The Piers Plowman Electronic Archive, Vol. 7: <list type="simple" TEIform="list"> <item TEIform="item">London, British Library, MS Lansdowne 398 </item> <item TEIform="item">Oxford, Bodleian Library, MS Rawlinson Poetry 38</item> </list> William Langland Edited by Robert Adams Associate Editors: Patricia R. Bart, M. Gail Duggan and Catherine A. Farley Technical Editor: Daniel Pitti Graduate Research Assistant Michael Blum, John Ivor Carlson, Carrie Lindley, Janice McCoy, Ashley Opps, Timothy L. Stinson, and Jordan Taylor. Computer Consultants and Programmers Robert Bingler, Shayne Brandon, Cynthia Girard, Chris Jessee, Daniel Pitti, David Seaman, and John Unsworth. The Medieval Academy of America and SEENET by Boydell and Brewer, LTD Woodbridge, Suffolk ISBN: (Individual) 9781843840947 (Institutional): 9781843840930

Commercially available:

copyright 2011, by SEENET

Combined facsimile & documentary edition. Identification of hands IRA = Robert Adams 1 computer optical disk : col. ; 4 3/4 in. The Medieval Academy of America and SEENET, by Boydell and Brewer, LTD. Woodbridge, Suffolk Source copy consulted: London, British Library, MS Lansdowne 398 and Oxford, Bodleian Library, MS Rawlinson Poetry 38


Attribute Values brown ink anglicana bastarda italic Lombard Cap ornamented capital, N lines high rubricated touched in red textura underlined underlined in red green ink blue ink

Latin French German November 2005 revise header, update DTD and ent files Hoyt N. Duggan New header created
Prologus de visione Petri Plowman R.P.0:

Fols. 77r and 80v are both very faded with considerable bleed-through from their opposite sides. In a few cases, the midline punctus elevatus must simply be inferred because the fading makes it indistinct though still visible. Moreover, the bottom and outer margins of all eight sides bound in Lansdowne 398 (77r-80v) are severely cropped.

More than one hundred lines at the beginning of the Prologue (= KD Prol.1-124) as well as the end of Passus 1 and the beginning of Passus 2 (= KD1.141 through KD2.40) have been lost from MS R since the early eighteenth century (at the latest). It is likely, but unproveable, that the event behind this loss also accounts for the fact that the surviving first few leaves of MS R (usually referenced as Oxford, Bodleian MS Rawlinson Poetry 38) are now bound into a completely different codex, London, British Library MS Landsowne 398. For more information, see Introduction, I.10."

Crist kepe þe sire kyng e  and þi kyngriche . And lene þe lede þi land  so lewete þe louye . And fore þi riȝtful rewelyng  be rewarded in heuene . And sithen in þe heir an hey  an aungel of heuene . Lowede to speke in latyne  fore lewede men ne cowede Iangele ne Iuge  þat iustifie hem schulde . Bute suffren an seruen for -þi seyde þe angel . Su m rex su m princeps · veutru m [n]eutru m neutrum R.P.8: The form of the <v> in veutrum is slightly blurred, but there is no possibility of transcribing it as <n>. This is a unique error in R. fortasse deinceps . A light brown brace in the right margin links RP.8-14. There is also some evidence of light brown underlining of the a-verses in RP.9-14, but it may be merely the ruling, which is quite visible on this side. O qui iura regis  christi specialia regis . Hoc quod agas melius  iustus es esto pius . Nudum vis ius R.P.11: Alpha reads vis. All beta family manuscripts read ius with the C archetype (a few C copies agree with alpha). a te  vestire vestiri R.P.11: Alpha is responsible for vestire, though the reading is shared by H. All authoritative Beta witnesses have vestiri, which is also the reading of the C version. wlt R.P.11: R's wlt is a common English scribal spelling for Latin vult; it appears to have resulted from an association between the customary phonetic value assigned to Latin initial <v> (= /w/) and a popular sense of the origin of the <w> graph itself as a "double v." Because many other European languages, as well as Latin, had no <w> graph, and foundries therefore omitted the form from standard typographic templates, English printers often resorted to using the digraph <vv> to represent English <w> till the end of the seventeenth century. Cf. R.7.65: ( wltis), R.12.71:, and R.15.43: (both wlt). pietate . Qualia vis metere  talia grana sere . Si vis ius R.P.13: Alpha reads vis. All beta family manuscripts read ius. nudatur  nudo de iure metatur . Si seritur pietur pietas pietas pietas de pietate metas . Þanne greuede hym a golyardas  a gloton of wordes . And to þe angel an heiȝ  answeres answerd R.P.16: Only L agrees with R in this reading; F and all other beta manuscripts agree on the preterite form, answerde. after . Dum rex a regere  dicatur nomen h abere . In both of these cases, the red boxing has been heavily rubbed. Nomen habet sine re  nisi studet iurea tenere . And þanne gan alle a þe R.P.19: R offers two unique readings in this line: (1) where the beta copies have þe comune(s) , R reads a comoune ; (2) where beta attests in vers, R has in a vers . F's reading of this line is completely erratic. comoune crie  in a nil vers of latyn . R.P.19: Immediately hereafter, alpha omits a line present in beta: To þe kynges conseille construe ho -so wolde . Precepta regis sunt nobis vincula legis . With þat ran þere a route of ratones  at ones . And smale mys with hem  mo þan a þousand . And comen to a conseyle  fore here comoune p rofitt e . For a cat of a courte  com whan hym lykede . And ouer -leep e hem liȝthliche  and lauȝt hem at his wille . And pleyed R.P.26: There is a small hole in the parchment here, causing the <d> of pleyed to be written slightly above the rest of the line. with hem periliousely  and possed hem aboute . For doute of diuerse dredes  we dare nouȝt wel loke . And ȝif we grucchen his of his ga mmen R.P.28: R's phrase here is unique; beta and the C version attest of his game(n) , while F reads his wille.  he will e greue vs alle . Cracche vs or clawe vs  and in his cloches halde . Þat vs lothes þe lyf  er he lete vs passe . Miȝt we with any witt  his wille with -stonde . We miȝte be lordes on lofte  and lyuen at oure eese . A raton of renou n  moste resonable R.P.33: R shares this reading with Hm and with the C version; beta and F read renable. of tonge . The rubricating scribe managed to touch each of the first characters on this side in red, at least so far as cropping permits us to judge. None of the initials survives in whole, but we have marked initials as rubricated where sufficient fragments remain to verify that fact. S eyde for a souereyne  helpe to hem alle hymselue(n) . R.P.34: In place of alpha's hem alle, beta reads hym-selue. I haue I -seyne segges q uod hee  in þe citee of london . B eren beȝes ful bryȝtte  abouten here nekkes . A nd somme coleres of crafty werke  vncoupled þei wenden . B oþe in warayne and in waste  whar e þaym lef lykeþ . R.P.38: There is considerable variation in attestation for the final phrase. F reads hem best lykeþ. Some of the beta copies show hem-self liked; others read hem leue liketh. A nd other while þei aren elles -where  as I here telle . W ere þere abelle a belle on here bye  by Ihesus Ihesu R.P.40: Beta witnesses have Ihesu. The C manuscripts are divided, a majority reading with beta but a significant minority, in both major branches, agreeing with alpha. as me thenketh . M en miȝt wite where þei wente  and away renne . A nd riȝt so q uod þat raton  reson me scheweth . T o bugge a belle of bras  or of briȝt siluer . A nd knetten it on a coler  for oure comoune p rofitt e . And hangen it vp -on þe cattes hals  þan here we mowen . R.P.45: Alpha lacks the following two lines attested by beta witnesses and by Cx: Where he ritt or rest or renneth to playe And ȝif him list for to laike þenne loke we mowen. And peren in his presens  þere while hym plei lyketh . And ȝif hym wraþheth ben y -war  and hys wey shonye . ¶ Alle þe þis route of þe nil R.P.48: The second þe in this line is R's unique addition. The first þe is an alpha variant (beta has þis). The C manuscripts are divided on this reading, a majority of the X family agreeing with beta, while a majority of the P family agrees with alpha. ratones  to þis resou n þei assented . Ac þo þe belle was I -bouȝt e  and on þe bey I -hanged hanged . Þere ne was raton in alle route þe route  for alle þe reme of france . Þat durste haue I -bounde þe belle  aboute þe cattes nekke . Ne hangen it aboute his þe cattes hals R.P.52: Beta reads þe cattes hals. Alpha's his hals is also the reading of the C version.  alle englond to wynne . And helden hem vn -hardy  and here conseyll e feble . And letu n here labour I -lost lost  and alle here longe studie . ¶ A mous þat muche goed  couthe as me thouȝte . Strok forth sternely  and stod by -forn hem alle . And to þe route of ratou ns  rehersed þis wordes . ¶ Þouȝ we had nil I -culled culled(en) R.P.58: In place of alpha's ( had I -culled ), beta witnesses read either kulled (LMCO) or killen (CrWHmG). The C version agrees on this reading with alpha. þe cat  yut schulde þer e cu m an -other e . To crachy vs and alle our e kynde  þouȝ we crope vnder benches . For -þi I conseill e alle þe comoune  to late þe catt I -worthe worthe . And be we neuer so bolde  þe belle hym to schewe . R.P.61: See note at RP.66. For I herde my sire seyen  is seuen ȝere I -passed . Þere þe cat ys a kytou n  þe court is ful elyng e . Þat witnesseth holy writt  who -so wile it rede . Ve terre vbi puer est rex rex est &c . For may no renk no þer R.P.65: F omits R's double negative and beta reads þer reste in place of R's no reste . reste haue  for ratones by nyȝte . It is unclear why the scribe's lineation went wrong here (running together the Latin quotation with half the following English line, so that the b-verse ended up relegated to a line of its own). It seems likely that the problem is related to a larger problem of textual order between KDProl.188 and 197 in Bx (See Kane-Donaldson, p. 176, for their editorial reasoning on re-ordering this passage). However, there is no evidence that the R scribe had any clue about the source of the problem. The <f> at the head of RP.66 is not capitalized, but it is touched in red, and the small + in the left margin is probably irrelevant; similar marks are found beside most lines that were to have been boxed in red (unnoted in this edition). Þe while he caccheth conynges  he coueyteth nauȝt our e caray ne But fet hym alle wyth veneson  defame we hym neuere . For better is a litel los  þen a longe sorwe . Þe mase a -monge vs alle  þeiȝ we misse a schrewe . For many mann us malt e  we muys wolde destruye . And also ȝe route of ratones  rende mennes clothes . Nere þat cat of þe courte  þat can ȝow ouer e -lepe . For hadde ȝe ratones ȝour e wille  ȝe couthe nouȝt rewele ȝour e -selue I seye it nil R.P.75: R's it is a unique addition to the line as read by the beta witnesses (at this same point, F uniquely adds þis). However, Cx agrees on this reading with R. for my -self me R.P.75: Beta reads me, a lection also supported by the C version. q uod þe mous  I see so muchel after . Schal neuere þe cat ne þe kyton  by my conseil be greued . Ne carpyng of þis colere  þat costed me neuere . And þouȝ it costed had cost(n)e(d) me R.P.78: R's it costed me agrees in substance with FG, which also read a simple preterite. Beta has it had cost(n)e(d) me. The alpha reading is also that of the C version. catel  beknowen it Inolde I nolde . But suffre as hym -self R.P.79: There is an ink dot (smaller than a punctus) immediately after hym-self. From the lack of spacing, it appears to be accidental. wolde  to do as hym liketh . Coupled and vn -coupled  to cacche what þei mowe . For Forþi R.P.81: R's For is unique; F has & while beta reads For-þi. vche a wyse man R.P.81: Cf. beta's wiȝte and F's Mows. I warne wite wel hys owene . What þis meteles by -mene  þe ȝe R.P.82: Unfortunately, R's hand often fails to distinguish þ from y. As was the judgment of Kane-Donaldson, our guess here, judging by proximate renderings of both characters, is that R intended þe; if that was, in fact, the scribe's intention, the resulting reading is unique error; the other B copies read ȝe. men þat ben myrie . Deuyne ye ne dar I nouȝt for I ne dar R.P.83: F rephrases this a-verse extensively; in place of R's ne dar I nouȝt, beta reads for I ne dar, a phrase that agrees completely with the reading of the C version.  by dere god in heuene  R.P.84: The paraph marker for this strophe was either partially erased or badly rubbed. Ȝeet houed þere an hundreth  in houues of silke  Seriauntz it semede  at þat R.P.85: R's at (cf. R2.11) is formally unique among the B witnesses, which uniformly attest the expected þat; one C manuscript (Dc) shows the same clipped form as R's. Otherwise all agree with Bx. s erueden at þe barre . Pletedden for penies  and pondres poundes R.P.86: Beta reads poundes (which is also the reading of Ax and Cx) while F has pownded. R's reading, which was misconstrued by Kane-Donaldson and Schmidt as poudres, is unique among the B witnesses (manuscript E in the A tradition also attests ponderys). MED, s. v. ponder, defines pondres as "balance scales" or "weights," but the variant is almost certainly scribal. þe lawe . And nouȝt for loue of oure lorde  vnlese here lippes ones . Þow miȝtest e better mete miste  on maluerne hulles . Þan gete a mu mme of here mouth  her but mony hem by were R.P.89: In place of alpha's er, beta copies read til or but. Only R adds hem after mony. Later in this half-line, most B witnesses read be, but LMCY attest were. Among the C manuscripts, most P family witnesses agree with beta's til, while a majority of the X family agrees with alpha's er. However, at the end of the half-line, the C families fragment their allegiances, most of the P group (and four of the X group — P 2CotLcBo) now aligning with the more widely attested B reading, be, while some of the best X witnesses read wer(e), agreeing with LMCY. Both sets of variants were already attested in various A copies. I -she wed Barons and burgeis  and bonde -men alse . I seiȝ in þis assemble  as ȝe schul here after Baxsteres and brewsteres  and bocheres manye . Wolle websteres  and weueres of lynnen . Taillours and tynkeres  and tolleres in marketes . Masons and mynours  and many other craftes . Of alle alkin libbynge R.P.96: Beta reads alkin libbyng. The entire line is missing from the A and C versions, but it is witnessed in Z (MS Bodley 851), where the phrasing in question is identical to that found in alpha. laboreres  lopen forthe so mme . As dikares and deluares  þat doth here dedes ille . The entire outer margin is cropped, but as far down as the passus division, all initials are visible and all are touched in red. After the rubricated heading for Passus 1, the cropping becomes severe. And dryueth forth here dayes here þe longe day R.P.98: Cf. R's here dayes here with F's þe fayre day and beta's þe longe day. Beta's reading agrees with that of the A version, while the reading of R closely parallels that of Cx (= here days). . with deu vous saue dam emme . Kokes and here knaues  cryeden hote pies hote . Gode gris and gees  gowe go we dyne gowe go we . Tau erneres vntil hem  tolde þe same . Whitt wyne of osey  and of gascoyne red wyn of Gascoigne R.P.102: R's of gascoyne is deficient in length and alliteration to complete a full half-line, but it is substantively the same as the reading of the P family of C manuscripts. F has wyn of Gaskoyne, which essentially agrees with both the X family reading in the C tradition and with Ax. Beta reads, more explicitly, red wyn of Gascoigne. . Of þe ryn and þe rochel R.P.103: Beta reads and of þe Rochel . This is also the reading of Ax and Cx.  þe rost to defye . Al þis seyȝI seyȝ I slepynge  and seuen sithes more .
Passus R.1.0: The <P> of Passus in the heading is partially visible, but its specific identity is clear only from context. prim us de visione petri plowman W hat þis montayne by -meneth  and þe merke dale . It is clear that the initials of line 1 and line 3 were oversized, and that they were executed in blue and red respectively. But their specific identities are inferrable only from context; hence, they are bracketed and marked as supplied. And þe felde ful of folke  I schal ȝow fayre schewe . A lufly lady of lere  in lynnen I -clothed . Com dou n from a castel  and cald me fayre . And seyde sone slepestow  sest þow þis poeple . How bysi þei ben  a -bouten þe mase . Þe moste partie of þis poeple  þat passeth on þis erthe . Haue þe þe[i] þei worchipp e in þis worlde  þei willne no better . Of other heuene þan here  holden þei no tale . ¶ I was aferd of here face  þeiȝ she faire were . And seyde mercy madame  what is þis to mene ¶ Þe toure vp þe tofte q uod sche  treuthe is þere -inne . And wolde þat ȝe wrouȝte  as his worde techeth . For he is fader of feyth  formede ȝow alle . Bothe with fel and with face  and ȝaf ȝow fyue wittes . For to worchipp e hym þere -with  þe while þat ȝe ben here . Þat And R.1.17: The cropping of R here makes any reading speculative, but it is obvious from the surviving <t> that R's reading was unique; we conjecture Þat, but no other manuscript, including those of the A version (which attests the same line), shows this reading. þer efore he hiȝtte R.1.17: Or hiȝcte? In either case the form is unique in the manuscript, and the scribe does not steadily distinguish these graphies. þe erthe  to helpe ȝow echone . Of wollen of ly nnen  of lyflode at nede . An old ownership stamp for the British Library (in red) is found in the right margin opposite these lines. In mesurable manere  to make ȝow at ese . ¶ And comanded of his curteisye  in co mmou n thre thynges . Aren none nedful bute þo  and nempne hem I thenke . And reken hem by resou n  reherse þou hem after . Þat on his R.1.23: Compare this erasure to a similar instance at RP.89, where her was changed to er. vesture  fram chele þe to saue . And mete at þe mele  for mysayse of thy -selue . And drynk whan þow dryst e  ac do nouȝt e oute of resou n . Þat þow worthe þe wors  whan þow werche schuldest e For loth in his lif -dayes  for lykyng of drinke . Dede by his dowtres  þat þe deuel lykede . Delyted hym in drynke  as þe deuel wolde . And leccherye hym lauȝte  and lay by hem bothe . And al he witte it it þe wyn  þat wykked dede . In the space between these strophes, there is red offset from the passus rubric on 78v. Inebriemus eu m vino  dormiamusq ue cu m eo . Vt s eruare possimus de patre n ostro semen Þorȝ wyn and þorȝ wymmen  þer e was loth a -combred . And þere gat in glotonie  gerles þat were cherles . Forthi drede delytable drynke  and þow schalt do þe bettre . Mesure is medicyne  þeiȝ þow muche ȝerne . It is nouȝt al gode to þe goste  þat þe gut axeth . Ne lyflode to þi licam  þat leef is is to þi þe R.1.39: R's reading here is unique among the B copies; the others attest is to þi. Ax agrees with the B majority (only Bodley 851 agrees with R) , but Cx is uncertain. Though a majority of C witnesses also supports the predominant reading, a significant minority of X family manuscripts (XICotBmBoUc) agrees with R in omitting to. soule . Lef nauȝt þi licame  for a lyeȝer hym techeth . Þat is þe wrecched world  wold þe bytraye . For þe fende and þi flesch  folwen þe to -gydere . This and that sueth seeþ R.1.43: Cf. F's sewe and beta's seeþ. Although three C witnesses (P 2UcDc) support the alpha verb, it seems clear that Cx had the same reading as beta. þi soule  and seith it in þin herte . And for þow schuldest ben Iwar  I wisse þe þe beste . Ma -dame mercy q uod I  me lyketh wel ȝour e wordes . Ac þe monee of þis molde  þat men so fast kepeth holdeth . R.1.46: For alpha's kepeth or kepe (= F), beta reads holdeth. Beta agrees here with the A version while alpha agrees with the C version. Telleth Telle R.1.47: R's verb inflection, Telleth, is unique among the B copies (F and beta agree on Tel(le)). However, both Ax and Cx agree with R's form. me ma -dame  to whom R.1.47: F omits madame from the a-verse phrase, while beta transposes the phrase and reads to whom Madame. F's reading agrees with that found in the other two versions of the poem. þat tresour appendeth . Goo to þe gospell e q uod sche  þat god seyde hym -selue . Þo þe poeple hym apposede  with a peny in þe temple . Wheither þei schulde þer e -with  worchipp e þe kyng sesar . And god axed of hem  of whom spak þe lettre . And þe ymage Ilyk  þat þere -inne stondeth The rubricating scribe failed to touch line-initial letters in red on this opening. . Cesaris þei seyden  we sen hym wel vchone . Reddite cesari q uod god  þat cesari byfalleth . Et que sunt dei deo  or elles ȝe don ille . For riȝtful resou n  schuld rewele ȝow alle . And kende witt be wardeyn  ȝoure welthe to kepe . And tutour of ȝour e tresor  and taken it ȝow at nede . For housbondrye and heo hij R.1.59: Heo, "she." R.1.59: Beta reads hij for alpha's he(o). However, both Ax and Cx support F's version of alpha's lection (= he).  halden to -gyderes . An old ownership stamp, in black ink, runs vertically downwards in the right margin opposite these lines; the label says MVSEVM BRITTANICVM. ¶ Þanne I frayned hire hir faire R.1.60: R uniquely omits faire at the end of the a-verse.  for hym þat hire made . Þat dongoen in þe dale  þat dredful is of siȝtte . What e may it be to mene  madame I ȝow by -seche . ¶ Þat is a castel R.1.63: Both F and beta read þe castel . This is also the reading of both Ax and Cx. Among the B manuscripts, only H agrees with R on this lection. of care  who -so cometh þere -Inne . May banne þat he borne was  to body or to soule . Þere -Inne wonyeth a wyȝth  þat wronge is I -hote . Fader of falsed  and founded it hym -selue . d d d e Adam and eue  he egged to Ille . Conseyled caym  to cullen his brother . Iudas he Iapede  with Iewes siluer . And sethen on an ellerne  honged hym after . He is lettere of loue  and lyeth hem alle Þat tristeth on his tresor  bytrayed aren R.1.72: Beta copies show considerable variation in this phrase, but beta itself probably read bitrayeþ he. This is a simple reversal of the archetypal C reading, which is he bytrayeth. By contrast, alpha's reading here is identical to that of Ax. sonnest e . ¶ Þanne haued I wonder in my witt  what wo mman sche it R.1.73: For alpha's sche, beta reads it. Beta's lection agrees exactly with Ax, but the alpha reading is semantically the same as that of C, which is he. wer e . Þat suche wyse wordes  of holy wrytt schewed . ¶ And hasked R.1.75: Cf. F's halsede and beta's asked. Though both variants are available in the A tradition, Ax almost certainly agreed with F's reading. The same is true for Cx, though a third variant, hanslede, is generated in some copies—apparently by metathesis—from a variant spelling of the original (= halsnede) still found in manuscripts EcQSc. The evidence in the B tradition seems easiest to explain on the supposition that R's verb form, hasked, was that of Bx and that beta merely normalized its spelling while F corrected the reading itself through his habitual use of an A manuscript for proofing. hyre on þe hiȝe name  er heo þennes ȝode . What he she R.1.76: He, "she." were witterly  þat wissed me so faire . ¶ Holy cherche I am q uod he she R.1.77: He, "she."  þow hauȝtest me to knowe . I vnderfong þe ferst  and þe faith tauȝte . And brouȝtest me borwes  my byddyng to fulfille . And to loue lely me me lelly R.1.80: Beta transposes alpha's lely me as me lelly. Beta's word order is also that of the A version.  þe while þi lif dureth . Þanne I courbed on my knes  and cryed hire of grace . And preyed hire pytously  to preye R.1.82: For alpha's to preye, beta reads simply prey. Alpha's reading is supported by that of Ax. for my synnes . And also kenne me kendely  on cryst to byleue . R.1.83: Alpha omits the following line found in beta: That I miȝte worchen his wille þat wrouȝte me to man. The line is not present in the C version but is attested in the A version. Teche me to no tresor  bute telle me þis ilke . How I may saue my soule  þat seint art I -holde . Whanne alle tresores aren tryed q uod he she R.1.86: He, "she."  trewthe is þe beste . I do it on deus caritas to deme þe sothe . Þat it It R.1.88: For alpha's Þat it, , beta and the other two versional archetypes read It. is as derworthe a drewery  as dere god hym -selue . He Whoso R.1.89: R's He is unique; F reads He þat while beta reads Whoso. The A-version reading agrees with beta's. The C-version has a revised b-verse, but for this variant the manuscripts divide into the two large families usually seen, with the P family supporting R ( For he) while the X family is closer to beta ( For who ). is trewe of his tonge  and telleth non other . R.1.89: Ordinarily, throughout this manuscript, the scribe indicated where rubricated paraph markers should be inserted—alternately blue and red—by leaving a small cc in the margin. Many of these survive in the latter part of the manuscript, where very few rubricated paraphs were ever completed, but some have been erased, as is the case on fol. 80, where only the blank line between strophes attests scribal intentions. And doth þe werkes þer e -with  and wilneth noman ille . He is a god by þe gospel  a -grownde and a -lofte . And I -lyke to oure lorde  by seint lukes wordes . Þe clerkes þat knowe þis  schulde kennen it a -boute . For cristene and vncristene  claymeth it vchone . Kynges and kniȝtes  schulde kepe it by resou n . BRyden and rappe adoun  in reumes aboute . And taken transgressores  and teyȝen hem fast . Til treweth hadde I -termined  here trespas to þende . And þat is þe p rofessiou n apertelyche  þat appendeth for R.1.99: Most B copies, including F, read to knyȝtes . The latter reading is also that of the A and C version. kniȝt es And nouȝt to fasten a friday  in fyue score winter . But holden with him and wyth hire  þat wolden alle tr ewthe Ne And R.1.102: Ne is an alpha variant; beta reads And. Beta's reading here is also that of Ax and Cx. neuere leue hem for loue  ne for lacchyng e of seluer For dauid in his dayȝes  dubbede kniȝtes  And dede hem swere on here swerdes swerde R.1.104: The plural form is unique to R; the other B witnesses show swerde. Although three copies from the A tradition and one from C agree with R's plural, it is clear that the archetype in both of the other versions agreed with the B majority.  