological notes are marked by a red superscriptedCSample codicological note..

     (b) Paleographic: These notes comment on letter forms, in particular ambiguous abbreviations, curls and other features. Paleographic notes are marked by a red superscriptedPSample paleographic note..

     (c) Lexical and Linguistic: These notes provide brief glosses for unusual, ambiguous, or difficult words or comment on items of linguistic interest. Both Lexical and Linguistic notes are marked by a red superscripted LXSample lexical/linguistic note..

     (d) Textual: These notes call attention to unique or shared readings which shed light on M's relationship to other manuscripts. See II.4 Classification of the Text. It must be emphasized that these notes are no more than an aid to the reader of the documentary text of M. They do not in any sense constitute a complete listing of variant readings nor anything beyond a first step in establishing the relationship of M to other manuscripts. They may imply that M's reading is not that of the B archetype, though we reserve all judgments about Bx until a later stage of our work, currently in progress. These notes are, then, an interim statement that will be of limited or no use once the B archive is complete and the variant listings can be electronically generated. The information for these notes is drawn from the listing of variants in the Kane-Donaldson edition which we have checked against those transcripts that are already available in the archive. Since it is not at this stage relevant which of the witnesses share the majority reading against M's unique variant, the majority readings are where possible presented in very simplified form, usually with the designation "other B manuscripts" or "most other manuscripts" or "all others." It is true that in most cases this means Bx, but it is important not to prejudge the issue. Textual notes are marked with an icon of a superscripted red TSample textual note..

IV. Linguistic Description:

In describing the forms of M we are faced with the problem that the underlying language has been overlaid with another, that of one or more correctors. As explained above, the aim was to bring the spelling system more in line with professional London practice, affecting both phonology and morphology. In describing the underlying language we have tried to choose examples that have not been altered, although alteration is not always detectable. This uncertainty will affect the statistics to some small degree, although we have taken particular care to note corrections in forms that we know to be subject to change. In order to show the nature of the corrections, we here display corrected forms within square brackets and erased forms within wavy brackets: thus song{e} indicates that <e> has been erased, whereas fe[et] means that the two last letters are corrections for a form beginning with fe; it is likely that the underlying form is fete. The consequences of alteration have been discussed above and will be noted below.

The underlying forms are strongly South-West Midland and it is usually impossible to distinguish relict forms representing Langland's own dialect which are found to some extent in all B text manuscripts. The only exceptions are forms necessary for the alliteration which must of course go back to the author: heo for "she" is necessary for the alliteration in 1.75 and 3.29, and in 5.647 where She has been written over a deletion (IV.2.3.1); hij appears in alliterating positions at 1.59 and 1.195, and is erased at P.66 (IV.2.3.4).NSee M. L. Samuels, "Langland's Dialect," Medium Ævum 54 (1985), 232-47; reprinted in The English of Chaucer and his Contemporaries, ed. J. J. Smith (Aberdeen, 1989). See also M. L. Samuels, "Dialect and Grammar," in A Companion to Piers Plowman, ed. John A. Alford (Berkeley and Los Angeles, 1988), 201-21.

The forms valuable for establishing the dialect of the M scribe are:

1. The spelling <u> for the vowel in unstressed syllables. This is characteristic of Western dialects and is many times corrected to <e>. It is used for genitives and plurals of nouns <-us> (IV.2.2.2, IV.2.2.4), in verbal endings, as occasionally infinitive and present plural <-un> (IV.2.5.1.1, IV.2.5.1.9), present third person singular <-uþ> or <-uth> (IV.2.5.1.8), weak past tense and participle forms <-ud> or <-ude> (IV.2.5.2), strong past participle <-un> (IV.2.5.3.6). Final <-ur> is common as in aftur, hungur, oþur, vndur, watur etc. Petur is the dominant spelling of the personal name.NLALME dot map 656 for weak past tense <-ud(e)> represents the northern area only (see also vol. 4 item 61), and charts a scattering of mainly Western examples; weak past participle <u> is plotted on dot maps 661 (Northern) and 1199 (Southern) to show a decidedly Western grouping with a scatter elsewhere (and see also vol. 4 item 63 for the northern area). LALME vol. 1, dot maps 642 and 595, vol. 4, item 54, lists <-us> as a majority form for noun plurals mainly in Western texts, with minority forms more widely scattered. A few examples of present plural <un> are listed for the northern area (LALME vol. 4, item 60).

2. The spellings <u> for OE /y/ and <u> or <uy> for OE /y:/ (IV.1.1.2.12, IV.1.1.2.14). These have a distinctively South-West Midland distribution, centering on Gloucestershire.

3. The use of ar, "before," beside er (IV.1.1.2.21).

4. The retention of rounding in spellings for leode, heo and burn (IV.1.1.2.23).

5. Instances of OE /hw/ as <w>, especially wuche (IV.1.2.1).NLALME, vol. iv, item 11 cites wuch and wuche from manuscripts located in Berkshire (2 manuscripts), Gloucestershire (8), Hampshire (1), Herefordshire (3, including MS Harley 2253), Oxfordshire (2), Warwickshire (3) Wiltshire (2) and Worcestershire (2). Two of the manuscripts are of Robert of Gloucester's Chronicle and no fewer than seven of the South English Legendary, both works of Gloucestershire provenance. Another two are C texts of Piers Plowman: MSS Trinity College, Dublin 212 and Digby 171 (where it is a minority form), placed close together on the Gloucestershire/Herefordshire border (LALME map, vol. i, p. 337).

