<div1>
<head><foreign>Passus quintus de visione vt supra</foreign></head>
<lb/>
<lg>
<l> The kyng and his knightes · to the kirke wente</l>
<l> To here matynes of þe day · and þe masse after</l>
<l> Þanne waked I of my wynkynge [·] and wo was with-alle</l>
Bx.5.4KD.5.4
<l> Þat I ne hadde sleped sadder · and yseiȝen more</l>
<l> Ac er I hadde faren a fourlonge · feyntise me hente</l>
<l> That I ne myȝte ferther a foot · for defaute of slepynge</l>
<l> And sat softly adown · and seide my bileue</l>
Bx.5.8KD.5.8
<l> And so I<note>Bx.5.8: <hi>so I</hi>: Omitted by R, but F has <hi>y bablede so</hi>, and beta is supported by <hi>Ax</hi>. The line is not in <hi>Cx</hi>.</note> babeled on my bedes · þei brouȝte me a-slepe</l>
</lg>
<lb/>
<lg>
<l> ¶ And þanne saw I moche more · þan I bifore tolde</l>
<l> For I say þe felde ful of folke · þat I bifore of seyde</l>
<l> And how resoun gan arrayen hym · alle þe reume to preche</l>
Bx.5.12KD.5.12
<l> And with a crosse afor<note>Bx.5.12: <hi>afor</hi>: On five occasions alpha and random beta mss. read the preposition as <hi>byfor</hi>; the others are <ref><hi>Bx</hi>.5.23</ref>, <ref>16.46</ref>, <ref>17.311</ref>, <ref>20.130</ref>. Probably this is regression to the commonplace, since <hi>afor</hi> is uncommon in the poem. Only in <ref>12.91</ref> does it have support from both branches. At <ref>14.145</ref> the adverb reads <hi>afore</hi> in beta, but <hi>tofore</hi> in alpha.</note> þe kynge · comsed þus to techen</l>
</lg>
<lb/>
<lg>
<l> ¶ He preued þat þise pestilences · was<note>Bx.5.13: <hi>was</hi>: LMCOR have this as the plural form. It is also the form in <hi>Cx</hi> (RK.5.115) and some <hi>A</hi> mss. For parallels see <ref><hi>Bx</hi>.13.368</ref>, <ref>17.11</ref>, <ref>19.47</ref>, <ref>19.96</ref>.</note> for pure synne</l>
<l> And þe southwest wynde · on saterday at euene</l>
<l> Was pertliche for pryde<note>Bx.5.15: <hi>pryde</hi>: Beta adds <hi>pure</hi> (from l. <ref>13</ref>); alpha is supported by <hi>AC</hi>.</note> · and for no poynt elles</l>
Bx.5.16KD.5.16
<l> Piries and plomtrees · were puffed to þe erthe</l>
<l> In<note>Bx.5.17: <hi>In</hi>: Alpha begins the line with <hi>And in</hi>, but <hi>AC</hi> support beta.</note> ensample ȝe segges · ȝe<note>Bx.5.17: <hi>ȝe segges ȝe</hi>: WHm have <hi>þat ȝe segges</hi>. F's <hi>we</hi> is also the reading of <hi>Cx</hi> (RK.5.119). <hi>A</hi> mss. show considerable variation.</note> shulden do þe bettere</l>
<l> Beches and brode okes · were blowen to þe grounde</l>
<l> [And]<note>Bx.5.19: <hi>And</hi>: Dropped in beta, but supported by <hi>AC</hi>.</note> torned vpward her taille<note>Bx.5.19: <hi>taille</hi>: As often, the distributive sg. is to be preferred. In this case alpha is supported by <hi>AC</hi>.</note> · in<note>Bx.5.19: <hi>in</hi>: Dropped by alpha, but supported by <hi>AC</hi>.</note> tokenynge of drede</l>
Bx.5.20KD.5.20
<l> Þat dedly synne ar domesday · shal fordon hem alle</l>
</lg>
<lb/>
<lg>
<l> ¶ Of þis matere I myȝte · mamely ful longe</l>
<l> Ac I shal seye as I saw · so me god helpe</l>
<l> How pertly afor<note>Bx.5.23: <hi>afor</hi>: See note to <ref><hi>Bx</hi>.5.12</ref> for variation with <hi>byfor</hi>.</note> þe poeple · resoun gan<note>Bx.5.23: <hi>gan</hi>: Strong support from LMR against <hi>bigan</hi>. F's word-order improves the alliteration of the b-verse, but cannot be archetypal. No doubt the order in <hi>Bx</hi> is a reminiscence of l. <ref>11</ref>. The line is not in the other versions.</note> to preche</l>
</lg>
<lb/>
<lg>
Bx.5.24KD.5.24
<l> ¶ He bad wastoure go<note>Bx.5.24: <hi>go</hi>: Only R has <hi>to</hi>, which is, however, the reading of the X family of <hi>C</hi>.</note> worche · what he best couthe</l>
<l> And wynnen his wastyng · with somme manere crafte<note>Bx.5.25: <hi>crafte</hi>: <hi>Ax</hi> also has the sg.</note></l>
</lg>
<lb/>
<lg>
<l> ¶ And<note>Bx.5.26: <hi>And</hi>: Clear support from LMR and HmCO, against <hi>He</hi> in CrWGF. <hi>A</hi> mss. vary, but <hi>Cx</hi> has <hi>He</hi>, possibly the source for CrG by contamination.</note> preyed peronelle · her purfyle to lete</l>
<l> And kepe it in hir cofre · for catel at hire nede</l>
</lg>
<lb/>
<lg>
Bx.5.28KD.5.28
<l> ¶ Thomme stowue he tauȝte · to take two staues</l>
<l> And fecche filice home · fro wyuen[e] pyne<note>Bx.5.29: <hi>wyuene pyne</hi>: Having lost -<hi>e</hi> (OE -<hi>ena</hi>), beta adds the article to correct a short b-verse. A few <hi>AC</hi> scribes do the same.</note></l>
</lg>
<lb/>
<lg>
<l> ¶ He warned watt · his wyf was to blame</l>
<l> Þat<note>Bx.5.31: <hi>Þat</hi>: Also the <hi>A</hi> reading. WHm <hi>For</hi> is also the <hi>C</hi> reading (RK.5.133).</note> hire hed was worth halue marke<note>Bx.5.31: <hi>halue marke</hi>: LCrCG include the indefinite article, and M is corrected to include it. But alpha and WHmO are without it, as is <hi>Cx</hi>. <hi>Ax</hi> has <hi>a mark</hi>.</note> · his<note>Bx.5.31: <hi>his</hi>: CrWHmF have <hi>and his</hi>, as also do <hi>AC</hi>. But probably <hi>Bx</hi> omitted <hi>and</hi>, since it is not attested by LR and CGO. In M it is first added by the corrector and then deleted.</note> hode nouȝte<note>Bx.5.31: <hi>nouȝte</hi>: The reading <hi>nouȝte worth</hi> looks suspiciously scribal. Six <hi>A</hi> mss. omit <hi>worth</hi>, as do all except one of the <hi>C</hi> mss. But attestation for <hi>Bx</hi> is uncertain: only HmG and R omit it. We follow the readings of R throughout this line.</note> a grote</l>
Bx.5.32KD.5.32
<l> And bad bette kut · a bow other tweyne</l>
<l> And bete betoun þer-with · but if she wolde worche</l>
<l> <note>Bx.5.34: WHm and alpha have a paraph, though it is not particularly appropriate.</note>And þanne he charged chapmen · to chasten<note>Bx.5.34: <hi>chasten</hi>: The form supported for <hi>Bx</hi> by LR and CrCG, as well as <hi>Cx</hi> and five <hi>A</hi> mss. Others have <hi>chastisen</hi>.</note> her childeren</l>
<l> Late no wynnynge [forweny hem]<note>Bx.5.35: <hi>forweny hem</hi>: Beta has <hi>hem forweny</hi>. The word-order of alpha is uncertain, since R omits the pronoun, but F's <hi>for-wayne hem</hi> is the order of <hi>AC</hi>.</note> · whil<note>Bx.5.35: <hi>whil</hi>: R alone has <hi>þe while</hi>, though it is in the X family of <hi>C</hi>.</note> þei be ȝonge</l>
Bx.5.36KD.5.36
<l> Ne for no pouste of pestilence · plese hem nouȝte out of resoun</l>
</lg>
<lb/>
<lg>
<l> ¶ My syre seyde so to me · and so did my dame</l>
<l> Þat þe leuere childe · þe more lore bihoueth</l>
<l> And Salamon seide þe same · þat sapience made</l>
Bx.5.40KD.5.39α
<l> <foreign>Qui parcit virge odit filium</foreign></l>
<l> Þe Englich of þis latyn is · who-so wil it knowe</l>
<l> Who-so spareth þe sprynge · spilleth<note>Bx.5.42: <hi>spilleth</hi>: Alpha includes the pronoun <hi>he</hi>, but beta is supported by <hi>Cx</hi>. The line is not in <hi>A</hi>.</note> his children</l>
</lg>
<lb/>
<lg>
<l> ¶ And sithen he preyed<note>Bx.5.43: <hi>preyed</hi>: Evidently the beta reading, supported against the variants by <hi>AC</hi>.</note> prelatz · and prestes togideres</l>
Bx.5.44KD.5.42
<l> Þat ȝe prechen to þe peple · preue it on ȝowre-seluen</l>
<l> And doth it in dede · it shal drawe ȝow to good</l>
<l> If ȝe lyuen as ȝe leren<note>Bx.5.46: <hi>leren</hi>: This varies with <hi>lerne</hi> also in <hi>A</hi>, but is supported by <hi>Cx</hi>.</note> vs [·] we shal leue ȝow þe bettere</l>
</lg>
<lb/>
<lg>
<l> ¶ And sithen he radde Religioun · here reule to holde</l>
Bx.5.48KD.5.46
<l> Leste þe kynge and his conseille · ȝowre comunes appayre</l>
<l> And ben stuwardes of ȝowre stedes · til ȝe be ruled<note>Bx.5.49: <hi>ruled</hi>: An obvious substitution in <hi>Bx</hi> for alliterating <hi>stewed</hi>, "governed", in <hi>AC</hi>.</note> bettre</l>
</lg>
<lb/>
<lg>
<l> ¶ And sithen he conseilled þe kynge · þe<note>Bx.5.50: <hi>þe</hi> (2): This is secure for <hi>Bx</hi>, though beta2 <hi>his</hi> is the <hi>Cx</hi> reading (RK.5.180).</note> comune to louye</l>
<l> It is þi tresore if tresoun ne were<note>Bx.5.51: <hi>ne were</hi>: R's loss of <hi>ne</hi> suggests that alpha may have read <hi>nere</hi>, as in F, which would be a little less clumsy. The line is not paralleled in <hi>AC</hi>.</note> · and triacle at þi nede</l>
</lg>
<lb/>
<lg>
Bx.5.52KD.5.50
<l> [¶]<note>Bx.5.52: <hi></hi>: The parallel with the paraphs for lines <ref>43</ref>, <ref>47</ref> and <ref>50</ref> also beginning <hi>And sithen</hi> supports the paraph recorded in WHmCOR (with a new line-group in M).</note> And sithen he prayed þe pope · haue pite on holicherche</l>
<l> And er he gyue any grace · gouerne firste hym-selue</l>
</lg>
<lb/>
<lg>
<l> ¶ And ȝe that han lawes to kepe · late treuthe be ȝowre coueytise</l>
<l> More þan golde or other<note>Bx.5.55: <hi>or other</hi>: M and beta2 (CrWHm) drop <hi>other</hi>, but it has support from LCHm and alpha. The line is not in <hi>AC</hi>.</note> gyftes · if ȝe wil god plese</l>
Bx.5.56KD.5.54
<l> For who-so contrarieth treuthe · he telleth in þe gospel</l>
<l> That god knoweth hym nouȝte · ne no seynte of heuene</l>
<l> <foreign>Amen dico vobis nescio vos ·</foreign></l>
</lg>
<lb/>
<lg>
<l> ¶ And ȝe þat seke seynte Iames · and seintes of Rome</l>
Bx.5.60KD.5.57