to serue trewth euere And who -so At this point there was an original diagonal tear in the parchment (repaired by stitching). The scribe has written around the tear, which also affects the spacing of R1.106 between crist and kyngene. passed þat poynt  was apostata in þa t ordre But crist kyngene kyng R.1.106: A partially erased <h> is visible, occupying the space between the <g> of king and the punctus elevatus. The <h> in question may connect with a partially erased ampersand above kyngene earlier in the line and indicate the occurrence of a fairly extensive erasure / correction in this passage.  kniȝtede ten . Cherubyn and seraphyn  suche seuene and othere . And ȝaf hem miȝt in his maieste  þe murgur hem tho uȝte And ouer his mene meyne  made hem archangeles Tauȝte R.1.110: The <e> of Tauȝte has been partially rubbed out but is still legible. hem by þe trinite  trewthe to knowe . And To R.1.111: For alpha's And, beta reads To, which is also the reading of the A version. be buxu m at his biddynge  he badde hem nauȝt ell es Lucifer with legiouns  lerned it in heuene . Cons t fol. 80. Til But for he brak buxu mnesse  his blisse gan he tyne And fel froo þat felaschypp e  in a fendes lyknesse . In -to a deep e derke helle  to dwelle þere fore euere . And mo thousantes with hym  þan man couthe nou mbre . Loupen ouȝt with lucifer  in lothelyche forme . For þei leueden vppon hym  þat lyede in þis manere . Ponam pedem in aquilone &c etera et similis ero altissimo R.1.119: R uniquely omits the end of this tag, as represented by the other B witnesses, as well as by C: et similis ero altissimo . . And alle þat hopede it miȝt be so  none heuene miȝt hem holde . But fellen outȝ in fendes lyknesse  nyne dayes to -gyderes . Til god of his goodenesse  gan stable and stynte . And garte þe heuene to stekye  and stonden in quiete . Whanne þis wykkede wente ouȝt  wonderwyse þei fellen . So mme in erthe so mme in eyre erthe R.1.125: Most beta copies (M being a notable exception) transpose the order of alpha's elements to Somme in eyre somme in erthe . This is also the order of Ax. However, Cx agrees with alpha's phrasing.  and so mme in helle deepe . Ac lucifer lowest  lyth of hem alle . For pruyde þat he pelt pult R.1.127: R's pelt is unique in spelling but substantively the same as the reading of Cr and L (= pult). F has pitte. Beta reads putte, which is also the reading of A. ouȝt  his peyne hath none ende . ¶ And alle þat worche with wronge  wende þei schulle . After here deth -day . to and R.1.129: For alpha's to, beta reads and, which is also the reading of the A and C versions. dwelle with þat schrewe ¶ Ac þoo þat werche wel  as holy writt telleth . And enden as I oer [e]r ere seyde  in trewthe þat is þe beste . Mowe be siker  þat here soule  schal wende to heuene . Þere trewthe is in trinite  for to and saue troneth R.1.133: In place of alpha's relatively vague saue(n), beta shows the properly alliterating troneth. The entire b-verse in beta ( and troneth hem alle ) is identical to that found in the A version . Although the C version is revised at this point, the appearance of trone in the P sub-archetype of its cognate line (X = tour) also suggests beta's superiority here. R's for to preceding the verb in question is unique; F has þat shal while beta reads and. hem alle . ¶ For -þi I seye as I seyde  ere by bi siȝte of þis tixtes . R.1.134: Evidently, alpha had dropped the third stave of this line, producing a short b-verse; beta, which agrees exactly with Ax here, reads bi siȝte of þise textis . Whan alle tresores ben R.1.135: Beta reads arn(e), which is also the probable reading of Ax. The C version is revised at this point. Itried  trewthe is þe R.1.135: The scribe has allocated a long space (approx. 1.8 cm.) between þe and beste to accommodate the original tear in the parchment described at R1.105. beste . ¶ Lereth R.1.136: Beta adds it after Lereth. This is also the reading of the A and C archetypes. this it þis lewed men  for lettrede me n it knoweth . Þat trewthe is tresor  þe triest on eerthe . ¶ I haue Ȝet haue I R.1.138: For alpha's I haue, beta reads Ȝet haue I, which is also the reading of Ax. However, the reading of alpha agrees with that of Cx. no kende knowyng quatz I  ȝette mote I ȝe lerne kenne me R.1.138: Once more, R's (and alpha's) third stave was defective in alliteration; cf. beta's b-verse, which offers ȝet mote ȝe kenne me to R's ȝette mote I lerne. Beta's reading agrees with that of the other two versions. bettre . By what craft in my corps  it comseth and where . ¶ Þow dotede daffe quatz sche  dulle arn þi wittes . R.1.140: Hereafter, R's text is lost for KD1.141 through KD2.40. Passus secundus de visione Petri Plowman . vt supra R.2.0: Here the lacuna in the manuscript (running from KD1.141-2.40) ends and text resumes. To on fals fikel of nil R.2.1: Beta omits of. tonge  a fendes byȝete . Fauel þoruȝ his faire speche  hath þis folke enchaunted . And alle is lyeres ledynge  þat sche is þus ywedded To -morwe worth ymad.  þe maydenes bruydale . And þere miȝt þow wite if þow wilt  whiche þei ben alle . Þat longeth to þat lordschippe  þe lasse and þe more . Knowe hem þere if þow canst  and kepen þi tonge . R.2.7: F and some beta copies read keep wel þyn tunge . Other beta witnesses show kepe þow þi tonge . Manuscripts CL agree with R's phrasing. None of the B witnesses agrees with Ax or Cx, which both attest kepe the fro(m) hem alle . And lak hem nauȝt but late hem worth  til lewte be iustic e And haue poer to puneschen hem  þanne putt forth thy r eson Now I bekenne þe crist q uod he she R.2.10: He, "she." R's he, repeatedly deployed for the feminine 3rd person sing. pronoun, is relatively uncommon among the B manuscripts in this role.  and his clene moder . And at lat R.2.11: At, "that." no consience acombre þe  for coueytise of mede . Þus left me þat lady  lyggynge a -slepe . And how mede was ymaried  in meteles me thouȝthe . Þat alle þe riche retenawans  þat regneth with þe false . Weren bede R.2.15: Though identical in meaning to the majority reading, R's form here is unique among the B copies; most of the others, including F, attest boden. Ax agrees with the B majority (although manuscripts VRa agree with R) , but Cx's verb form is uncertain. Though a majority of C witnesses also supports the predominant reading, most P family manuscripts (PVcAcQScFcGc ~ ibede McNc) agree with R. to þis þe R.2.15: F reads his; beta shows þe. Two A witnesses agree with R (manuscripts LaK), but Ax agrees with beta. Cx is uncertain: the X family reading is identical to that of beta, but the P family reading is þat. bridale  on bothe to sydes . In the right margin, in black ink, there is an early ownership stamp for the Bodleian Library. Of alle maner of men  þe mene and þe riche . To marie þis mayde  was many man ensembled assembled R.2.17: For alpha's ensembled (F has ensemblyd þere), beta reads assembled. . As of kniȝtes and of clerkes  and other comou ne poeple . As sysoures and sompnoures  scheryues and .her e clerke s Bedeles and bayliues  and brokoures of chafare . Forgoeres and vitayleres  and vocates of þe arches  I can nouȝt rekene þe route  þat ran a -boute mede . Ac symonye and cyuile  and sysoures of courtes . Were moste priue with mede  of any men þouȝte me þouȝte . R.2.24: R's reading here, men þouȝte, is unique error. R is probably reproducing an omission that had already occurred in alpha. As is often the case, F 's homologous reading, þere owte, has more the appearance of an attempted repair than a faithful reproduction of the sub-archetype. The Cx reading here is identical to beta's; however, manuscript X, the single best witness of this version, agrees with R in omitting me but then reinserts it in another hand! X's corrected error may attest to an early documentary parent of both B and C traditions in which the word occurred only as an interlinear correction. Ac fauel was þe furste  þat fette hire oute of boure . And as a brokour e brouȝt hire  to be with fals enioyned . Whanne symonye and cyuile  seiȝ here bether e beire R.2.27: The variant genitive forms bether, boþer and beire all appear in the manuscripts. wille . And Thei R.2.28: For alpha's And, beta reads Thei. Although the line is revised in the C version, its opening is identical to beta's phrase. assented for seluer  to seggen seye R.2.28: R's verb form here is unique; F has seyn while beta reads seye. as bothe wolde . Þanne leup lyere forth  and seyde lo here a chartre . Þat gyle with his grete othes  gaf hem to -gyderes . And preyede cyuile to se  and symony to rede it . ¶ Þanne symony and cyuile  stondeth R.2.32: Although R's meaning here is identical to that found in beta (presumably reading stonden), R's verb form is unique at this point in the B tradition; F agrees with YM in reading a preterite, stoden castel. Nevertheless, R's form is probably also the reading of Ax and is certainly the form attested in the X family of C. Among the P manuscripts, the preferred reading is that of FcYcMc (which can also be found among some A witnesses). forth bothe . And vnfoldeth þe feffement  þat fals hath y -maked . And þus bygyneth þis gomes  at to R.2.34: R's at (cf. R2.11 above) = standard þat and is a unique reading here. The other witnesses attest to. greden ful hiȝe . S ciant presentes R.2.35: R uniquely omits the end of this citation: & futuri &c . The C version of this citation agrees with that found in F/beta. &c etera & futuri . ¶ Wyteth and witnesseth  þat wonyeth vppon þis eerthe . Þat mede is I -maried  more for hire godes . Þan for any vertu or fairenesse  or any fre kende . Falsenesse is fayne of hire  for he wote hire riche . And fauel with his fikel speche  feffeth by þis chartre . To be princes in pruyde  and pouerte to despise . To bagbiten and to bosten  and bere fals witnesse . To scorne and to scolde  and s.clander to make . Bold and vnbuxu m Vnboxome and bolde R.2.44: In beta, the phrasal order of this a-verse is reversed: Vnbuxome and bolde . The beta rendering is, however, the presumptive original since it matches the order found in C.  to breke þe ten hestes . ¶ And þe eerldam of enuye  and wrathe to -gyderes . With þe chastelett of gestes chest R.2.46: Cf. F's Ieestys and beta's chest. The C reading agrees with that of beta.  and chaterynge out of tyme resoun . R.2.46: F's b-verse is unique ( & þe Iangelynge of synne). For R's tyme, beta reads resoun. The C reading agrees with that of beta. Þe contee of coueytise  and alle þe costes aboute . Þat is vsure and auarice  alle I hem graunte . In bargeynes and in brokages  with alle þe borgh e of thefte . ¶ And alle þe lordeschipp e of lecherie  in lenthe and in brede . As in werkes and in wordes  and waytynges with eiȝes . In wedynges wedes R.2.52: Cf. F's wenyngis, which is probably the reading of Bx as well as that of alpha; beta reads wedes. The Cx reading is uncertain; the P family agrees with beta, but the X group rejects all the B variants, revising to woldes. and in wischynges  and with ydel þouȝtes . Þere as wille wolde  and werkmanschipe fayleth . ¶ Glotonye he gaf hem eeke  and grete othes to -gydere . And alle day to drynke  at dyuerse tauernes . And þere to iangle and to iape  and iuge her e euencristene . And with in R.2.57: R's with is unique; Bx reads in. fastyng -dawes to frete  ar ful tyme were . And þan to sitten and soupen  til slepe hem assaile . And bredun as a as R.2.59: Beta does not attest a. burgh swyne  and bedden hym hem esely . Til slewth and slepe  slyken his sydes . Robart Bente you shalbe w ith vs at Budworthe and there to Testyfy youre knowlegh in a mater R.2.61: At the top of fol. 2r, a secretary hand has written in black, Robart Bente you shalbe w ith vs at Budworthe and there to Testyfy youre knowlegh in a mater . This note may explain the cropped margins of the first few pages of the manuscript: someone appears to have been using them for occasional short notes. Budworth probably refers to one of two ancient parishes in Cheshire, Great Budworth and Little Budworth. And þanne wanhope to a -wake hym so  with no wille t o amende For he leueth be yloste lost  þis is his here R.2.62: R's his is unique (but preferable to the F and beta alternatives, which are rejected in its favor by Kane-Donaldson and Schmidt); F reads þe while beta reads here. laste ende . And þei to haue and to holde  and here heyres after . A dwellynge with þe deuel  and dampned be fo for R.2.64: Neither MED nor OED2, s. v. for, cites an example of fo as viable for the preposition signified here, but it occurs in R in four widely separated contexts (cf. R14.60, R15.379, and R20.224) and probably represents an instance of idiolect apocope. euere . With alle þe .p urtenaunces purtenaunce of purgatorie  in -to þe pyne of hel le Ȝelding for þis thinge  at one ȝeres ende . Here soules to sathan  and to R.2.67: R's and is unique; Bx reads to. suffre with hym peynes . And with hym to wonyen in wo  while god is in heuene . In wytnesse of þis which R.2.69: For alpha's non-alliterating þis, beta properly reads which. Unfortunately, the problem appears to be, at some level, authorial rather than merely scribal. That is, although Ax clearly agrees with beta in alliterating this line on /w/ (using the exact same variants), Cx agrees with alpha just as emphatically in ignoring the normal alliterative pattern. Russell-Kane emend their C text back to the norm, but that seems pointlessly meliorative. At the very least, the aforementioned variant array (which is fairly typical) suggests a cavalier attitude toward such small metrical issues on the part of the C author. þing  wronge was þe furste . And peres þe pardonere  of paulynes doctrine . Bette þe bedel of bokyngh am -schire . Reynald þe reue  of rokelond Rotland sokene . Munde þe mellere  and many mo other e . In date of þe deuel  þis dede I asele . By siȝt of sir e symonye  at and R.2.75: This is a unique reading in R; Bx reads and. Cx agrees with the B majority. cyueyles leue . Þenne teneth tened R.2.76: Only R deploys a seemingly present-tense form of this verb (but the difference may be illusory; see the Introduction III.2.2.10. Beta reads tened while F has was teenyd. Both Ax and Cx agree here with beta. hym teologye  whan he þis tale yherde herde . And seyde to cyuile  now sorwe mote þow haue . Swiche weddynges to wurche  to wrathe with trewth e And ar þis weddyng be y -wroȝt wrouȝte  wo the be -tyde . For mede is moylere  of amendes engendreth . R.2.80: R shares an apparently nonsensical verb inflection ( engendreth for engendred) with beta witnesses LCY. Nevertheless, any RL shared form, however odd, is intrinsically likely to be archetypal, albeit perhaps non-authorial—because of their extraordinary accuracy as well as their definitive stemmatic positions. If this lection is not merely a blatant archetypal error (one "corrected" by most later copyists to the expected form), it may be that the R and L scribes (or the Bx scribe) understood the -eth suffix in this word as allomorphic with the past participle suffix -ed / -et attested in other B copies. The final phone of engendreth would then probably have been construed by L and R as /t/ (not the /θ/ which the spelling would suggest to us). Cf. the 1408 London will of John Plot. Twice in this brief document, Plot uses a phonologically identical verbal suffix <-yth> to denote the past participle form usually spelled as <-ed>. In the first instance, Plot requests that "thyr be Spendyth among my Nyebourus in Mete & in drynke" a certain amount of money; in the second, he requests that some of his assets be used for road repairs, or, as he phrases it, "be yspendyth betwene London and ware, of fowle weys, . . . there most nede ys" ( The Fifty Earliest English Wills, ed. Frederick J. Furnivall (London: Trübner, 1882), 14-15. A few pieces of evidence scattered throughout manuscripts L and R may support such a conclusion. One wonders, for example, whether the strong preference in manuscripts L and R for the ON-derived spelling of the cardinal number 100 (= hundreth) over the OE-derived form (= hundred) indicates that these scribes, or their models, would have pronounced that word with /θ/ as the final phone, rather than /t/. Such a conclusion seems doubtful. Rather, this spelling preference for the number 100 probably attests the same trivial orthographic anomaly hypothesized above concerning engendreth. For fuller discussion see Introduction III.2.2.10. And god granteth to gyue  mede to trewthe . And þow hast gyue hire to a gyloure  now god gyue þe sorwe Þe Thi tixit tixt tixt telleth þe nouȝt so  trewthe wote þe sothe . For dignus est operarius  hys huyre to haue . And þow hast fast hire to fals  fy on þi lawe . For alle by lesynges þow lyuest  and lechores lecherouse werkes Symonye and þi -selue  schenden holy cherche . Þe notaries and ȝee  nuyȝet R.2.88: Some beta copies and F agree in reading noyen but beta itself probably read as LMHm, noyeth. Only R shows what appears to be a preterite form of the verb (but see the Introduction III.2.2.10 regarding the possibility of morphological ambiguity in R's verb-tense marking). In a similar version of this line, Ax agrees with beta and F in using a present-tense form. þe poeple . Ȝe schulle a -byggen it bothe  by god þat me made . Wel ȝe wyten wernardes  but if ȝour e witt fayle . Þat fals is faithles  and fals fikel R.2.91: R's fals is unique; Bx reads fikel. in his werkes . And as was R.2.92: Beta reads was. Both Ax and Cx agree here with alpha. a bastarde ybore  of belsabubbes kynne . And mede is moylere  a mayden of goode . And miȝte kysse þe kyng  for cosyn and he she R.2.94: He, "she." On this variant of the feminine, 3rd person sing. pronoun, see the note at R.2.10:. wolde . For -thi wercheth by wysdom  and by witt al -so . And ledeth hire to londou n  þere lawe is yschewed . If any lawe wil loke  þei ligge to -gyderes . And þouȝ iustices iuge hire  to be ioyned with fals . Ȝut beth war of þe nil R.2.99: R's þe is owed to alpha, though it is shared by GH. Beta omits it. Many A witnesses, as well as Cx, agree with alpha on the presence of this determiner. weddynge  for witty is trewthe . And consience is of his conseyle  and knoweth ȝow echeone . And if he fynde ȝow in defaute  and with þe fals holde . It schal be -sitte ȝoure soules  ful soure atte laste . ¶ Hereto assenteth cyuile  ac symonye ne wolde . Til he hadde siluer for this his R.2.104: R's this is unique; beta reads his while F rephrases the entire line (which appears to be corrupt at the archetypal level when compared to the rendering in A and C). s eruise  and al -so þe notaries . ¶ Þanne fette fauel forth  floreyns ynowe . And bad gyle to gyue  gold al aboute . And nameliche to þe notaries  þat hem non ne fayle . And feffe fals witnesse  with floreyns ynowe . For he þei R.2.109: Only L, among beta copies, agrees with alpha's reading here; the other beta witnesses attest þei. However, the LRF reading is almost certainly authorial, matching the reading found in Ax and Cx. may mede a -maystrie  and maken at my wille . ¶ Þo þis gold was Igyue  grete was þe thonkyng e . To fals and to fauel  for here faire ȝiftes . And comen to conforte  fram care þe fals . And seyden certes sir e  sese schul we neuere . Til mede be þi wedded wyf  þorȝ wittes of vs alle . For we haue mede amaysterud  thorȝ with R.2.115: Beta reads with here, agreeing with the apparent reading of Ax (two A copies, manuscripts LaE, agree with alpha); the C manuscripts support alpha almost unanimously. oure merie speche . Þat heo graunteth goo to gon R.2.116: For R's goo, F reads for to goo while beta has to gon. Both Ax and Cx agree with beta's phrasing.  with a good wille . To londou n to loke  if þat þat þe R.2.117: R's if þat lawe is unique; F and some beta copies read ȝif þat þe Lawe ; other beta witnesses have if þe lawe. The reading of Ax is uncertain (because of a wide variety of variants here) but may have agreed with R's. The reading of Cx is if lawe . lawe walde iuge . R.2.117: R uniquely divides this line after iuge; all other B witnesses divide the line before this word. Ȝow ioyntly  in ioye for euere . ¶ Þanne was falsenesse fayne  and fauel as blythe . And leten sompne alle þe R.2.120: R is joined by G in attesting þe (and Ax agrees with the R variant), but the word is not found in F ( alle me n ) or beta ( alle segges). The P family of C agrees with beta's rendering of this phrase, but the X family has alle his segges . segges  in schires aboute . And bad hem alle be bown  beggeres and otheres othere . R.2.121: R is the only witness for the plural form; all other B copies (except Hm, which shows ellis) attest the singular othere. Both Ax and Cx agree with the F/beta phrasing. M M Butte N N Nell XXX CX God from Sam R.2.122: At the top of fol. 3r are the remains of various sixteenth-century pen trials, scribbled in jagged fashion horizontally, from left to right. To wenden with hym hem R.2.122: Though RF opt for the apparently singular form (a reading endorsed by Schmidt), beta agrees with Ax in reading hem, which seems more appropriate to the plural referents named in previous lines; the P family of C also attests this reading, but the X family agrees with alpha. However, MED, s.v "hem," documents hym as an available but rare spelling, especially in the fifteenth century, for the objective case of the third-person plural pronoun. So the difference noted above may be only a clutter of scribal / authorial accidentals. to westmenstre  to witnesse þe þis R.2.122: For alpha's þe, beta reads þis. A majority of A witnesses supports beta, but a sizeable minority agrees with alpha. dede . Ac þanne cared þei for caplus  to cayren hem þidur . And fauel fette forth þenne  folus Inowe . And sette mede vpp e vpon R.2.125: For R's vppe, F has on while beta reads vpon. The same line occurs in the A version, where the reading agrees with F's. a schyriue  I -schoud shodde alle newe . The terminal punctus for each of these lines has been rubbed or partially erased long ago, but they are both sufficiently visible for detection in normal light. And fals satt on a sysoure  þat softlyche trotted . And fauel on a flaterere  fetysliche atyred . Þo haued R.2.128: Only manuscript L agrees with R's unusual verb form haued (common in the thirteenth century but nowhere cited in MED, s. v. haven, later than about 1330, except for Piers Plowman). Both F and the majority of beta copies read had(de). The same line appears in A, but Kane was not concerned to record such morphemic variations, regarding them all as accidentals. notaries none  anuyed þei were . For symonye and cyuile  schulden on here feet gange . Ac þanne swore symonie  and cyuile bothe . Þat sompnoures schulde be sadeled  and s erue hem vchone . And lat apparayle þe þis R.2.132: FGH omit any determiner, while beta reads þis. The same array of variants is present at this point in the A witnesses , with no strong evidence for originality. p rouisoures  in palfreyes wyse . Syre symonie hym -selue  schal sitten vppon here bakke s T om th Between these two lines is what seems to be a pen trial of uncertain age: T om th . Denes and southdenes  drawe ȝow to -gyderes . Erchedeknes and deknes nil R.2.135: R's deknes is a unique addition unattested in any other B manuscript.  officiales and alle ȝour e regestreres Lat sadle hem with siluer  oure synne to suffre . As deuoutrie auoutrie R.2.137: This is the alpha variant. Beta offers a more common synonym, but one that fails to alliterate: auoutrie. F reverses the noun phrasal order of the a-verse. and deuoses  and derne vsure . To bere byschoppes aboute  a -brode in visitynge . Paulines priues  for pleyntes in þe consistorye Schule serue my -selue  þat cyuile is nempned . And carte -sadele þe comyssary  oure carte schal he lede And fecche vs vitailes  at fornicatores . And makeþ of lyare a lange karte  to leden alle þes oth ere As fobberes Freres R.2.144: Beta has Freres. A majority of A witnesses attests folis at this point (though three, manuscripts LaWaN, agree with beta). Cx has fobbes. That support, as well as the word's relative rarity, suggests alpha's variant is likely to be authorial in B. Schmidt accepts fobberes at face value. MED, s. v. fobben, hypothesizes that fobberes was derived from "fobben," v., but cites merely two Piers Plowman manuscripts for the form's existence. Conversely, Kane-Donaldson view alpha's variant as scribal and emend to the C form. and faytoures  þat on here feet rennen . And þus fals and fauel  fareth forth to -gyderes . And mede in þe myddes  and alle þese men after . I haue no tome to telle  þe taile þat hem folweth . Of many maner man  þat on þis molde libbeth . John naylle R.2.149: A sixteenth-century signature written vertically upwards appears in the lower left margin of fol. 3r: John naylle . Ac gyle was forgoer e  and gyede hem alle . Sothnesse seiȝ hem wel  and seyde bute a litel . And prikede his palefrey  and passed hem alle . And com to þe kynges courte  and consience it tolde . And consience to þe kynge  carpud it after . Now by crist q uod þe kyng e  and I cacche miȝte . Fals other or fauel  other or R.2.155: Beta reads this set of correlative conjunctions as or . . . or. F agrees with R for the first ( eyþ ir) but reverts to the beta variant for the second . A majority of A witnesses agrees with beta (but Kane chooses the alpha set, exemplified in manuscript T, as a "harder reading"). Cx's reading is also somewhat ambiguous, with a majority of witnesses (mostly of the P family) attesting oþur in the first case and a minority (again mainly P family manuscripts) offering the same variant in the second instance. any of his feres . I wolde be wroke of þoo wrecches  þat wurcheth so ille . And do hem hange by þe hals  and alle þat hem maynteneth . Schal neu ere man of molde  meyntprise þe leste . But riȝt as þe lawe wil loke  lat falle on hem alle . ¶ And comanded a constable  þat com atte þe furste . Goo R.2.161: Beta reads To, which is also the reading of Ax. Four C witnesses (IP 2FcNc) agree with beta, but Cx clearly agrees with alpha's Goo. atache þo tirantes  for eny thyng I hote . And fetreth fast falsnesse  for eny kynnes ȝiftes . And gerdeth of gyles heued  and lete hym goo no further e . And ȝif ȝe laccheth lacche(n) R.2.164: R's form is unique; F and most beta copies read lacche(n). Both Ax and Cx appear to agree with the F/beta reading. lyere  lat hym nouȝt asckapen . Ar he be put on þe pillerie  for eny preyere I hote . And brengeth mede to me  maugre hem alle . ¶ Drede atte dore stode  and þe dome herde . And how þe kynge demede comaunded R.2.168: R uniquely fails in alliteration here; F and beta read comaunded.  