6. Addition or loss of <h> before an initial vowel (IV.1.2.6).

7. The form hij in non-alliterating positions in 5.116, 6.15, and 10.321 (IV.2.3.4).

8. The present plural verb ending <-th>/<-þ> (IV.2.5.1.9)

9. The survival of OE -ian weak verbs in the infinitive and present first singular ending <-ye(n)>/<-ie(n)> (IV.2.5.1.1, IV.2.5.1.6), and the present plural endings <-ieth>, <-yen> (IV.2.5.1.9).

The scribal dialect is demonstrably South-West Midlands. Samuels' statement that it is from Herefordshire NSee M. L. Samuels, "Langland's Dialect," Medium Ævum 54 (1985), 232-47, note 59; reprinted in The English of Chaucer and his Contemporaries, ed. J. J. Smith (Aberdeen, 1989). The evidence cited is the weak preterite ending <-ude>. seems to us to rest on slender evidence. Indeed, point 6 above, the addition or loss of <h> before an initial vowel (IV.1.2.6) would, Samuels himself claims, exclude Herefordshire. NM. L. Samuels, "Langland's Dialect," Medium Ævum 54 (1985), 237, map on p. 236; reprinted in The English of Chaucer and his Contemporaries, ed. J. J. Smith (Aberdeen, 1989). It seems more probable that the scribe was from a little further east. Gloucestershire would seem most likely, but south Worcestershire or east Herefordshire would also fit the profile.

IV.1 Phonology:

IV.1.1 Vowels:

IV.1.1.1 Quantity:

Vowel length of <a>, <e> and <o> is sometimes marked by doubling in open syllables and when the vowel is lengthened before consonant groups. It may alternatively be marked by final <-e>, but see comment below.

For /a:/: <a> ~ <aa>

For /e:/ and /ε:/: <e> ~ <ee>

For /o:/ and /ɔ:/: <o> ~ <oo>

Where final <-e> might be misunderstood as a false inflection, the spelling is sometimes corrected.

IV.1.1.2 Quality:

IV.1.1.2.1 OE, ON /a/: <a>

IV.1.1.2.2 OE, ON /a/ before a nasal: <a> ~ <o>

IV.1.1.2.3 OE, ON /a/ before lengthening consonant groups: <a> ~ <o>

IV.1.1.2.4 OE, ON /a:/: <o> ~ <oo>

IV.1.1.2.5 OE, ON /a:/ + w: <ow> ~ <ou> ~ <(a)>

IV.1.1.2.6 OE, ON, OF /o/: <o>

IV.1.1.2.7 OE, ON /o/ + lengthening consonant group: <o> ~ <oo>

IV.1.1.2.8 OE, ON /o:/: <o> ~ <oo>

IV.1.1.2.9 OE, ON, OF /u/: <u> ~ <o>

The <o> spelling is used predominantly in proximity to minims.

IV.1.1.2.10 OE, ON, OF /u/ with lengthening: <ou> ~ <o> ~ <u>

The <ou> spelling is an indication of length, as below.

IV.1.1.2.11 OE, ON /u:/: <ou> ~ <ow>

<ow> is mostly in final position.

IV.1.1.2.12 OE, ON /y/: <i> ~ <y> ~ <u> ~ (<e>)

The frequent <u> spellings are Western.NSee M. L. Samuels, "Langland's Dialect," Medium Ævum 54 (1985), 241, 243; reprinted in The English of Chaucer and his Contemporaries, ed. J. J. Smith (Aberdeen, 1989). A particular example is þuder, "thither," corrected to þyder (5.533, 581). The <u> form of the word is mapped by LALME, dot map 532 and 4.265 (recorded particularly in Gloucestershire).

IV.1.1.2.13 OE, ON /y/ before lengthening clusters: <y> ~ <u> ~ (<e>)

IV.1.1.2.14 OE, ON /y:/: <i> ~ <y> ~ <u> ~ <uy>

The <u> ~ <uy> spellings are Western. NSee M. L. Samuels, "Langland's Dialect," Medium Ævum 54 (1985), 241, 243; reprinted in The English of Chaucer and his Contemporaries, ed. J. J. Smith (Aberdeen, 1989).

IV.1.1.2.15 OE, ON /i/: <i> ~ <y>

IV.1.1.2.16 OE, ON /i:/: <i> ~ <y>

IV.1.1.2.17 OE, ON, OF /e/: <e>

IV.1.1.2.18 OE, ON, OF /e/ before lengthening clusters: <e> ~ <ee> ~ <ie>

The spelling <ie> for /e:/ is ascribed to French influence and is common in Gower.NFernand Mossé, A Handbook of Middle English, trans. James A. Walker (Baltimore, 1952), §11; G. C. Macaulay, ed. John Gower's English Works EETS, ES 81 (London, 1901), 1.xcv.