<l> Seketh seynt treuthe · for he may saue ȝow alle</l>
<l> <foreign>Qui cum patre & filio ·</foreign> þat feire hem bifalle</l>
<l> Þat suweth my sermon · and þus seyde resoun</l>
</lg>
<lb/>
<lg>
<l> ¶ Thanne ran repentance · and reherced his teme</l>
Bx.5.64KD.5.61
<l> And gert wille to wepe · water with his eyen</l>
</lg>
<lb/>
<lg>
<l> ¶ Peronelle proude-herte · platte hir to þe erthe</l>
<l> And lay longe ar she loked · and lorde mercy cryed</l>
<foreign>Superbia ·</foreign><note>Bx.5.67: The beta scribes have Latin titles for the sins in the main hand in the margin, and marked out by rubrication from other marginal annotations, but the alpha scribes do not have them.</note>
<l> And byhiȝte to<note>Bx.5.67: <hi>to</hi>: Supported by <hi>Ax</hi> (K.5.47) and <hi>Cx</hi> (RK.6.5) against MO <hi>vn-to</hi>.</note> hym · þat vs alle made</l>
Bx.5.68KD.5.65
<l> She shulde vnsowen hir serke · and sette þere an heyre</l>
<l> To affaiten hire flessh · þat fierce was to synne</l>
<l> Shal neuere heiȝe herte me hente · but holde me lowe</l>
<l> And suffre to be myssayde · and so did I neuere</l>
Bx.5.72KD.5.69
<l> But now wil I meke me · and mercy biseche</l>
<l> For al<note>Bx.5.73: <hi>al</hi>: Dropped by alpha, but supported by <hi>AC</hi> and alliterating.</note> þis I haue · hated in myne herte</l>
</lg>
<lb/>
<lg>
<l> ¶ Þanne lecchoure seyde allas · and on owre lady he cryed</l>
<foreign>Luxuria ·</foreign>
<l> To make mercy for his mis-dedes · bitwene god and his soule</l>
Bx.5.76KD.5.73
<l> With þat he shulde þe saterday<note>Bx.5.76: <hi>þe saterday</hi>: R's <hi>on þe day</hi> probably represents alpha. Beta is supported by <hi>Ax</hi> and alliteration.</note> · seuene ȝere þere-after</l>
<l> Drynke but [with]<note>Bx.5.77: <hi>with</hi>: Alpha is supported by <hi>A</hi> and <hi>C</hi> (RK.6.174); elsewhere (e.g. <ref><hi>Bx</hi>.4.79</ref>) Langland uses <hi>myd</hi> to provide alliteration. For discussion of this variation, see note to <ref><hi>Bx</hi>.7.26</ref>.</note> þe doke · and dyne but ones</l>
</lg>
<lb/>
<lg>
<foreign>Inuidia</foreign>
<l> ¶ Enuye with heuy herte · asked after scrifte</l>
<l> And carefullich <foreign>mea culpa</foreign> · he comsed to shewe<note>Bx.5.79: <hi>shewe</hi>: R has <hi>schrewe</hi>, F has <hi>shryue</hi>, but beta is supported by <hi>Ax</hi>. The line is revised in <hi>Cx</hi>.</note></l>
Bx.5.80KD.5.77
<l> He was as pale as a pelet · in þe palsye he semed</l>
<l> And clothed in a caurimaury · I couthe it nouȝte discreue</l>
<l> In kirtel and kourteby · and a knyf bi his syde</l>
<l> Of a freres frokke · were þe<note>Bx.5.83: <hi>þe</hi>: Beta is supported by the majority of <hi>A</hi> mss., but six read <hi>his</hi> with alpha.</note> forsleues</l>
Bx.5.84KD.5.81
<l> And as a leke hadde<note>Bx.5.84: <hi>hadde</hi>: Beta2 (CrWHm) read <hi>þat hadde</hi>, supported by the majority of <hi>A</hi> mss., though four omit <hi>þat</hi>. It is more probably supplied than dropped. The line is not in <hi>Cx</hi>.</note> yleye · longe in þe sonne</l>
<l> So loked he with lene chekes · lourynge foule</l>
</lg>
<lb/>
<lg>
<l> ¶ His body was to-bolle for wratthe · þat he bote his lippes</l>
<l> And wryngy[ed]<note>Bx.5.87: <hi>wryngyed</hi>: R (= alpha?). <hi>A</hi> has <hi>wroþliche he wroþ his fest</hi> (K.5.67) and <hi>C</hi> reads <hi>A wroth his fuste vppon wrath</hi> (RK.6.66). Apparently <hi>Bx</hi> misunderstood the pun (<hi>wroth</hi> from <hi>writhen</hi>, and <hi>wrath</hi>), and substituted the past tense of <hi>wringen</hi>. Beta's <hi>wryngynge he ȝede</hi> is an expansion. Elsewhere in <hi>B</hi> the past tense of <hi>wring</hi> is <hi>wrong</hi>, as in F here.</note> with þe fiste · to wreke hym-self he þouȝte</l>
Bx.5.88KD.5.85
<l> With werkes or with wordes · whan he seighe his tyme</l>
<l> Eche a worde þat he warpe · was of an addres tonge</l>
<l> Of chydynge and of chalangynge · was his chief lyflode</l>
<l> With bakbitynge and bismer · and beryng of fals witnesse</l>
Bx.5.92KD.5.89
<l> Þis was al his curteisye · where þat euere he shewed hym<note>Bx.5.92: F entirely rewrites the line with much more emphatic alliteration. W omits. Lines 91-5 are not in <hi>AC</hi>, and lines 96-121 are not in <hi>C</hi>.</note></l>
</lg>
<lb/>
<lg>
<l> ¶ I wolde ben yshryue quod þis schrewe · and I for shame durst</l>
<l> I wolde be gladder bi god · þat gybbe had meschaunce</l>
<l> Than þouȝe I had þis woke ywonne · a weye of essex chese</l>
</lg>
<lb/>
<lg>
Bx.5.96KD.5.93
<l> ¶ I haue a neighbore neyȝe me · I haue ennuyed<note>Bx.5.96: <hi>ennuyed</hi>: The spellings <hi>anoyed</hi>, <hi>noyed</hi>, <hi>enuyed</hi> are all forms of "annoyed".</note> hym ofte</l>
<l> And lowen on hym to lordes · to don hym lese his siluer</l>
<l> And made his frendes ben his foon · thorw my false tonge</l>
<l> His grace and his good happes · greueth me ful sore</l>
Bx.5.100KD.5.99
<l> Bitwene m[eyne] and m[eyne]<note>Bx.5.100: <hi>meyne and meyne</hi>: We take R's reading ("retinue") to represent alpha, supported by <hi>Ax</hi> <hi>hym & his meyne</hi>. F thus revises, perhaps partly on the basis of <hi>A</hi>; beta regresses to the commonplace.</note> · I make debate ofte</l>
<l> Þat bothe lyf and lyme · is lost þorw my speche</l>
</lg>
<lb/>
<lg>
<l> ¶ And whan I mete him in market · þat I moste hate</l>
<l> I hailse hym hendeliche · as I his frende were</l>
Bx.5.104KD.5.103
<l> For he is douȝtier þan I · I dar do non other</l>
<l> Ac hadde I maystrye and myȝte · god wote my wille</l>
</lg>
<lb/>
<lg>
<l> ¶ And whan I come to þe kirke<note>Bx.5.106: <hi>kirke</hi>: Hm and alpha have non-alliterating <hi>cherche</hi>. See Introduction <xref>IV.1</xref>.</note> · and sholde knele to þe Rode</l>
<l> And preye for þe pople · as þe prest techeth</l>
Bx.5.108KD.5.107
<l> For pilgrimes and for<note>Bx.5.108: <hi>and for</hi>: Support from LR and CrWCO, though MHmG lack <hi>for</hi>. F instead lacks <hi>and</hi>, as in <hi>Ax</hi>.</note> palmers · for alle þe poeple after</l>
<l> Þanne I crye on my knees · þat cryste ȝif hem sorwe</l>
<l> Þat bar awey my bolle · and my broke schete</l>
</lg>
<lb/>
<lg>
<l> ¶ Awey fro þe auter þanne ·<note>Bx.5.111: The alpha scribes punctuate the line before <hi>þanne</hi>.</note> turne I myn eyghen</l>
Bx.5.112KD.5.111
<l> And biholde how [H]eleyne<note>Bx.5.112: <hi>Heleyne</hi>: F perhaps notices the unlikelihood of Envy envying a woman's coat, and substitutes <hi>Hervy</hi>. Initial <hi>H</hi>- is presumably archetypal: <hi>Ax</hi> has the male name, <hi>Heyne</hi> (cf. <ref><hi>Bx</hi>.5.114-15</ref>).</note> · hath<note>Bx.5.112: <hi>hath</hi>: Alpha's <hi>hath on</hi> is not supported by <hi>Ax</hi>. Following this line F adds rather a vivid line.</note> a newe cote</l>
<l> I wisshe þanne it were myne · and al þe webbe after</l>
</lg>
<lb/>
<lg>
<l> ¶ And of [his]<note>Bx.5.114: <hi>his</hi>: <hi>Bx</hi> alters the coatwearer from male (<hi>Ax</hi> has <hi>Heyne</hi>) to female <hi>Heleyne</hi>, and faces scribes with a problem, which they deal with characteristically. Beta and F revise to <hi>mennes</hi>, but R copies his exemplar regardless, ignoring the inconsistency. <hi>Ax</hi> also reads <hi>his</hi>, but there is nowhere evidence that R is contaminated by an <hi>A</hi> text. See notes to ll. <ref>115</ref> and <ref>116</ref>.</note> lesynge I laughe · þat liketh myn herte</l>
<l> A[c]<note>Bx.5.115: <hi>Ac</hi>: There is the same common variation with <hi>But</hi> in <hi>A</hi> mss., with three reading <hi>And</hi>, as beta does here.</note> for h[is]<note>Bx.5.115: <hi>his</hi>: Again R follows his copy-text, and on this occasion F is content to do the same, whereas beta logically replaces with <hi>her</hi>, "their". <hi>Ax</hi> again reads <hi>his</hi>.</note> wynnynge I wepe · and waille þe tyme</l>
</lg>
<lb/>
<lg>
Bx.5.116KD.5.115
<l><note>Bx.5.116: <hi></hi>: The inappropriate paraph is in LR, with a new line-group in M, and so is apparently a <hi>Bx</hi> error.</note> And deme [men]<note>Bx.5.116: <hi>men</hi>: Again R is supported by <hi>A</hi>, with F supplying the object <hi>hem</hi> instead. Since beta already has a plural referent, <hi>men</hi> is unnecessary and is therefore dropped.</note> þat hij don ille · þere I do wel worse</l>
<l> Who-so vndernymeth me here-of · I hate hym dedly after</l>
<l> I wolde þat vche a wyght · were my knaue</l>
<l> For who-so hath more þan I · þat angreth me sore</l>
Bx.5.120KD.5.119
<l> And þus I lyue louelees · lyke a luther dogge</l>
<l> That al my body bolneth · for bitter of<note>Bx.5.121: <hi>of</hi>: Probably R's <hi>in</hi> represents alpha (F has <hi>ys</hi>). <hi>Ax</hi> supports <hi>of</hi>.</note> my galle</l>
</lg>
<lb/>
<lg>
<l> ¶ I myȝte nouȝte eet many ȝeres · as a man ouȝte</l>
<l> For enuye and yuel wille · is yuel to defye</l>
Bx.5.124KD.5.123
<l> May no sugre ne swete þinge · asswage my swellynge</l>
<l> Ne no diapenidion · dryue it fro myne herte</l>
<l> Ne noyther schrifte ne shame · but ho-so schrape my mawe</l>
</lg>
<lb/>
<lg>
<l> ¶ Ȝus redili quod repentaunce · and radde hym to þe beste</l>
Bx.5.128KD.5.127
<l> Sorwe of synnes · is sauacioun of soules</l>
</lg>
<lb/>
<lg>