constables and s eriantz . Falsnesse and his felaschipe  to fetturen and to bynden . Þanne drede wente wiȝtlyche  and warnede þe fals . And bad hym flee for fere  and his felawes alle . ¶ Falsenesse for fere þanne  fleiȝt fleiȝ R.2.172: R's form is unique; F and many beta copies read fleiȝ. Other beta witnesses have fledde. Both Ax and Cx appear to agree with the F/beta reading. to þe freres . And gyle doth hym to go  agast for to deye . Ac marchantz mette with hym  and made hym abyde . And by -schetten hym in here schoppe  to schewen her e war e . And apparayled hym as a prentise  þe poeple to s erue . ¶ Liȝtliche lyere  leep awey þenne  Lurkynge þorȝ lanes  to -lugged of many . He was nawer e welcome . for his many tales . Ouer al I -howted  and I -hote trusse . Til pardoneres haued pite  and pulled hym into house . Þei weschen hym and wyped hym  and wonden hym in cloth es cloutes R.2.182: Although Hm agrees with alpha, beta itself must have read cloutes. It is unclear what the reading of Ax was since the alpha / beta disjunction is mirrored by nearly equal numbers of A witnesses. Cx, however, agrees with beta on this form. And sent hym with seales  on sonedayes to cherches . And gaf pardon fore pans  pound -mel aboute . Þanne lowrede leches  and lettres þei sente . Þat he schulde wonye with hem  watres to loke . Spiceres speken to with R.2.187: Beta has with. Although both readings are available in the A tradition, a large majority of A witnesses agrees with beta. On the contrary, Cx clearly agrees with alpha's to. hym  to spien here ware . For he couthe on of R.2.188: Although Hm supports alpha, beta itself must have read of. Both Ax and Cx agree with alpha's reading. here craft  and knew many gu mmes . Ac minestrales and mesageres  metten with hym ones . And helden hym half a an half R.2.190: For alpha's half a, beta reads an half. Cx probably read simply half, a variant also found in the A tradition (manuscripts TRaUNMa). However, a plurality of A witnesses agrees with alpha's phrasing (though two agree with beta). ȝere  and elleuene dayes . Freres with fayre speche  Fetten hym þennes . And for knowyng of comeres  coped hym as a frere . Ac he hath leue to lepe out  as ofte as hym liketh . And is welcome whan he wille  and wonyeth with hem o fte Alle fledden for fere  and flowen into hyrnes . Saue mede þe mayde  no man na mo R.2.196: R's no man is unique in the B tradition (four A copies share the reading); F and beta read na mo, which is the reading of Cx and of a majority of the A manuscripts. durste abyde . Ac trewly to telle  heo tremeled for fere drede . And al -so ek R.2.198: Beta reads ek, which is also the reading of Ax ( Cx has bothe). wepte and wrong wronge whan heo she was R.2.198: Where R has heo was, F reads for she was while beta reads whan she was. Beta's rendering is that attested by the other two versions for this passage. a -thachud . Passus tercius de visione petri plowman . vt sup ra &c etera . N ow is mede þe mayde  and namo of hem alle . With bedelus and with bayliues  y -brouȝt by -for þ e kynge Þe kyng called aclerke a clerke  can I nouȝt his name . To take mede þe mayde  and maken hire at ese . I schal assay hire my -selue  and couthliche sothelich R.3.5: R's couthliche is a unique variant, probably reflecting an alpha error (beta = sothelich); F smooths this mistake to sotilly hire . appose . What man of þis world molde R.3.6: The beta reading is molde. However, the entire a-verse represents a problem. F's rendering of this half-line, What wyȝe of þis world, may be the original, descended vertically from alpha. If that were the case, R (which is usually much more faithful to copytext than F) would have slipped into the same modernized form for the first stave word as beta did ( man) but preserved the second as rendered in alpha and Bx, (i.e., worlde), creating an instance of faulty alliteration. By contrast, having modernized wyȝe out of this line, beta would then have gone on, in the interest of alliterative regularity, to invoke the archaism molde as a replacement for the original worlde. Unfortunately for this theory, the evidence from the other versions fails to provide it with much support. The evidence of the A witnesses overwhelmingly favors man as the first stave of this half-line. And a significant majority of the same manuscripts supports worlde as the second stave (RaUHaEKWa = molde), agreeing exactly with the testimony of manuscript R in the B tradition (against both F and beta). Among the C copies, the agreement with R's non-alliterating phrase is universal. If Langland's metrics seem too embarrassed by this evidence, it is still possible to theorize that he intended the first stave word of the line to be What rather than man, but this hypothesis remains awkward since it would require an elevation of were in the b-verse from its apparent position in a dip.  þat hire were leuest . And if heo she wurche be my witte  and my wille folwe . I wile for -gyue hire þis gultes gilte R.3.8: The beta reading is gilte. Ax agrees with beta, but Cx agrees with alpha.  so me god helpe . Curteysliche þe clerke þanne  as þe kyng e hiȝte . Tok mede by þe middel  and brouȝte hire in -to chaumbre . Ac þere was murthe and ministracie mynstralcye  mede to plese . Þ ei þat wonyeth at westmenstre  worschipped hire alle . Gentiliche with ioye  þe iustices somme . Busked hem to þe bowr e  þere þe buyrde dwelte dwelled . To conforte hire kendeliche  by clergyse leue . And seyden mourne nouȝt mede  ne make þow no sorwe . For we wil wyse wisse(n) þe kyng  and þi weye schape . To be wedded at þi wille  and where þe leef lyketh . For alle consience crafte caste  and cast craft R.3.19: R's phrase, crafte and cast, represents the alpha order for this pair (cf. F's craft or cast); in beta manuscripts they are reversed. Both Ax and Cx agree with beta's order. as I trowe . ¶ Mildliche mede Mede þanne R.3.20: Here R uniquely omits þanne (F attests this adverb's presence at the head of the a-verse rather than the tail, as in beta). Both Ax and the X family of C agree with beta's reading. The P family of C positions the adverb as in beta, but reads it as þo.  merciede hem alle . Of þeire grete godenesse  and gaf hem vcheone . Coupes of clene gold  and cuppes of siluer . Rynges with rubies  and ricchesse ricchesses R.3.23: Beta shows the plural, ricchesses. Both variants are available in the A tradition, but a significant majority of A witnesses agrees with beta's reading. manye . Þe lest man of here meyne  a motou n of goolde . Þanne lauȝte þei leue  þis lordes at mede . ¶ With þat comen clerkes  to conforte hire þe same . And beden hire be blithe  for we beth þin owene . For to wurche þi wille  þe whyle þow miȝt last . ¶ Hendeliche heo þanne  be -hiȝte hem þe same . To loue ȝow R.3.30: F and most beta manuscripts show hem here (M has been erased and overwritten to match this majority reading), but L uniquely agrees with R's ȝow. Although two A manuscripts (RaU) agree with the LR reading, it seems clear that Ax supports the F/beta variant. However, the A majority continues in the following line with third-person plural references (RU again dissenting), but a large cluster of fairly reliable B manuscripts (MCOF, as well as the BmBoCot set) all suddenly shift to second-person plural pronouns at this point, coming back into alignment with LR. It appears likely, then, in the light of this attestational pattern, that LR are not randomly agreeing in error but witnessing, at R3.30, what actually appeared in Bx, that the other B copies, by contrast, are merely drifting along with the third-person references established in previous lines of this passage and fail to notice, until the next line, the sudden shift in viewpoint caught here by the two most careful B scribes. lely  and lordes to make . And in þe constorie at courte  do calle ȝoure names . Schal no lewednesse lette  þe clerkes leode R.3.32: Although alpha's third stave shows defective alliteration (cf. beta's leode, which is also the reading of Ax), alpha's clerkes is supported by Cr and universally by the C version. The possibility that alpha and C have randomly converged in error here, both chancing upon the same word that neither alliterates in its line nor is an equivalent for the word replaced, seems unlikely—especially since this particular pattern of alpha / C agreement in editorial change is one that recurs frequently throughout the developing narrative. Only two explanations seem plausible for this array of variants and for many similar ones; however, at this distance the two explanations are almost indistinguishable: (1) While he was working on B, Langland began to be much more concerned about his London readers not understanding obsolescent words like leode than about small metrical lapses and therefore entered a series of marginal "updatings" into Bx's exemplar, moving it away, at times awkwardly, from original A readings in order to meet his changed perception of audience needs. When confronted with such evidence of authorial ambiguity in his exemplar, the scribe of Bx usually hedged his bets by copying the text unaltered, with the authorial change reproduced in his own margin (perhaps thinking it a gloss). In the final stage of this process, alpha and beta followed their respective proclivities, with alpha normally taking such an entry as authorial revision and using it to supplant the original text while beta usually took it as a mere gloss, ignored it, and copied what he saw in the body of the line of Bx. Or, (2) like its many anomalous relatives in other "revised" lines, clerke(s) was indeed a purely scribal gloss in the immediate ancestor of Bx and had no warrant from the author, but still seemed sufficiently ambiguous to the Bx scribe to deserve exact reproduction. The roles of alpha and beta in this scenario remain the same as in the first. But what is painfully evident is that, by the time he began using the exemplar of Bx to create the C text, Langland either didn't care anymore about such small aesthetic matters or had completely forgotten who had authored clerkes—or both! þat I louye . Þat he he ne R.3.33: Only the two alpha manuscripts here omit ne before the verb. The A version here agrees completely with beta. worth furst vaunsed auanced R.3.33: R's vaunsed is a unique form among the B witnesses but dominates the X family of C. The P family tends to agree with the F/beta reading, auaunced. As would be expected, MED, s. v. vauncen, categorizes vaunsed as a clipped form derived from the verb avauncen. Citations for this form are mostly from the fifteenth century (but the passage from manuscript R is also cited as well as one from Robert Manning).  for ich am by -knowe . Þere connynge clerkes  schulle clokke by -hynde . ¶ Þanne com þer e a confessoure  ycoped coped as a frere . To mede þis þe mayde  he mellud þis R.3.36: R's phrase þis mayde is unique. F's has þat Mayde while beta reads þe mayde. Both Ax and Cx agree with beta. wordes . And seyde ful sotely softly R.3.37: R's sotely is unique; the other B witnesses read softly; the C version is revised at this point, but Ax agrees with the B majority and no A copy agrees with R .  in schrift as it were . Þeiȝ lewed men and lered men  had leyne be þe bothe . And falsede falsenesse R.3.39: Cf. F's Fals and beta's falsenesse. Though all three variants are available in the A tradition, Ax almost certainly agreed with R. As for C, the manuscript support in that version for R's lection is almost universal. hauede I -folwed þe  al þis fifty wynter . I schal assoyle þe my -selue  for a seme of whete . And al -so be þi bedeman  and bere wel þi message . Amonges kniȝtes and clerkes  consience to turne . Þanne mede for here misdedes  to þat man kneled . And schrof hire of hire schrewednesse  schameles I trowe . Tolde hym a tale  and tok e hym a noble . For to ben hire bedeman  and hire brocour e alse . Þanne he assoyled hyre sone  and sytthen a he R.3.47: A, "he." sayde . We han a wyndow a -werchynge  wil sytten vs ful hiȝe . Waldestow glase þat gable  and graue þer e þere -inne R.3.49: R's þer e is unique; the other B manuscripts have þere-inne. However, almost all C witnesses agree with R. þi name . Syker schulde þi soule be  heuene to haue . Wist I þat quatz mede  þere nys is wyndow no wowȝ ne awhter . R.3.51: Beta's version of the b-verse reads I wolde nouȝt spare . It seems likely that the long passage of damaged or missing text in alpha after this point (cf. R3.52) may actually begin with this b-verse. Þat I ne nil wolde make and amende do mende it or make  it with of myne at þe fulle . These two lines are offered by alpha instead of eleven lines preserved in beta (= KD3.52-62); it appears that the beta passage is archetypal but was somehow overlooked or unavailable to alpha and that the two lines in their stead were introduced to patch the resulting incoherence, adapted from two cognate lines in A (= Kane 3.50-51). Beta's cognate lines are as follows: Wist I that quod þat womman I wolde nouȝt spare For to be ȝowre frende frere and faille ȝow neure Whil ȝe loue lordes þat lechery haunteþ And lakkeþ nouȝt ladis þat loueþ wel þe same It is a frelete of flesche ȝe fynde it in bokes And a course of kynde wher -of we komen alle Who may scape þe sklaundre þe skaþe is sone amended It is synne of seuene sonnest relessed Haue mercy quod Mede of men þat it haunte And I shal keure ȝowre kirke ȝowre cloystre do maken Wowes do whitten and wyndowes glasen . And my name write & do peynte me & myn name openliche apertly þere -Inne . Þat vch a eury R.3.54: For alpha's vch a, beta reads eury. Alpha's reading is supported by a majority of A witnesses, but beta's variant agrees with C. segg e schal Ise seyn R.3.54: This form (cf. F's see) is an alpha reading that agrees with both Ax and Cx against beta's seye(n) (= modern "say").  Ich am suster to ȝow alle of ȝowre hous . R.3.54: R's to ȝow alle is from alpha; beta reads of ȝowre hous. Both Ax and Cx here support beta, the former by complete agreement, the latter by being an obvious revision of the beta phrase (C = of ȝoure ordre). Ac god to alle goed folke  suche grauynge defendeth . To writen in wyndowes  of here wel -dedes . And An aunter pruyde be ypeynted peynted þere  and pompe of þe wo rld For god crist R.3.58: R's god fails to alliterate properly (cf. the F/beta reading, crist, which does alliterate). Nevertheless, all the C witnesses agree with R against F and beta. knoweth þi consience  and þi kende wille . Þi And þi R.3.59: The opening of this line in F reads Boþe þe coost; cf. beta's And þi coste. However, the P family of C agrees with R's line opening (the X family has no reading here at all as a result of an earlier error that merged two lines in its subarchetype). cost and þi coueytyse  and who þe catel owȝte . For -thi I lere ȝow lordes  leueth suche werkes . To writen in wyndowes  of ȝour e wel -dedes . Or to grede after goddes men  whan ȝe delen doles . An aunter ȝe han ȝoure hir e here  and ȝour e heuene al se Nesciat sinistra quid faciat dextera  R.3.64: The right end of this red boxing has been cropped. Lat nouȝt þi left half  late ne rathe . Wite what þow worchest  with þi riȝte syde . For þus bitt þe gospel  goede men do here almesse . Meyres and maceeres  þat menes ben by -twene . Þe kyng and þe comou ne  to kepe þe lawes . To ponysch vppon on R.3.70: Beta reads on. Ax clearly agrees with beta here, but the C families are divided, the P group supporting beta while the X group agrees with alpha's vppon. pilaries  and pynyng e stoles . Brewsteres and bakesteres  bocheres and kokes . For þese aren men vppou n on R.3.72: Beta reads on. C is revised at this point, but the A version attests an identical line, in which witnesses are mostly divided between beta's reading and of þis molde  þat moste harme werchen . To þe pouere poeple  þat parcelmel buggen . And also For they R.3.74: R's And also is unique but may derive from alpha; cf. beta's For they and F's &. C is revised at this point, but the A version attests an almost identical line, in which nearly all manuscripts agree with beta's opening phrase. poysene þe poeple  priuelich and ofte . Þei richen thorȝ regratrie  and rentes hem buggeth . With that þe pouer poeple  schulde putte in here wombe . For toke þei on -trewely  þei tymbrede nouȝt so heyȝe . Ne bouȝte none burgeage burgages R.3.78: R's use of the singular burgage is unique but may reflect alpha (beta attests the plural); cf. F's bargayn. Although manuscripts EWa of the A version agree with R's singular, it seems clear that both Ax and Cx read as beta does here.  be ye R.3.78: Beta has ȝe. F and G clearly have þe, but R's y and þ show sufficient overlap that the scribe's intentions here are ambiguous. ful certeyne . ¶ Ac mede þe mayde  þe mayre hath be -souȝte . Of alle suche selleres  siluer to take . Or presentz with -oute pans  as peces of siluer . Rynges other other e or other ricchesse R.3.82: This awkward phrase, other othere richesse, almost certainly derives from alpha; cf. F's smoothing omission, or rycchesse. Beta has or other ricchesse.  þe reg ratoures to meyntene . ¶ For my lord loue R.3.83: Beta reads loue, which is also the reading of Ax in an identical line. The C version is revised at this point, but an analogous line ( Loue hem for my loue quod this lady mede) confirms the originality of beta's variant. q uod þat lady  loue hem vchone . And suffre hem to selle  somdel aȝeynes resou n . ¶ Salamon þe sage  a sarmou n he made . For to amende meyres  and men þat kepe lawes . And tolde hem þis teeme  þat I telle thenke . I gnis deuorabit tabernacula eoru m qui libent er libenter accipiunt munera . R.3.88: Beta finishes the citation with a phrase omitted by alpha: accipiunt munera. &c etera . Among e þis letterede lede ledes R.3.89: R's lede (a unique reading) is to be construed as a collective singular (="these lettered folk"); the other B manuscripts witness the plural ledes. Both Ax and Cx attest a completely different word for this alliterative stave: lordes.  þis latyn is to mene . Þat feer schal falle and beerne  al to bloo askes . Þe houses and þe homes  of hem þat desireth . Ȝiftes or ȝereȝiftes  be -cause of here office offices . R.3.92: Beta has the plural, Offices. ¶ Þe kyng fram conseyle cam  and called after mede . And ofsent hire alswithe R.3.94: Only L (= alswythe) and M (= alsswythe) agree precisely with R in attesting this exact phrase and writing it as a single word (cf. manuscript O = als wythe, WC = as swiþe , and F = swyþe). However, both OED2 s. v. alswith and MED in a quotation from the early fourteenth-century King Alexander s. v. certe and smert(e), attest the phrase's occurrence in this merged form in the fourteenth century, the former in Kyng Alisaunder ( He þonked Kyng Alisaundre þerof, certe, And starf alswiþe, wiþouten smerte ) and the latter in Barbour's Bruce ( His ansuer he tald alswith VIII. 153). The same a-verse occurs in the A version, where the archetypal reading appears to be that found in manuscripts WC of B, but one of the oldest of the A copies, Vernon, merges the words, reading aswiþe. The LMR form is likely to be that of Bx.  with seriantes manye . Þat brouȝten hire to bowre  with blisse and with ioye . ¶ Curteysliche þe kyng þanne  cumseth comsed to telle . To mede þe moayde  melleth þis wordes . Vnwittiliche wo mman  wrouȝt hastow ofte . Ac R.3.99: The cropped word cannot be supplied with confidence, since F has synonymous But in place of beta's Ac. wors wrouȝtestow neuere  þan þo þow fals toke . But I for -gyue þe þat gult e  and g raunte þe my grace . Hennes to þi deth -day  to do do R.3.101: R's to do is a unique variant. Most betas read simply do; cf. F's yf þou do. The same phrase occurs in A, where its form agrees with that found in beta. so namore . I haue a kniȝt consience  cam late fram by -ȝunde . Ȝif he wilneth þe to wyue  wil -tow hym haue . Ȝee lord quatz þat lady  lord it me for -bede forbede elles . R.3.104: R's b-verse is unique in the B tradition ; F and beta read lorde forbede elles, which agrees with Ax. However, R's phrasing is identical to that found in Cx. But I be holy at ȝour e heste  lete hange me sone . And þanne was consience I -called calde  to comen and appere . By -forn þe kyng and his conseyll e  as clerkes and other e . Knelynge consience to þe kyng lowtede . To wite what his wille were  and what he do schulde . Wiltow wedde þis wo mman q uod þe kyng  ȝif I wole assente n . For sche is of is fayne of R.3.111: Here R omits a key word; beta reads is fayne of þi felawship . The fact that F appears to try smoothing this passage ( is of fair shap ) suggests that R's omission was inherited from alpha. Both Ax and Cx read the phrase as beta does. þi felaschipp e  for to be þi make . quatz consience to to [þe] to þe kyng e  crist it me for -bede . Er I wedde swich a wif  wo me bytyde . For sche is frele of hire feyth  and fykel R.3.114: R's and fykel is paralleled by Hm and H in the beta tradition; however, beta itself omitted the conjunction. F reads & fals. The beta reading agrees with that of Ax, but R's phrasing is identical to that found in Cx (emended out of the Athlone text by Russell and Kane). of hire speche . And maketh men mys -do  many score tymes . Trust of hire tresor  treyeth ful manye . Wyues and wydewes  wantownnesse heo techeth . And lereth hem lecherie  þat loueth hire ȝiftes . Ȝoure fader he she R.3.119: He, "she." R's he, repeatedly deployed by the scribe in this passage, is uncommon in form among the B manuscripts but not substantive; OED2, s. v. heo, and MED, s. v. he (pron. 2), indicate that he is a variant for heo between the twelfth and fifteenth centuries. feelde  þoruȝ faire fals R.3.119: R's faire is an alpha variant. Cf. beta's fals. The same line occurs in A, where the original reading is clearly the same as in beta; however, three A witnesses, VHaMa, agree with alpha's variant. beheste . And hath appoysond popes  and appayreth peired R.3.120: R's present tense for this verb is unique among the B copies ; the others show its form as a preterite. However, Cx clearly agrees with R's verb form and the A manuscripts are split between present and past-tense forms. Cf. see the Introduction III.2.2.10 on a potential ambiguity of tense marking in R. holy cherche Is nouȝt a better e baude  by hym þat me made . By -twene heuene and helle  and R.3.122: Beta reads in, but Ax and Cx both agree here with the reading of alpha. eerthe þeyȝ men sou ȝte For sche is tykil of hire taile  tale -wise of of hir tonge R.3.123: R's phrase at the end of this line represents the alpha reading; beta has talwis of hir tonge . However, Cx clearly agrees with alpha's phrasing while the A manuscripts are split between the alpha and beta lections. . As comoun as a carte -wey  to ech a knaue þat walketh To monkes and to menstrales  to mesels in hegges . Sysours and sompnoures  suche men hire preyseth . Schirryues of schires  were schent ȝif sche nere . For sche doth men lese her londe  and here lif bothe . Sche lat leteth R.3.129: R's lat is unique in form and appears, at first glance, to be a preterite; the other B witnesses show a common present-tense inflection (e.g., LMCrWHmOGF = leteth); however, MED attests this form as 3rd sing. pres., and it seems clear that R intends the same meaning as the other B manuscripts rather than a preterite (cf. payeth later in this line). In the C version, lat is actually the predominant reading among the X family. A similarly inflected form also appears in several A manuscripts. See the Introduction III.2.2.10 for a full discussion of ambiguities in R's tense marking. passe prisons prisoneres R.3.129: R's prisons is uniquely spelled (F and beta read prisoneres); however, MED, s. v. prisoun, verifies that the R form is viable as a variant of the Bx word. The plural, prisons (= "prisoners"), actually appears in Bx at R7.30: Pore poeple or prisones  fynden hem her e fode . R's form is also found in some A witnesses at this point, and is the predominant form in the cognate line of the C version.  and payeth for hem ofte . And gyueth þe gayleres gold  and grotes to -gyderes . To vnfettere þe false  fle where hym lyketh . And taketh trewthe R.3.132: R's trewthe is an alpha variant. Beta has þe trewe. Both phrasings occur in the other two versions. They seem almost equally distributed in the A copies, while a majority of C manuscripts, including the most important members of the X family, agrees with alpha (but a significant minority, especially among the P family, agrees with beta). by þe toppe  and teyeth hym faste . And hangeth hym for hatrede  þat harme dede neu ere . To be cursed in constorie  he she R.3.134: He, "she." conteth nouȝt a rusche . For he she R.3.135: He, "she." copeth þe comissarie  and coteth his clerkes . He She R.3.136: He, "she." is assoyled assone as sone  as hire -selue liketh . And may neiȝ as muche do  in amoneth a moneth ones  As ȝoure secret seal  in six score dayes . For he she R.3.139: He, "she." is priue with þe pope  prouisoures it knoweth . For sire symonye and hir e -selue  sealeth her e bulles . He She blisseth þes bisshopes  and prestes am.eynteneth a m.eynteneth meynteneth . R.3.141: R's phrasing in this line represents a unique compression of two lines from Bx, apparently necessitated by textual loss in alpha since F also improvises at this point. Beta, which here appears to reflect Bx better, reads: She blesseth þise Bisshopes þeiȝe þey be lewed Prouendreth persones, and prestes meynteneth . Nevertheless, when R and F witness an alpha variant in this last half-line, with their inclusion of a reasserted pronoun (cf. F's & prestys she meynteneþ ), they are presumably attesting Bx accurately since both the A and C versions agree on this point. To haue lotebies and lemmanes lemmannes and lotebies R.3.142: R here offers a reversal of the beta phrase, lemmannes and lotebies; F uniquely omits and lotebies. Both the A and C versions agree with beta on this phrase.  