IV.1.1.2.19 OE, ON, OF /e:/: <e> ~ <ee> ~ <ie>

IV.1.1.2.20 OE /æ/: <a>

IV.1.1.2.21 OE /æ:/: <a> <e> ~ <ee>

IV.1.1.2.22 OE /ēa/: <e> ~ <ee> ~ <ie>

IV.1.1.2.23 OE /eo/, /ēo/ (and OF /ue/): <e> ~ <ee> ~ (<eo>) ~ (<u>) ~ (<ui>) ~ (<ie>)

The three words of Western distribution, burn, leode and heo, retain Western rounding, as also does lurne, though the form is corrected 7x to lern-.NFor the <eo> forms, see M. L. Samuels, "Langland's Dialect," Medium Ævum 54 (1985), 241-43; reprinted in The English of Chaucer and his Contemporaries, ed. J. J. Smith (Aberdeen, 1989).

Before dentals and <l> the short /e/ arising from various OE sources including /ēo/ occasionally goes to <i> in fille "fell" (16.111), yit (9x) vs. yet (42x), to-gidre(s) (usually by correction, in 1.42 from togadre). These forms are especially associated with London English.

OF /ue/ gives rise to a few /oe/ spellings: doel 5.390; moebles 3.274; moeue (6x) ~ moue- (5x); peple (75x, 70 by correction, probably of puple) ~ poeple (16x) ~ puple 14.209.

IV.1.2 Consonants:

IV.1.2.1 OE /hw/: <wh> ~ <w>

The the usual scribal spelling is <wh>: whan, what, where, while etc. However, the preferred spelling of "which" is the South-West Midland form wuch(e), always altered to which(e) except where it was missed in the catchword on fol. 88v. In 8.62 a corrector has inserted <h> above the line in were, and the corrected forms where (5.665) and while (5.414) may conceal original <w> for <wh>. Reverse spellings are whasshen (15.199) and whisshede (20.193).

IV.1.2.2 OE, ON <þ, ð>: <þ> ~ <th>

The scribe uses <þ> ~ <th> without patterning.

IV.1.2.3 OE /š/: <sh> ~ <ssh> ~ (<sch>)

The most common spelling is <sh>, with <ssh> medially and finally after a short vowel: bisshop; childissh; englissh; fissh; flessh; punisshe; shame; shafte; sheep; ship; sholde. The spelling <sch> is rare: schendeth (6.176), lordeschipe (7.173). The loan-word "parish" is once spelled parisch (P.83).

IV.1.2.4 OE, ON /sk/: <sk> ~ (<sc>)

Spellings with initial <sc> are mainly from OF; e.g. scorne (2.84); scrippe (6.63).

IV.1.2.5 OE /xt/: <ȝt> ~ <ght> ~ <ȝth>

The writing with <ȝt> predominates.

The scribe prefers <ȝ> to <gh> as the spelling for the velar spirant, but also writes it with both or without either.

IV.1.2.6 Initial <h>

Addition of <h> is sporadic and generally corrected, as are {h}armes "arms" (5.421); {h}ere "ear" (12.230); {h}oure "our" (17.102). The corrector overlooked initial <h> in ham "am" (16.245). The reverse, loss of <h>, is found in is for "his" (5.247, 12.166, 16.102, 20.331), and oueth (for houeth) (3.209).NM. L. Samuels claims that in this area this feature "is limited to a patch running from Warwickshire through south Worcestershire to Gloucestershire," and concludes that it excludes Herefordshire: "Langland's Dialect," Medium Ævum 54 (1985), 237, map on 236, reprinted in The English of Chaucer and his Contemporaries, ed. J. J. Smith (Aberdeen, 1989).

IV.1.2.7 OE /j/

In initial position scribal <ȝ> is regularly corrected, by replacing it with <y>, or removing it in the case of ȝif, since in London the initial palatal was lost before /i/. One instance remains of the spelling ȝif in the text (5.261), and one in the catchword on fol. 24v. The other words beginning with yogh are respelled by erasing <ȝ> and replacing with <y>, and only very occasionally overlooked, as in An auenture ȝe haue ȝoure hire here and ȝoure heuen als (3.72). There is a tendency to alter <ȝ> to <y> within the word; for example words with initial for- (forȝefnesse, forȝete) are rewritten with <y> twelve times out of twenty. See above II.2 The Nature of the Corrections: Spelling.

IV.2 Morphology:

IV.2.1 Metrical Considerations: The Status of Final <-e>:

The scribe's writing of final <-e> is not entirely random, but probably to some degree reflects the usage of his exemplar. It is clear that the M scribe had no concern for, and perhaps no knowledge of, the grammatical and etymological considerations that the W scribe represents so accurately in his usage of final <-e>. What is remarkable is that a corrector (or correctors) expended great efforts on imposing a system on the text by, for the most part, erasing <-e> where it is not justified historically. The corrected usage closely matches that of W, and so in what follows comparisons are made with that manuscript.NGeorge Kane and E. Talbot Donaldson remark on W's "careful handling of final e" in Piers Plowman: The B Version rev. ed. (London, 1988) pp. 214-16, note 184, though in note 179 (p. 215) they overstate W's consistency.

In nouns, scribal <-e> is retained where the OE noun ends in a vowel: e.g. churche (OE cyrice), erthe (OE eorþe), herte (OE heorte); tonge (OE tunge), and occasionally <-e> is added as in wrath[e] (OE wræððu) 11.4. It is also retained as a dative singular ending (see IV.2.2.3). Otherwise it is regularly erased: e.g. in clerk{e}, court{e}, craft{e}, faith{e}, folk{e}, kyng{e} (at least 70x), lord{e}, and very many others.