<l> ¶ I am <app><rdg>sori</rdg><rdg>euere sory</rdg></app><note>Bx.5.129: <hi>sory / euere sory</hi>: Beta agrees with <hi>Ax</hi>, while alpha's <hi>euere sory</hi> agrees with <hi>Cx</hi>.</note> quod þat segge · I am but selde other</l>
<l> And þat maketh me þus megre · for I ne may me venge</l>
<l> Amonges Burgeyses haue I be · dwellynge at Londoun</l>
Bx.5.132KD.5.131
<l> And gert bakbitinge be a brocoure · to blame mennes ware</l>
<l> Whan he solde and I nouȝte · þanne was I redy<note>Bx.5.133: <hi>redy</hi>: R has <hi>aredy</hi> which may be right. It is the <hi>Cx</hi> form (RK.6.97). The passage <hi>Bx</hi>.5.131-89 is not in <hi>A</hi>.</note></l>
<l> To lye and to loure on my neighbore · and to lakke his chaffare<note>Bx.5.134: <hi>chaffare</hi>: Alpha's <hi>ware</hi> presumably arises by eyeskip from l. <ref>132</ref>, where F actually reads <hi>to lakke mennys ware</hi>. For the reverse situation see l. <ref>209</ref> and note. <hi>Cx</hi> is rewritten.</note></l>
<l> I wil amende þis ȝif I may · þorw myȝte of god almyȝty</l>
</lg>
<lb/>
<lg>
<foreign>Ira ·</foreign>
Bx.5.136KD.5.135
<l> ¶ Now awaketh wratthe · with two whyte eyen</l>
<l> And nyuelynge with þe nose · and his nekke<space></space> hangynge</l>
</lg>
<lb/>
<lg>
<l> ¶ I am wrath<note>Bx.5.138: <hi>wrath</hi>: Alpha's <hi>wrothe</hi> is perhaps here the adjective, "I am angry"; alternatively it is the adjective used as a noun (see <title>MED</title> <hi>wroth</hi>), as in R's spelling in l. <ref>136</ref>. The X family of <hi>C</hi> also has this form.</note> quod he · I was sum-tyme a frere</l>
<l> And þe couentes Gardyner · for to graffe ympes</l>
Bx.5.140KD.5.139
<l> On limitoures and listres · lesynges I ymped </l>
<l> Tyl þei bere leues of low speche · lordes to plese</l>
<l> And sithen þei blosmed obrode<note>Bx.5.142: <hi>obrode</hi>: Dropped by alpha, but supported by alliteration.</note> · in boure to here shriftes</l>
<l> And now is fallen þer-of a frute · þat folke han wel leuere</l>
Bx.5.144KD.5.143
<l> Schewen her schriftes to hem · þan shryue hem to her persones</l>
</lg>
<lb/>
<lg>
<l> ¶ And now persones [han]<note>Bx.5.145: <hi>persones han</hi>: "Now that parsons have perceived" (as in beta2 and G), rather than, with the word-order in CO and alpha, "Now have parsons perceived". Either order could be <hi>Bx</hi>. L drops the auxiliary, while M drops the phrase, and <hi>persons han</hi> is added above the line in another hand. The line is not in <hi>AC</hi>.</note> parceyued · þat Freres parte with hem</l>
<l> Þise possessioneres preche · and depraue freres</l>
<l> And freres<note>Bx.5.147: <hi>freres</hi>: Alpha drops the word, but the following line refers to the peripatetic activity of friars.</note> fyndeth hem in defaute · as folke bereth witnes</l>
Bx.5.148KD.5.147
<l> That whan þei preche þe poeple · in many place<note>Bx.5.148: <hi>place</hi>: This is probably the beta form. As a romance loan ending in a sibilant, it may be singular or plural.</note> aboute</l>
<l> I wrath walke with hem · and wisse hem of my bokes</l>
<l> Þus þei speken of spiritualte<note>Bx.5.150: <hi>spiritualte</hi>: Beta2 (and M by correction) and GR supply <hi>my</hi>, by contamination with the next line. It makes no sense here. See note on l. <ref>151</ref>.</note> · þat eyther despiseth other</l>
<l> Til þei be bothe beggers · and by my spiritualte<note>Bx.5.151: <hi>by my spiritualte</hi>: Hm and alpha drop <hi>my</hi>, by confusion with the previous line. F, observing nonsense, revises to <hi>by almesse</hi>.</note> libben</l>
Bx.5.152KD.5.151
<l> Or elles alle riche [·] and riden aboute</l>
<l> I wrath rest neuere · þat I ne moste folwe</l>
<l> This wykked folke · for suche is my grace</l>
</lg>
<lb/>
<lg>
<l> ¶ I haue an aunte to nonne · and an abbesse<note>Bx.5.155: <hi>abbesse</hi>: R is supported by <hi>Cx</hi>. Beta's addition of <hi>bothe</hi> and F's of <hi>eek</hi> are responses to the unusually light b-verse.</note></l>
Bx.5.156KD.5.154
<l> Hir were leuere swowe or swelte · þan soeffre any peyne</l>
<l> I haue be cook in hir kichyne · and þe couent serued</l>
<l> Many monthes with hem · and with monkes bothe<note>Bx.5.158: <hi>bothe</hi>: R's <hi>alse</hi> probably represents alpha (F rewrites), but beta is supported by <hi>Cx</hi> (RK.6.131).</note></l>
<l> I was þe priouresses<note>Bx.5.159: <hi>priouresses</hi>: LMR and Hm have this form of the genitive, whereas other mss. have the unmarked form, as do most <hi>C</hi> mss.</note> potagere · and other poure ladyes</l>
Bx.5.160KD.5.158
<l> And made hem ioutes of iangelynge · þat dame Io[n]e<note>Bx.5.160: <hi>Ione</hi>: So MCrO and alpha, as well as <hi>C</hi> mss. It is the usual feminine form of the name in Middle English.</note> was a bastard</l>
<l> And dame Clarice a kniȝtes douȝter · ac a kokewolde was hire syre</l>
<l> And dame Peronelle a prestes file · Priouresse worth she neuere</l>
<l> For she had childe in chirityme · al owre chapitere it wiste</l>
</lg>
<lb/>
<lg>
Bx.5.164KD.5.162
<l> ¶ Of wykked wordes I wrath · here wortes I-made<note>Bx.5.164: <hi>I-made</hi>: Past tense; <title>MED</title> <hi>imaken</hi>. The prefix recorded in LOR makes the b-verse regular.</note></l>
<l> Til þow lixte and þow lixte · lopen oute at ones</l>
<l> And eyther hitte other · vnder þe cheke</l>
<l> Hadde þei had knyues bi cryst · her eyther<note>Bx.5.167: <hi>her eyther</hi>: "each of them". LMWR include the genitive pronoun.</note> had killed other</l>
</lg>
<lb/>
<lg>
Bx.5.168KD.5.166
<l> ¶ Seynt Gregorie was a gode pope · and had a gode forwit</l>
<l> Þat no Priouresse were prest · for þat he ordeigned</l>
<l> Þei had þanne ben <foreign>infamis</foreign> þe firste day<note>Bx.5.170: <hi>þe firste day</hi>: Dropped by M, G (in a rewritten line) and O to shorten a long a-verse. Cr replaces with two invented lines. The line is not in <hi>AC</hi>.</note> · þei can so yuel hele conseille</l>
</lg>
<lb/>
<lg>
<l> ¶ Amonge monkes I miȝte be · ac many tyme I shonye</l>
Bx.5.172KD.5.170
<l> For þere ben many felle frekis · my feres to aspye</l>
<l> Bothe Prioure an supprioure · and owre <foreign>pater abbas</foreign></l>
<l> And if I telle any tales · þei taken hem togyderes</l>
<l> And do me faste frydayes · to bred and to water</l>
Bx.5.176KD.5.174
<l> And am<note>Bx.5.176: <hi>And am</hi>: R's <hi>ȝeet</hi>, "furthermore", may be right, even though not in F. It gives good sense and has some support from <hi>Cx</hi> <hi>Ȝut am I</hi>.</note> chalanged in þe chapitelhous · as I a childe were</l>
<l> And baleised on þe bare ers<note>Bx.5.177: <hi>ers</hi>: Beta is confirmed by <hi>Cx</hi>. R initially read <hi>hers</hi>, altered to the euphemism <hi>bak</hi> by the correcting hand. F also reads <hi>bak</hi>.</note> · and no breche bitwene</l>
<l> For-þi haue I no lykyng · with þo leodes to wonye</l>
<l> I ete there vnthende fisshe · and fieble ale drynke</l>
Bx.5.180KD.5.178
<l> <note>Bx.5.180: Alpha has a paraph.</note>Ac otherwhile whan wyn cometh · whan<note>Bx.5.180: <hi>whan</hi> (2): So LMR and G; W has <hi>þanne</hi>, the others read <hi>and</hi> or omit.</note> I drynke w[el]<note>Bx.5.180: <hi>wel</hi>: Beta's repetition of <hi>wyn</hi> gives weaker sense, though <hi>wel</hi> is repeated in the following line. Perhaps both alpha and beta are making up for a loss in <hi>Bx</hi>, since <hi>Cx</hi> has <hi>late</hi> instead (RK.6.160).</note> at eue</l>
<l> I haue a fluxe of a foule mouthe · wel fyue dayes after</l>
<l> Al þe wikkednesse þat I wote · bi any of owre bretheren</l>
<l> I couth it in owre cloistre · þat al owre<note>Bx.5.183: <hi>owre</hi> (2): The distribution of variants is odd here. LM, the two most reliable beta witnesses, are joined by CrW with <hi>owre</hi>; the other mss., including F, have <hi>þe</hi>, except for G with <hi>your</hi> and R which has nothing. It may be that R is faithfully copying a <hi>Bx</hi> error, with other scribes supplying the obvious omission. <hi>Cx</hi> has <hi>þe</hi>.</note> couent wote it</l>
</lg>
<lb/>
<lg>
Bx.5.184KD.5.182
<l> ¶ Now repent þe quod Repentaunce · and reherce þow neure</l>
<l> Conseille þat þow cnowest · bi contenaunce ne bi [speche]<note>Bx.5.185: <hi>speche</hi>: Alpha's reading is obviously correct and supported by <hi>Cx</hi>.</note></l>
<l> And drynke nouȝte ouer-delicatly · ne to depe noyther</l>
<l> Þat þi wille bi cause þer-of [·] to wrath myȝte torne</l>
Bx.5.188KD.5.186
<l> <foreign>Esto sobrius</foreign> he seyde · and assoilled me<note>Bx.5.188: <hi>me</hi>: Here and in l. <ref>189</ref> Cr has third person pronouns in line with <hi>Cx</hi>, and M is altered to the same readings.</note> after</l>
<l> And bad me wilne to wepe · my wikkednesse to amende</l>
</lg>
<lb/>
<lg>
<foreign>Auaricia ·</foreign>
<l> ¶ And þanne cam coueytise · [I can]<note>Bx.5.190: <hi>I can</hi>: Alpha's word-order is supported by <hi>AC</hi>.</note> hym nouȝte descryue</l>
<l> So hungriliche and holwe · sire henri<note>Bx.5.191: <hi>henri</hi>: There is reasonable support for this as <hi>Bx</hi> from LR and CG, and possibly M before correction, but <hi>AC</hi> have <hi>heruy</hi>, agreeing with WHmOF. Although F may have derived the correction from <hi>A</hi>, it is difficult to see how the other three mss. got it, although Bennett (1972), 163 quotes a parallel from Skelton, possibly an indication that Hervy was a traditional name for a miser.</note> hym loked</l>