alle here lif -dayes . And bringen forth barnes  aȝeyne forbode lawes . Þere he she R.3.144: He, "she." is wel with þe kynge  wo is þe reume . For he she R.3.145: He, "she." is fau ourable to þe fals  and fouleth trewth ofte . ¶ By ihesus with hire ieweles  þe ȝowre R.3.146: Beta reads ȝowre, which agrees with Ax. The alpha reading is supported by Cx. iustices heo schendeth . And lith aȝeyne þe lawe  and let letteth R.3.147: R's let is unique; cf. Bx's letteth. hym þe gate . Þat feyth may nouȝt haue his forth  hir e floreynes go so thikke . Heo She let ledeth R.3.149: R's let is unique; Bx has ledeth. Cx agrees with Bx here. Cf. R's lat at R.3.129:. þe lawe as hire leste  and louedayes maketh . And doth men lese thoruȝ hir e loue  þat lawe miȝt wynne . Þe mase for a mene man  þouȝ he mote hir e euere . Lawe is so lordliche  and loth to make ende . With -owten presentz other or R.3.153: R's other is unique; the other B witnesses read or. Nevertheless, R's reading may well be original here since it agrees with the majority of C witnesses, including the best copies from the X family. pans  he she R.3.153: He, "she." pleseth ful R.3.153: R's ful is an alpha variant shared with F; beta has wel, which agrees with the lection found in half of the A manuscripts. Most of the other A copies agree with RF, as does the archetype of the C version. fewe . ¶ Barouns and burgeys  heo bringeth in sorwe . And alle þe comou n in care  þat coueyten lyf R.3.155: R's lyf is unique among the manuscript witnesses of the B version but is shared with Cr 2-3. The other B copies read lyue. in trewthe . For clergise R.3.156: R shares this reading ( clergise for Bx's clergye) with Hm by convergence. The A version agrees with the majority B reading here. and coueytise  he she R.3.156: He, "she." coupleth to -gyderes . Þis is þe lyf of þat lady  now lord ȝif hire sorwe . And alle þat meyteneth her e men  meschaunce hem bytyde . For pore men mow haue no powere  to pleyne he m þoȝ þei sm erte . Suche a maister is mede  amonges men of goode . R.3.161: Because the left margin has been lost to cropping, it is not possible to be confident that it lacked the parasign that appears in LMWHmO. The scribe frequently fails to skip a line between strophes when the last line falls at the foot of a page. Þanne mornede mede  and mened hir e to þe kynge . To haue space to speke  spede ȝif a she R.3.162: A, "she." miȝte . Þe kynge graunted hire grace  with a goede wille . Excuse þe if ȝow canst  I can namore seggen . For consience acuseth þe  to congey þe for euere . Nay lord quatz þat lady  leue hym þe worse . When ȝe witen witterly  where þe wronge lyggeth . Þere þat mischief is grete  mede may helpe . And þat nil R.3.169: Only RF attest þat; beta omits it. The A version agrees with beta; by contrast, a majority of C manuscripts attests the presence of þat, but most of these omit þow. XIFc read þat knowestou. þow knowest consience  I cam nouȝt to chyde . Ne depraue þi persone  with a proude herte . Wel þow wost wernard  but if þow wilt gabe . Þow hast hangen on myn half  elleuen tyme tymes R.3.172: R's tyme is unique in form, an unmarked plural. Compare R.3.89:, R.10.224:, and R.15.407:). Both Ax and Cx agree with the ordinary plural of the other B manuscripts. . And al -so I -gripe griped R.3.173: Beta reads griped. my gold  gyue it where þe liked . And whi þow wrathest þe now  wonder me thenketh . Ȝet I may as I miȝte  menske þe with ȝiftes . And mayntene þi manhode . more þanne þow knoweste . Ac þow haste famed me foule  by -fore þe kyng here . For kuld I neu ere no kniȝt kynge R.3.178: R's kniȝt is an alpha reading; cf. beta's kynge. Both Ax and Cx agree with beta.  ne conseyled þere -after . Ne dede as þow demest  I do it on þe kynge . In normandy was he nouȝt  I -nuyed noyed for my sake . Ac þow þi -selue sothelich  schamedest hym ofte . Crope in -to a caban  for cold of þi nayles . Wendest þat wynter  wold a last R.3.183: R's last is matched only by Hm; F and beta read (y)lasted. Nevertheless, the Ax form agrees with RHm. euere . And draddest to be dede  for a dym clowde . And heȝedest homward  for hunger of þi wombe . With -oute pite pyloure  pouer men þow robbedest . And bere here bras at þi bak  to caleys to selle . Þere I lefte with my lord  his lif for to saue . I made hys men merie  and murnyge murny[n]ge mornyng lette . I batered hem on þe bak  and bolded here hertes And dede hem hoppe fore hope  to haue me at wille Hadde I be marschal of his men  be marie of heue ne I durste haue leid my lyf  and no lesse wedde  He schuld haue be lord of þat lond  a lenthe and a brede . And al -so kynge of þat kyth  hys kyn for to helpe . Þe leste brol of his lond blode R.3.196: Here alpha's alliteration fails; cf. beta's blode, a reading which is also found in Ax and Cx.  a barounes pere . ¶ Cowardlich þow consience  consayledest hym þennes . To leuen his lordschipp e  for a litel siluer . Þat is þe riccheste rewme  þat reyn ouer -houeth . ¶ It bycometh to a kuyng e  þat kepeth a rewme . To ȝiue mede to men  þat mekeliche hym s erueth . To alyenes and to alle men  to honour e hem with ȝiftes . Mede maketh hym by -loued  and for a man holden . ¶ Emperoures and erles  and alle maner lordes . Þurȝ for R.3.205: Beta reads For, but both Ax and Cx agree with alpha. ȝiftes han ȝoumen ȝonge men R.3.205: Beta's phrase is ȝonge men, a reading also found in Ax; however, Cx agrees with alpha.  to ȝernen renne(n) R.3.205: Here beta reads synonymous renne. Ax agrees with this non-alliterating reading, as does the P family of the C tradition, but the X family clearly agrees with alpha's ȝernen, a choice endorsed by both Russell-Kane and Schmidt. and to ride . ¶ Þe pope and alle p relates  presentes vnderfongen . And medeth men hem -selue  to meyntene her e lawes . ¶ Seruantz for here s eruise  we seth wel þe sothe . Taken mede of her meystres  as þei mowen acorde . Beggeres for here beggyng e biddynge  bydden men mede . Minstrales for here murthe  mede þei aske  ¶ Þe kynge hath mede of his men  to make pes i n londe . Men þat techet teche(n) R.3.213: Here most B witnesses (including F) read teche(n). Manuscript C reads techeþ, which may be what was intended by R (C's form is used by R on many other occasions). MED, s. v. techen, acknowledges techet as a possible 3rd sing. pres. inflection of techen, but the few citations are almost all from the twelfth and thirteenth centuries. children  crauen of hem mede . ¶ Prestes þat p reche þe poeple  to god to gode R.3.214: Here beta reads gode, but Hm agrees with alpha, specifically with F (= of god). Ax agrees with beta. asken mede . And masse -pans and here mete  at at þe R.3.215: Beta reads at þe while F has at here . meal -tymes . ¶ Alle kynnes crafty men  craue n mede for her e prentyces . Marchauntz and mede  most nede go to -gyderes . No wiȝt as I wene  with -oute mede may lybbe . ¶ Quatȝ þe kyng to consience  by crist as me thenketh . Mede is is wel worthy R.3.220: Beta reads wel worþi while F revises the entire a-verse, producing Now ys Meede worthy. Ax reads this verse exactly as R does, and though the C version introduces a phrasal revision in mid-line, it also omits beta's qualifier before worthi, supporting the originality of R's reading.  þe maystrie to haue . R.3.220: The eighth leaf of Rawlinson is slit (part of the same act as the cropping of leaves 1-7) at the right edge of the writing area for a space of some 12.5 cm, from R3.229-49. Nay quatz consience to þe kyng e  and kneled to þe erthe . Þere arn to maner e of medes  my lord by with R.3.222: Beta reads with, but Ax agrees with alpha. ȝoure leue . to .....ratio bonor um qui . operans iusticiam In the right margin, in a brown contemporary hand, an early reader left behind this partially legible note. Þat on god of his grace  graunteth in his blisse . To þo þat wel werchen  while þei ben here . Þe p rophite p recheth þere -offe  and putte it in þe sauter R.3.225: In the right margin, just inside the ruling, someone has added an oversized punctus for this line in a darker shade of ink than that used by the original scribe. D omine quis h abitabit in tabernaculo tuo . &c etera . Lord who schal wonie in þi wones  with and with R.3.227: Beta reads and with; the line does not appear in either A or C. þi holy seyntes . Or resten on in R.3.228: The alpha reading is here confirmed by LM, but the other beta copies show in. þin holy hilles  þis asked asketh R.3.228: Beta shows the present form, asketh. dauid . And dauid assoyleth it hym -selue  as þe sauter telleth . psal xiv R.3.230: The appropriate scriptural reference has been inscribed in the right margin by an early reader. Qui ingreditur sine macula  & operatur iusticiam . Þoo þat entren of o colour e  and of one wille . And haue I -wrouȝt werkes  with riȝt and with resou n . And he þat ne vseth nauȝt  þe lyue of vsurie . And enformeth pouer e men  and pursueth trewthe . mor wm R.3.235: In the right margin are the remains of what appears to have been a pen trial. Qui pecunia m sua m no n dedit ad vsura m & mun era sup er innoc entem . &c etera . And alle þat helpeth þe innocent  and holt with þe riȝtful . With -oute mede doth hem goed  and þe trewe trewthe R.3.237: Beta reads trewthe. helpeth : Richard A sixteenth-century reader has recorded a pen trial, written vertically downwards in the right margin. Suche man er men my lorde  schal haue þis furste mede . Of god at a grete nede  whanne þei gon hennes . Þere is an -other e mede mesurles  þat maystres desireth . To meytene misdoeres  mede þei take . And þere -offe seith þe sauter  in a salmes ende . In quor um manib us iniquitates sunt  dext ra eor um repleta e st mun erib us . And he þat gripeth her e gold  so me god helpe . Schal abye it ful bitter e  or þe book lyeth . Prestes and persones  þat plesinge desireth . Þat taked take(n) mede and mone  for masses þat þei songen syngeth . R.3.247: R's past-tense verbs in this line are unique; all other B manuscripts use present-tense forms. Cf. see the Introduction III.2.2.10 on R's problematic tense marking. Both the A and C versions of this line agree with Bx on these verbs being in the present tense. Taketh here mede her e  as mathew vs techeth . Amen amen recipiebant mercedem suam . Þat laboreres and lewed lowe folke  taketh of here meystres . It nis is R.3.251: The negative verb is unique to R; all other B manuscripts read is. no maner mede  but a mesurable hire . In marchaundise is no mede  I may it wel a -vowe . It is R.3.253: Beta reads a permutacioun . This is also the reading found in the other two versions. permutacion a permutacioun apertly  a peny -worth for another . Ac reddestow neuer e Regu m  þow recrayede mede . Whi þe vengaunce fel  on saul and on his children . God sente to saul  by samuel þe p rophete . Þat agag of amalec  and and al R.3.257: In the B tradition, R uniquely omits al before the phrase his poeple. However, though Ax reads the phrase exactly as the B majority, a significant, closely interrelated set of A manuscripts (TDChH 2) also omits al at this point. The same phenomenon can be seen among the C witnesses, a small minority reading with R while most agree with F/beta. his poeple after . Schulde deye for a dede  þat don hadde here elderes . For -thi sayde samuel to saul  god hym -selue hoteth . Þe be buxu m at my his R.3.260: Beta reads his biddynge , which is also the reading found in the A version, but the C text agrees with alpha's my. byddynge  his will e to fulfille . Wend to amalec with þin host  and what þow fynst þer e slee it . Bernes and bestes  brenne hem to dede . Wydewes and wyues  wommen and chyldren . Mebles and vnmebles  and al þat þow miȝt fynde . Brenne it bere it nauȝt a -way  be it neu ere so riche . For mede ne for mone  loke þow destruye it . Spille it and spare it nauȝt  þow schalt spede þe bett ere . And for he coueytede her e catel  and þe kyng spared . For -bare hym and his bestes bothe  as þe bible wytnesseth . Otherwise þan he was  warned of þe p rophete . God seyde to samuel  þat saul schulde deye . And alle his seed for þat synne  schentfulliche endede ende . R.3.272:The RF usage of past tense here (beta = ende) reflects alpha and is shared by convergence with Hm. Though both A and C traditions attest—in isolated manuscripts—preterite forms for this verb, it is obvious that the beta uninflected base form is archetypal in both of the poem's other versions. Suche a meschief mede made  saul þe kyng to haue . Þat god hated hym for euere  and alle his heires after . Þe culoru m of þis cas  kepe I nouȝt to schewe . Aunter An auenture R.3.276: Beta opens this line with An auenture; F begins it with For hap . The witness of the other two versions is in favor of An aunter . it nuyed men  non eende wille I make . For so is þis worlde went  with hem þat han power e . Þat who -so seyth hem sothes  is sonnest I -blamed . I consience knowe þis  for kynde witt me it tauuȝte . Þat rosoun r[e]soun resoun schal regne  and rewmes gouerne . And riȝt as agag hadde And riȝt as agag hadde  happe schulle so mme . The ninth leaf of Rawlinson is slit (part of the same act as the cropping of leaves 1-7) at the right edge of the writing area for a space of some 10 cm, from R3.296-309. Samuel schal scleen hym  and saul schal be blamed . And dauid schal be diademed  and dau nten hem alle . And on cristene kyng  kepen hem alle . Schal namore mede  be mayster as he she R.3.285: Alpha's he does not differ substantively from beta; he is a variant of heo, "she." See note at R.3.119:. is nowthe . I R.3.286: In the extreme right margin, beyond the pricking, is what appears to a modern eye to be a barred 9, probably an <I> or <Q> in this hand, though it is unclear what it signifies. Ac loue and lowenesse  and lewte to -gyderes . Þese schul be maystres on molde  trewthe to saue . And hoo -so trespaseth aȝeynes trewthe  or taketh aȝeynes his wille . Lewte schal don hym lawe  and no lif elles . Schal no seriaunt for here s eruise  were a silk howue . Ne non pelure in his clook  for pledyng atte barre . Mede of mysdoeres  maketh many lordes . And ouer lordes lawes  reuleth þe realmes . Ac kende loue schal come ȝut  and consience to -gydere . And make of lawe a laborer e  suche loue schal aryse . And suche pees a pees R.3.296: Beta reads such a pees ; the C version agrees with alpha. amonge þe poeple  and a parfit trewth . Þat Iuwes schal wene in her e witt  and wexen wonder glade . Þat moyses or messye  be come in -to þis eerthe . And haue wonder in her e hertes  þat men beth so trewe . Alle þat bereth baslard  brode -swerde or launce . Ax other hachet  or eny wepne elles . Schal be demed to þe deth  but ȝif he do it smythye . Into sycul or to sythe  to schare or to culter e . Conflabunt gladios suos in vomeres &c etera . Ich man to pleye with aplow a plow  pycoys or spade . Spynne other or R.3.306: R's other is unique; the remaining B witnesses attest or. sprede dong  or lese spille R.3.306: This non-alliterating third stave ( lese for Bx's spille) is an alpha reading. hym -self with slewthe . Prestes and persones  with placebo to hunte . And dyngen vppon dauid  vch a day til eue . Huntynge or haukynge  ȝif any of hem vse . His boste of his benefice  worth by -no mme hym after . Schal neyther kyng ne kniȝt  constable ne meyre . Ouer -lede þe comune  ne to þe court sompne . Ne putte hem in panel  to don hem pliȝt here trewthe . But after þe dede þat is don  on dom schal rewarde . Mercy or no mercy or no mercy or no mercy or no mercy  as trewthe wil acorde . Kynges court and comune court  constorie and chapitele . Alle schal be but on court  and on baron be iustice . Þanne worth trewe tonge  a tydy man þat tened me neuere Batayles schul non be  ne noman ber e wepne . And what smyth þat any smithie smyteth R.3.320: R's smithie is the alpha variant; most of the beta copies attest either smyþeþ (e.g., WO) or smyteth (LC).  be smiten þer e -with to dede dethe R.3.320: R's dede is a unique variant, but its meaning is identical to the common reading ( dethe) found in other witnesses, both from the B and the C traditions. . Non leuabit gens contra gentem gladiu m &c etera . And er e þis fortune falle  fynde men schal þe werste . By six sonnes and a schipp e  and half a schef of arwes . And þe meddel of a mone  schal make þe Iewes torne to torne R.3.324: Cf. beta's Iewes to torne ; but the C reading clearly agrees with alpha's omission of to. . And sarasyns for þe siȝte þer e -offe þat siȝte R.3.325: R's þe siȝte þere -offe reflects alpha; cf. beta's þat siȝte. The C reading is identical to beta's.  schul synge gl oria i n excelsis . For makomet and mede  myshappe schal þat tyme . For melius est bonu m nome n q uam diuicie multe . Also wroth as þe wynde  wex mede in a while  I can can no latyn R.3.329: R's omission of the negative in this phrase is unique among the extant B manuscripts but is shared by convergence with Cr 3. quatz sche  clerkes wote þe sothe . See what salomon seith  in sapiences sapience R.3.330: R's genitive sapiences is unique; F and beta both attest the unmarked possessive form. bokes . Þat hij þat ȝiueth ȝiftes  þe victorie wynneth . And muche wurchip haad þer ewith  as holy writt telleth . Honorem adquiret qui dat mun era munera &c etera . I leue wel lady q uod consience  þat þi latyn be trowe tr[e]we trewe . Ac þow art lyk a lady  þat radde a lesson ones . Was o mnia p robate  and þat plesede hire herte . For þat lyne was no lenger e  atte þe leuesende leues ende . Had sche loked þat other half  and þe lef I -turned torned R.3.338: R's I -turned reflects an alpha reading; cf. beta's torned. The Cx reading is identical to beta's. . Heo She schulde a haue R.3.339: R's a is unique in form but represents the same semantic element as the majority's haue. founde fele wordes  folwyng þer e -after . Quod bonu m est tenete  trewthe þat tixt made . And so ferd ȝe ma -dame  ȝe coude namore fynde . Þo ȝe loked on sapience  sittynge in ȝour e stodie . Þis tixt þat ȝe haue Itolde tolde  were goed for lordes . The tenth leaf of Rawlinson is slit (part of the same act as the cropping of leaves 1-7) at the right edge of the writing area for a space of some 4 cm, from R4.12-18. Ac ȝow failede a cu nnyng clerk  þat coude þe lef attorned a torned R.3.344: a, "have." . And if ȝe seke sapience eft  fynde schal ȝe þat foleweth . A ful teneful tixt  to hem þat taketh mede . And þat is a ni mam aute m aufert accipientiu m accip ientium &c R.3.347: After this Latin tag (completely omitted by F), beta adds &c.. Cx agrees with R in omitting it. . And þat is þe taile of þe tixt  of þat þ at ȝe schede schewed R.3.348: R's schede is unique in form but probably represents the same intention as beta's schewed (F omits the entire line). Nevertheless, neither MED, s. v. sheuen (v. 1), nor OED2, s. v. show, acknowledges R's form as a viable inflected spelling for the verb in question. . Þat þow we wynne worchipe  and with me mede R.3.349: Alpha introduced the error of me for original mede. haue victorie . Þe soule þat þe souȝd soude taketh  be so muche is bounde . p R.4.0: At the extreme right margin, on the same line as the passus rubric, there is a small, brown <p>, presumably the remains of a cropped guide. R.4.0: Possibly other material was erased along with the parasign. Passus quartus de visione petri plowman . vt s upra . Sesseth R.4.1: There is no blue ornamental capital at the head of this line; instead, the <S> is the same size and color as all line initials. seyde þe kyng  I suffre ȝow no lenger e . Ȝe schal sauȝtne for -sothe  and s erue me bothe . Kysse hire quatz þe kyng  consience I hote . Nay by crist quatz consience  congeye me for euere . But reson rede me þere -to  rather wil I deye . And I comande þe q uod þe kyng  to consience þenne . Rape þe to ride  and resou n þow fecche . Comaunde hym þat he come  my conseill e to here . For he schal reule my reume  and rede me þe beste . R.4.9: Like most B manuscripts, R here lacks a following line attested in YOC 2 as well as in the Crowley imprints and in versions A and C: Of Mede and of mo oþere what man shal hire wedde. And acounte with þe consience  so me crist helpe . How þow lernest þe poeple  lered þe lered and lewede þe lewede . R.4.11: R is unique in dropping two determiners from the Bx b-verse ( þe lered and þe lewede ). Cf. F's boþe lerede & þe lewede . I am fayn of þat forward  seyde þe freke þanne . And ritt riȝt to resou n  and rou nneth in his ere . And seyde as þe kyng bad  and sithes tok his leue . I schal araie me to ride quatz resou n  reste þe a while . And called caton his knaue  curteyse of speche . And al -so tho mme trewe tonge  telle me no tales . Ne lesyng to lawe of  for I louede hem neuere . And sette my sadel vppe vppon R.4.19: R's vppe is unique; Bx reads vppon, the same reading found in Ax and Cx. suffre  til I se my tyme . And let warrok it wel  with witti wordes gerthes . And hange on hym þe heuy bridel  to helde holde R.4.21: R's helde is a unique form; the other B copies read holde. his hede lowe For he wil make wehe  tweye R.4.22: Only LM agree with this R reading. Cf. F's to trewþe and the predominant beta reading twies. ere he be þere . R.4.23: This paraph marker is blotted and smeared. Þanne consience vppon his capel  cayreth forth faste And resoun with hym rit  rounyng e togyderes . Whiche maystries on eerth  mede þe mayde maketh Mede maketh on þis erthe R.4.25: R's phrasing here reflects an alpha error; cf. beta's Mede maketh on þis erthe . Not only is the alliterative pattern of beta's a-verse appropriate (where alpha's is not); in its favor there is also the fact that Cx's a-verse (in a revised line) is identical to beta's. On waryn wisdome  and witty his fere . Folwed hem faste  R.4.27: R shares with L an omission ( for þei ) at the head of this b-verse. The reading of Ax agrees with that of the B majority. However, because of the extraordinary accuracy that the R and L scribes regularly demonstrate, and because these copies are at the top of any credible B stemma, it seems likelier that their shared mistake here attests the presence of a marginal correction in Bx (which both scribes overlooked) than a purely random convergence in ordinary omission. haued to done . In þe cheker cheker and R.4.28: After cheker, R uniquely omits and. L and R then agree against all the other B copies in reading at(te) þe chauncerie where the majority reads in þe chauncerie with W (and, probably, Ax). F simply omits the prepositional phrase altogether, reading & Chawncerye. atte þe chauncerie  to be descharged of thinges . And riden fast for resoun  schulde rede hem þe beste . For to saue hem fro for siluer R.4.30: R's fro siluer is a unique error but may reflect alpha; most other B manuscripts here read for siluer. Cf. F's & hem.  fram schame and fram harmes . And consience knewe hem wel  þei louede coueytise . And bad bad resou n R.4.32: R uniquely omits resoun from the phrase, And bad resoun ride . ride faste  and recche of her e nother . Þere arn wiles in here wordes  and with mede þei dwelleth . Þere -as wrathe and wranglyng is  þer e wynne þei siluer . R.4.35: R uniquely omits the line-head conjunction witnessed in other B manuscripts as Ac (a majority), And or But. Cx reads Ac. Þere Ac þere is loue and lewte  þei wole nouȝte come þere . Contric io & infelicitas in vijs eor um . &c etera . Þei ne gyueth nouȝt of god  on gose wynge . Non est timor dei ante oculos eor um . For wote god þei wold do more  for a doseyne chikenes . Or as many capounes  or for a sem of otes . Þanne for loue of oure lorde  or alle hys leue seyntes . For -thi R.4.42: R's omission of the vocative resoun after For -thi is derived from alpha and shared with F. late resoun lete hem ride  þo riche be hem -selue . For consience knoweth hem nouȝt  ne crist as I trowe . And þanne resou n roed fast  þe riȝt hye gate . As consience hym kennede  til þei come to þe kynge . Curteiselich þe kynge þanne  come aȝeyne resoun . R.4.46: There is a superfluous bar over the <n> of resoun. And bytwene hym -self and his sone  sette hym on benche . And wordeddu n wel wyseliche  a grete whyle to -gyderes . And þanne come pees  in -to parlement  and putte forth a bille . How wronge aȝeynes his wille  haued his wif taken . And how he rauesched rose  raynaldes loue . And margarete of hir e maydenhode  mawgre hire chekes . Bothe my gees and my gris  his gedelynges feccheth . I dar nauȝt for fere of hym  fiȝt ne chide . He borwed of me bayard  and he brouȝt it hym R.4.55: Where the other B manuscripts read the pronoun reference to Bayard as hym, R uniquely reads it. Both Ax and Cx agree with the B majority. home neu ere . Ne no ferthyng þere -fore  for nauȝt auȝte I couthe plede . He meynteneth his men  to murther myn heuues R.4.57: The variant heuues derives from alpha. F's spelling here is heuuys ; cf. beta's hewen = "servants." Kane-Donaldson transcribe this variant in R as hennes and F's as hennys). What the R scribe actually thought he was writing is open to doubt since neither he nor the F scribe again uses <uu> for <w> after <e>. However, R writes the name "Stowe" as stouue at R5.28, and it is reasonable to assume that is what he intended here. . He nil forstalleth R.4.58: R uniquely adds He at line-head position. Cf. F's & to. Beta simply begins the line with Forstalleth. The Ax reading agrees with beta's, but Cx begins the line And forstalleth. my feyres  and fiȝteth in my chepynge . And breketh vppe my bernesdore bernes dore  and bereth a -wey my whete . And taketh me but a tayle  for ten quarteres of otes . And ȝet he bet me þer eto  and lyth by my mayde . I am nauȝt hardy for hym  vnnethe to loke . Þe kynge knewe he seyde sothe  For consience hym tolde . Þat wronge was a wikked luste R.4.64: R shares this obvious transcriptional error ( luste for luft) with G and BmBoCot by convergent variation. F's wyght has the appearance of a typical smoothing, so R may have derived the error from alpha.  