When it is not a plural, the genitive pronoun þine is generally altered to þy or þin (see IV.2.3.3).

The definite and indefinite adjectival inflections are unknown to the M scribe, but are imposed with some regularity by the final <-e> eraser. For details see IV.2.4.1.

In verbs, the frequent final <-e> for present 2nd and 3rd person singular (IV.2.5.1.7 and IV.2.5.1.8), for the preterite 2nd and 3rd person singular of weak verbs (IV.2.5.2.2 and IV.2.5.2.3) and the preterite 1st and 3rd person singular of strong verbs (IV.2.5.3.1, IV.2.5.3.3) is generally erased, and the removal of final <-e> on weak past participles is assiduously pursued (see IV.2.5.2.5).

IV.2.2 Nouns:

IV.2.2.1 Nominative/Accusative Singular: nil

IV.2.2.2 Genitive Singular: <-es> ~ <-s> ~ <-us> ~ nil

The ending is usually <-es>, but <-us>, sporadically corrected to <-es>, is not at all uncommon. LALME 1, dot maps 642 (Northern) and 958 (Southern) record the predominantly West Midland distribution of <-us> as a plural ending. LALME 4, lists substantive plurals as item 56, with <-us> as a majority form mainly in Western texts.

IV.2.2.3 Dative Singular: nil ~ <-e>

The distinction between inflected and non-inflected uses is often obscured by the scribe's tendency to write <-e> for both. However, the <-e> is often erased when it is not appropriate.

The inflected form of "wife" is never used; so bi his furste wiff (9.19) (cf. W's bi his firste wyue).

All fifteen occurrences of "ground" are spelled with final <-e> whatever their grammatical function. Prepositional uses are not altered (ten occurrences). The remaining five examples are corrected to ground{e}; four of these are not prepositional, but 12.201 is a hyper-correction since the noun both follows a preposition and is at line-end: sitte ... on þe ground{e}.NThe form at this point in W is the expected grounde. It may be noted, however, that in W.10.235 the uninflected form is used with a preposition: vp-on þis ground where M has the expected form with <-e>.

IV.2.2.4 Nominative/Accusative Plural: <-es> ~ <-s> ~ <-us> ~ (<-z>) ~ <-en> ~ <-n> ~ nil

As in W, words ending in <t> add plural <-z>, which is used nowhere else as a plural ending. The only example of plural <-ys> is saint[y]s (P.47), itself a correction.

IV.2.2.5 Genitive Plural: <-es> ~ <-s> ~ <-us> ~ (<-ene>) ~ (<-en>)

IV.2.3 Pronouns:

IV.2.3.1 Nominative Singular:

1st Person: I ~ (y) ~ ich

The form ich occurs seven times, twice before <ȝ-> corrected to <y-> (ich yede 7.157 and ich yeue 12.145), and otherwise before <h-> (e.g. ich hatte 15.24). The form y occurs just twice: 5.171, 9.128. Archetypal ik occurs in the phrase so theik (5.230), where Langland's joke is at the expense of the Norfolk dialect of Sir Hervey, as in Chaucer's Reeve's Tale.

2nd Person: þow ~ þou ~ thow ~ (Thou)

Thou is used twice, at the beginning of lines 5.277 and 16.163.

3rd Person:

     Masculine: he

     Feminine: she ~ (heo)

The two occurrences of heo (1.75, 3.29) are in alliterating positions. At 5.647 where She has been written over a deletion, Heo is required for the alliteration. NSee M. L. Samuels, "Langland's Dialect," Medium Ævum 54 (1985), 241-43; reprinted in The English of Chaucer and his Contemporaries, ed. J. J. Smith (Aberdeen, 1989), 70-85.

     Neuter: it ~ (hit)

The spelling hit remains only at 19.328.

IV.2.3.2 Accusative and Dative Singular:

1st Person: me

2nd Person: þe ~ the ~ (þee)

There are three instances of þee (3.174, 6.275, 8.52), all alterations.

3rd Person:

     Masculine: hym ~ him ~ (hem)

The form hem occurs in an unambiguous context in 13.219, but has been corrected to him or hym at 16.86 and 19.227, and likely as well at 4.55, 6.323, 6.329, 7.6, 11.423, 13.177, and 19.74.

     Feminine: hire ~ (hir) ~ (here)

The main form is hire. The form Hir is found just once as an alteration (2.15), and here is rare (e.g. 5.65, 18.438).

     Neuter: it

IV.2.3.3 Genitive Singular:

1st Person: my ~ myn ~ myne

Myn occurs before vowel or <h>, both singular and plural. Its use before a consonant is often corrected by the deletion of <-n>, e.g. 5.124 my{n} swellynge.

The eight occurrences of myne are before vowel or <h>, or are disjunctive or used as an absolute (18.337), and all are plural except for 6.149 myne almesse where the noun was perhaps understood as plural. In 5.651 the corrector has deleted <-e> in myn{e} he[ed].