Bx.5.192KD.5.190
<l> He was bitelbrowed · and baberlipped also</l>
<l> With two blered eyghen · as a blynde hagge</l>
<l> And as a letheren purs · lolled his chekes</l>
<l> Wel sydder þan his chyn · þei chiueled for elde</l>
Bx.5.196KD.5.193
<l> And as a bondman of his bacoun · his berde was bidraueled</l>
<l> With [his]<note>Bx.5.197: <hi>his</hi>: Alpha's reading is in agreement with <hi>Cx</hi> against beta's indefinite article.</note> hode on his hed · a lousi hatte aboue</l>
<l> And in a tauny tabarde · of twelue wynter age</l>
<l> Al totorne and baudy · and ful of lys crepynge</l>
Bx.5.200KD.5.196
<l> But if<note>Bx.5.200: <hi>if</hi>: Beta has <hi>if þat</hi>; <hi>AC</hi> agree with alpha.</note> a lous couthe · l[epe]<note>Bx.5.200: <hi>lepe</hi>: Comparison with <hi>AC</hi> suggests that R represents <hi>Bx</hi>, with F and beta expanding an uncomfortably short line. <hi>A</hi> mss. show considerable variation in the b-verse, and <hi>Cx</hi> offers a desperate repair: <hi>But ȝif a lous couthe lepe y leue it as y trowe</hi> (quoting Skeat C.7.204).</note> þe bettre</l>
<l> [H]e<note>Bx.5.201: <hi>He</hi>: Whether masc. or fem., alpha's reading is supported by <hi>AC</hi>.</note> sholde nouȝte walk[e]<note>Bx.5.201: <hi>walke</hi>: Beta's <hi>haue walked</hi> follows on from its reading of the previous line. Alpha's infinitive is supported by <hi>AC</hi>, which, however, have <hi>wandre</hi>.</note> on þat welche<note>Bx.5.201: <hi>welche</hi>: "Welsh cloth", the reading of LR and <hi>AC</hi>, but mistranscribed or avoided by the other <hi>B</hi> scribes.</note> · so was it thredebare</l>
</lg>
<lb/>
<lg>
<l> ¶ I haue ben coueytouse quod þis caityue · I biknowe it here</l>
<l> For some-tyme I serued · symme atte stile</l>
Bx.5.204KD.5.200
<l> And was his prentis ypliȝte · his profit to wayte</l>
<l> <note>Bx.5.205: WHm and R have a paraph.</note>First I lerned to lye · a leef other tweyne</l>
<l> Wikkedlich to weye · was my furst lessoun</l>
<l> To wy and to wynchestre · I went to þe faire</l>
Bx.5.208KD.5.204
<l> With many manere marchandise · as my Maistre me hiȝte</l>
<l> Ne had þe grace of gyle [·] ygo<note>Bx.5.209: Only LW punctuate the line after <hi>ygo</hi>.</note> amonge my ware<note>Bx.5.209: <hi>ware</hi>: A good example of the superiority of LMR (joined by F here), supported in this reading by <hi>AC</hi>, against <hi>chaffare</hi> in other beta mss. (i.e. beta1). See Adams (2000), 176, and for the same variation see l. <ref>134</ref>.</note></l>
<l> It had be vnsolde þis seuene ȝere · so me god helpe</l>
<l> ¶ Thanne drowe I me amonges draperes · my donet to lerne</l>
Bx.5.212KD.5.208
<l> To drawe þe lyser alonge [·] þe lenger it semed</l>
<l> Amonge þe riche rayes · I rendred a lessoun</l>
<l> To broche hem with a bat-nedle<note>Bx.5.214: <hi>bat-nedle</hi>: The form in L and alpha, with M corrected to the majority beta form, <hi>paknedle</hi>. Adams (2000), 181. <hi>A</hi> mss. have the same variation, with the majority reading <hi>p</hi>- (K.5.126).The P family of <hi>C</hi> also has <hi>p</hi>-, but the X family has <hi>b</hi>-, and <hi>plaited</hi> is revised to <hi>bande</hi> so that the line alliterates on /b/ (RK.6.218). <title>MED</title> lists only this instance for <hi>bat-nedle</hi>.</note> · and plaited hem togyderes</l>
<l> And put hem in a presse · and pyned<note>Bx.5.215: <hi>pyned</hi>: "subjected to stress" in the <hi>presse</hi>, which is apparently a stretching frame. Schmidt (1995), 72, glosses "tortured". The reading is supported by LW and alpha against the easier and less appropriate <hi>pynned</hi>, glossed as "kept fastened" in Kane (2005). <hi>A</hi> mss. vary, but <hi>Cx</hi> has <hi>pynned</hi>.</note> hem þerinne</l>
Bx.5.216KD.5.212
<l> Tyl ten ȝerdes or twelue · tolled<note>Bx.5.216: <hi>tolled</hi>: LM and alpha, supported by <hi>AC</hi>; other scribes supply <hi>hadde</hi>. <title>MED</title> v.(1) interprets <hi>tollen</hi> as "stretch out to" (OE *<hi>tollian</hi>), while Kane (2005) glosses "tease, coax". It varies with <hi>tolde</hi> in <hi>A</hi>; in <hi>C</hi> the X family reads <hi>tolde</hi>, the P family has <hi>tilled</hi>.</note> out threttene</l>
</lg>
<lb/>
<lg>
<l> ¶ My wyf was a webbe · and wollen cloth made</l>
<l> She spak to spynnesteres<note>Bx.5.218: <hi>spynnesteres</hi>: Alpha has the sg. <hi>A</hi> mss. are split, but <hi>Cx</hi> has the plural. Athlone adopts the sg. in all versions.</note> · to spynnen it oute</l>
<l> Ac þe pounde þat she payed by · poised<note>Bx.5.219: <hi>poised</hi>: A variant of <hi>peised</hi>, though given a separate entry by <title>MED</title>.</note> a quarter<note>Bx.5.219: <hi>quarter</hi>: Beta has <hi>quateroun</hi> (<title>MED</title> <hi>quartroun</hi>), but alpha is supported by <hi>AC</hi>.</note> more</l>
Bx.5.220KD.5.216
<l> Than myne owne auncere · who-so weyȝed treuthe</l>
</lg>
<lb/>
<lg>
<l> ¶ I bouȝte hir barly<note>Bx.5.221: <hi>barly</hi>: Though all <hi>B</hi> mss. except for R read <hi>barly malte</hi>, R is supported by <hi>AC</hi>. We assume that the easy addition was made by beta and F, with R as usual reproducing his exemplar.</note> · she brewe it to selle</l>
<l> Peny ale and podyng ale · she poured togideres</l>
<l> For laboreres and for<note>Bx.5.223: <hi>for</hi> (2): Not in MCrCOF, or in most <hi>AC</hi> mss.</note> low folke · þat lay by hym-<note>Bx.5.223: <hi>hym-</hi>: LWHmR have sg.: "was set aside on its own", So in <hi>Cx</hi>, while <hi>A</hi> mss. vary with the plural. The odd use of <hi>hym</hi> to refer to a barrel of ale (cf. Cr <hi>it</hi>) perhaps confused scribes.</note>selue</l>
</lg>
<lb/>
<lg>
Bx.5.224KD.5.220
<l> ¶ The best ale lay in my boure · or in my bedchambre</l>
<l> And who-so bummed þer-of · <app><rdg>bouȝte</rdg><rdg>he bouȝte</rdg></app><note>Bx.5.225: <hi>bouȝte / he bouȝte</hi>: Beta, without the pronoun, agrees with <hi>Ax</hi>; alpha and <hi>Cx</hi>, add the pronoun <hi>he</hi>.</note> it þer-after</l>
<l> A galoun for a grote · god wote na lesse</l>
<l> And ȝit it cam in cupmel · þis crafte my wyf vsed<note>Bx.5.227: <hi>vsed</hi>: Beta's past tense is supported by <hi>AC</hi>.</note></l>
Bx.5.228KD.5.224
<l> Rose þe regratere · was<note>Bx.5.228: <hi>was</hi>: Beta's past tense is supported by <hi>AC</hi>.</note> hir riȝte name</l>
<l> She hath holden hokkerye · al hire lyf-tyme</l>
</lg>
<lb/>
<lg>
<l> ¶ Ac I swere now so the ik · þat synne wil I lete</l>
<l> And neuere wikkedliche weye · ne wikke<note>Bx.5.231: <hi>wikke</hi> "dishonest". So LMW. Particularly in context, the <hi>wikked</hi> of alpha and others is likely to be a scribal substitution. <hi>A</hi> mss. exhibit some uncertainty over the word; the line is not in <hi>C</hi>.</note> chaffare vse</l>
Bx.5.232KD.5.228
<l> But wenden to walsyngham · and my wyf als</l>
<l> And bidde þe Rode of bromeholme · brynge me oute of dette</l>
</lg>
<lb/>
<lg>
<l> ¶ Repentestow<note>Bx.5.234: <hi>Repentestow</hi>: Representing the past tense, as more "properly" in WHmR. (See <title>MED</title> <hi>repenten</hi> for pa.t. <hi>repente</hi>). <hi>C</hi> mss. show the same forms. In <ref><hi>Bx</hi>.5.453</ref> WHm again "improve" to <hi>Repentedestow</hi>, falsely on that occasion. LM (= beta?) treat the verb as reflexive, but <hi>þe</hi> is not in R (= alpha?) or <hi>C</hi>. HmF include <hi>þe</hi> but drop the subject pronoun. For a similar set of variants see <ref><hi>Bx</hi>.5.453</ref> and note. The passage <ref><hi>Bx</hi>.5.234-308</ref> is not in <hi>A</hi>.</note> euere quod repentance · ne<note>Bx.5.234: <hi>ne</hi>: Another example of LMR presenting a superior reading, here agreeing with <hi>Cx</hi>. See Adams (2000), 176. The syntax is, however, odd.</note> restitucioun madest</l>
</lg>
<lb/>
<lg>
<l> ¶ Ȝus ones I was herberwed quod he · with an hep of chapmen</l>
Bx.5.236KD.5.232
<l> I roos whan þei were arest · and yrifled<note>Bx.5.236: <hi>yrifled</hi>: The form, though unmetrical (x x / x x / x), is supported by LR.</note> here males</l>
</lg>
<lb/>
<lg>
<l> ¶ That was no restitucioun quod repentance · but a robberes thefte</l>
<l> Þow haddest better<note>Bx.5.238: <hi>better</hi>: Although most mss. have <hi>be better</hi>, LR and original M give secure evidence for <hi>Bx</hi>, with <hi>be</hi> in the b-verse governing both <hi>worthy</hi> and <hi>hanged</hi>. <hi>Bx</hi> might have lost <hi>be</hi> in this context, with other scribes restoring it conjecturally, just as the M corrector did. Adams (2000), 183 judges this instead to be an accidental omission in RL. <hi>Cx</hi> revises the a-verse to "Thow wolt be hanged heye þerfore" (RK.6.238).</note> worthy · be hanged þerfore</l>
<l> Þan for al þat · þat þow hast here shewed</l>
</lg>
<lb/>
<lg>
Bx.5.240KD.5.235
<l> ¶ I wende ryflynge were restitucioun quod he · for<note>Bx.5.240: <hi>quod he for</hi>: Probably in response to the length of the line, MHmGF drop <hi>quod he</hi>, while CGO drop <hi>for</hi>. The line is not in <hi>AC</hi>.</note> I lerned neuere rede on boke</l>
<l> And I can no frenche in feith · but of þe ferthest ende of norfolke</l>
</lg>
<lb/>
<lg>
<l> ¶ Vsedestow euere vsurie quod repentaunce · in alle þi lyf-tyme</l>
</lg>
<lb/>
<lg>