and muche sorwe wrouȝte . R.4.64: R's b-verse involves a transposition (uniquely shared with H) of the Bx phrase, which reads and wrouȝte moche sorwe. Wronge was aferd þanne  and wysdom he souȝte . To make pees with his pans  and profered hym manye . And seyde hadde I loue of R.4.67: Here alpha and Cr omit a stave; in beta the phrase would have read my lorde þe kynge . þe kynge my lorde þe kynge  lytel wolde I recche . Þeiȝ pees and his powere  pleyned hem eu ere . Þo wan R.4.69: Only L and C agree with R's wan; F omits the entire line, while most beta copies read wente. The b-verse of this line is unmetrical and the entire line is a revision of the A version (K4.53), but the uniformity of attestation for its text from L and R (with C's concurrence) make it clearly a genuine Bx line. wysdom  and sire waryn þe witty . For þat wronge hadde I -wrouȝte  so wikked a dede . And warned wronge þo  with swyche a wise tale . Hoo -so wercheth by wille  wreth maketh ofte . I sey it by thi -self  þow schalt it wel fynde . But if mede it make  þi meschief is vppe . The eleventh leaf of Rawlinson is slit (part of the same act as the cropping of leaves 1-7) at the right edge of the writing area for a space of some 3.5 cm, from R4.74-77. For bothe þi lif and þi londe  lith in his grace . Þanne wowede wronge  wisdom ful ȝerne . To make his pees with his pans  handi -dandi payed . Wisdom and wit þanne  wenten to -gyderes And token mede myde hem  mercy to wynne . Pees putte forth his hed  and his panne blody . c iij With -oute gult god it wote  gatt I þis skathe . Consience and þe co mmune  knowen þe sothe . Ac wisdom and witt  were aboute faste . To ouercome þe kyng e  with catel if þei myȝte . Þe kyng swore by crist  and by his crowne bothe . Þat wronge for his werkes  schuld wo thole . And comanded a conestable  to casten hym in yrenes . And lat hym nauȝte þis seuen ȝeer e  seen his feet ones . God wote quatz wisdom  þat were nauȝt þe best . And he amendes mowe make  lat meynprise hym haue . And be boruȝ for his bale  and buggen hym bote . And so amende þat is mysdo  and euer e be euermore R.4.92: R's phrase, euere be, is unique, but alpha and Cr agree on euere; cf. beta's euermore. F reads the entire phrase as he shal do euere þe bettre. Both Ax and Cx agree with beta. þe bett ere . Witt acorded þere -with  and seyde þe same . Þat nil R.4.94: Þat is an alpha addition unattested by any beta manuscript. Both Ax and Cx agree with beta on the phrasing of this a-verse. bett ere is þat bote  bale adoun brynge . Þanne bale be I -bete  and bote neuer e þe bett ere . And þanne gan mede to meken mengen R.4.96: Beta reads mengen, but clear majorities of both the A and C witnesses support alpha's reading. hire  and m ercy he she R.4.96: He, variant of heo, "she." See note at R.3.119:. bysouȝte . And profered pees a present  alle of pure golde . Haue þis man of me quatz sche  to amende þi scathe . For I wil wage for wronge  he wil do so na more . R.4.100: This paraph marker was mistakenly executed in red and then erased, in anticipation of being redrawn in blue. Pytousliche pees þanne  preyed to þe kyng e . To haue mercy on þat man  þat mysdede hym so ofte . For he hath waged me wel  as wisdom hym tauȝte . And I forgyue hym þat gult  with a goed wille . So þat þe kyng assente  I can sey no bett ere . For mede hath made myn me R.4.105: For alpha's myn, beta reads me. Both Ax and Cx attest possessive case for this pronoun rather than objective case. amendes  I may namore axe . Nay quatz þe kyng þo  so me god cryst R.4.106: R's god is an alpha reading; cf. beta's cryst. helpe . Wronge wendeth nauȝt so awey  arst wil I wyte more . For loupe he so liȝtly  leyȝhen he wolde . And efte þe bolder be  to beten myn hewes . But reson haue reuthe on hym  he schal rest in my stokkes . And þat as longe as he lyueth  but lowenesse hy hy[m] hym borwe . Su mme men radde reson þo  to haue reuth on þat schrewe . And for to consayle þe kyng  and consience after . Þat mede moste be meynp ernour e  reson þei besouȝte . Rede me nauȝt q uod reson  no reuth to haue  Til lordes and ladyes  louyen alle treuthe . And haten alle harlotrie  to heren it or to mouthe it . Til peronelles purfil  be putte in hire hucche . And childerne chirissynge  be chastyng with ȝerdes . And harlotes holynesse  be holden for nauȝte an hyne . R.4.120: R's nauȝte is unique; cf. F's vanyte and beta's an hyne. The C version completely revises this b-verse, but the Ax version agrees exactly with beta's rendering. Til clerkene coueytise  be to clothe þe pore and to fede . And religiouse romares  recordare in here cloystres . As seynt benet hem bad  bernard and Franceys . And til p rechoures p rechyng e  be proued on hem -selue . Til þe kynges conseyll e  be þe co mmune p rofit Til bisshopes bayardes  ben beggere beggeres R.4.126: Beta shows the genitive plural beggeres; F reads beggerys in. chaumbres . Here haukes and here houndes  help to pouer e religiouses Religious . R.4.127: R's plural form is unique; beta reads Religious, while F has relygyouse howsys. And til seynt Iames be souȝt  þere I schal assygne . Þat no man go to galys  but if he goo for euere . And alle rome renneres  for robberes of by -ȝende . Bere no siluer ouer see  þat signe of kyng scheweth . Nother graue ne vn -graue  gold nother seluer . Vppon forfeture of þat fee  ho -so fynt hym at douer e . But if it be marchant or his man  or messager e with l ettres . Prouisoure or preest  or penaunt for his synnes . The twelfth leaf of Rawlinson is slit (part of the same act as the cropping of leaves 1-7) at the right edge of the writing area for a space of some 2 cm, from R4.134-37. R.4.136: This paraph marker was mistakenly executed in red and then erased, in anticipation of being redrawn in blue. And ȝet q uod resou n be þe rode  I schal no reuthe haue . While mede hath þe maystrie  in þis moot halle . Ac I may schewe ensaumples  as I se other e otherwhile otherwhile otherwhile . I sey it be my -selue q uod he  and it so were . Þat I were kyng with crowne  to kepen a rewme . Schuld neu ere wrong in þis world  þat I witt miȝte  c iiij Ben vnpunesched in my powere  for peril of my soule . Ne gete my grace þoruȝ for R.4.143: R's þoruȝ is an alpha variant shared exclusively with F. Beta reads for. Cx is revised immediately beyond this point in the line, but it agrees with alpha on the identity of this preposition. Ax reads the entire a-verse exactly as alpha does. gyftes  so me god saue . Ne for no mede haue mercy  but mekenesse it made make . For nullu m malu m þe man  mette with inpunitu m . And bad nullu m bonu m  be irremuneratu m . Lat ȝoure confessour e sir e kyng  construe þis vn -glosed . And ȝif ȝe werkyn it in werke  I wedde myn eres . Þat lawe schal ben a laborere  and lede afelde donge . A hand in the left margin points to these lines. And loue schal lede þi londe  as þe leef lyketh . Clerkes þat were confessores  couplede hem to -gyderes . Alle to construe þis clause  and for þe kynges p rofit . Ac nauȝt for conforte of þe co mmune  ne for þe kynges soule . For I seyȝ mede in þe moot halle  on men of lawe wynken . And þei lawhynge lope to hir e  and left reson manye . Waryn wysdom  wynkede vppon mede . And seyde madame I am ȝour e man what -so my mouth iangle . I falle in floreyns q uod þe þat R.4.158: R's þe is an alpha variant (cf. beta's þat), shared by convergence with Cr 2-3. freke  and faile speche ofte . Alle riȝtful recordede  þat resou n R.4.159: The scribe's rendering of this word is ambiguous; he has placed a dot over the final character rather than the customary nasal bar, but his typical spelling suggests that he intended this word to be construed here as resoun. trewthe tolde . And witt acordede þere -with  and comended his wordes . And þe most poeple in þis þe R.4.161: R's þis is unique. The other B manuscripts read þe. halle  and many of þe grete . And leten mekenesse a mayster e  and mede a mansed schrewe . Loue loue Loue Loue lett of hire lyȝte  and leute wel ȝit R.4.163: R's wel is an alpha variant shared with Hm by convergence. Beta reads ȝit. Cx agrees with beta. lasse . And seyde it so hye  þat alle þe halle it herde . Ho -so wilneth hire to wyue  for welth of hire godes . But he be knowe for a kokewolde  bitte kut R.4.166: R's bitte is unique error. The other B manuscripts read kut, which matches the line's alliterative pattern. Both Ax and Cx agree with the B majority's reading. of my nose . Mede morned þo  and made heuy chere . For þe most co mmune of þat courte  called hire an hore . Ac a sysour e and a sompnour e  sued hire faste . And a schirreuesclerke schirreues clerke  by -schrewed alle þe route . For ofte haue I q uod he  I -hulpe holpe(n) ȝow atte barre . And ȝet ȝeue ȝe me neu ere  þe worth of a rusche . Þe kyng called consience  and afterwardes resou n . And recordede þat resou n  hadde riȝtfulliche yschewed schewed . And modiliche vppon mede miȝte with myȝte þat þe kyng loke loked . R.4.175: Alpha wrongly construed myȝte as a modal instead of a noun. And gan wax wroth with lawe  for mede almost had schent it . And seyde þurȝ ȝour e lawe as I leue  I lese many chetes . Mede ouer -maystrieth lawe  and muche treuthe letteth . Ac resou n schal rekne with ȝow  ȝif I regne any while . And deme ȝow be þis daye  as ȝe haue deseruede . Mede schal nouȝt meynprise ȝow  be þe marie of heuene . I wil haue lewte in lawe  and let be al ȝoure iangelynge . And as alle moste R.4.183: R's alle is an alpha reading. Cf. beta's moste. If beta is original at this point, the omission of the second alliterating stave wel at the end of the a-verse is also derived from an alpha error but one shared with Cr, G, and H. However, it is quite conceivable that F's reading for this half-line is the authorial one (though unlikely to have been derived from alpha): & as alle wyȝes witnesse . folke witnesseth  wronge schal be ydemed demed . quatz consience to þe kyng e  but þe co mmune wil assente . It is ful hard be my hed  herto to brynge it . Alle ȝoure ȝowre lige R.4.186: R uniquely omits beta's lige from the phrase lige leodes ; cf. F's phrase, londys leedys. In the C tradition, the phrase is either identical to beta's (a majority of C witnesses) or more closely resembles beta's (XYcP 2 = lege lordes ). Most A witnesses also agree with beta. leedes  to lede þus euene . By hym þat rauȝte on þe rode  q uod resou n to þe kyng e . But if I rewle þus ȝoure rewme  rend oute my guttes . If ȝe bydden buxu mnesse  be of myn assente . And I assente quatz R.4.190: Here beta probably read seith (the reading of LMCrW), but several beta witnesses (HmGOH) support alpha's quod (F) or quatz (R). Ax clearly agrees with the alpha reading, despite its deficient alliteration. þe kyng  by seint marie my lady . Be my conseill e I -come  of clerkes and of erles . Ac redily resou n  þow schalt e nouȝt ride fro me . In the right margin, in black ink, there is an early ownership stamp for the Bodleian Library. The thirteenth leaf of Rawlinson is slit (part of the same act as the cropping of leaves 1-7) at the right edge of the writing area for a space of some 2 cm, from R4.192-94. For as longe as I lyue  leet þe Inelle I nelle . I am aredy q uod resou n  to rest with ȝow euere . So consience be of oure conseil  I kepe no bett ere . And I graunt q uod þe kyng  godesforbode godes forbode he it R.4.196: For R's he, cf. F's þou and beta's it. Ax agrees with R's reading. faile . As longe as oure lif last lasteth R.4.197: Beta reads lasteth.  leue we to -gyderes . s us Passus quintus de visione petri plowman . vt sup ra . t T he kyng and his knyȝtes  to þe kerke wente . To here matynes of þe day  and þe masse after Þanne awaked I of my wynkyng  and wo was with -all e . Þat I ne hadde sleped sadder e  and I -seyȝen more . Ac er I hadde faren a furȝlonge  feyntise me hente . Þat I ne miȝte forther e a foot  for defaut of slepynge . And sat softly adoune  and sayde my beleue . Nota And R.5.8: R uniquely omits so I before babeled and replaces the presumably original on with vppon ; cf. F's y bablede so on. C omits this passage, but in the A version it reads as in the beta manuscripts of B. babeled so I babeled vppon on my bedes  þei brouȝt me a -slepe . And þanne saw I muche more  þan I befor tolde . For I say þe felde ful of folke  þat I be -fore of seyde And how resoun gan arayen hym  alle þe rewme to p reche . And with a crosse by -for þe kyng  cumsede þus to techen . He p reued þat þis pestilensez  was for pure synne . And þe south -west wynde  on saterday at eue . Was perteliche for pruyde pure pryde R.5.15: R's phrase here is that of alpha (cf. F's virtually indentical wording); the beta copies read pure pryde . The A and C versions both agree with alpha.  and for no poynt elles . Pyries and plumtres  wer e puffedde to þe erthe . And In R.5.17: And is an alpha variant (cf. F's somewhat different rendering of this line), but it is not present in the beta manuscripts, nor is it attested in the manuscripts of the other versions at this point. in ensaumple ȝee segges  ȝee schulden do þe bett ere . Beches and brode okes  were blowe to þe grounde . And nil R.5.19: And is an alpha variant completely unattested in beta copies (which begin the line with Torned. However, the alpha reading of this line opening agrees exactly with that of the other two versions. turned vpward here taile tailles in R.5.19: Alpha omits beta's in before tokenynge. Both the A and C versions agree at this point with beta. to kenynge tokenynge in tokenynge of drede . Þat dedly synne ar domesday  schal for -don hem alle . Of þis matere I miȝt  mamely ful longe . Ac I schal seye as I sawe  so me god helpe . How perteliche be -for þe poeple  reson gan to p reche . He bad wastour to go R.5.24: R's to is unique among the B manuscripts. The others read go, which is also the reading of the A version and of the P family of C witnesses. However, the X family agrees with R's reading. werche  what he best coude . And wynnen his wastinge  with so m maner e crastys cra[f]tys crafte . R.5.25: R's crastys, is probably an alpha error (cf. beta's crafte and F's reconstructed werkys). The phrase is omitted from C, but the A witnesses support beta's rendering. And preied pernele  hire purfile to lete . And kepe it in hire coffre  for catel at hire nede . Thomme stouue he tauȝte  to take to stones staues . R.5.28: R's stones is unique. The other B manuscripts, as well as the other versions, read staues. And fecche felice home  fram wyuene þe wyuen pyne . He warned watte  his wif was to blame . Þat hire hed was worth half marke  his hode nauȝt a grote R.5.31: HmGCotH join R in omitting worth from the final phrase of this line (witnessed by F and most beta copies as nouȝte worth a grote . The majority of A witnesses agrees with this B majority in attesting the word, but RaUChJEK agree with Rawlinson 38 in omitting it. So do all but two of the C witnesses. . And And bad R.5.32: R uniquely omits the verb bad before bet here. bet cutte  a bow other tweye . And bete betou n þere -with  but ȝif heo she wolde werche . And þanne he charged chapmen  to chaste R.5.34: The uncovered final <e> makes R's reading unmetrical. R's verb form is unique among the B witnesses; some beta manuscripts have chastize(n). However, R's form is attested in some copies of both A and C. Likewise, some of the A witnesses agree with beta. The Cx form appears to have been chasten, quite possibly the reading of Bx, since it is also the reading of LCrCG. hire childerne . Late no wynnynge for -wanyen  þe hem forweny R.5.35: R uniquely omits Bx's hem at the end of the a-verse and uniquely adds þe at the head of the b-verse. However, the addition of þe is paralleled in the X family of C manuscripts. while þei ben ȝonge . Ne for no pouste of pestilence  plese hem nauȝt oute of resou n . My sire seyde so to me  and so dede my dame . Þat þe leuer childe  þe more lore byhoueth . And salomon seyde þe same  þat sapience made . Qui parcit virge odit filiu m . Þe englisch of þis latyn is  ho -so wil it knowe . Ho -so spareth þe sprynge  he nil spilleth R.5.42: Beta omits he. Although three C manuscripts include the pronoun, it seems clear that Cx read here as beta does. his childern . And sitthen he p roued preyed R.5.43: R's p roued is unique and presumably results from scribal anticipation of p roue in the next line; beta reads preyed (which is confirmed by both the A and C versions) while F completely rewrites the line. p relatz  and prestes to -gyderes . Þat ȝee prechen to þe poeple  p roue it on ȝow -selue . And doth it in dede  it schal drawe ȝow to gode . If ȝe lyuen as ȝe lerne vs  we schal leue ȝow þe bett ere . And sitthe he radde religiou n  here rewle to holde . Lest þe kyng and his conseyle  ȝour e comunes appeyre . And ben stwardes of ȝoure stedes  til ȝe be rewled bett ere . And siþþen he conseyled þe kynge  þe co mmune to louye . It is þi tresor if treson were treson [n]ere tresoun ne were R.5.51: R uniquely drops the negative.  and triacle at þi nede . And siþþen he preyed þe pope R.5.52: The word pope has been partially erased, though the original reading is visible. The erasure is very old but not the work of the original scribe.  haue pite on holy cherche . And er he gyue any grace  gouerne furst hym -selue . And ȝe þat haue lawes to kepe  lat trewthe be ȝour e coueytise . More þanne gold or other giftes  if ȝe wil god plese . For ho -so contrarieth trewthe  he telth in þe gospel . Þat god knoweth hem nouȝt  ne no seint of heuene . Amen dico vobis nescio vos  . And ȝe þat seke seint Iames  and seintes of rome . Seketh seint trewthe  for he may saue ȝow alle . Qui cu m p atre & filio  þat fair e hem befalle . Þat sueth my sarmou n  and þus seyde resou n . Þanne ran repentaunce  and reherced his teme . And gerte wille to wepe  water with his eyȝes . Pernele proude -herte  platte hire to þe erthe . And lay longe ar he she R.5.66: He, "she." loked  and lord mercy he criede cryed R.5.66: R alone reiterates the feminine pronoun (in R's typical form, he) in the b-verse. The correctness of the majority reading is confirmed by the text of Ax and Cx. . And by -hiȝte to hym  þat vs alle made . He She R.5.68: He, "she." sclulde sc[h]ulde shulde vnsowen hire serke  and setten þere an haire . To affaiten hire flesche  þat fers was to synne . Schal neu ere heyȝ herte me hente  but holde me lowe . And suffre to be myssayde  and so dede I neu ere . But now wil I meke me  and mercy byseche . For For al þis R.5.73: Beta's phrase is For al þis . F has For þ(a)t ." I haue  I -hated hated in myn herte . Þanne lecchour seyde allas  and on our e lady he cried To make m ercy for his misdedes  bitwene god and his soule . With þat he schulde on þe day þe saterday R.5.76: Cf. beta's þe saterday and F's euery day. The C version is completely revised at this point, but the A reading agrees completely with that of beta.  seuen ȝer þer e -after . Drinke but with þe doke  and dyne but ones . Enuye with heuy herte  askede after schrifte . And carfulliche mea culpa  he cumsed to schrewe schre[u]e shewe . R.5.79: R's error, schrewe, was not a misreading for shewe (the beta variant) but for schreue, the alpha reading (cf. F's shryve). The A reading agrees with that of beta. He was as pale as a pelete  in þe palsey he semed . And cluted clothed R.5.81: This is a unique R reading ( Bx = clothed). According to MED, s. v. clouten (v. 1), the form is the past participle of clouten, which usually means "to mend" but here and in a few other documented instances clearly signifies "to wear patched or ragged clothes." in a tauri -mauri [c]auri -mauri R.5.81: R, probably by coincidence, shares the <c/t> confusion with Bm.  I coude nauȝt it it nouȝte R.5.81: R reverses this phrase, which in the other B manuscripts (as well as the A version) reads it nouȝte. descriue . In a kertel and curteby  and a knyf be his side . Of a frere freres R.5.83: R's uninflected form, frere, is unique in the B version. The other copies have freres. However, five A-version manuscripts (DJLaEN) agree with R's unmarked genitive. frokke  were his þe R.5.83: In place of alpha's his, beta reads þe. Ax is unclear on this point, a majority agreeing with beta, but a large minority (HaLaEAKWa) agreeing with alpha. fore -sleues . And as a leek hadde I -leye  longe in þe sonne . So loked he with lene chekes  louring foule . His body was to -bolle for wrathe  þat he bot his lyppes . And wryngyed with wryngynge he ȝede with R.5.87: Beta reads wryngynge he ȝede. F has hise hondis he wrong . þe fist  to wreke hym -self he thouȝte . With werkes or with wordes  whan he seyȝ his time . Eche word þat he warp  was of an addre an addres R.5.89: Beta shows a genitive, addres or Neddres. tonge . Of chydynge and of chalengynge  was his chief lyflode With bagbityng and with nil R.5.91: This line's second with is a unique addition in R. bysmere  and berynge of fals wytnesse . Þis was alle his curteysye  where þat euere he schewed hym . I wolde be Ischryue q uod þis schrewe  and I for schame durste . I wolde be gladder e by god  þat gybbe hadde mischaunce . Þan þouȝ I hadde þis woke I -wonne  a weye of essex chese . I haue a neyȝbore neyȝ me  I haue enuyed hym ofte . And lowen on hym to lordes  to don hym lese his siluer . And al -so made R.5.98: R's al-so is unique; cf. Bx's made. his frendes ben his fon  þoruȝ my fals tonge . His grace and his good happes  greueth me ful sore . By -twene mayne and mayne many and many R.5.100: The form mayne is the R scribe's spelling for meine, "household" (see also R.16.247:). In F the a-verse reads By-twixe hym & manye me n ; most beta manuscripts have Bitwene many and many. Though MED lists the head form as meine, OED notes that by the opening of the fifteenth century the word was sometimes spelled many, which appears to have been beta's intention. Its authenticity is also supported by a cognate line from the A version, where the phrase reads Betwyn hym & his meyne .  I make debate ofte . Þat both lyf and lyme  is lost þoruȝ my speche . And whan I mete hym in market e  þat I most hatye . I hayls hym hendelich  as I his frende were . For he is douȝtier þan I  I dar do non other . Ac hadde I maystrie and miȝt  god wote my wille . And whan I come to þecherche þe cherche þe kirke R.5.106: HmF agree with R in reading cherche, but they omit the article. The beta reading, kirke (also the reading of the A tradition), fits the alliterative pattern of the line; cherche was the alpha reading, shared by convergence with Hm.  and schulde knele to þe rode . And preye for þe poeple  as þe prest techeth . For pylgrimes and for palmeres  for alle þe poeple after Þanne I crie on my knes  þat crist ȝif hem sorwe . Þat bare away my bolle  and my broke schete . Awey fro þe auter  þanne turne I myn eyȝes . And beholde how heleyne  hath on nil R.5.112: R's on is an alpha addition unattested in beta or in the A version. a newe cote I wysche þenne it were myn  and alle þe web after . R.5.113: Here the scribe omits his usual line break before a new verse paragraph, presumably because he has reached the end of a side. And of his mennes R.5.114: R's his is a unique reading among the B manuscripts (both F and beta attest mennes). However, it is clear that Ax reads as R does. lesynge I lawhe  þat lyketh myn herte . Ac And for his wynnyge wynny[n]ge wynnynge R.5.115: At the beginning of this phrase, R's Ac is unique among the B manuscripts but is also the reading of Ax; by contrast, F has But and beta reads And). As for his (an alpha variant contrasting to beta's hir), a majority of A witnesses agree with RF. I wepe  and wayle þe tyme . And deme men nil R.5.116: R's men is unique; F substitutes hem while beta omits it altogether. However, in a slightly different phrase found in the cognate A line ( I deme men þere hy don ille), we find unambiguous support for R's reading. þat hij don ylle  þere I do wel worse . Who -so vndernymeth me her eoffe  Ich hate hym dedly after . I wolde þat vch a wyȝt  were my knaue . For ho -so hath more þanne I  þat angreth me sore . And þus I lyue loue -lees  lyche a lyther dogge . Þat alle my body bolneth  for bytter in of R.5.121: Cf. R's in to F's ys and beta's of; it is unclear what the alpha reading was. The A reading agrees with beta. my galle . I myȝte nauȝt ete many ȝeres  as a man ouȝte . For enuye and euel wille  is yuel to defye . May no sucre ne swete thynge  aswage my swelynge . Ne no diapenidion  driue it fro myn herte . Ne noyther schrifte no ne schame  but ho -so schrape my mawe . Ȝis redily q uod repentaunce  and radde hym to þe beste . Sorwe of synnes  is sauac iou n of soules . I am euer e nil R.5.129: Beta omits euere. This omission is also found in the A version, but the C version agrees with alpha and includes the qualifier. sory q uod þat segg e  I am but selde other . And þat maketh me þus megre  for I ne may me venge . Amonges burgeys haue I be  dwellynge atte londou n . And gert bagbytyng be a brokoure  to blame me nnes ware . Whan he solde and I nauȝt  þanne was I aredy redy R.5.133: Beta reads redy, and F has ful redy. Some C copies agree with beta, but Cx agrees with R's form, aredy. . To lye and to loure on my neyȝbore  and to lakken his ware chaffare R.5.134: R's ware is the alpha reading; cf. beta's chaffare. . I wil amende þis if I may  þorȝ miȝt of god al -miȝti . Now waketh wrothe awaketh wratthe R.5.136: R's waketh is unique (most of the other witnesses have awaketh). Likewise, R's spelling of the following noun (= wrothe, but rendered as Wraþe or wratthe by most of the others) is unique among the B copies—cf. the same spelling at R5.138 (at which point F and the X family of C concur with R's form). According to OED2, s. v. wrath, and MED, s. v. wroth, this spelling is a late adaptation from the adjective wroth, = "angry." Nominal usage is also found in a manuscript of Gower's Confessio (at 3.217) and in the Trinity manuscript of the A-version (at 5.66).  with to white eyȝes . And nyuelyng with þe nose  and his nekke hangynge . I am wrothe wrath R.5.138: R's wrothe is a relatively uncommon spelling for this word (cf. R5.136 above), but it is also attested in F and among the X family of C. Beta and the P family of C witness the more usual spellings (e.g, W's wraþe). quatz he  I was su mtyme a frere . And þe couent couentes R.5.139: R's uninflected form is unique; F and beta read a normal genitive, couentes. On the other hand, the R scribe may have taken the phrase couent gardiner e as a compound noun. gardiner e  for to graffe ympes . On lymitoures and listres  lesynges I ymped . Til þei bere leues  of lowe speche  lordes to plese . And sitthe þei blosmed blosmed obrode R.5.142: The beta phrase, which has the advantage of alliterating properly, is blosmed obrode .  