2nd Person: þi ~ (thi) ~ þy ~ thy ~ þin ~ þine ~ (þyn) ~ (þyne) ~ (thyne)

Of the 145 instances of þi, only one is before <h-> (3.238) and never before a vowel. The spelling þy (including þy-selue(n)) is used 39 times, only once before passus 5 and only three times after passus 9. In all but five instances it is a correction, since in the practice of the original scribe þine is both singular and plural, and is used before vowel or consonant. For the corrector, however, þine is properly plural, and so the singular is generally altered to þy before a noun beginning with a consonant (e.g. 5.493) and to þin before vowel or <h> (e.g. 1.144). The corrector can be seen at work in But {ȝ}if it were with þ{ine}[y] tonge or with þine two handus (5.299).

3rd Person:

     Masculine: his ~ hise ~ (is)

The general form is his used with singular and plural nouns. The inflected form hise, developed by analogy with myne and þine, is used with plural nouns and also in absolute use to mean "his people." It is twice used with singular nouns (12.290, 19.170) and is not corrected. For examples of is see IV.1.2.6

     Feminine: hire ~ (here)

The usual form is hire, but there are occasional instances of here (e.g. 3.43, 3.46)

     Neuter: his (12.259 [in a correction], 17.250)

IV.2.3.4 Nominative Plural:

1st Person: we

2nd Person: ye ~ ȝe

The scribe's form ȝe has been consistently corrected and is preserved only 6 times. Line 3.72, An auenture ȝe haue ȝoure hire here and ȝoure heuen als, is an obvious example of a line the corrector missed.

3rd Person: þei ~ (þey) ~ þai ~ (þay) ~ thei ~ (They) ~ (hij) ~ (he)

The predominant spelling is þei. They is in line-initial position only (4x); otherwise (4x) it is a correction. There are seven instances where the scribe has written hij (P.66,NThis, clearly an archetypal reading, is crossed through and overwritten in a later hand. 1.59, 1.195, 3.341, 5.116, 6.15, 10.321). Of these the first three are also in W and are in alliterating positions. It is not clear whether the scribe recognized the plural form in he at 4.27.

IV.2.3.5 Accusative and Dative Plural:

1st Person: vs

2nd Person: yow ~ ȝow

The spelling ȝow is preserved only three times.

3rd Person: hem

IV.2.3.6 Genitive Plural:

1st Person: oure ~ (owre)

2nd Person: youre ~ yowre ~ (ȝoure) ~ (ȝowre)

The only two surviving instances of ȝoure are both in 3.72 quoted above (IV.2.3.4). ȝowre is preserved once (10.152).

3rd Person: here ~ (her) ~ hire ~ (hir)

There are two occurrences of her (P.30, 10.99). The nine instances of hir are all corrections. There are no oblique plural forms beginning with <þ->.

IV.2.3.7 Personal pronoun with "self":

Forms with -self (-selue, -seluen) are: my, þi/þy/thi, hym/him hire/here, oure and vs (7.141), your and yow (object 16.129),NIn the previous line, 16.128, yow-self is subject. Probably the syntax has become confused. hem.

For the line-terminal position the -selue(n) forms are always used. There are some corrections in this respect: line-terminal þi-self is altered to þi-sel[ue] in 16.164; cf. my-sel[ue] (18.338), and within the line my-sel[f] (15.168).

IV.2.4 Adjectives and Adverbs:

IV.2.4.1 Definite and Indefinite Adjectives:

In the language of W, monosyllabic adjectives ending in a consonant follow definite and indefinite inflections; i.e. <-e> is added in the plural, and also in the singular when used with the definite article, a demonstrative adjective or a possessive pronoun. In other cases the adjective is without inflection. So there is a distinction in W's usage between indefinite singular gret (or greet), and grete, "great." It is clear that the original scribe of M makes no such distinction. With the exception of gret (4.48), the scribal form of the adjective is grete. This form is retained where the adjective is definite singular (e.g. þe grete god [2.30]) or plural (e.g. Grete lobyes [P.55]). It is corrected on twelve occasions by the erasure of <-e> when the adjective is indefinite singular (e.g. 3.178, 5.390, 7.92). On a further nine occasions the appropriate correction is not made (2.151, 5.364, 8.9, 11.260, 13.361, 15.151, 18.314, 19.339, 20.20). In this total of twenty-one cases the spelling in W is without <-e> with only one exception, men of grete witte (8.9), where the form of the adjective is perhaps influenced by the preceding plural noun. The <-e> is also retained in M in this instance.

"Long" is a more complex case. The most frequent use is as an adverb, which is always longe in M as in W. There is also the dative of time and extentNDiscussed by Tauno F. Mustanoja, A Middle English Syntax, Part 1: Parts of Speech (Helsinki, 1960), 107-8; noted in J. D. Burnley, "Inflexion in Chaucer's Adjectives," Neuphilologische Mitteilungen 83 (1982), 169-77. going back to the OE dative, and examples of this in both M and W are longe tyme in 18.5 and 18.65, though in 5.580 where M properly has <e>, W does not. For this reason a longe lengthe may be right in 14.248 (W has a different reading). Both correctly have longe wille in 15.158, since it defines a proper name. French and Anglo-Norman monosyllabic adjectives follow the same pattern. For example, proud(e) is used correctly in M eleven times and incorrectly once (15.173). Among many adjectives that are corrected in this way are: blinde, bolde, colde, olde, depe, derke, faire, furste, glade, gode, ȝonge, nexte, harde, sadde, somme, suche, and warme.