<l> ¶ Nay sothly he seyde · saue in my ȝouthe</l>
Bx.5.244KD.5.239
<l> I lerned amonge lumbardes · [a lessoun and of iewes]<note>Bx.5.244: <hi>a lessoun and of iewes</hi>: The word-order in alpha is alliteratively correct, and R's reading is shared with <hi>Cx</hi>. Beta simplifies; F revises.</note></l>
<l> To wey pens with a peys · and pare þe heuyest</l>
<l> And lene it for loue of þe crosse · to legge a wedde and lese it</l>
<l> Suche dedes I did wryte · ȝif he his day breke</l>
Bx.5.248KD.5.243
<l> I haue mo maneres þorw rerages<note>Bx.5.248: <hi>rerages</hi>: Alpha's <hi>regages</hi> is an obvious error.</note> · þan þorw <foreign>miseretur & comodat</foreign></l>
</lg>
<lb/>
<lg>
<l> ¶ I haue lent lordes · and ladyes my chaffare</l>
<l> And ben her brocour after · and bouȝte it my-self</l>
<l> Eschaunges and cheuesances · with suche chaffare I dele</l>
Bx.5.252KD.5.247
<l> And lene folke þat lese wol · a lyppe at euery noble</l>
<l> And with lumbardes lettres · I ladde golde to Rome</l>
<l> And toke it by taille here<note>Bx.5.254: <hi>here</hi>: R alone reads <hi>þere</hi>, presumably by anticipation. Lines 247-51 and 254 have no parallel in <hi>AC</hi>.</note> · and tolde hem þere lasse</l>
</lg>
<lb/>
<lg>
<l> [¶]<note>Bx.5.255: <hi></hi>: The line is at the top of the page in L, and the rubricator presumably missed the paraph which is in WHmC and alpha, with a new line-group in M, to mark the opening of the speech.</note> Len[t]estow<note>Bx.5.255: <hi>Lentestow</hi>: L's present tense <hi>Lenestow</hi> is not supported by <hi>Cx</hi>.</note> euere lordes · for loue of her mayntenaunce</l>
</lg>
<lb/>
<lg>
Bx.5.256KD.5.251
<l> ¶ Ȝe I haue lent lordes · loued<note>Bx.5.256: <hi>loued</hi>: Alpha alone has <hi>quod he</hi>, but it is not in <hi>Cx</hi>. CrHmOF supply the subject pronoun <hi>that</hi>, as in <hi>Cx</hi>, but its omission, supported by LMWCGR, is idiomatic.</note> me neuere after</l>
<l> And haue ymade many a<note>Bx.5.257: <hi>a</hi>: Not in COR, but in <hi>Cx</hi>.</note> knyȝte · bothe mercere & drapere</l>
<l> Þat payed neuere for his prentishode · nouȝte a peire gloues</l>
</lg>
<lb/>
<lg>
<l> ¶ Hastow pite on pore men · þat mote nedes borwe</l>
</lg>
<lb/>
<lg>
Bx.5.260KD.5.255
<l> ¶ I haue as moche pite of<note>Bx.5.260: <hi>of</hi> (1): The reading of LR and WC, against <hi>on</hi>, likely to have been prompted by the phrase in the previous line.</note> pore men · as pedlere<note>Bx.5.260: <hi>pedlere</hi>: Alpha has <hi>þe pedlere</hi>, which is perhaps over-specific. For want of guidance from other versions, we follow copy-text.</note> hath of cattes</l>
<l> Þat wolde kille hem yf he cacche hem myȝte · for coueitise of here skynnes</l>
</lg>
<lb/>
<lg>
<l> ¶ Artow manlyche amonge þi neiȝbores · of þi mete and drynke</l>
</lg>
<lb/>
<lg>
<l> ¶ I am holden quod he as hende · as hounde is in kychyne</l>
Bx.5.264KD.5.259
<l> Amonges my neighbores namelich · such a name ich haue</l>
</lg>
<lb/>
<lg>
<l> ¶ Now god lene neure quod repentance · but þow repent þe rather</l>
<l> Þe grace on þis grounde · þi good wel to bisette</l>
<l> Ne þine [v]sue<note>Bx.5.267: <hi>vsue</hi>: This is the initial spelling in L, then corrected to <hi>ysue</hi>. R and possibly uncorrected M share the word, against <hi>heires</hi> in all others mss. except F with <hi>houswif</hi>! R spelling <hi>vssue</hi> explains F's reading as an attempt to make sense of <hi>vssue</hi> via <hi>hussie</hi>. Schmidt (1995), lxvii-iii, supposes this is shared error in LR; and see Adams (2000), 181-2. We take it to represent the reading of <hi>Bx</hi>, despite the fact that <hi>Cx</hi> has <hi>heyres</hi>, by error or revision.</note> after þe · haue ioye of þat þow wynnest</l>
Bx.5.268KD.5.263
<l> Ne þi excecutours wel bisett · þe siluer þat þow hem leuest</l>
<l> And þat was wonne with wronge · with wikked men be despended</l>
<l> For were I frere of þat hous · þere gode faith and charite is</l>
<l> I nolde<note>Bx.5.271: <hi>nolde</hi>: Beta shares this reading with the P family of <hi>C</hi>, against alpha's <hi>wolde nouȝt</hi>. At RK.7.202 the P family again has <hi>nolde</hi>, as do all <hi>B</hi> mss. (<ref><hi>Bx</hi>.5.578</ref>) against <hi>ne wol(d)</hi> in the X family.</note> cope vs with þi catel · ne owre kyrke<note>Bx.5.271: <hi>kyrke</hi>: Alpha has the non-alliterating <hi>cherche</hi>.</note> amende</l>
Bx.5.272KD.5.267
<l> Ne haue a peny to my pitaunce of þyne<note>Bx.5.272: <hi>of þyne</hi>: Well supported by LMCOR, though omitted by beta2 and GF, presumably because of its metrical clumsiness. The <hi>C</hi> reviser's text evidently had the phrase, for the line is rewritten as <hi>Ne take a meles mete of þyn and myn herte hit wiste</hi> (RK.6.289). LC, confused by its clumsiness, punctuate before the phrase; MOR punctuate after it.</note> [·] bi my soule hele<note>Bx.5.272: <hi>bi my soule hele</hi>: Clearly <hi>Bx</hi>, despite its lack of alliteration. Beta2 and G have <hi>so God my soul help</hi>, with W increasing alliteration by altering the verb to <hi>saue</hi>.</note></l>
<l> For þe best boke in owre hous · þeiȝe brent golde were þe leues</l>
<l> And I wyst wytterly · þow were suche as þow telle[st]<note>Bx.5.274: <hi>tellest</hi>: L has <hi>telleth</hi>, though the line is marked for correction. Following this line CO have the line <hi>Or elles þat I kouþe knowe it by any kynnes wise</hi>, and it is supplied in Cr<hi>23</hi>. It is the first of four spurious lines recorded by CO in this passus; see notes to <ref><hi>Bx</hi>.5.308</ref>, <ref>342</ref>, <ref>580</ref>.</note></l>
<l> <foreign>Seruus es alterius · cum fercula pinguia queris</foreign></l>
Bx.5.276KD.5.269β
<l> <foreign>Pane tuo pocius · vescere liber eris ·</foreign></l>
</lg>
<lb/>
<lg>
<l> ¶ Thow art an vnkynde creature · I can þe nouȝte assoille</l>
<l> Til þow make restitucioun<note>Bx.5.278: <hi>restitucioun</hi>: Only R adds <hi>quod repentaunce</hi> as a second alliterating stave in the a-verse. The <hi>Bx</hi> reading without it seems to have been the basis for the revision in <hi>C</hi> alliterating on <hi>ymad ... myhte ... men</hi>.</note> · and rekne with hem alle</l>
<l> And sithen þat resoun rolle it · in þe regystre of heuene</l>
Bx.5.280KD.5.273
<l> That þow hast made vche man good · I may þe nouȝte assoille</l>
<l> <foreign>Non dimittitur peccatum · donec<note>Bx.5.281: <hi><foreign>donec</foreign></hi>: Perhaps alpha and Cr alter the reading to <hi>nisi</hi> on the basis of a well-known maxim of canon law; see Alford (1992), 46. In <ref><hi>Bx</hi>.17.316</ref>, RK.6.257a and RK.19.290a the quotation is in many mss. abbreviated to <hi>Numquam dimittitur peccatum &c</hi>. (the form recorded in the <title>Speculum Christiani</title>), with some mss. expanding to <hi>donec</hi>.... and others to <hi>nisi</hi>.... On this occasion the former is marginally more appropriate than the latter ("until" vs. "unless"; cf. l. <ref>278</ref> <title>Til</title>). Not in <hi>AC</hi>.</note> restituatur ablatum &c</foreign></l>
</lg>
<lb/>
<lg>
<l> ¶ For alle þat hath<note>Bx.5.282: <hi>hath</hi>: LMR have the sg., "everyone who has", smoothed to the pl. after <hi>alle</hi> in other mss. See next line.</note> of þi good · haue god my trouthe</l>
<l> Is<note>Bx.5.283: <hi>Is</hi>: LMR continue with the sg., as does F, although rewriting the line. Many <hi>C</hi> mss. have <hi>Is</hi> here, though they have the pl. in the previous line.</note> hold[ynge]<note>Bx.5.283: <hi>holdynge</hi>: "obliged". R's curious pres. ppl. form, revised out by F, must represent alpha, since it is also the form in the X family of <hi>C</hi>. Indeed R's whole line is reproduced by that family. See next note.</note> at þe heighe dome · to helpe þe restitue<note>Bx.5.283: <hi>restitue</hi>: Beta has <hi>to</hi> + infinitive, but R is supported by <hi>Cx</hi>.</note></l>
Bx.5.284KD.5.276
<l> And who-so leueth nouȝte þis be soth · loke in þe<note>Bx.5.284: <hi>þe</hi>: Beta supported by <hi>Cx</hi>, against R's <hi>a</hi>. F has four unsupported lines in place of this one.</note> sauter glose</l>
<l> In <foreign>miserere mei deus</foreign> · where I mene treuthe</l>
<l> <foreign>Ecce enim veritatem dilexisti &c</foreign><note>Bx.5.286: The line that follows this in alpha, <hi>Þere is no laborere wolde leue with hem . þat knoweth peres þe plowman</hi>, appears to be spurious. Its reference to <hi>peres þe plowman</hi> who has not yet been introduced "relates the immediate discussion to a consideration earlier and subsequently expressed" (KD, p. 193).</note></l>
</lg>
<lb/>
<lg>
<l> ¶ Shal<note>Bx.5.287: <hi>Shal</hi>: Alpha begins <hi>For schal</hi>, tying this line to the added line preceding it. There is no parallel for this passage in <hi>AC</hi>.</note> neuere werkman in þis worlde · þryue wyth þat þow wynnest</l>
Bx.5.288KD.5.278
<l> <foreign>Cum sancto sanctus eris</foreign> · construe me þat on englische</l>
</lg>
<lb/>
<lg>
<l> ¶ Thanne wex þat<note>Bx.5.289: <hi>þat</hi>: Beta, against alpha's <hi>þe</hi>. We follow copy-text.</note> shrewe in wanhope · and walde haue hanged himself</l>
<l> Ne hadde repentaunce þe rather · reconforted<note>Bx.5.290: <hi>reconforted</hi>: R (= alpha?) loses alliteration with <hi>conforted</hi> (as do HmG independently). F repairs with <hi>reersyd</hi>.</note> hym in þis manere</l>