in boure to here schriftes . And now is falle þere -offe a fruyt  þat folk haue wel leuer e . Schewen her e schriftes til to hem  þan schriuen hem til to R.5.144: Both of R's uses of til in this line are unique; F and beta read to in the first instance. F revises the b-verse substantially (so as to be unsuitable for comparison), but beta again deploys to. her e p ersones . And now haue p ersones ap erceyued parceyued  þat freres parte with hem . Þes possessioneres p rechen  and dep rauen freres . And fynden freres fyndeth R.5.147: The agreement of RF in omitting the first stave of this line ( freres in beta) indicates that the error derives from alpha. hem in defaute  as folke bereth witnesse . Þat whanne þei p reche þe poeple  in many places aboute . I wrathe walke with hym hem R.5.149: R's hym is unique and obviously an error; F and beta read the plural hem, which agrees with all of the surrounding context, including another pronoun reference later in this same line.  and wisse hem of my bokes . Þus þei speken of my spiritualte  þat eyther despi.seth other . Til þei be bothe beggeres  and by by my spiritualte R.5.151: Beta reads by my spiritualte . F has by almesse. libben . R.5.151: There is a black, vertical line from this point to R5.160 in the right margin. Or elles alle riche  and riden aboute I wrathe reste neu ere  þat Ine I ne mot folwe . Þis wikked folke  for swich is my grace . I haue anaunte an aunte to nonne  and an abbesse abbesse bothe R.5.155: Beta adds bothe at the end of this line. The C version, however, agrees with alpha in omitting it. . Hire were leu ere swowe or swelte  þan suffre any peyne . I haue be cook in hire kychyne  and þe couent serued . Many monthes with hem  and with monkes alse bothe . R.5.158: R's alse is unique; the other B manuscripts read bothe. The C reading agrees with that of the B majority. I was þe prioresses potager e  and other pouer ladyes . And made hem ioutes of iangelyng  þat dame ione was a bastard . And dame claris a kniȝtes douȝter  ac a cokewolde was hir e sire . And dame peronel a prestes fyle  prioresse worthe heo neu ere . For heo hadde childe in chirityme  al our e chapitere it wiste . Of wikked wordes I wrathe  here wortes I -made R.5.164: Only RLOC 2 have the metrically necessary dissyllabic form from OE gemacian. F has a recomposed line, and other B manuscripts have made. . Til þow lixt and þow lixt e  lopen oute at ones . And eyther hitte other  vnder þe cheke . Hadde þei had knyues by crist  her e eyther hadde kulled other e . Gregorius Seynt gregorie was a goed pope  and hadde a goed forwitte R.5.168: In this verse paragraph and the next, the scribal hand becomes noticeably smaller, and yet the 36-line ruling is unchanged from the previous leaf. Þat no prioresse were prest  for þat he ordeyned . Þei hadden þanne be infamis þe firste day  þei cu nne so euel hele conseyle . Amonges monkes I miȝt be  and many tyme I schonie . For þere ben many felle frekes  my feres to aspie . Bothe priour and suppriour  and oure pater abbas . And ȝif I telle any tales  þei taken hem to -gyderes . And do me faste fridayes  to brede and to water . And ȝeet am am R.5.176: R's ȝeet is a unique addition to this line, as witnessed in the other B manuscripts. However, it is also clearly attested in the C version. chalenged  in þe chapitelhous . As I achild a child were R.5.177: R's line division here is unique and obviously an error; F, beta, and the C version read this phrase as the end of the preceding line . And baleysed on þe bare hers bak ers R.5.178: R's cancelled reading, hers, is the Bx original. F agrees with R's "corrected" and euphemized reading, bak, but the C version agrees with the Bx original.  and no breche by -twene . For -thi haue I no likyng  with þo ledes to wonye . I ete þere vnthende fissh  and feble R.5.180: The final <e> of feble is blotted. ale drinke . Ac other -while whan wyn cometh  whan I drynke wel wyn R.5.181: Beta reads wyn. at euen . I haue a flix of a foul mouth  wel fyue dayes after . Alle þe wikkednesse þat I wote  by any of oure bretheren . I couthe it in oure cloystre  þat alle R.5.184: R uniquely omits a determiner after alle. A majority of beta copies, and F, read þe here while LMCrW attest owre. Cx agrees with F and the beta majority. couent owre couent wot it . Now repente þe q uod repentance  and reherce þow neu ere . Conseill e þat þow knoweste  by contenance ne by speche riȝte R.5.186: R and F agree with the C version in attesting speche as this line's final stave. By contrast, beta reads riȝte at this point. . And drink nauȝt ouerdelicatly  ne to depe neyther . Þat þi wille be cause þere -offe  to wrathe miȝt turne . Esto sobrius  he seyde  and so he nil R.5.189: R's so he is a unique addition to the text witnessed by both Bx and Cx. assoyled me after . And badde me wilne to wepe  my wikkedenesse to amende . descripcio avaritie And þanne come coueytyse  I can hym nauȝt descriue . So hungrilyche and holewe  sire henry hym lokede . He was bittel -browed  and baber -lipped alse . With to blered eyȝes  as a blynde hagge . And as a letheren purce  lolled his chekes . Wel sydder þanne his chyn  þei chyueld for elde . And as a bonde -man of his bakun  his berd was bydraueled . With his an R.5.198: For alpha's his, beta reads an. The C reading agrees with alpha. hode on his hed  a lousy hatt aboue . And in a tawne tabbarde  of twelue wynter age . Al to -torne and baudy  and ful of lys crepynge . But if a lous coude  lepe haue lopen þe bett ere . R.5.201: All the B manuscripts are corrupt, presumably losing most of the original b-verse: But if (þat) a lous couthe (haue lopen / lepen) þe bettre . RF omit þat and avoid the perfect tense. F's reading for this line is unique in other ways as well. The A-version reading for this line's second half is uncertain, with considerable variation between witnesses. Kane chose I may it nouȝt leue. The C b-verse has the appearance of a feeble patch rather than a common original: y leue and y trowe. He She ne nil schulde nouȝt walke haue walked R.5.202: Cf. beta's She sholde nouȝte haue walked. Evidence from the A and C versions suggests that archetypal B was already misreading the first verb in this line (= wandre in A and C). Though most C manuscripts attest the line's opening as He sholde, manuscripts X and P 2 here agree with R's version of the opening phrase, He ne schulde. Among the A copies, the same pattern is apparent, with most opting for some form of he shulde but with ChRaU paralleling R's double-negative syntax. on þat welsch  so was it thredebar e . I haue be coueytouse q uod þis caytyf  I be -knowe it here . For su m -tyme I seruede  symme at þe style . And was his prentis I -pliȝt  his p rofit to wayte . R.5.205: R omits his customary blank line between strophes at the juncture of ll. 205-06, presumably because the latter is to fill the last line ruled for this side. First I lerned to lye  a leef or other tweyne . wikedlyche to weye . R.5.206:These catchwords are partially cropped. Wikkedliche to weye  was my furst lessou n . To wy and to wynchestre  I wente to þe feyre . With many maner marchandise  as my mayster me hiȝte . Ne hadde þe grace of gyle  I -go amonge my ware chaffare . R.5.210: The alpha variant ware is supported by LM, but most beta copies read chaffare. However, as is often the case with such splits, both Ax and Cx support the LMRF variant. It hadde be vnsold þis seuen ȝer e  so me god helpe . Þanne drow I me amonges draperes  my donet to lerne . To drawe þe lyser a -longe  þe lenger it semed . Amonge þe riche rayeres rayes R.5.214: R's rayeres, "a maker or seller of striped cloth," is a unique variant among B witnesses; Bx reads rayes. The same variant occurs in manuscript Uc of the C version, but both Ax and Cx clearly attest the same word here as the B majority. For other citations of this R form, see MED, s. v. raier.  I rendred a lessou n . To brochen hem with a batnedel paknedle R.5.215: Most beta manuscripts read paknedle, but L (and perhaps M originally, which has been corrected to pak by erasure and writeover) supports alpha's batnedel. The majority of A witnesses agrees with beta, but manuscripts AE agree with alpha's lection. Batnedel is also the reading of the best C manuscripts (though most of the P family copies agree with the common beta reading).  and playted hem to -gyderes . And putte hem in a presse  and pyned pynned hem þere -Inne . Til ten ȝerdes or twelue  tolled hadde tolled oute threttene . My wif was a webbe  and wolene cloth made . Heo She spak to a spinnester spynnesteres R.5.219: Beta reads spynnesteres, which is also the reading of Cx. The Ax reading is uncertain since the singular and plural forms are both well attested among extant copies.  to spynne n it oute . Ac þe pou nd þat heo payed by  peysed a quarter quarteroun R.5.220: Beta reads quarteroun, but Ax and Cx confirm alpha's lection. more . N ota Þan myn owenauncer owen auncer  ho -so weyȝed trewthe . I bouȝte hire barly barly malte R.5.222: Beta and F read barly malte, but both Ax and Cx confirm R's reading.  heo brewe it to selle . Peny -ale and puddynge -ale  heo poured to -gyderes . For laboreres and for low folke  þat lay bi hym -selue . Þe best of alle ale R.5.225: R's of alle is unique; F and beta read ale. Among the C manuscripts, the P family omits this lection entirely (as does the cognate line in A), while the X family agrees with the reading of F and beta. lay in my boure  or in my bedde -chaumbre . And ho -so bu mmeth bummed R.5.226: R is the only B manuscript to render this verb in the present tense (but see the Introduction III.2.2.10 on R's—and alpha's—possibly ambiguous tense marking); the others read bummed. Both Ax and Cx agree with the majority B reading. þere -offe  he bouȝte it þere -after . A galoun for a grote  god wote no lesse . And ȝet it com in coppe -mele  þis crafte my wif vseth vsed . R.5.228: The present-tense marking represents alpha's reading (but cf. see the Introduction III.2.2.10 on R's—and alpha's—possibly ambiguous tense marking); cf. beta's vsed. Both Ax and Cx agree with beta. Rose þe regrater  is was R.5.229: The present-tense marking represents alpha's reading; cf. beta's was. Both Ax and Cx agree with beta. hire riȝte name  . Heo She hath I -holde holden R.5.230: Cf. beta's holden. A and C witnesses show a mixture of verb forms here, but the P family of C agrees with alpha's form. hokkarie  alle hire lif -tyme . Ac I swere now so theich the ich  þat synne wil I lete . And neu ere wikkedlich weye  ne wikked chaffare vse . But wenden to walsyngh am  and my wif alse . And bidde þe rode of bromeholme  brynge me oute of dette . Repentedest þow euere q uod repentau nce  ne R.5.235: Among the B manuscripts, only LM support R's ne; most beta copies read or and F has &. However, Cx clearly agrees with the LMR reading. restitucion madest . Ȝus ones I was herberwed q uod he  with an hepe of chapmen . I ros whan þei were a -reste  and I -rifled R.5.237: LR alone have unmetrical I -rifled . Other B manuscripts have riflede. here males . Þat was no restituc iou n q uod repentance  but a robberes thefte . Þow haddest bett ere R.5.239: R shares with LM alone the omission of be in the phrase, be bettere. Their reading is, however, likely to be the original. M later was "corrected" to the majority reading. worthi  be hanged þere -fore . Þan for alle þat  þat þow hast here schewed . I wende riflynge wer e restituc iou n q uod heo he  for I lerned neu ere rede on boke . And Ican I can no french in feyth  but of þe ferþest ende of norfolke . Vsedestow eu ere vsurie q uod repentance  in alle þi lif -tyme . Nay sothly heo he seyde  saue in my ȝouthe . I lerned amonges lumbardes  a lessou n and of iewes and iewes a lessoun R.5.245: Cf. beta's and iewes a lessoun. F reads a lessoun be herte . The C reading agrees exactly with R's. . To weye pans with a peys  and pare þe heuiest . And lene it for loue of þe cros  to legge a wedde and lese it . Swiche dedes I dede write  if he his day broke . Ich haue mo maneres þorȝ regages rerages R.5.249: Beta has rerages; alpha's reading, regages, is unrecorded in both OED2 and MED s. v. rerage, and arrearage, and is presumably nonsense generated by the misreading of a single graph, an anglicana r.  þan þorȝ miseret ur & comodat . I haue I -lente lent lordes  and ladies my chaffare . And ben here brokour after  and brouȝt R.5.251: R shares this error ( brouȝt for bouȝte) by convergence with Cot alone. it my -selue . Exchaunges and cheuysaunces  with suche chaffares chaffare R.5.252: R uniquely deploys the plural form here. I dele . And lenefolke lene folke þat lese wole  a lippe at eueri noble . And with lumbardes l ettres  I ladde golde to rome . And toke it be taille þere here R.5.255: Cf. R's þere with beta's here; F omits the adverb entirely.  and tolde hem ther e lasse . Lenedest þow Lentestow eu ere lordes  for loue of here mayntenance . A design is scratched in drypoint in the left margin beside these lines; its shape is that of three pillars of approximately the same length, a vertical pillar with two supporting pillars on its left side; the higher supporting pillar is a flat horizontal, joining the vertical midway along its length, the lower one running diagonally upwards to join the other two at the same point. The effect is almost that of a reversed capital <K>. Ȝe I haue lent lordes q uod heo  loued loued R.5.257: The attributive phrase, quod heo, is an alpha reading not witnessed in any beta manuscript; cf. F's quod he. me neu ere after . And haue I -mad many kniȝt  bothe m ercer and draper . Þat payed neu ere for his prentishode  nauȝt a peyre gloues . Hastow pyte on pore men  þat mote nedes borwe . I haue as muche pyte of pore me n  as þe nil R.5.261: R's determiner, þe, is an alpha variant; it is not present in beta manuscripts. pedler e hath of cattes . Þat wolde kulle he m and yf R.5.262: and, "if." Beta reads yf. he cacche he m miȝte  for coueytise of her e skynnes . R.5.262: At the bottom center margin of 17v, there is a drypoint figure almost identical to the one noted at R5.256 for the left margin. Now, however, the <K> figure faces downwards, and the top is trapezoidal rather than a simple rectangular pillar. Art þow manliche amonge þi neyȝbores  of þi mete and drinke . I am holden q uod he as hende  as hound in his nil R.5.264: R's in his kychyne is a metathesis of beta's presumably original text (= is in kychyne). Cf. F's in þe kycchene. kychyne . Amonges my neyȝbores nameliche  suche a name Ich haue . Now god lene neu ere q uod repentaunce  but þow repent þe rather . Þe grace on þis grounde  þi good wel to bi -sette . Ne þin vssue R.5.268: R reads vssue, agreeing with L alone (= ysue); M has been erased and overwritten to match the other beta manuscripts' reading, heires. F reads houswif. after þe  haue ioye of þat þow wynneste . Ne þi seketoures excecutours R.5.269: RF's seketoures is an aphetic form of beta's excecutours. wel bi -sette  þe seluer þat þow hem leuest And þat was wonne with wronge  with wykked men be despe nded For were I frer e of þat hows  þer e good faith and charite is . I wolde nouȝt nolde cope vs with þi catel  ne our e cherche kyrke R.5.272: Two alpha variants are attested here, the second inadequate for the line's alliterative pattern; cf. RF's wolde nouȝt and cherche with beta's nolde and kyrke. amende Ne haue a peny to my pitance of þine  be my soule hele . For þe beste boke in our e hows  þouȝ brend gold wer e þe leues . And I wiste witt erly  þow wer e swiche as þow telleste . Seruus es alterius cu m fercula pinguia queris . Pane tuo pocius  vescere liber eris . Þow art an vnkende creatur e  I kan þe nauȝt assoile . Til þow make restitucion q uod repentance nil R.5.279: Only R records quod repentance. The other B witnesses, including F, are here content with a line having only two alliterating staves. The B manuscript that Langland used in creating C appears to have shared this faulty alliteration since the alliterative key is there shifted from /r/ to /m/ in order to make use of make (= B) / ymad (= C) in the first stave position.  and rekene with he m alle . And sitthen þat resou n rolle it  in þe registre of heuene . N ota Þat þow hast made vch man goed  I may þe nauȝt saue assoille . R.5.281: R's saue is unique. The other manuscripts read assoille. Non dimittit ur peccatu m nisi donec R.5.282: The predominant beta variant here is donec, but Cr agrees with alpha. restituat ur ablatu m ablatum &c . For alle þat hath of þi good  haue god my trewthe . Is haldynge holde(n) R.5.284: Only LMF agree with R's Is (but F's verb occurs in a completely rewritten line). The other copies show be(n). R's haldynge is unique; the other B copies all show holde(n) as do many P manuscripts in the C tradition. However, the X family of C agrees with R's lection. at þe heyȝ dome  to helpe þe restitue to restitue R.5.284: With reference to beta, R's lection here appears at first glance to involve a unique omission. Beta reads þe to restitue . However, the supposition of omission collapses when we notice that the nearly unanimous C reading at this point is identical to R's (F rewrites the line completely). . And ho -so leueth nauȝt þis be soth  loke in a þe R.5.285: R's lection is unique; the other manuscripts read þe, as does the C version. saut er glose . In miserer e mei deus  where I mene trewthe . Ecce enim veritatem dilexisti &c etera . Þer e is no laborer e wolde þat wil leue with he m þe  þ at knoweth longiþ to peres þe nil plowma n . R.5.288: Kane-Donaldson dismiss this alpha line as spurious because of its reference to Piers Plowman, who has not yet been introduced into the narrative. For nil R.5.289: For is an alpha variant; beta omits it. schal neu ere werkman in þis world e  thriue with þ at þow wynnest . Cu m s anc to sanctus eris  construe me þat on englisch e . Þanne wex þe þat schrewe in wanhope  and walde haue hanged hy m -selue . Ne hadde repentance þe rather e  co nforted R.5.292: Although Hm and G agree with R (presumably by convergence), beta itself had a compound of this verb, needed for alliteration: reconforted; F offers, in a rewritten b-verse, reersyd, which looks like an attempted repair. hy m in þis maner e . d ij Haue mercy in þi mende  and with þi mouthe byseche it . For godes mercy is more  þan alle his other werkes . Misericordia d omini eius R.5.295: This is a unique variant in R; the beta copies that contain this citation read eius, the accurate Vulgate form. Approximately half the beta copies and F omit the entire citation. super o mnia opera eius . &c etera . And alle þe wikkednesse in þis world  þ at man miȝt worche or þenke Ne is namore to þe mercy of god  þan in þe see a glede O mnis iniquitas qu antu m ad m isericordiam dei  est quasi sintilla in medio maris . For -thi haue mercy in thi mynde  and marchandise leue it . For þow hast no goed grounde  to gete þe with a wastel . But if it wer e with þi tonge  or elles with þi to handes . For þe goed þat þow haste gete  bygan alle with falshede . And as longe as þow lyuest þer e -with  þow ȝeldest nouȝt but borwest . And if þow wite neu ere to whiche  ne wham to restitute . Bere it to þe bischopp e  and bydde hym of his grace . By -sette it hym -selue  as best is for thi soule . For he schal answere for þe  at þe heyȝ dome . For þe and for many mo  þat man schal ȝiue a rekenynge . What he lerned ȝow in lente  leue þow non other . And what he lente ȝow of our e lordes goed  to lette ȝow fro synne . Now bygynneth glotou n  for to go to schrifte And cayres hym to cherchewarde kirkeward R.5.312: As is frequently the case, R's unique reading here is defective in alliteration (cf. Bx's kirke-ward).  his coupe to schewe . Ac betou n þe brewester e  bad hym goed morewe . And asked of hym with þat  whyderward he wolde . To holy cherche q uod he  for to here masse . And sitthen I wil be schriue  and synne namore . I haue goed ale gossib q uod sche  glotou n wiltow assaye . Hastow auȝt in þi purce  any hote spices . R.5.318: Although the scribe customarily enters a blank line between paragraphs, following this line he seems to have forgotten to do so because of the heavy prevalence of paraph markers to indicate dialogue. I haue peper and pioyne piones R.5.319: R's singular is unique among the B manuscripts but agrees with the reading found in Ax and in the X family of C. The P family of C shows the same plural as the B majority. q uod sche  and a pound of garlek e . A ferthyngworth of fenel seed  for fastynge -dayes . Þanne goth glotou n Inne  and grete othes after . Sesse þe souteres  sate on þe benche . Watte þe warner  and his wif bothe . Symme R.5.324: R's non-alliterating Symme is shared with HmYCBoCot among the B manuscripts but also agrees, oddly, with the reading found in A manuscripts TRaDH 2H and with the reading of C manuscript P 2. Presumably the cause of these overlapping errors is the mutual resemblance of the relevant capitals. þe tynker e  and tweyne of his p rentys . Hikke þe hakeney -man  and howe þe nedeler e . Clares of cokkeslane  and þe clerk e of þe cherche . And nil R.5.327: And is an alpha variant; beta begins the line with Sire. The C version of this line agrees exactly with beta. sire peres of pridie  and p eronele of flaundres Dawe þe dyker e  and a doseyne other . A ribybour e a ratoner e  a rakyer of chepe . A roper a redyng -kyng e  and rose þe dissheres douȝter nil . R.5.330: The phrase dissheres douȝter is an alpha variant; cf. beta's disshere(s). Both the A version and the C version agree with beta on this reading. Godefrey of garlek -hethe  and grifyth gryfin R.5.331: F reads Geffrey while beta has gryfin. Both the A version (some copies omit the line completely) and the P family of the C version agree with beta's name form here (P family = griffyng), but the X family of C agrees with R's otherwise unique form. þe walsh e . And vpholders an hepe  herly by þe morwe . Geue glotou n with gladde chere  goed ale to hansall e . Clement þe coblere  cast of his cloke . And to atte R.5.335: Beta reads atte; F has in. The A version reading agrees with beta while the C version agrees with R's to. þe newfeyre new feyre new faire he  nempned R.5.335: F and beta both read he nempned ; however, both the A archetype and the unanimous C reading agree with R in omitting he. it to selle . Hikke þe hakeneyman  hit his hood after . And badde bette þe bocher e to nil R.5.337: The verbal particle to is an alpha variant completely absent from the beta manuscripts. It does, however, appear in a few A manuscripts (RaUDJ) and in approximately half of the C manuscripts. It may, in fact, be the family reading of the P group. ben on his side . Þere Þere were(n) R.5.338: R uniquely omits the Bx verb were at this point, treating the participle Ichose at the end of the a-verse as the verb. chapmen Ichose  þis chaffare to preyse . N ota Ho -so haueth þe hood  schul haue amendes of þe cloke . Þo R.5.340:R and Bm are the only witnesses in the B tradition beginning this line with Þo (= not the common adverb but rather the plural demonstrative pronoun, equivalent to Modern English, "Those." See Kane, Glossary, pp. 207-8). The other beta copies all attest Two while F begins the line with Þan. The P family of C is joined by manuscripts Uc and Dc in support of the beta reading, but manuscripts X, I and P2 (probably reflecting the X subarchetype) agree with R and Bm in reading Þo. The A archetype also supports the RBm variant. Presumably R attests this reading by descent from alpha while Bm has it through correction, perhaps from the C exemplar used for its opening passus. Schmidt is probably correct in hypothesizing (II, 353) that Bx actually spelled Þo in this line as To, thus inviting the common subsequent scribal error of Two. risen risen vp in R.5.340: R uniquely omits a word from this phrase. Beta reads risen vp in while F has ryse þey ful. The reading of beta is also found in many A manuscripts and is clearly archetypal in that tradition; among the C witnesses, the P family mostly agrees with R (omitting vp from the phrase) while the X family attests its presence. However, many C manuscripts in both families omit in. rape  and rowned to -gyderes . And preysed þe þese R.5.341: Beta reads þese in place of alpha's þe. The A reading agrees with alpha, but the C version is split by families, with the P family supporting alpha while the X family agrees with beta. penyworthes  a -part by hym -selue hemselue(n) . Þei coude nauȝt by her e consience  acorden in trewthe . Til robyn þe ropere  arise þe arose bi þe southe . R.5.343: R may well represent Bx here, but the b-verse is plainly wrong. It is obvious that the F redactor has completely recast the b-verse because of its archtypally defective sense; in beta, the b-verse is similarly lacking, reading arose bi þe southe. Both Kane-Donaldson and Schmidt emend this verse conjecturally to conform to the C version: aryse they bisouhte . And nempned hym for a noumper  þat no debate were R.5.344: Beta and F read nere in place of R's were, but both G and Hm agree with R (presumably by correction). The A witnesses are split, but a majority agrees with R; the C version reading is unambiguously the same as R's. . Hikke þe hosteler e  hadde þe cloke . In the right margin, a note was written horizontally here, covering three lines, but was later scraped. It appears to be in the same hand (later than that of the scribe) as a number of other erased notes, such as that found on fol. 94r. In couenaunt þat clement  schulde þe cuppe fille . And haue hikkes hood hostelere  and holde hym I -s erued . And ho -so repented rathest  schuld arise after . And grete sire glotou n  with a galoun ale . Þer e was lawhyng e and lowryng e  and late go þe cuppe . And sitten so til euesonge  and syngen songe(n) R.5.351: The present tense, syngen, is unique to R; cf. F's sunge and beta's songen. R also shows a present-tense form earlier in this line, sitten, where most B witnesses record a past-tense form (but Cr and G agree with R). Context alone would suggest that the past tenses are correct, but the unanimous witness of C manuscripts as well as the attestation of a clear majority of A manuscripts in favor of these forms decides the question. Cf. see the Introduction III.2.2.10 for a discussion of R's problematic tense marking. vmb -while . Til glotou n haued I -globbed  a galoun and a gille . His guttes gonne to godly R.5.353: R's godly is shared exclusively with LM. As odd as it first seems, this was almost certainly the form of the verb in Bx. F has gowle while most beta manuscripts read goþelen. The same phrase occurs in the C version, where it reads His gottes gan to gothly. Significantly, a sizeable number of C witnesses agree exactly with manuscripts LMR of B on the spelling of the verb form, and MED, s. v. gothelen, acknowledges both -dly and -þly forms as viable for gothelen, but citations are solely to Piers Plowman. The same limited acknowledgment of these forms is found in OED2, s. v. gothele, and godele(n), -y.  as to gnedy g[r]edy gredy sowes . He pyssed a potel  in a pater noster while . He And R.5.355: He is a unique variant; the other B manuscripts read And. However, R's reading is also that of four A manuscripts and of the X family of the C version. Both Kane-Donaldson and Schmidt prefer the F/beta reading, presumably on stylistic grounds since And avoids a syntactic repetition (the previous line begins with He) that modern tastes find clumsy. blew his round rowet  at his rigges bones R.5.355: Most manuscripts have the compound riggebone, but both rigges and bones are genitives. Manuscript C has the same reading. ende . Þat alle þat alle [þat herde] þat alle þat herde þat R.5.356: Here R uniquely omits a key phrase from the archetypal text. horne  held here nose after . And wischedun it hadde be wexed  with a wips of firses . He miȝte neyther steppe ne stonde  er e he his staf hadde . And þanne gan he go  liche a glwemannes bicche . Sum -tyme a -syde  and sum -tyme arere . As ho -so leith lynes  for to lacche foules . Ac And R.5.362: Ac is unique to R; the other manuscripts read And. The A archetype seems to omit any connective here, but the C version agrees with F / beta. whan he drouȝ to þe dore  þanne dymmed hys eyȝes . He tremled R.5.363: R's tremled is uniquely supported by L (M having been altered once more to conform to the typical beta reading—for which, see below); F reads tripplid while most betas offer the non-alliterating stombled. Among the A manuscripts, only Vernon offers a viable reading, þrompelde, while the others attest various forms of stombled. As for the C version, though a few opt for stumblet, the majority attest thromblede. Based on its alliterative pattern, this is presumably what Langland wrote originally, but the LR form is almost certainly what it became in Bx. on þe threswolde  and threw to þe erthe . Clement þe cobbler e  cauȝt hym by þe myddel . For to lifte hym a -lofte  and leyde hym on his knowes . Ac glotou n was a grete cherle  and a grym in þe liftyng . And cowede vppe a caudel  in clementes lappe . Is non so hungri hound  in hertford e -schire . Durst lape of þe leuyng leuynges R.5.369: Beta shows the plural form, leuynges, while F rewrites the a-verse beyond recognition. The C version agrees here with R's singular.  