Adverbs with the same form as adjectives retain their <-e>, e.g. faste (P.40); harde (17.159); etc.

IV.2.4.2 Polysyllabic Adjectives:

Polysyllabic adjectives of French derivation ending in <-ous> generally follow the same pattern:

But unlike the practice in W, religious used as a noun only once has <-e> for the plural (20.58).

IV.2.4.3 "All" and "both":

"All" has the following inflections: sg. al, pl. alle, gen. pl. aller (19.477). On a number of occasions when the scribe has written alle for the singular, the corrector erases <-e> and puts a bar through <-ll> to signal the end of the word (e.g. P.186, P.187, 2.90 etc.). Within corrected phrases this form all with barred <-ll> is used in 3.39, 17.196.

"Both" as an adjective is boþe or bothe, with genitive boþere (2.68) and boþes (18.39).

IV.2.4.4 Comparative:

IV.2.4.4.1 Adjectives: <-er> ~ <-re> ~ <-ere> ~ (<-ur>)

Both for adjective and adverb the scribal spellings for "better" are bettre ~ better ~ bettur. The corrected form is bettre, especially frequent in the first half of the text.

IV.2.4.4.2 Adverbs: <-er> ~ <ere>

Scribal <-e> is generally erased.

IV.2.4.5 Superlative:

IV.2.4.5.1 Adjectives: <-est> ~ <-este>

Scribal <-e> is often erased when the adjective is indefinite singular, following the practice of W.

IV.2.4.5.1 Adverbs: <-est> ~ <este>

The original scribe's <este>, including Dobest{e}, is regularly corrected.

IV.2.4.6 Adjectives in <-ly>:

The ending <-ly> varies with <-lich> and <-liche> (there are no examples of <-lye> or <-lie>). There seems to be no clear pattern of usage. As in W (and L), the spelling louely is always used for the attributive adjective, and louelich(e), vnlouelich(e) for the predicative. NIn the identical lines 5.573 and 8.85, the final <-e> on loueliche is erased on the second occasion. But this pattern does not apply to other adjectives, and presumably the usage simply reflects the archetype.

IV.2.4.7 Adverbs in <-ly>:

The same ending <-ly>, <-lich>, and <-liche> are used as in adjectives, and are equally unpatterned. NSee Hoyt N. Duggan, "Langland's Dialect and Final e," Studies in the Age of Chaucer 12 (1990), 157-91.

The comparative ending is <-loker> and once <-liker>: frendeloker 10.236; liȝtloker 15.450 ~ liȝtliker 12.156; wisloker 13.341. The ending <-lier> does not occur. Superlatives end once in <-lokest> and once <-iest{e}>: hastlokest 19.475; wikkedliest{e} 10.437.

IV.2.5 Verbs:

IV.2.5.1 Present Tense:

IV.2.5.1.1 Infinitive: <-e> ~ <-en> ~ (<-un>) ~ <-n> ~ nil

Final <-e> is often added: put[e] (19.63); shew[e] (15.558); wassh[e] (13.32); etc.

Endings derived from OE <-ian> verbs are frequently but not always preserved; hence the following infinitive forms with <-i-> or <-y->: erie (6.109) ~ erien (6.4); louye (5.50) ~ louyen (11.111); prikye (18.11); swerye (14.40); wanye (7.58). This is a feature of South-Western dialects.NSee M. L. Samuels, "Dialect and Grammar," in A Companion to Piers Plowman, ed. John A. Alford (Berkeley, 1988), p. 217. Oddly, it is more frequent in W, perhaps as a relict form. However, forms such as haten (10.99) and sweren (5.578) show the lack of consistency in this respect.

IV.2.5.1.2 Gerund: <-yng(e)> ~ <-ing(e)> ~ (<-enge>)

In both the gerund and the pres. ppl. the predominant ending is <-yng(e)>, though <-inge> is not uncommon.

After <y>:

The ending <-enge> occurs once only after <i> in burienge (11.80).

IV.2.5.1.3 Present participle: <-yng(e)> ~ (<-inge>)

There are no examples of <-ing>, and <-yng> is rare:

Other forms such as <-ande>, <-ende>, <-inde> or <-enge> do not appear.

IV.2.5.1.4. Imperative Singular: nil ~ <-e>

IV.2.5.1.5 Imperative Plural: <-(e)th> ~ <-e> ~ nil

The form with <-e> or without ending is commonly used before a subject pronoun: be ye 3.87; dyuyne ye P.210; leue ye 5.307; stonde ye 5.598; worche ye 9.187.

IV.2.5.1.6 Present 1st Singular: <-e> ~ nil

With ending derived from OE <-ian> verbs:

As in OE, stems ending in a vowel have no inflection: do (5.116); see (P.202).

IV.2.5.1.7 Present 2nd Singular: <-est(e)> ~ <-st(e)> ~ (<-xte>)

The usual ending is <-(e)st>, with scribal <-e> often erased.

Apart from miȝt(e) etc., "may" also has pres. 2 sg. may (19.485); etc.