<l> Haue mercye in þi mynde · and with þi mouth biseche it</l>
Bx.5.292KD.5.281.1
<l> For goddes mercye is more · þan alle hise other werkes</l>
<l> <foreign>Misericordia eius super omnia opera eius &c ·<note>Bx.5.293: The Latin line is only in LMOR. It is translated in l. <ref>292</ref> above (which KD exclude), but that does not seem sufficient explanation for other mss. to lose it. For the first <hi>eius</hi> (as in the Vulgate) R reads <hi>domini</hi>, which gives a precise translation of l. <ref>292</ref> and may be right. The verse is quoted again as <ref><hi>Bx</hi>.17.325</ref>, where both R and F read <hi>domini</hi>.</note></foreign></l>
</lg>
<lb/>
<lg>
<l> ¶ And al þe wikkednesse in þis worlde · þat man myȝte worche or thynke</l>
<l> Ne is no more to þe mercye of god · þan in þe see a glede</l>
Bx.5.296KD.5.283α
<l> <foreign>Omnis iniquitas quantum ad misericordiam dei · est [quasi]<note>Bx.5.296: <hi><foreign>quasi</foreign></hi>: Omitted in L and added in M, but in all other <hi>B</hi> and <hi>C</hi> mss. For the source see Alford (1992), 47.</note> sintilla in medio maris</foreign></l>
</lg>
<lb/>
<lg>
<l> ¶ For-þi haue mercy in þi mynde · and marchandise leue it</l>
<l> For þow hast no good grounde · to gete þe with a wastel</l>
<l> But if it were with thi tonge · or ellis with þi two hondes</l>
Bx.5.300KD.5.287
<l> For þe good þat þow hast geten · bigan al with falsehede</l>
<l> And as longe as þow lyuest þer-with · þow ȝeldest nouȝte but borwest</l>
</lg>
<lb/>
<lg>
<l> ¶ And if þow wite neuere to whiche · ne whom to restitue<note>Bx.5.302: <hi>restitue</hi>: The form is supported by <hi>Cx</hi> (RK.6.344) against GR <hi>restitute</hi> and CrF <hi>restore</hi>. See l. <ref>283</ref>.</note></l>
<l> Bere it to þe bisschop · and bidde hym of his grace</l>
Bx.5.304KD.5.291
<l> Bisette it hym-selue · as best is for þi soule</l>
<l> For he shal answere for þe · at þe heygh dome</l>
<l> For þe and for many mo · þat man shal ȝif a rekenynge</l>
<l> What he lerned ȝow in lente · leue þow none other</l>
Bx.5.308KD.5.295
<l> And what he lent ȝow of owre lordes good · to lette ȝow fro synne<note>Bx.5.308: The line following this in CO is obviously scribal, to compensate for the error <hi>his goode</hi> for <hi>owre lordes goode</hi>.</note></l>
</lg>
<lb/>
<lg>
<foreign>Gula ·</foreign>
<l> ¶ Now bigynneth glotoun · for to go to schrifte</l>
<l> And kaires hym to kirke-ward · his coupe to schewe</l>
</lg>
<lb/>
<lg>
<l> ¶ Ac Beton þe brewestere · bad hym good morwe</l>
Bx.5.312KD.5.299
<l> And axed of hym with þat · whiderward he wolde</l>
</lg>
<lb/>
<lg>
<l> ¶ To holicherche quod he · forto here masse</l>
<l> And sithen I wil be shryuen · and synne namore</l>
</lg>
<lb/>
<lg>
<l> ¶ I haue gode ale gossib quod she · glotown wiltow assaye</l>
Bx.5.316KD.5.303
<l> <note>Bx.5.316: W and alpha have a paraph here to mark the start of direct speech.</note>Hastow auȝte in þi purs · any hote spices</l>
</lg>
<lb/>
<lg>
<l> ¶ I haue peper and piones quod he<note>Bx.5.317: <hi>he</hi>: We preserve copy-text's form for "she". It was beta, though, not alpha, since it is the original form in M, and also survives in CrCG (by misunderstanding?).</note> · and a pounde of garlike</l>
<l> A ferthyngworth of fenel-seed · for fastyngdayes</l>
</lg>
<lb/>
<lg>
<l> ¶ Þanne goth glotoun in · and grete othes after</l>
Bx.5.320KD.5.307
<l> Cesse þe souteresse · sat on þe benche</l>
<l> Watte þe warner · and his wyf bothe</l>
<l> Tymme<note>Bx.5.322: <hi>Tymme</hi>: The probable source of the variants <hi>Tomme</hi> and <hi>Symme</hi>. <hi>A</hi> and <hi>C</hi> mss. vary similarly; see note to Kane's <title>A Version</title>, p. 443.</note> þe tynkere · and tweyne of his prentis<note>Bx.5.322: <hi>prentis</hi>: The plural form of a romance loan ending in a sibilant.</note></l>
<l> Hikke þe hakeneyman · and hugh þe nedeler</l>
Bx.5.324KD.5.311
<l> Clarice of cokkeslane · and þe clerke of þe cherche</l>
<l> <note>Bx.5.325-6: We follow the order of alpha; beta reverses the two lines. Line 325 is not in <hi>Ax</hi>; <hi>Cx</hi> has the alpha order, although adding a line between these two.</note> [Sire<note>Bx.5.325: <hi>Sire</hi>: Alpha begins <hi>And sire</hi>, but without support from <hi>Cx</hi>.</note> Piers of Pridie · and Peronelle of Flaundres</l>
<l> Dawe þe dykere [·] and a dozeine other]</l>
<l> A ribibour a ratonere · a rakyer of chepe</l>
Bx.5.328KD.5.315
<l> A ropere a redyngkyng · and Rose þe dissheres<note>Bx.5.328: <hi>dissheres</hi>: See <title>MED</title> <hi>disheresse</hi>. Alpha supposes it is genitive of <hi>dishere</hi>, and so adds <hi>douȝter</hi>. <hi>AC</hi> have <hi>disshere</hi>.</note></l>
<l> Godfrey of garlekehithe · and gryfin<note>Bx.5.329: <hi>gryfin</hi>: Both this and R's <hi>grifyth</hi> are forms of Welsh Gruffudd. <hi>Ax</hi> and the P family of <hi>C</hi> have the former, while the X family has the latter.</note> þe walsh</l>
<l> And vpholderes an hepe · erly bi þe morwe</l>
<l> Geuen glotoun with glad chere · good ale to hansel</l>
</lg>
<lb/>
<lg>
Bx.5.332KD.5.319
<l> ¶ Clement þe cobelere · cast of his cloke</l>
<l> And <app><rdg>atte</rdg><rdg>to þe</rdg></app><note>Bx.5.333: <hi>atte / to þe</hi>: Beta's <hi>atte</hi> is the reading of <hi>Ax</hi> (K.5.169). R's <hi>to þe</hi> is the reading of <hi>Cx</hi> (RK.6.377).</note> new faire · nempned<note>Bx.5.333: <hi>nempned</hi>: The reading of R. Beta and F (by convergence?) add the pronoun <hi>he</hi>, but it is not supported by <hi>AC</hi>.</note> it to selle</l>
<l> Hikke þe hakeneyman · hitte his hood after</l>
<l> And badde bette þe bochere · ben<note>Bx.5.335: <hi>ben</hi>: Alpha's <hi>to ben</hi> is found in some <hi>AC</hi> mss.</note> on his side</l>
Bx.5.336KD.5.323
<l> Þere were chapmen ychose · þis chaffare to preise</l>
<l> Who-so haueth<note>Bx.5.337: <hi>haueth</hi>: WHmO have the past tense, as in <hi>AC</hi>.</note> þe hood · shuld<note>Bx.5.337: <hi>shuld</hi>: M has <hi>shal</hi>, though it is an addition above the line; R has <hi>schul</hi>, which is probably present tense, though <title>LALME</title> 4, 41, records examples of this as a past-tense form. <hi>AC</hi> have the past.</note> haue amendes of þe cloke</l>
</lg>
<lb/>
<lg>
<l> ¶ [Þo]<note>Bx.5.338: <hi>Þo</hi>: The reading of R (= alpha, with F glossing as <hi>Þan</hi>.) Beta reads <hi>Two</hi>. Choice is difficult, since <hi>A</hi> and <hi>C</hi> mss. also have both readings, with <hi>Tho</hi> the majority <hi>A</hi> reading, and <hi>Two</hi> the majority <hi>C</hi> reading. Perhaps scribes felt the lack of a subject; note that F supplies a pronoun.</note> risen vp in rape · and rouned togideres</l>
<l> And preised þese<note>Bx.5.339: <hi>þese</hi>: So beta and the X family of <hi>C</hi>, but alpha's <hi>þe</hi> is the reading of <hi>Ax</hi> and the P family. We follow copy-text.</note> penyworthes · apart bi hem-selue<note>Bx.5.339: <hi>hem-selue</hi>: Original M, Hm and alpha have <hi>hym-</hi>, as do a few <hi>C</hi> mss.</note></l>
Bx.5.340KD.5.328
<l> Þei couth nouȝte bi her conscience · acorden in treuthe</l>
<l> Tyl Robyn þe ropere · [arise] þe southe<note>Bx.5.341: <hi>arise þe southe</hi>: R's nonsensical b-verse explains the attempts by F and beta to correct it, but it is a corruption of <hi>Cx</hi> <hi>aryse they bisouhte</hi> (RK.6.387). <hi>Ax</hi> has <hi>was red to arisen</hi> (K.5.178). See KD, p. 90; Hanna (1996), 217.</note></l>
<l> And nempned hym for a noumpere · þat no debate [w]ere<note>Bx.5.342: <hi>were</hi>: The reading of HmGR, and probably the alpha reading (F rewrites). It agrees with the majority of <hi>A</hi> mss. and all <hi>C</hi> mss. but one. Beta's <hi>nere</hi> is perhaps due to alliterative attraction.</note><note>Bx.5.342: Following this line, a non-alliterating line occurs in Cr<hi>23</hi> and OC.</note></l>
</lg>
<lb/>
<lg>
<l> ¶ Hikke þe hostellere · hadde þe cloke</l>
Bx.5.344KD.5.332
<l> In couenaunte þat Clement [·] shulde þe cuppe fille</l>
<l> And haue hikkes hode hostellere · and holde hym yserued</l>
<l> And who-so repented rathest · shulde arise after</l>
<l> And grete sire glotoun [·] with a galoun ale</l>
</lg>
<lb/>
<lg>
Bx.5.348KD.5.336
<l> ¶ Þere was laughyng and louryng · and let go þe cuppe</l>
<l> And seten so til euensonge · and songen vmwhile</l>
<l> Tyl glotoun had yglobbed · a galoun an a Iille</l>
<l> His guttis gunne to godly<note>Bx.5.351: <hi>godly</hi>: The form is supported for <hi>Bx</hi> by LMR, and 11 <hi>C</hi> mss. have -d- forms. See <title>MED</title> <hi>gothelen</hi>, and the variants for <ref><hi>Bx</hi>.13.95</ref>, where LR again have <hi>god(e)le</hi>.</note> · as two gredy sowes</l>
Bx.5.352KD.5.341
<l> He pissed a potel · in a pater noster while</l>
<l> And blew his rounde ruwet · at his rigge-bon[es]<note>Bx.5.353: <hi>rigge-bones</hi>: Scribes vary in treating one or both parts of the compound as appositive or genitive. Most <hi>AC</hi> mss. support <hi>rygbones</hi>.</note> ende</l>
<l> That alle þat herde þat horne · held her nose after</l>
<l> And wissheden it had be wexed · with a wispe of firses</l>
</lg>
<lb/>
<lg>
Bx.5.356KD.5.345
<l> ¶ He myȝte neither steppe ne stonde · er he his staffe hadde</l>