so vnlouely it þei smauȝte . With alle þe wo of þis world  his wif and his wenche . Baren hym to home to R.5.371: Beta reads home to; although three A copies (VHaN) agree with beta here, both Ax and Cx concur with alpha's omission of hom. his bed  and brouȝte hym þere -inne . And after alle þis excesse  he hadde an accidie . Þat he slepe seterday and soneday  til sonne ȝede to rest . Þanne waked he of his wynkyng  and wiped his eyȝes . Þe furst word þat he warpp e  was wher e is þe bolle . His witt gan edewyte hym þo  how wikkedliche he leuede . And repentance riȝt so  rebuked hym þat tyme . As þow with wordes and werkes  hast wrouȝt euele in þi lyue . Schriue þe and be aschamed þere -offe  and schewe it with þi mouthe . I glotou n q uod þe gome  gulty me ȝelde . Of nil þ at R.5.381: R's Of is unique; the other B witnesses begin the line with Þat. However, the C reading here is identical to R's. I haue trespased with my tonge  I can nauȝt telle how ofte . Sworen godes soule and his sydes nil R.5.382: Beta omits the entire R phrase for the end of the a-verse ( and his sydes); F reads an abbreviated version, & side. The C reading is probably that of the X family (which agrees exactly with R's). The P family reading ( & sides) agrees with F's omission of the possessive but with R's plural number.  and so help me god & holy -dome god me help and halidom R.5.382: Beta appears to reverse a key alpha phrase (the latter being more colloquial), rendering alpha's so help me god as so god me help. The C reading for the end of this b-verse is revised but its opening agrees exactly with alpha's phrasing, so helpe me god. Beta's motive for revision was probably metrical, but manuscript M somehow still agrees with alpha and displays a form that may explain both Langland's intention and the apparent lapse in alliteration. Alpha and M read the line as aliterating on /s/ (hence M's selpe) but beta judges that it must alliterate on /g/ and generates the aformentioned phrase reversal to highlight that possibility. . Þere no nede ne was  nyne hundreth tymes . And ou er -seye me at my soper  and sum -tymes sometyme R.5.384: R's -s form here is unique. The other B copies show some -tyme , a reading shared with C. at nones . Þat I glotou n gurt it vppe  ar I hadde gon a myle . And I -spilt þat myȝt be spared  and spend spended R.5.386: R's spend is an alpha variant; cf. F's spend it and beta's spended. on sum hungry . Ou er delicatly R.5.387: There is a gap in the parchment between de and licatly caused by a pre-existing tear; the same gap occurs on the other side at R5.417. on fastyng -dayes  dronken and eten bothe . And sat sum -tyme so longe þer e and þat I slep R.5.388: R's and slep is unique; beta appears to reflect Bx accurately with its þat I slepe . Cf. F's þere -at y sleep . and eet at ones . For loue of tales in tau ernes  to ete drynke R.5.389: R's ete is unique and redundant with the preceding line; cf. Bx's drynke. þe more I dyned . And hied to þe mete er none  when fastyng -dayes were . Þis schewyng schrifte  q uod repentance  schal be merite to þe . And þanne gan glotou n to grete  and muche grete R.5.392: R's muche is unique and may have been adopted to avoid the apparent repetition of the archetype, which reads grete here. doel to make . For his lither lif  þat he leued hadde . And auowed R.5.394: Most beta manuscripts read to fast , as does F. GLM agree with R's omission of to. M's corrector adds to above the line. On the other hand, like the beta majority, the A version attests the presence of to in this phrase. faste  for hunger or for thurst . Schal neu ere fisch e on þe friday  defien in my wombe . Til abstinence myn aunte  hath haue R.5.396: R's hath is the alpha form (shared exclusively with F); beta copies show haue, a reading shared with both Ax and Cx. Iȝeue ȝiue(n) me leue . And ȝet haue I hated hire  al my lif -tyme . Þanne come sleuthe al by -slobred  with to slymed slymy R.5.398: Beta reads slymy. A significant majority of C witnesses (XIP 2PRcVcAcQScKcGc) agrees with the alpha reading. eyȝes . I most sitte seyde þe seg e R.5.399: R's seg e is an unusual spelling of an old-fashioned word (= "man"). More commonly, the word appears in Middle English as segge, but this form is cited by MED, s. v. segge, in one manuscript of the Awntyrs of Arthur, Lincoln Cathedral Library 91: A knyghte carelyone be crownede for kynge; That sege salle be sesede at a sesone (289). The spelling is likewise attested in OED2, s. v. segge, for the fourteenth through the sixteenth century, and a passage from Dunbar is cited.  or elles schulde I nappe R.5.399: There is a superfluous bar over the final <-pe> of nappe. . I may nouȝt stonde ne stoupe  ne with -oute stole a stole knele . Were I brouȝt a -bedde  but if my taylende it made . Schulde no ryngyng do me rise  ere I were ripe to dyne . He bygan b en edicite with a bolke  and his brest knokked . And roxede and rored  and rutte atte laste . What awake renk q uod repentance  and rape þe to schrifte . If I schulde deye bi þis daye  q uod he nil R.5.406: Beta omits quod he. The P family of C manuscripts agrees with the alpha reading, but the X family agrees with beta and omits the phrase.  me lest nouȝt to loke . I can nouȝt p arfitlyche my pat ern oster  as þe prest it syngeth . But I kan rymes of robyn hood  and randolf erl of chestre Ac noþer of our e lor R.5.409: R's clipped form for lord is unique among the B manuscripts both here and in R18.61. See Richard Jordan, Handbook of Middle English Grammar: Phonology, translated and revised by Eugene Joseph Crook (The Hague: Mouton, 1974): §199, remark 3. ne of our e lady  þe leste þ at eu ere was maked . R.5.409: Here the scribe omits his customary line break before a new verse paragraph, presumably because the next line will be the final one of this side. I haue mad vowes fourti  and forȝete hem or on morwe R.5.410: This occurrence of or is unique; the full phrase in most B witnesses reads on (þe) morne. C has amorwen. . d iiij I parforned neu ere penance  as þe prest me hiȝte . Ne riȝt sori for my synnes  ȝet was I neuere . And ȝif I bedde any bedes  but if it be in wrathe . Þat I telle with my tonge  is to myle fro my herte . I am ocuped vche a day  halyday and other . With Idel tales at þe ale  and other -while in cherche cherches . Godes peyne and his passiou n  ful selde thenke I þer e -on . R.5.417: A repaired tear in the parchment (existing before its use by the scribe) partially obscures the loop from the <r> in þer e-on by which the following <e> is signified. I visitede neu ere seke fieble R.5.418: R's seke is a non-alliterating alpha variant; cf. beta's fieble, which is also the reading of the C version.. men  ne fetered folke in puttes . I hadde haue R.5.419: Cf. alpha's hadde to beta's haue; alpha's reading is also that of the C version. leu ere here an herlotrie  or a som er game of souteres . Or lesynges to lawhe of at R.5.420: R's of is an alpha variant; beta reads at, but Cx agrees with alpha here.  and bilye my neyȝbore . R.5.421: R uniquely omits a transitional word here in linehead position; cf. F's & and beta's Þan. The C reading agrees with beta's. Al Þan al þat eu ere marke made  mathew Ioh an and lucas . And vigilies and fastyngdayes  alle þis I late þise late I R.5.422: Beta transposes this phrase as late I. F reads the entire phrase as y leet hem passe . And ligge a -bedde in lente  and my le mman in myn armes . Til matynes and masse be do  and þan go to þe freres . Kome I to Ite missa est  I hold me I -s erued . I am nauȝt schriue su mtime  but ȝif sekenesse it make . Nouȝt twyes in to ȝere  and þanne vpe gesse I schriue me . I haue be prest and p ersone  passyng thretty wynter . Ȝet can I neyther solue ne synge  ne seyntes lyues rede . But I can fynde in a felde  or in a furlonge an hare . Better þan in beatus vir  or in beati om nes . Construe it clausemel oon clause wel R.5.432: R's it clausemel is an alpha variant; cf. beta's oon clause wel .  and kenne it to my parochiens . I can holde louedayes  or and R.5.433: Beta reads and. here a reues rekkenynge . Ac in canon ne in þe decretales  I can nouȝt rede a lyne . Ȝif I begge and borwe it  but if it be I -tayled . I for -ȝete it as ȝerne  and ȝif men me it aske . Sixe sithes or seuene  I forsake it wit with R.5.437: R's spelling of with here is unique among the Piers Plowman manuscripts and also fairly unusual in the Middle English corpus. The same spelling is found extensively in Cursor Mundi; it also occurs several times in the Prose Alexander (ca. 1440), once in Sir Gawain and the Green Knight, line 113 (where Andrew and Waldron emend it away), and once in Havelok the Dane, line 2489. othes . And þus I tene tene I R.5.438: Beta transposes this phrase as tene I. trewe men  te te[n] ten hundreth tymes . And my s eruantes su m -tyme  her e salarie is byhynde . Reuthe is to her e here þe rekkenynge  when we schul rede acou ntes . So with wikked wille and wrathe  my werk -men I paye . R.5.441: R omits his customary blank line between strophes at the juncture of ll. 441-42, presumably because the latter is to fill the last line ruled for this side. Ȝif anyman any man doth me a benfeet  or helpeth me at nede . I am vnkende aȝeyn his curteysie  and can nauȝt vndersta nde it . For I haue and haue hadde  sumdel haukes maneres . I am nauȝt lured with loue  but þere ligge auȝt vnder thou mbe þe thombe R.5.445: Here the beta manuscripts have a determiner, making the phrase þe thombe. This discrepancy is also apparent in the two major families of C: the P family here agrees with alpha and the X family supports beta. . Þe kendenesse þat myn euencristene  kudde me ferneȝer e . Sixty sithe sythes R.5.447: This R form is unique; both F and beta read sythes. Once more, the P family of C manuscripts here agrees with R and the X family supports beta (with F's collateral attestation). I slewthe  haue foreȝete it setthe . In speche and in sparinge of speche  I -spelt many a tyme . Bothe flesch e and fisch e  and myn many R.5.449: Cf. R's erroneous myn to F's fele and beta's many. other vitailes Bothe brede and ale  butter e melke and chese . For -sleuthed in my s eruise  til it miȝte s erue noman . I ran aboute in ȝouthe  and ȝaf me nouȝt to lerne . And eu ere sitthe be beggere  for my foule sleuthe . Heu michi q uod quia R.5.454: This alpha variant is supported by BoCot, but beta reads quia. The C reading at this point agrees with alpha's. sterilem  vitam duxi iuuenilem . Repentest Repentestow þe nil nouȝt q uod repentance  and riȝt e with þat he swhounede . Til vigilate þe veile  fette water at his eyȝes . And flatte it on his face  and on faste on R.5.457: R uniquely omits the third stave; the other B manuscripts read and faste on . hym cried . And seyde war þe fram wanhope  wolde þe bytraye . Ich am sori for my synnes  sey so R.5.459: Although most beta copies omit so and F reads now so, LM support R's so to. Cx shows the same reading as the beta majority. Among the A manuscripts, only EAH attest so in this phrase. to þi -selue . And bete þi -selue on þe breste  and bidde hym of grace . For is no gult here so grete  þat his goednesse is more . Þanne sate sleuthe vppe  and seyned hym swythe . And made a vow to -for god  for his foule sleuthe . Schal no soneday be þis seuen ȝere  but ȝif sekenesse me it R.5.464: R's me is unique; the other manuscripts read it. Both Ax and Cx read (h)it make . lette . Þat I ne schal do me er daye  to þe dere cherche . And heren matynes and masse  as I a monke were . Schal non ale after mete  halde me thennes . Til I haue euensonge herde  I be -hote to þe rode . What I nam And ȝete wil I R.5.469: The defective opening phrase of R's a-verse ( What I nam) may well reflect alpha accurately; cf. F's & what y have take to and beta's And ȝete wil I. ȝelde aȝen  if I so muche haue . Alle þat I wikkedliche wan  sitthen I witt hadde . And þouȝ me my R.5.471: R's me is unique among the B witnesses; the other manuscripts read my, which is also the reading of most manuscripts for the other two versions of Piers. However, manuscripts X and Yc in the C tradition, two of the most authoritative copies, agree at this point with R. lyflode lakke  leten I nelle . Þat vch man schal haue his  ar I hennes wende . And with þe residue and þe remenant  by þe rode of chestre . I schal seke treuthe  arst er I se rome . Roberd þe robbere  on reddite lokede . And for þere was nauȝt where -of  he wepe swythe sore . Ac ȝet þe synful schrewe  seyde to hym -selue . Crist þat on caluar e  vppon þe croys deyedeste . Þo bymas [d]y[s]mas dismas R.5.479: Kane-Donaldson commit a rare mistranscription here, rendering this R variant as bysmas. my brother  by -souȝte ȝow of grace . And haddest mercy on þat man  for memento sake . So rewe on þis robber e  þat redder e ne haue . Ne neu ere wene to wynne  with craft þat I knowe owe . R.5.482: For alpha's knowe, all but Cr among the beta copies read owe. However, both Ax and Cx agree with alpha's reading. But for thi michel m ercy  mitigac ion I be -seche . Ne dampne me nauȝt at domes -day  for þ at I dede so ille . What be -fel of þis felou n  I can nauȝt fare faire R.5.485: This form is unique and Kane-Donaldson treat it as a possibly substantive variant, but it is likelier to be merely an unusual spelling variation for standard faire found in the other B witnesses, as well as in Ax. Cf. MED, s. v. fair (adj.), where this spelling is documented, sporadically, for c. 13-15. schewe . Wel I wote he wepte faste  water with bothe his eyȝes . R.5.486: Immediately after this line, the alpha tradition omits a line present in beta (and also present with slight differences in Ax and Cx): And knowleched his gult to cryst ȝete eftsones. Þat penitencia his pik  he scholde polsche newe . And lepe with hym ouer londe  al his lif -tyme . For he hath had R.5.489: F omits this verb entirely and the beta copies render it in the past tense as had. Among the A and C manuscripts, only Vernon agrees with R's reading. The others support beta. Cf. the Introduction III.2.2.10 regarding R's problematic tense marking. leyne be latro  luciferes aunte . And þanne hadde repentance reuthe  and redde he m alle to knele . For I schal by -seche for alle synful  our e saueour e of grace . To amende vs of oure mysdedes  and do m ercy to vs alle . R.5.493: The rubricator failed to notice the "cc" paraph marker. Now god q uod quod he R.5.493: R agrees with the beta manuscripts about most of this phrase but uniquely omits he at this point (F completely reworks the phrase as & seid god ). þat of þi goednesse  gonne þe worlde make . And of nauȝt madest auȝte  and man most liche to þi -selue And sitthen suffredest hym for R.5.495: In place of alpha's hym, beta reads for, completely misconstruing the meaning of the preceding verb. The C reading is identical to R's form of the alpha phrase. to synne  a sekenesse to vs alle . And alle for þe best as I be -leue  what -euere þe boke telleth . O felix culpa . O nessariu m ne[ce]ssariu m necessarium R.5.497: The correcting scribe marked the line for correction with a marginal <+>. peccatu m ade ade &c . For thorȝ þat synne þi sone  sent was to þis erthe . And bi -cam man of a mayde  man -kende to saue . And madest þi -selue with þi sone  and vs synful I -liche . Faciamus ho mi nem ad ymaginem & similitudinem n ostram . Et alibi  qui manet in caritate in deo manet & deus in eo . And sitthe with þe þi R.5.503: R's þe selue sone (= "the Son Himself") is unique and probably signifies nothing more than a casual error by the scribe for Bx's þi -self sone ("Thine own Son"). selue sone  in oure seute dyedest . On godefriday for mannes sake  at fulle tyme of day þe daye . R.5.504: The other B witnesses read of þe daye . Þere þi -selue and R.5.505: Only G agrees with R here; beta and F read ne. þi sone  no sorewe in deth feledest . But in oure secte was þat þe R.5.506: Beta reads þe. sorewe  and þi sone it ladde . Captiuam duxit captiuitatem . Þe sonne for sorewe þer e -offe  les siȝte for a tyme . Aboute myddaye  whan most liȝt is  and mel -tyme of seyntes . Feddest þo Feddest R.5.510: Beta omits þo while F transposes the phrase to Þo feddyst . The C reading here is identical to R's. with þi fresch blode  oure forfadres in derkenesse . Populus qui ambulabat in tenebris vidit lucem magnam . And thorȝ þe liȝt þat lepe oute of þe  lucifer it was R.5.512: Beta reads was while F omits the word entirely. The C reading agrees with R's. blent . And blew alle þin blissed þennes nil R.5.513: Alpha's þennes (in F = þens boldely) is omitted by beta. However, in a revised version of this line, C includes it.  in -to þe blisse of paradys . Þe thridde day after  þow ȝedest in oure seute . A synful marie þe seiȝ  er seynte marie þi dame . And alle to solace synful  þow suffredest it so were . R.5.516: There is an ink blot over the final <e> of were. Non veni vocare iustos  sed peccatores ad penitenciam . And alle þat mark hath I -mad  matheu Iohan and lucas . Of þin douȝtyest douȝty dedes  were don in oure armes . Verbu m caro factu m est . & habitauit in nobis . And by so m.yche it me R.5.521: Cx agrees with alpha. semeth  þe sykerloker sikerere R.5.521: This is a unique form in R; F has sikere while beta reads sikerere; nevertheless, Cx agrees exactly with R's comparative, an odd form that appears restricted to texts from the West Midlands (as evidenced by citations from both MED, s. v. sikerli [adv. 3], and OED2, s. v. sickerly.) we mowe . Byde and bi -seche  if it be þi wille . Þat art our e fader and oure brother e  be m ercyable to vs . And haue mercy reuthe R.5.524: Cf. the beta variant, reuthe, which properly alliterates and agrees with the reading of the C version. on þes ribaudes  þat repenten hem sore here sore Þat euere þei wrathed þe þe in þis worlde R.5.525: Beta here attests an additional phrase: in þis worlde .  in worde thouȝt or dedes . Þanne hent hope an horn  of deus tu co nu ersus viuificabis no s nil . And he nil R.5.527: Before this verb, R uniquely adds he; the C version reading agrees with the B majority in omitting the pronoun. blew it with beati quor um remisse sunt iniquitates . Þat alle seyntes in heuene  songen at ones . Ho mi nes & iumenta saluabis que madmodu m multiplicasti m isericordiam tua m d eus . A thousend of men þo  throngen to -gyderes . Cride vppward to crist  and to his clene moder . Grace to god To haue grace to go R.5.532: This b-verse of this line appears to have been defective in Bx, but beta's version of the a-verse ( To haue grace to go . . .) seems more likely to be original than either F ( To graunte swich grace . . .) or R's truncated and semantically deficient Grace to god. The C version reading for the a-verse agrees with that of beta. with hem  trewthe to seke . Ac þere was wiȝt non so wis  þe weye þider coude . But blostereden forth as bestes  ouer bankes and hilles . Til late was and longe  þat þei a lede mette . Apparayled as a paynym  in pilgrimes wise . He bare a bordou n I -bounde  with a brode liste . In a wythewyndes wyse  I -wonden aboute . A bulle bolle R.5.539: Beta reads bolle which is clearly the intended meaning. Alpha's form is not documented as a variant spelling for this word (= mod. bowl. The C reading agrees with beta's. and a bagge  he bare by hys syde . An hundred of ampulles  on his hat seten . Sygnes of a -sise synay R.5.541: Cf. F's seyntys and beta's Synay. R's reading is supported by Cx and by several A-version manuscripts (including Bodley 851, i.e., Z). Most A copies agree with beta's reading.  and schelles of gales . And many cruche on his cloke  and þe nil R.5.542: The beta manuscripts do not attest þe. Both the A and C versions also omit this determiner. cayes of rome . And þe vernicle before  for men schulde I -knowe knowe . And se by his seynes R.5.544: Seynes, "signs."  wham he souȝt hadde . Þis folke frayned hym furst : fro whennes he come . Fram synay he seyde  and fram oure lordes sepulcre . In bethlem and in babiloyne  I haue ben in bothe . In ermony and in in R.5.548: R's phrase, and in, is unique. Cr includes and but omits in; both F and the majority of beta copies omit and. The Ax and Cx reading here appears to agree with the F/beta lection. alisaundre  in many other places . Ȝe may se be my sygnes  þat sitten on myn hatte . Þat I haue walked ful wyde  in wete and in drye . And souȝte gode seyntes  for my soule helth . Knowest þow auȝt a corseint  þat men calleth treuthe . Koudest þow auȝt wissen vs þe wey  þer e where R.5.553: R's þere is unique in the B tradition (LaAMaH among the A manuscripts agree with R); F and beta read where, which is also the reading of the A majority. þ at wyȝte R.5.553: R's wyȝte agrees only with Cr; F and the beta majority read wy(e), which is also the reading of the A version. dwelleth . Nay so me god helpe  seyde þe gome þanne . I seyȝ neu ere palmer e  with pik ne with scrippe . Axen after hym er e  til now in þis place . Peter q uod a plowman  and put forth his hed . I knowe hym as kendely  as clerk doth his bokes . Consience and kende witt  kenned me to his place . And deden me suren he m hym R.5.560: R's hem is an alpha reading, shared exclusively with F, and the plural reference is presumably to the two figures mentioned in the previous line. By contrast, beta deploys a third-person singular accusative, presumably understanding treuthe as the referent. The P family of C witnesses supports the beta reading, but the X subarchetype appears to have omitted the pronoun altogether. sikerly  to s erue hym for eu ere . Bothe to sowe and to sette  þe while I swynke miȝte . I haue ben his folwar  al þis fourty fifty wynter . Bothe I -sowe his sede  and sued his bestes . With -Inne and with -oute  I -wayted his profite . I dike and I delue  I do þat he treuthe R.5.565: Beta reads treuthe for alpha's he. hoteth . Sum -tyme I sowe  and sum -tyme I thresche . R.5.566: There is a superfluous bar over the final <e> of thresche. In tayloures crafte in and R.5.567: F omits this entire line, but beta reads and here. tynkeres crafte  what treuthe ca n deuise . I weue and I wynde  and do what treuthe hoteth . For þouȝ I seye it my -selue  I serue hy hy[m] hym to paye . Ich haue my huyre of hym nil R.5.570: Beta omits of hym, but both Ax and Cx witness this alpha phrase. wel  and otherwhiles more . He is þe presteste payer  þat pore men knoweth . He with -halt no men hewe(n) R.5.572: R's men is unique (cf. beta's hewe and F's man) and forces the following pronoun choice, here, shared by convergence only with Cot. At first glance, F's reading seems closer to alpha ( man his), because of the pronoun agreement with beta, but the plural forms shared by R and F later in this line (R = þei ne haue it; F = þat þey haue it) indicate that R is probably, as usual, closer to their common parent. Either way, the beta phrasing, hewe his hyre þat he ne hath it , is almost certainly the original since it preserves the line's alliteration and exactly parallels the text of Ax (and of the a-verse of Cx). her e huyre  þat þei he ne haue hath it anone at eue(n) . R.5.572: For R's anone, F has soone while beta reads at euen. Both Ax and Cx support the beta reading. He is as lowe as a lombe  and louelich of speche . And if ȝe wilneth to wite  where þat he dwelleth . I schal wise ȝow witterly  þe weye to his place . Ȝe leue peres q uod þes pilgrymes  and profered hym huyre . For to wende with hem  to treuthes dwellynge place . Nay by my soule perel soules helth R.5.578: R's perel is unique but may well represent Bx. The alliterative pattern is no more adequate than that of its only competitor, helth (= WCrGLM), but both Ax and Cx attest R's lection. q uod peres  and gan for to swerie . I nolde fonge a ferthyng  for seint Thom as schrine Treuthe wolde loue me þe lesse  a longe tyme þere -after . Ac if ȝe wilneth to wende wel  þis is þe weye thider Ȝe mote go thorȝ mekenesse  bothe men and wyues . Til ȝe come in -to consience  þat crist wite þe sothe . Þat ȝe louen oure lorde god  leuest of alle thynges . And þanne ȝour e neyȝbores nexst  in none wise appaire Otherwise þan þow woldest  he wrouȝte to þi -selue . R.5.586: In the right margin, beginning here and written vertically up the page