IV.2.5.1.8 Present 3rd Singular: <-eth(e)>NThe ending <-eh> at 10.208, 11.98, 13.157, and 14.307 is presumably an error. ~ <-eþ> ~ <-th> ~ <-þ> ~ (<-uth>) ~ (<-uþ>) ~ <-t>

With ending derived from OE <-ian> verbs:NIn some dialects of OE the <-i-> was extended by analogy to the 3rd person. See A. Campbell, Old English Grammar (Oxford, 1959), para 757. The corrector sometimes alters to this form.

OE preterite-present verbs without inflection in the present 1st and 3rd sg. are, e.g.: dar (P.210); can (P.200); May (1.64); Shal (2.34); wot (5.226) ~ wot{e} (5.182).

IV.2.5.1.9 Present Plural (and Subjunctive): <-e> ~ <-en> ~ (<-un>) ~ <-eþ> ~ <-eth> ~ <-uþ>

With ending derived from OE <-ian> verbs:

The form in <-eþ> (or <-eth>, <-uþ>) is not uncommon. Samuels points out that this plural form is very rare in the London English of Chaucer, but is retained in Southern and South-Western areas until after Langland's death. He also comments on the form aren in alliterating position as evidence for Langland's West Midland dialect. NM. L. Samuels, "Dialect and Grammar," in A Companion to Piers Plowman, ed. John A. Alford (Berkeley, 1988), pp. 209 and 216. Some of the <-e> ~ <-en> forms will historically be subjunctives since they occur in contexts where a subjunctive is to be expected.

The plural forms of preterite-present verbs are, for example: can ~ kan ~ conne ~ conneth; may ~ mowe ~ mowen; shal ~ shul ~ shulle ~ shullen ~ shullun.

IV.2.5.1.10 Subjunctive Singular: <-e> ~ nil

The forms are the same as those of the 1st indic. sg.

IV.2.5.2 Preterite: Weak Verbs:

IV.2.5.2.1 Preterite 1st Singular: <-ed> ~ <-ede> ~ <-de> ~ <-te> ~ <-ud>

The forms with and without final <-e> occur in free variation. The correctors sometimes correct <-ud> to <-ed> (e.g. courb[e]d [2.1]; wip[e]d [16.175]) but often let it stand. The endings <-id(e)> ~ <-yd> do not occur.

IV.2.5.2.2 Preterite 2nd Singular: <-dest> ~ <-edest> ~ (<-edeste>)

IV.2.5.2.3 Preterite 3rd Singular: <-ed> ~ <-ede> ~ <-de> ~ <-ud> ~ <-ude> ~ <-te>

The forms are of course the same as those for the 1st singular.

IV.2.5.2.4 Preterite Plural: <-ed> ~ <-ede> ~ <-eden> ~ <-ud> ~ <-ude> ~ (<-uden>) ~ <-de> ~ <-den> ~ <-t> ~ <-ten>

IV.2.5.2.5 Past Participle: <-ed> ~ <-ede> ~ <ud> ~ <-ud{e}> ~ <-t> (with or without <y-> prefix)

The scribe's final <-e> is consistently corrected.

IV.2.5.3 Preterite: Strong Verbs:

IV.2.5.3.1 Preterite 1st Singular: <-e> ~ nil

Final <-e> is often deleted.

IV.2.5.3.2 Preterite 2nd Singular: <-e> (often with vowel gradation)

In the case of took (20.7), the appropriate final <-e> has not been added by the corrector.

IV.2.5.3.3 Preterite 3rd Singular: <-e> ~ nil

The forms are of course the same as those for the 1st singular, though final <-e> has been more consistently corrected. It is perhaps accepted as a marker of vowel length in forms such as stode and toke.

IV.2.5.3.4 Preterite Plural: <-e> ~ <-en> ~ nil

IV.2.5.3.5 Preterite Subjunctive Singular: <-e> (often with vowel gradation)

The forms are the same as the 2nd singular.

IV.2.5.3.6 Past Participle: <-e> ~ <-(e)n> ~ (<-un>) (with and without <y-> prefix)

V. List of Manuscript Sigils:

The Piers Plowman Electronic Archive uses a set of sigils that departs in some respects from the sigils used since Skeat's editions. The traditional set uses identical sigils to represent different manuscripts and different sigils to identify single manuscripts. For example, British Library Additional 10574 has no sigil at all for the A text, is B's Bm, and C's L. To avoid such confusion, the archive represents each manuscript with a unique sigil.

V.1 A Manuscripts:

V.2 B Manuscripts:

V.3 C Manuscripts:

V.4 AB Splice:

V.5 AC Splices:

V.6 ABC Splices:

VI. Bibliography:

VI.1 Editions:

Adams, Robert, ed. The Piers Plowman Electronic Archive, vol. 6: Oxford, Bodleian Library, MS Rawlinson Poetry 38, and London, British Library, MS Lansdowne 398 (R), SEENET, Series A (forthcoming).

Duggan, Hoyt N., and Ralph Hanna, III, eds. The Piers Plowman Electronic Archive, vol. 4: Oxford, Bodleian Library, MS Laud Misc. 581 (L), SEENET, Series A.6. Cambridge: Published for the Medieval Academy of America and The Society for Early English and Norse Electronic Texts (SEENET) by Boydell and Brewer Ltd. 2005.

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

Kane, George, 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: Athlone Press; Berkeley and Los Angeles: University of California Press, 1988.