<l> And þanne gan he go<note>Bx.5.357: <hi>go</hi>: The infinitive without <hi>to</hi>, as in LMC and alpha, is supported by <hi>Cx</hi> and most <hi>A</hi> mss.</note> · liche a glewmannes bicche</l>
<l> Somme-tyme aside · and somme-tyme arrere</l>
<l> As who-so leyth lynes · forto lacche foules</l>
</lg>
<lb/>
<lg>
Bx.5.360KD.5.349
<l> ¶ And<note>Bx.5.360: <hi>And</hi>: R's <hi>Ac</hi> has no support from <hi>AC</hi>, and may be picked up from l. <ref>364</ref>.</note> whan he drowgh to þe dore · þanne dymmed his eighen</l>
<l> He trembled<note>Bx.5.361: <hi>trembled</hi>: The word is supported for <hi>Bx</hi> by LR, with the M corrector revising to <hi>stombled</hi> in line with beta1 to make better sense in context. The word that lies behind this must be the rare verb <hi>thromblede</hi>, "tripped", as in the X family of <hi>C</hi>, which causes confusion in all versions. <title>OED</title> needlessly divides <hi>thrumble</hi> into two verbs, though it is true that the predominant sense recorded is "to crowd in, to jostle". The line in L is marked for correction, but probably this refers to <hi>an</hi> for <hi>and</hi>.</note> on þe thresshewolde · an threwe to þe erthe</l>
<l> Clement þe cobelere · cauȝte hym bi þe myddel</l>
<l> For to lifte hym alofte · and leyde him on his knowes</l>
Bx.5.364KD.5.353
<l> Ac glotoun was a gret cherle · and a grym in þe liftynge</l>
<l> And coughed vp a caudel · in clementis lappe</l>
<l> Is non so hungri hounde · in Hertford-schire</l>
<l> Durst lape of þe<note>Bx.5.367: <hi>þe</hi>: LMR against <hi>þat</hi> or <hi>þo</hi> in other mss. (F rewrites). But <hi>Cx</hi> has <hi>þat</hi>.</note> leuynges<note>Bx.5.367: <hi>leuynges</hi>: R has the sg., as does <hi>Cx</hi>, and may be right. See next note.</note> · so vnlouely [it]<note>Bx.5.367: <hi>it</hi>: Supported by MHm and alpha as well as <hi>Cx</hi>. Presumably other beta scribes altered to plural in concord with the plural subject. See <title>MED</title> <hi>hit</hi> 4a(b) for its use as the grammatical subject relating to an abstract plural.</note> smauȝte</l>
</lg>
<lb/>
<lg>
Bx.5.368KD.5.357
<l> ¶ With al þe wo of þis worlde · his wyf and his wenche</l>
<l> Baren hym<note>Bx.5.369: <hi>hym</hi> (1): Beta has <hi>hym home</hi>, but alpha is supported by <hi>AC</hi>.</note> to his bedde · and brouȝte hym þer-inne</l>
<l> And after al þis excesse · he had an accidie</l>
<l> Þat he slepe saterday and sonday · til sonne ȝede to reste</l>
Bx.5.372KD.5.361
<l> <note>Bx.5.372: A paraph would be appropriate here, but only WR have one.</note>Þanne waked he of his wynkyng · and wiped his eyghen</l>
<l> Þe fyrste worde þat he warpe · was where is þe bolle</l>
<l> His witte<note>Bx.5.374: <hi>witte</hi>: The reading of all <hi>B</hi> mss. except <hi>wif</hi> in CrW and revised M. <hi>Ax</hi> has <hi>wife</hi>, but <hi>Cx</hi> has <hi>His wyf and his inwit</hi>. See Schmidt (1995), 371, who suggests that <hi>Bx</hi> corrupted the reading represented in <hi>Cx</hi>.</note> gan edwite hym þo · how wikkedlich he lyued</l>
<l> And repentance riȝte so · rebuked hym þat tyme</l>
</lg>
<lb/>
<lg>
Bx.5.376KD.5.365
<l> ¶ As þow with wordes and werkes · hast wrouȝte yuel in þi lyue</l>
<l> Shryue þe and be [a]shamed<note>Bx.5.377: <hi>ashamed</hi>: CrHmO and alpha have the prefix, as do the equivalent lines in <hi>AC</hi> (K.5.206, RK.6.422). There are only two other examples of either form: <ref><hi>Bx</hi>.3.191</ref> <hi>shamedest</hi>, and <ref><hi>Bx</hi>.20.284</ref> <hi>ashamed</hi>.</note> þer-of · and shewe it with þi mouth</l>
</lg>
<lb/>
<lg>
<l> ¶ I glotoun quod þe gome · gylti me ȝelde</l>
<l> [Of]<note>Bx.5.379: <hi>Of</hi>: In the absence of F which misses the line, the presumed alpha reading which is also that of <hi>Cx</hi>. The passage <ref><hi>Bx</hi>.5.378-89</ref> is not in <hi>A</hi>.</note> þat I haue trespassed with my tonge · I can nouȝte telle how ofte</l>
Bx.5.380KD.5.369
<l> Sworen goddes soule [and his sydes]<note>Bx.5.380: <hi>and his sydes</hi>: R is supported by the X family of <hi>C</hi>. The P family drops the pronoun as does F, while beta omits the phrase altogether.</note> · and so [help me god]<note>Bx.5.380: <hi>so help me god</hi>: Alpha's word-order is also that of <hi>Cx</hi>. M's agreement, with <hi>selpe me god</hi>, is probably coincidental. Otherwise beta, having lost the second alliterating stave in the a-verse, brings the noun forward so that the line alliterates on /g/ rather than /s/.</note> and halidom</l>
<l> Þere no nede ne<note>Bx.5.381: <hi>ne</hi>: Dropped by beta2 (CrWHm), and GF. <hi>C</hi> mss. vary.</note> was · nyne hundreth tymes</l>
</lg>
<lb/>
<lg>
<l> ¶ And ouer-seye me at my sopere · and some-tyme at nones</l>
<l> Þat I glotoun girt it<note>Bx.5.383: <hi>it</hi>: LR and WHmG, but omitted in MCrCO (F rewrites as <hi>vp a-geyn</hi>). The line is not in <hi>C</hi>.</note> vp · er I hadde gone a myle</l>
Bx.5.384KD.5.373
<l> And yspilte þat myȝte be spared · and spended on somme hungrie</l>
<l> Ouerdelicatly on fastyng-dayes · drunken and eten bothe</l>
<l> And sat some-tyme so longe þere · þat I slepe and ete at ones</l>
<l> For loue of tales in tauernes · to drynke þe more I dyned<note>Bx.5.387: <hi>to drynke þe more I dyned</hi>: R and F are flummoxed by the sense. Bennett (1972) translates the line: "I dined in taverns, where I could hear tales, in order to satisfy my thirst more easily". Lines 386-89 have no parallel in <hi>AC</hi>.</note></l>
Bx.5.388KD.5.377
<l> And hyed to þe mete er none · whan fastyng-dayes were</l>
</lg>
<lb/>
<lg>
<l> ¶ This shewyng shrifte quod repentance · shal be meryte to þe</l>
</lg>
<lb/>
<lg>
<l> ¶ And þanne gan glotoun grete<note>Bx.5.390: <hi>grete</hi>: HmR have <hi>to grete</hi>.</note> · and gret doel to make</l>
<l> For his lither lyf · þat he lyued hadde</l>
Bx.5.392KD.5.381
<l> And avowed fast<note>Bx.5.392: <hi>fast</hi>: The idiomatic use of infinitive without <hi>to</hi> is found in LRG and original M. Nevertheless, <hi>Ax</hi> has <hi>to faste</hi>. Donaldson (1955), 198, takes <hi>faste</hi> as adverb. There is no parallel in <hi>C</hi>. See Adams (2000), 182.</note> · for hunger or for thurst</l>
<l> Shal neuere fisshe on þe<note>Bx.5.393: <hi>þe</hi>: Supported by LR only. <hi>AC</hi> mss. are split.</note> fryday · defien in my wombe</l>
<l> Tyl abstinence myn aunte · haue ȝiue me leue</l>
<l> And ȝit haue I hated hir · al my lyf-tyme</l>
</lg>
<lb/>
<lg>
<foreign>Accidia</foreign>
Bx.5.396KD.5.385
<l> ¶ Þanne come sleuthe al bislabered · with two slym[ed]<note>Bx.5.396: <hi>slymed</hi>: The form in alpha and many <hi>C</hi> mss., with beta adopting the commoner form <hi>slymy</hi>. The passage <hi>Bx</hi>.5.396-453 is not in <hi>A</hi>.</note> eiȝen</l>
<l> I most sitte seyde þe segge · or elles shulde I nappe</l>
<l> I may nouȝte stonde ne stoupe · ne with-oute a<note>Bx.5.398: <hi>a</hi>: HmC and alpha drop the article. <hi>C</hi> mss. are split (RK.7.3).</note> stole knele</l>
<l> Were I brouȝte abedde · but if my taille-ende it made</l>
Bx.5.400KD.5.389
<l> Sholde no ryngynge do me ryse · ar I were rype to dyne</l>
<l> <note>Bx.5.401: Alpha has a paraph.</note>He bygan benedicite with a bolke · and his brest knocked</l>
<l> And roxed and rored · and rutte atte laste</l>
</lg>
<lb/>
<lg>
<l> [¶]<note>Bx.5.403: <hi></hi>: The paraph in WHmC (with a new line-group in M) and alpha marks the speech-opening. Cf. l. <ref>255</ref>.</note> What awake renke quod repentance · and rape þe to shrifte</l>
</lg>
<lb/>
<lg>
Bx.5.404KD.5.393
<l> ¶ If I shulde deye bi þis day<note>Bx.5.404: <hi>day</hi>: Alpha adds <hi>quod he</hi>, as in the P family of <hi>C</hi>. Beta reads as the X family. Schmidt (1995), 372, argues that the phrase is required for its liaison alliteration on /d/. <hi>Cx</hi> has a revised b-verse.</note> · me liste nouȝte to loke</l>
<l> I can nouȝte parfitly my pater noster · as þe prest it syngeth</l>
<l> But I can rymes of Robyn hood · and Randolf erle of Chestre</l>
<l> Ac neither of owre lorde ne of owre lady · þe leste þat euere was made</l>
</lg>
<lb/>
<lg>
Bx.5.408KD.5.397
<l> ¶ I haue made vowes<note>Bx.5.408: <hi>vowes</hi>: This is the form of <hi>Cx</hi>, against <hi>avowes</hi> in WHmF.</note> fourty · and forȝete hem on [morwe]<note>Bx.5.408: <hi>on morwe</hi>: O, F (= alpha?) and corrected M have this form, which receives support from <hi>Cx</hi> <hi>amorwe(n)</hi>. R has uniquely <hi>or morwe</hi>, C has <hi>at morwe</hi>, LCrWHmG supply the definite article.</note></l>
<l> I parfourned neure penaunce · as þe prest me hiȝte</l>
<l> Ne ryȝte sori for my synnes · ȝet<note>Bx.5.410: <hi>ȝet</hi>: F supplies <hi>soþly</hi> for the alliteration of the b-verse, which <hi>Cx</hi> rewrites.</note> was I neuere</l>
<l> And ȝif I bidde any bedes · but if it be in wrath</l>
Bx.5.412KD.5.401
<l> Þat I telle with my tonge · is two myle fro myne herte</l>
<l> I am occupied eche day · haliday and other</l>
<l> With ydel tales atte ale · and otherwhile in cherches<note>Bx.5.414: <hi>cherches</hi>: Beta2, G and alpha have the sg., but the plural is supported by <hi>Cx</hi>.</note></l>
<l> Goddes peyne and his passioun · ful selde þynke I þere-on</l>
</lg>
<lb/>
<lg>
Bx.5.416KD.5.405