to R5.580, there is a signature which has been erased. It appears to be the same hand as on fol. 94r. Still discernible are traces of an initial and a surname in full. And so bouh forth by a broke  be buxu m of þi nil R.5.587: R's þi (cf. F's þy n ) is not attested in beta nor in the A version. speche . Til ȝe fynden a forthe  ȝour e faderes honoureth . Honora patrem et matrem matrem &c R.5.589: R's Latin tag omits a final &c found in beta (also, cf. F's unique added phrase ut sis longeu us sup er t erra m ). Most C witnesses agree on this point with beta. Wadeth in þat water  and wascheth ȝow wel þere . And ȝe schul lepe þe lyȝtloker e  alle ȝour e lif -tyme . And so schalt þow se swere nauȝt  but if it be for nede . And namliche an Idel  þe name of god almiȝti . Þanne schaltow come bi a crofte  but come þow nauȝt þer eInne . Þe That crofte hate coueyte nauȝt  men R.5.595: R uniquely omits a word after men, but the omission probably goes back to alpha. F rewrites by dropping men altogether and uniquely substituting the phrase neyþir maydins , while beta deploys a phrase that parallels Ax : mennes catel. The Cx phrase, menne catel , involves an unmarked genitive that might help account for the error in alpha. ne catel ne here wyues . Ne none of her s eruantes  þat nuyen hem miȝte . Loke þow ye breke no bowes þere  but if it be on nil R.5.597: R's on is a unique addition. ȝoure owen . To stokkes þere stondeth  ac stynte ȝe nauȝt þere . Þei hatte stele nauȝt ne slee nauȝt  strike forthe by bothe . And leue hem on þi left halfe  and loke nauȝt þere -after And hold wel þin haliday  heye til euene . Þan schalt þow blenche at a beruh  bere no fals wittnesse . R.5.603: Among the B witnesses, R uniquely omits the subject of this line, which is He in beta (as well as in most A manuscripts) and It in F. However, R's reading (with its ellipsis) is also attested in Cx Is He is frythed in with floreynes  and other fees manye . Loke ȝe þow plukke no plante þere  for perel of ȝoure soules þi soule R.5.604: The plural pronouns in this line represent alpha readings, the first of which is also attested in F; cf. beta's Loke þow and þi soule . However, R's soules is a unique reading since F here goes his own way, rendering the final phrase of this line as þe heyward. Both Ax and Cx agree with beta. . Þanne schal ȝe sei sothe  so it be to done . In none maner e elles nauȝt  for no mannes byddynge . Þanne schaltow come to a courte  as cler e as þe sonne . Þe mote is of mercy  þe maner e aboute . And alle þe walles ben of witt  to halden wille oute . And Icarneled kerneled with cristendome  þat kende mankynde R.5.610: In the B tradition, R's þat kende is unique; both F and beta read mankynde. However, both Ax and Cx agree with R and verify that its reading is original. to saue . I -buterased Boterased with by -leue so  or þow best nauȝt I -saued . And alle þe houses ben I -hyled hiled  halles and chaumbres . With no lede but with loue  and lowe speche as bretheren . Þe brugge is of bidde wel  þe bette may þow spede . Iche pyler is of penau nce  of preyeres to seyntes . Of almesdedes aren þe hokes  þat þe gates hangen on . Grace hatte þe gatewarde  a goed man for -sothe . His man hatte amende ȝow  many for many man men R.5.618: Most beta manuscripts read men here, but LM support alpha's singular, which also happens to be the reading of both Ax and Cx. hym knoweth . Telleth hym þis tokene  þat truthe wyte þe sothe . I parfourmed þe penau nce  þe prest me enioyned . And am ful sori of for R.5.621: Beta reads for here, which is also the reading of the A version; though a majority of C manuscripts also agrees with beta's preposition, some of the most reliable X family witnesses (XYcUcDc) agree with alpha's of. my synnes  and so I schal euere . Whan I thenke þere -on  þei I were a pope . Byddeth amende ȝow meken hy m  to his meister ones . To wayue vppe þe wykat  þat þe wo mman schette . Þo adam and eue  eten appeles vn -rosted . Per euam cunctis clausa est  & p er mariam virginem it eru m nil &c etera patefacta est R.5.626: Beta omits iterum and finishes the citation with the phrase patefacta est. The C version includes both elements, reading iterum patefacta est . For he hath þe keye and þe clikat  þowe þe kynge slepe . And if grace graunt þe  to go in þis wise . Þow schalt se in þi -selue  treuthe sitte in þin herte . In a cheyne of charite  as þow a childe were . To suffre hym and to R.5.631: R shares this repeated verbal particle solely with G, presumably by convergence. segge segge nouȝte R.5.631: Here R uniquely omits a word; most of the other manuscripts read segge nouȝte . F rewrites the line but also omits the negative, suggesting that the error originated in alpha.  aȝeynes þi sires wille . Ac be war þanne of wrathe þe  þat is a wikked schrewe . He hath enuye to hym  þat in þin herte sitteth . And pucketh pukketh forþ R.5.634: In the beta tradition, as in many of the A manuscripts and in Cx, this phrase reads pukket forþ. pruyde  to preyse þi -selue . Þe boldenesse of þi benefetes  maketh þe blynd þanne . And þanne worstou driue n out as deuh  and þe dore I -closed closed . I -cayed Kayed and I -clycated clikated  to kepe þe with -oute n . Happely an hundreth wynt er  ar þow efte entre . Þus miȝtow lesen his loue  to lete wel by þi -selue . And neu ere happeliche efte entre  but grace þow haue . Ac þere aren seuen ȝiftes sustren R.5.641: The Bx reading here was sustren (as in Ax and Cx), and alpha obviously was in error; cf. F's reading, seruauntys, which is probably a characteristic smoothing rather than the original alpha variant.  þat s eruen treuthe eu ere . And aren porteres ouer of R.5.642: Cf. F's at and beta's of. The beta reading agrees with the predominant reading among A witnesses, while the reading of R is that of Cx. þe post ernes  þat to þe place longeth . Þat on hatte abstinence  and vmblete humilite R.5.643: R's form here is obviously synonymous with the F/beta humilite (which is also the lection found in Ax). However, R's form is treated by MED and OED2, s. v. humblete, not as a spelling variant but as a different, rarer noun, derived from humble by suffixing. MED cites examples from manuscripts of Chaucer's Physician's Tale, Parson's Tale, and the Tale of Melibee. The P family of the C version agrees with the prevalent F/beta reading, but the X family shows the same rare form as R, suggesting that it may well have been the reading in both Bx and Cx. an -other . Charite and chastite  ben his chief maydenes . Pacience and pees  miche poeple þei helpeth . Largenesse þe lady  he she R.5.646: He, "she." R.5.646: Cf. L's heo; F and the majority of beta copies read she (though Cr omits any pronoun reference). The A and C manuscripts agree with Cr. lete in ful manye . Heo she hath hulpe a thousend oute  of þe deueles pondfolde . And ho R.5.648: ho, "who, whoever." is sib to þis seuene  so me god helpe . Heo He is wonderlich welcome  and faire vnderfongen . And but if þat nil R.5.650: R's þat is a unique addition to this phrase, which reads in Ax exactly as it does in beta. ȝe be sib  to su mme of þis seuene . It is ful hard be my hed q uod peres  for eny of ȝow alle . To geten Ingonge at any gate þer e  but grace be þe more . Now be crist q uod a cuttpurs  I haue no kynne þere . Ne Nor I q uod an apewarde  by auȝt þat I knowe . Wite god q uod a wafrester e  wiste I þis sothe for sothe . R.5.655: F reads this b-verse as þat y wiste þe sothe. Beta has wist I þis for sothe . Obviously, R is closer to beta here, differing only in omitting for. Ax appears to have had the same reading as beta. Cx is uncertain (a major group of P manuscripts reading þe soþe ), but the X family clearly supports beta's version of the phrase. Schulde I neuer forther e a foot  for no frere freres R.5.656: Among the B copies, R's frere is a unique reading, an unmarked genitive; many C witnesses agree with the B majority, but most of the X family reads this lection in agreement with R. p rechynge . Ȝus q uod peres þe plowman  and poked hem alle to goed . Mercy is a mayde maydene R.5.658: R's mayde is unique in the B tradition; all other witnesses attest maydene. However, witnesses in the C tradition are split on this reading, most of the X family agreeing with the B majority while most of the P family (as well as several of the X grouping) concur with R's form. þere  hath miȝt ouer hem nil alle . And sche is sib to alle synful  and hire sone alse . And thorȝ þe helpe of hem to  hope þow non other . Þow miȝt gete grace þere  be bi so þow go by -tyme . R.5.661: There is an apparently random ink blot, partially erased, immediately below the punctus, giving it, at first glance, the appearance of a modern semicolon. By seint poule q uod a pardonere  par -aunter I be nauȝt welcome knowe(n) þere . R.5.662: Where the alpha manuscripts have welcome, the beta copies read knowe þere. I wil go feche my box  with my breuettes  & a bull e with bisshop es l ettres . By crist q uod a comune woman  þi company wil I folwe . Þow schalt sey I am þi suster  Ine I ne wote wher e þei by -come . Passus Sextus de visione vt sup ra . t T his were a wikked way  but ho -so hadde a gyde . Þat wolde folewe vs ech a fote  þus þis folke he m mened . R.6.3: There is no paraph marker here because there was no space for one (the passus initial extends down slightly beyond this line in the left margin). Quatȝ perkyn þe plowma n  by seint peter of rome . I haue an half acre to erie  by þe heyȝ weye . Hadde I hered þis half acre  and I -sowen sowe(n) it after . I wil R.6.6: Among the other B copies, only LCY share R's wil (the rest read wolde). That the latter form of the verb was original in the A version seems obvious (though three of its manuscripts—Ra, U, and La—support a future meaning in place of the conditional); but the reading of Cx is more problematic. Most C copies agree with the AB majorities on wolde, but three of the most authoritative X-family witnesses, XYcI, support the RLCY reading. wenden with ȝow  and þe weye teche . Þis were a longe lettyng  q uod a lady in a sclayre . What schuld we wo mmen  werche þere þe R.6.8: Most beta manuscripts of all other sub-groups read þe, but the LM pair agrees with alpha. whiles . Su mme schal sowe þe sak q uod peres  for scheding of þe whete . And ȝe loueliche ladyes  with ȝour e longe fyngeres . Þat ȝe haue selke and sendel  to sowe whan tyme is . Chesibles for chapeleynes  cherches to honoure . wyues and wydewes . Wyues and wydewes  wolle and flex spynneth . Maketh cloth I conseile ȝow  ad a[n]d and kenneth so ȝour e douȝtres . Þe nedy and þe naked  nemeth hede how a þei R.6.15: a, "they." lyggeth . And casteth hym hem hym, "them." For <a> and hym as spellings of "they" and "them," see MED, s.v. he pron.(3) and hem clothes  for so comaundeth treuthe . For I schal lene hem lyflode  but if þe londe fayle . Flesche and brede bothe  to riche antd to pouere . As longe as I lyue  for þe lordes loue of heuene . And alle maner e of men  þat thorȝ mete and drynke libbeth . Helpeth hym to werche wyȝtliche  þat wynneth ȝour e fode . By crist q uod a kniȝt þo  he kenneth vs þe beste . Ac on þe teem treuly  tauȝt was I neuere . Ac kenne me q uod þe kniȝt  and by crist I wil assay . By seint poule q uod perkyn  ȝe p rofre ȝow so faire . Þat I schal swynke and swete  and sowe for vs bothe . And other laboreres laboures do for þi loue  al my lif -tyme . In R.6.28: There is a mark above the <I> of In. It appears to be a backwards <c> in brown, touched with red. couenau nt þat þow kepe  holy cherche and my -selue . Fro wastoures and fro wikked men  þat þis world strueth . And go hunte hardiliche  to hares and to foxes . To bores and to bukkes brockes R.6.31: Cf. beta's brockes; F rewrites the a-verse. Both Ax and Cx agree with R's lection.  þat breketh a -dou n myn hegges . And go afayte þi faukones  wilde foules to kulle . For swyche cometh to my crofte  and croppeth my whete . Curteiseliche þe kniȝt þanne  comsed þes wordes . By my power peres q uod he  I pliȝt þe my treuthe . To fulfille þi þis R.6.36: R's þi is unique; Bx reads þis, which is also the reading of a plurality of A witnesses in a very similar rendering of this line. forwarde  þouȝ I fiȝte schulde . As longe as I lyue  I schal þe meyntene . Ȝe and ȝet a poynt q uod peres  I preye ȝow of more . Loke ȝe tene no tenaunt  but treuthe wil assente . And þouȝ ȝe mowe am ercy men hem R.6.40: Beta reads hem. In a slightly revised line, the C tradition agrees here with alpha, reading men.  lat mercy be taxour e . And mekenesse ȝour e þi R.6.41: Both F and beta attest þi. The same is true of Cx. mayster e  maugre mede medes R.6.41: R's uninflected form is unique among the B manuscripts; the others have medes. However, a majority of C witnesses, including the best X-family copies, agree with R's form. chekes . R.6.41: Here the scribe again omits his usual insertion of a blank line to mark a new paragraph, presumably because the next line is the last of the side. And þouȝ pore men profre ȝow  p resentes and ȝiftes . e j Nyme it nauȝt an aunter  þow ȝe R.6.43: R's aunter þow is an alpha reading; beta has auenture ȝe ; however, though five A witnesses support auenture, it seems clear that the entire phrase, in both Ax and Cx was rendered as in alpha. mowe it nauȝt deserue For þow schalt ȝelde it aȝen  at on ȝeres ende . In a ful periliouse place  purgatorie it hatteth . And mysbede nauȝt þi bonde -man bondemen  þe bett ere may þow spede . Þouȝ he be þin vnderlyng here  wel may happe in heuene . Þat he worth worthier sett  and with more blisse . Amice ascende superius . For in a nil chanel charnel R.6.50: chanel, "charnel house." For the spelling, see MED, s.v. charnel n.(1). at cherche  clerkes cherles R.6.50: Beta reads cherles, which clearly reflects Bx (and agrees with C). Alpha's variant (F = a clerk) is deficient in alliteration as well as less suitable to the context (levelling of social classes beyond death). ben euel to knowe . Or a kniȝt fram a knaue þere  knowe þis in þin herte . And þat þow be trewe of tonge þi tonge R.6.52: Beta reads of þi tonge . Most A witnesses agree on this reading with beta, as does the X family of C manuscripts. However, four A copies (TDH 2V) agree with alpha's omission of any determiner. Most of the P family in the C tradition attest the presence of hys at this point.  and tales þat þow hatie . But if þei ben of wisdom  or of witt þi werkmen to chastie . Holde nauȝt nil R.6.54: No beta manuscript attests nauȝt. Two A copies agree with alpha at this point, but most support beta. By contrast almost all C witnesses support alpha's use of the double negative in this a-verse, although the Russell-Kane edition emends this evidence away. with non herlotes  ne here nauȝt here tales . And nameliche at þe mete  swyche men eschewe . For it beth þe deueles disoures  I do þe to vnderstonde . I assente by seint Iame  seyde þe kniȝt þanne . For to werch by thy wordes  þe while my lyf dureth . And I schal apparaille me q uod perkyn  in pilgrimes wise . And wende with ȝow I wil  til we fynde treuthe . And caste on me R.6.61: Only L agrees with R in attesting me my. F and the other beta copies simply show my. Ax and Cx agree against B in reading this phrase as third-person description: He caste on ( his = A) / ( hym his = C) cloþis . It is interesting to note that the one difference between the A and C phrases parallels the grammatical difference between LR and the other B witnesses. my clothes  I -clouted and I -hole hole . My cokeres and my cuffes  for colde of my nayles . And hange myn hoper at myn hals  in stede of a sc rippe . A buschel of brede -corne  bryng me þer e -Inne . For I wil sowen it my -selue  and sytthen wil I wende . To pilgrimage as palmeres dou n  pardon for to haue . Ac ho -so helpeth me to erye  or sowen her e er e I wende . Schal haue leue be oure lord  to lese here in herueste . And maken hym hem R.6.69: Beta reads hem here, but both Ax and Cx agree with alpha's hym. merie þer emyde  maugre ho -so it nil be -grucche . R.6.69: Beta transposes the final phrase as bigruccheth it. This word order parallels that found in the X family of C (the P family omits it completely, as do several of the less authoritative copies in the B tradition). However, the A witnesses strongly support alpha's phrasing at this point. And alle kynne crafty men  þ at conne lyuen in treuthe . I schal fynde hem fode  þat feithfullyche libbeth . R.6.71: Here the scribe again omits his usual insertion of a blank line to mark a new paragraph, presumably because the next line is the last of the side. Saue Iakke þe iogelour  and Ihonete of þe stiues stues . And daniel þe dys pleyere  and denote þe bawde . And frere þe faitoure  and folke of hys ordre . And robyn þe ribauder  for his rousty wordes . Treuthe tolde me ones  and badde me tellen it after Deleantur de libro viuenciu m  I schulde nauȝt dele with hem . For holy cherche is hote  of hem no tythe to aske take . R.6.78: R's aske is unique; both beta and F read take. However, both Ax and Cx agree with R's reading. Quia cu m iustis non scribantur . Þei ben aschaped good auntour  now nil R.6.80: Beta reads auenture in place of alpha's auntour, but Ax and Cx agree with alpha's form. Also, most beta copies omit alpha's now, but its occurrence in LM clearly attests its authenticity, both in beta and Bx. Almost all C witnesses—and a majority of A copies—attest the presence of this adverb. god hem amende . Dame worche whan tyme is  peres wif hyȝte . His douȝte douȝte[r] hiȝt do riȝt so  or þi dame schal þe bete . His sone hiȝt suffre  þi souereynes haue to haue(n) R.6.83: Beta reads to hauen . A majority of the A witnesses supports beta (none supports alpha), but Cx clearly agrees with alpha in omitting to. here wille . Deme hem nauȝt for if þow dost  þow schalt it der e a -bugge . Lat god I -worth with al  for so his worde techeth . For now I am holde and hore  and haue of myn owene . To penaunce and to pilgrimage  I wil passe with þis other e . For -thy I wil ar I wende  do write my queste biqueste . R.6.88: R's queste is unique in form but identical in sense to beta's reading (See MED, s. v. quiste [n.]); cf. F's enqweste and beta's biqueste. Both Ax and Cx agree with beta's form. In dei no mi ne amen  I make it my -selue  he schal haue my soule . R.6.89: Here alpha differs from beta by merging the opening phrase from the archetype's next line (= He shal haue my soule ) with this one, and then truncating the third line by deleting its final phrase (= for so I bileue ), so that the three lines in Langland's presumptive original are reduced to two non-alliterating ones. Þat best hath I -serued it  and fro þe fende I -keped it it defende R.6.90: R's I -keped is unique; F rephrases this verse, with his own unique verb phrase ( weyvid fram yt); the cognate beta reading is it defende. Both Ax and Cx substantially agree with beta's phrasing in this entire passage, where alpha was clearly corrupt (cf. note at R6.89). . Til I come til to R.6.91: R's til before his acountes is unique among the B witnesses. Ax, like beta and F, shows to here, but Cx agrees with R. his acountes  as my crede telleth me telleth R.6.91: Beta adds me before telleth. A majority of A manuscripts agrees with beta on this addition, but a minority (ChHaLaEWaMa) supports alpha's omission of the pronoun. Moreover, Cx clearly supports alpha in omitting me. . To haue a reles and a remissiou n  on þat rental I leue . Þe kyrke schal haue my carayne  and kepe myn bones . For of my corne and catel  he crauede tithe þe tythe . R.6.94: R uniquely omits a determiner before tithe. Most B witnesses agree with Ax in attesting þe before the noun; however, Cr agrees with Cx in reading my at this point. I payede it hym prestly  for peryl of my soule . For -thi is he holden I hope  to haue me in his messe . And mengen in his memorie  amonge alle cristene My wif schal haue of þat I wan  with treuthe and namore . And dele amonge my douȝteres  and my der e childres children R.6.99: R's form is unique; all other witnesses in the B tradition have children. Two C manuscripts (XI) agree with R, but most agree with the B majority. . For þouȝ I deyede deye R.6.100: Beta's form agrees with the reading of both Ax and Cx. to -day  my dette dettes is Iquited ar quitte . R.6.100: R's singular is unique; F shows a plural, dettys ben alle qwitte , and beta attests a slightly different plural phrasing, dettes ar quitte . The A tradition clearly supports the F/beta plural format here, but the C tradition is divided, much of the P family concurring with F/beta while the X family (and a few P copies) support R's singular. I bare hom þat I borwede  ar I to bedde ȝede . And with þe residue and þe remenau nt  be þe rode of lukes . I wil worschipe þer ewith  treuthe by my lyue . e ij And ben his pilgrime at þe plow  for pore menne sake . My plow -pote plow(es)fote R.6.105: Beta reads plow -fote , but Ax clearly agrees with the alpha reading. schal be my pyk -staf  and picchen at atwo R.6.105: Cf. F's a -wey and beta's atwo. Most of the A witnesses agree with R. þe rotes . And helpe my culter to kerue  and clenese clense þe forewes . Now is p erkyn and þese his R.6.107: Beta reads his; F has þe. Most A copies support F's reading. Among the C manuscripts, the P family also agrees with F, but the X family reading, þis, concurs with R's þese. pilgrimes  to þe plow faren . To herye þis haf hacre  holpen hym manye . Dikeres and delueres  digged vp þe balkes . Þer ewith was p erkyn apayd  and preised hem faste . Other werkmen þer e were  þat woruten ful ȝerne . Vch man in his maner e  made hym -self to done . And somme to plese p erkyn  piked vp þe wedes . At heȝ prime peres  lete þe plow stande . And ȝeed nil R.6.115: R's And ȝeed to is unique; otherwise R reads the line as beta does (which begins the line simply with To . F completely rephrases the line, but the fact that F's rewritten line begins & wente indicates that R's unique opening probably derives from alpha. Both the A and C versions begin this line by omitting R's opening (the former reading the phrase exactly as beta does). to ouer -sen hem hy m -self  and ho -so best wrouȝte . He schuld be huyred þer e -after  whan heruest -tyme come . And þanne sete so mme  and songen at þe ale . And ho nil R.6.118: ho, "they." See MED, s. v. he pron.(3). helpen to erye R.6.118: Only G agrees with R's assignment of present tense to this verb; only F agrees with R's inclusion of to after the verb. Beta reads hulpen (cf. F's hulpen). R's ho is unique error. his half acre  with how trolly lolly . Now be þe perill e of my soule q uod peres  alle i n puer tene . But ȝe arise þe rather e  and rape ȝow to werche . Schal no greyn þat groweth  glade ȝow at nede . And þouȝ ȝe deyede deye(n) R.6.122: R's deyede is unique. All other B manuscripts show a present-tense form here, as do the A and C versions. for deul  þe deuel haue þat recche reccheth . Þo were faitoures aferde  and feyned hem blynde . So mme leyde here legges a -lyry  as swiche loseles co nneth . And made here mone to peres  and preyed hym of grace . For we haue no lymes to labore with  lord graced be ȝe . Ac we preye for ȝow peres  and for ȝour e plow bothe . Þat god of his grace  ȝoure greyne multiplye . And ȝelde ȝow of ȝour e almesse  þat ȝe ȝiuen vs here . For we may nouȝt swynk ne swete  suche seknesse vs ayleth . If it be soth q uod peres þ at ȝe seyne  I schal it sone aspie . Þo Ȝe R.6.132: R's reading is unique and clearly defective; beta has Ȝe and F reads Þat ȝee. Both the A and C versions confirm the correctness of beta's reading. ben wastoures I wot wel  treuthe and treuthe R.6.132: R's omission of and at the beginning of the b-verse is unique. The A version confirms the presence and location of this conjunction, as found in beta and F. wot wel nil R.6.132: R's redundancy (reiterated wel in the b-verse) is unique. þe sothe . And I am his olde hyne  and hiȝte hym to warne . Whiche þei weren in þis world  his werkemen apayreth appeyred(en) . R.6.134: Beta reads appeyred. F has a-peyre. Ȝe wasten þat men wynnen  with trauayle and with tene . Ac treuthe schal teche ȝow  his teme to driue . Or ȝe schal ete barly brede  and of þe broke drynke . But if he blynde be blynde R.6.138: R's omission of the verb here is unique; cf. F's he þat is blynd and beta's he be blynde. or brokelegged  or bolted with yrenes . . He schal ete whete brede  and drinke with me -selue . Til god of his goednesse  amendement hym sende . Ac ȝe miȝt trauaile as treuthe wolde  & take mete & huyre . To kepe ken in þe felde  þe corne fro þe bestes . Diken or deluen  or dyngen vppon scheues . Or helpe make morter  or bere muk a -felde . In lecherie and in losyngrie  lyue [ȝe] lyue ȝe lyue(n) and in sleuthe . R.6.145: There is an ink blot, probably the result of offset, that has the appearance of a second punctus far to the right of the intended line terminal punctus and just inside the ruling margin. And al is thorȝ suffraunce  þ at vengance ȝow ne taketh . Ac ancres and heremites  þat eten but at nones . And namore ar morewe  myn almesse schul þei haue . And of my nil catel to cope hem with  þat haue cloystres i n kerkes and cherche(s) R.6.149: R's in kerkes is unique; the other B witnesses read and cherche(s). . Ac robert renneaboute  schal nauȝt haue of myne . Ne postles but þei p reche co nne  and ȝut nil R.6.151: R's ȝut is a unique addition to the text of Bx. haue pouer of þe bisshopp e . Þei schal haue payn and potage  and make hem -self at ese . For it is an vnresonable religion  þat hath riȝt nauȝt of certeyne . And þanne gan a wastour to wra.the hy m  & wold haue I -fouȝte . And to peres þe plowman  he profered his gloue . A bretoner e a braggere  abosted a bosted R.6.156: A, "he." peres alse . And bad hym go pissen on with R.6.157: Beta reads with. Both Ax and Cx agree with beta. his plow  forpyned schrewe . Wil þow or nell e þow  we wil haue oure wille . And nil of R.6.159: R's And of is an alpha phrase (F omits of). The X family of C also begins the line with And. Though two A manuscripts (KMa) also begin the line exactly as R does, Ax, like beta, omits And. þi flour e and of þi flesch e  fecche whanne vs liketh . And make vs merie þer e -myde  maugre þi chekes . Þanne peres þe plowman  pleyned hym to þe kniȝte . eiij To kepe hym as couenant was  fram cursed schrewes . And fram þes wastoures wolues -kynnes  þat maketh þis R.6.163: Although G agrees with R, beta reads þe; F o