Russell, George, and George Kane, eds. Piers Plowman: The C Version. Will's Visions of Piers Plowman, Do-well, Do-Better and Do-Best. An Edition in the Form of Huntington Library MS Hm 143, Corrected and Restored from the Known Evidence, with Variant Readings. London: Athlone Press; Berkeley and Los Angeles: University of California Press, 1997.

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. 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, Walter W., ed. The Vision of William Concerning Piers the Plowman together with Vita de Dowel, Dobet, et Dobest secundum Wit et Resoun by William Langland: Part II. The "Crowley" Text; or Text B. EETS OS 38. London: N. Truebner, 1869.

—, ed. The Vision of William Concerning Piers the Plowman, in Three Parallel Texts Together With Richard the Redeless By William Langland (About 1362-1399 A.D.). 2 vols. Oxford: Clarendon Press, 1886.

Turville-Petre, Thorlac, and Hoyt N. Duggan, eds. The Piers Plowman Electronic Archive, vol. 2: Cambridge, Trinity College, MS B.15.17 (W), SEENET, Series A.2. Ann Arbor: SEENET and The University of Michigan Press, 2000.

VI.2 Studies:

Adams, Robert. "The Reliability of the Rubrics in the B-Text of Piers Plowman." Medium Ævum 54 (1985): 208-31.

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

—. "Evidence for the Stemma of the Piers Plowman B Manuscripts." Studies in Bibliography 53 (2000): 173-194.

Benson, C. David, and Lynne S. Blanchfield with acknowledgements to the work of Marie-Claire Uhart. The Manuscripts of Piers Plowman: The B-Version. Cambridge: D. S. Brewer, 1997.

Burnley, David. "Inflexion in Chaucer's Adjectives." Neuphilologische Mitteilungen 83 (1982): 169-77.

Campbell, A. Old English Grammar. Oxford: Clarendon Press, 1959.

Clopper, Lawrence M. "A Response to Robert Adams, 'Langland's Ordinatio.'" Yearbook of Langland Studies 9 (1995): 141-146.

Doyle, A. I. "Remarks on Surviving Manuscripts of Piers Plowman." In Medieval English Religious and Ethical Literature: Essays in Honour of G. H. Russell. Edited by Gregory Kratzmann and James Simpson, 35-48. Cambridge, Eng.: D. S. Brewer, 1986.

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

Hailey, Robert Carter. "Giving light to the reader: Robert Crowley's editions of Piers Plowman." Ph.D. dissertation, University of Virginia, 2001.

Hanna III, Ralph. William Langland. Authors of the Middle Ages: English Writers of the Late Middle Ages 3. Aldershot: Variorum, 1993.

Horobin, Simon. The Language of the Chaucer Tradition. Cambridge: D. S. Brewer, 2003.

Jordan, Richard. Handbook of Middle English Grammar: Phonology. Trans. and rev. Eugene J. Crook. The Hague: Mouton, 1974.

Macaulay, G. C., ed. The English Works of John Gower. 2 vols. EETS ES 81-82. Oxford: Oxford University Press, 1900-1901.

McIntosh, Angus, M. L. Samuels and Michael Benskin, with the assistance of Margaret Laing and Keith Williamson, eds. A Linguistic Atlas of Late Mediaeval English. 4 volumes. Aberdeen: Aberdeen University Press, 1986.

Mossé, Fernand. A Handbook of Middle English. Trans. James A. Walker. Baltimore: Johns Hopkins Press, 1952.

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.

Samuels, M. L. "Some Applications of Middle English Dialectology." English Studies 44 (1963): 81-94. Reprinted in Middle English Dialectology: Essays on Some Principles and Problems by Angus McIntosh, M. L. Samuels and Margaret Laing. Edited by Margaret Laing, 64-80. Aberdeen: Aberdeen University Press, 1989.

—. "Chaucer's Spelling." In Middle English Studies Presented to Norman Davis in Honour of his Seventieth Birthday. Edited by Douglas Gray and E. G. Stanley, 17-37. Oxford: Oxford University Press, 1983. Reprinted in The English of Chaucer and his Contemporaries: Essays by M. L. Samuels and J. J. Smith. Edited by J. J. Smith, 23-37. Aberdeen: Aberdeen University Press, 1988.

—. "Langland's Dialect." Medium Ævum 54 (1985): 232-47 with corrections in Medium Ævum 55 (1986): 40. Reprinted in The English of Chaucer and his Contemporaries: Essays by M. L. Samuels and J. J. Smith. Edited by J. J. Smith, 70-85. Aberdeen: Aberdeen University Press, 1988.

—. "Dialect and Grammar." In A Companion to Piers Plowman. Edited by John A. Alford, 201-221. Berkeley and Los Angeles: University of California Press, 1988.

Smith, Jeremy J. "The Language of the Ellesmere Manuscript." In The Ellesmere Chaucer: Essays in Interpretation. Edited by Martin Stevens and Daniel Woodward, 69-86. San Marino, California: Huntington Library and Tokyo: Yushodo, 1995.

Turville-Petre, Thorlac. "Putting it Right: The Corrections of Huntington Library MS. Hm 128 and BL Additional MS. 35287." Yearbook of Langland Studies 16 (2002): 41-65.