<l> ¶ I visited neuere fieble<note>Bx.5.416: <hi>fieble</hi>: There is no support for alpha's non-alliterating <hi>seke</hi>. <hi>Cx</hi> agrees with beta.</note> men · ne fettered folke in puttes</l>
<l> I ha[dd]e<note>Bx.5.417: <hi>hadde</hi>: Alpha is supported by <hi>Cx</hi> against beta's present tense.</note> leuere here an harlotrie · or a somer game of souteres</l>
<l> Or lesynges to laughe [of]<note>Bx.5.418: <hi>of</hi>: Alpha is supported by <hi>Cx</hi>, though in a revised line in which the phrase <hi>to lauhen of</hi> comes finally (RK.7.22).</note> · and belye my neighbore<note>Bx.5.418: <hi>neighbore</hi>: LM and alpha agree on the sg. (though F has another noun).</note></l>
<l> Þan<note>Bx.5.419: <hi>Þan</hi>: Beta's conjunction (supported by <hi>Cx</hi>) is necessary to the sense.</note> al þat euere Marke made · Mathew Iohan & lucas</l>
Bx.5.420KD.5.409
<l> And vigilies and fastyng-dayes · alle þise late I passe</l>
<l> And ligge abedde in lenten · an my lemman in myn armes</l>
<l> Tyl matynes and masse be do · and þanne go to þe freres</l>
<l> Come I to <foreign>ite missa est</foreign> · I holde me yserued</l>
Bx.5.424KD.5.413
<l> I [am]<note>Bx.5.424: <hi>am</hi>: Alpha and <hi>Cx</hi> against <hi>nam</hi>, probably beta.</note> nouȝte shryuen some-tyme · but if<note>Bx.5.424: <hi>if</hi>: LR and beta2 (CrWHm) and <hi>Cx</hi>, but omitted by MCGOF.</note> sekenesse it make</l>
<l> Nouȝt tweies in two ȝere · and þanne vp gesse I s[ch]ryue me</l>
</lg>
<lb/>
<lg>
<l> ¶ I haue be prest and parsoun · passynge thretti wynter</l>
<l> Ȝete can I neither solfe ne synge · ne seyntes lyues rede</l>
Bx.5.428KD.5.417
<l> But I can fynde in a felde · or in a fourlonge an hare</l>
<l> Better þan in <foreign>beatus vir ·</foreign> or in <foreign>beati omnes</foreign></l>
<l> Construe [it] clause[mel]<note>Bx.5.430: <hi>it clausemel</hi>: Beta's <hi>oon clause wel</hi> is prompted by the unusual compound, "clause by clause". KD omit <hi>it</hi>, on the grounds that it lacks a referent (p. 186). Note that the L scribe left spaces for <hi>oon</hi> and <hi>wel</hi> which were later written in, suggesting that he was unable to read or make sense of his copy. <hi>Cx</hi> rewrites, blending <hi>Bx</hi>.5.430 and 432 as one line, RK.7.34.</note> · and kenne it to my parochienes</l>
<l> I can holde louedayes · and<note>Bx.5.431: <hi>and</hi>: Alpha reads <hi>or</hi>. <hi>Cx</hi> has <hi>and</hi> in a revised line.</note> here a Reues rekenynge</l>
Bx.5.432KD.5.421
<l> Ac in canoun ne in þe<note>Bx.5.432: <hi>þe</hi>: LMR and G, but omitted in others. <hi>Cx</hi>'s revised line offers no evidence.</note> decretales · I can nouȝte rede a lyne</l>
</lg>
<lb/>
<lg>
<l> ¶ Ȝif I bigge and borwe it<note>Bx.5.433: <hi>it</hi> (1): Despite the <hi>auȝt</hi> of beta2 (CrWHm) and G, <hi>Bx</hi> appears to have read <hi>it</hi>, and shares the reading with the X family of <hi>C</hi>. The P family has the more specific <hi>ouht</hi>.</note> · but ȝif it be ytailled</l>
<l> I forȝete it as ȝerne · and ȝif men me it axe</l>
<l> Sixe sithes or seuene · I forsake it with othes</l>
Bx.5.436KD.5.425
<l> And þus tene I<note>Bx.5.436: <hi>tene I</hi>: So beta; the phrase is reversed in alpha. <hi>C</hi> has <hi>haue y tened</hi> and so offers no grounds for choice.</note> trewe men · ten hundreth tymes</l>
</lg>
<lb/>
<lg>
<l> ¶ And my seruauntz some-tyme · her salarye is bihynde</l>
<l> Reuthe is to here rekenynge<note>Bx.5.438: <hi>rekenynge</hi>: So LM and alpha, supported by the X family of <hi>C</hi>. Other <hi>B</hi> mss. and the P family add the definite article.</note> · whan we shal rede acomptes</l>
<l> So with wikked wille and wraththe · my werkmen I paye</l>
</lg>
<lb/>
<lg>
Bx.5.440KD.5.429
<l> ¶ Ȝif any man doth me a benfait · or helpeth me at nede</l>
<l> I am vnkynde aȝein his curteisye · and can nouȝte vnderstonde it</l>
<l> For I haue and haue hadde · some-dele haukes maneres</l>
<l> I [am]<note>Bx.5.443: <hi>am</hi>: The form is supported by <hi>Cx</hi>. LMO have <hi>nam</hi>. Cf. ll. <ref>424</ref>, <ref>459</ref>.</note> nouȝte lured with loue · but þere ligge auȝte vnder þe<note>Bx.5.443: <hi>þe</hi>: Alpha omits, as does the P family of <hi>C</hi>. The X family, however, has the article.</note> thombe</l>
</lg>
<lb/>
<lg>
Bx.5.444KD.5.433
<l> ¶ The kyndenesse þat myne euene-cristene · kidde me farnere</l>
<l> Sixty sythes I sleuthe · haue fo[r]ȝete it sith</l>
<l> In speche and in sparynge of speche · yspilte many a tyme</l>
<l> Bothe flesche & fissche · and many other vitailles</l>
Bx.5.448KD.5.437
<l> Bothe bred and ale · butter melke and chese</l>
<l> Forsleuthed in my seruyse · til it myȝte serue noman</l>
</lg>
<lb/>
<lg>
<l> ¶ I ran aboute in ȝouthe · and ȝaf me nouȝte to lerne</l>
<l> And euere sith be<note>Bx.5.451: <hi>be</hi>: So LR and original M. The M corrector inserts <hi>haue</hi>, in line with other mss. (<hi>haue I</hi> in WHmO). <hi>Cx</hi> includes <hi>haue I be</hi> in a reordered a-verse.</note> beggere · for my foule sleuthe</l>
Bx.5.452KD.5.440α
<l> <foreign>Heu michi qu[od]<note>Bx.5.452: <hi><foreign>quod</foreign></hi>: For the same quotation see <ref><hi>Bx</hi>.1.143</ref>. In both cases alpha's <hi>quod</hi> for beta's <hi>quia</hi> is also the form in <hi>Cx</hi>. See Alford (1992), 35-6.</note> sterilem vitam duxi Iuuenilem</foreign></l>
</lg>
<lb/>
<lg>
<l> ¶ Repentest þe<note>Bx.5.453: <hi>Repentest þe</hi>: The reading of alpha and <hi>Cx</hi>. LM presumably record the form of beta, <hi>Repentestow þe</hi>, which the other witnesses simplify by dropping the reflexive pronoun. Cf. the variants at <ref><hi>Bx</hi>.5.234</ref>, where we again follow alpha and <hi>Cx</hi>.</note> nauȝte quod repentance · and riȝte with þat he swowned</l>
<l> Til <foreign>vigilate</foreign> þe veille · fette water at his eyȝen</l>
<l> And flatte it on his face · and faste on hym criede</l>
Bx.5.456KD.5.444
<l> And seide ware þe fram<note>Bx.5.456: <hi>fram</hi>: CrWO read <hi>for</hi>, and CrW punctuate after <hi>þe</hi> (1), taking <hi>for</hi> as a conjunction. <hi>AC</hi> mss. are divided.</note> wanhope · wolde þe bitraye</l>
<l> I am sori for my synnes · sey so<note>Bx.5.457: <hi>so</hi>: LM and alpha, thus secure for <hi>Bx</hi>, but <hi>AC</hi> are without it.</note> to þi-selue</l>
<l> And bete þi-selue on þe breste · and bidde hym of grace</l>
<l> For is no gult here so grete · þat his goodnesse [i]s<note>Bx.5.459: <hi>is</hi> (2): Supported by <hi>AC</hi> against L, corrected M, and W <hi>nys</hi>. See ll. <ref>424</ref> and <ref>443</ref> for LM's preference for the negative forms of the verb. In l. <ref>474</ref> M alone has the negative.</note> more</l>
</lg>
<lb/>
<lg>
Bx.5.460KD.5.448
<l> ¶ Þanne sat sleuthe vp · and seyned hym swithe</l>
<l> And made avowe to-fore god · for his foule sleuthe</l>
<l> Shal no sondaye be þis seuene ȝere · but<note>Bx.5.462: <hi>but</hi>: MHmR have <hi>but ȝif</hi>, as in the X family of <hi>C</hi>, but not in <hi>Ax</hi>. F omits the line.</note> sykenesse it lette</l>
<l> Þat I ne shal do me er day · to þe dere cherche</l>
Bx.5.464KD.5.452
<l> And heren matines and masse · as I a monke were</l>
<l> Shal none ale after mete · holde me þennes</l>
<l> Tyl I haue euensonge herde · I behote to þe Rode</l>
<l> And ȝete wil I<note>Bx.5.467: <hi>And ȝete wil I</hi>: Beta is supported by <hi>Ax</hi>. The a-verse is rewritten in <hi>C</hi> (RK.6.309). R's reading, <hi>What I nam</hi>, apparently represents alpha, revised by F.</note> ȝelde aȝein · if I so moche haue</l>
Bx.5.468KD.5.456
<l> Al þat I wikkedly wan · sithen I wytte hadde</l>
</lg>
<lb/>
<lg>
<l> ¶ And þough my liflode lakke · leten I nelle</l>
<l> Þat eche man shal<note>Bx.5.470: <hi>shal</hi>: Alpha and HmGO; the presumed beta reading is <hi>ne shal</hi> in LMCrWC, but with little support from <hi>AC</hi>. Cf. ll. <ref>424</ref>, <ref>443</ref>, <ref>459</ref> for the preference for an extra negative.</note> haue his · ar I hennes wende</l>
<l> And with þe residue and þe remenaunt · bi þe Rode of chestre</l>
Bx.5.472KD.5.460
<l> I shal seke treuthe arst · ar I se Rome</l>
</lg>
<lb/>
<lg>
<l> ¶ Robert þe robbere · on <foreign>reddite</foreign> lokede</l>
<l> And for þer was nouȝte where-of [·]<note>Bx.5.474: All except L punctuate the line after <hi>where-of</hi>.</note> he wepe<note>Bx.5.474: <hi>wepe</hi>: LM and alpha use the strong form of the past tense. (In M it is respelt.) Contrast l. <ref>484</ref>, where all mss. have the weak form.</note> swithe sore</l>
<l> Ac ȝet þe synful shrewe · seyde to hym-selue</l>
Bx.5.476KD.5.464
<l> Cryst þat on caluarye · vppon þe crosse deydest</l>
<l> Tho dismas my brother · bisouȝte ȝow of grace</l>
<l> And haddest mercy on þat man · for <foreign>memento</foreign> sake</l>
<l> So rewe on þis robbere<note>Bx.5.479: <hi>þis robbere</hi>: Certainly <hi>Bx</hi>, though F has <hi>me Robbere</hi>, and Hm is altered to that reading. Most mss. of <hi>A</hi> have <hi>þis Robert</hi>, though four <hi>A</hi> mss. and the P family of <hi>C</hi> have <hi>me Robert</hi>, while the X family has simply <hi>Robert</hi> (K.5.241, RK.6.321). See Schmidt (2008), 351.</note> · þat <foreign>reddere</foreign> ne haue</l>
Bx.5.480KD.5.468
<l> Ne neuere wene to wynne · with crafte þat I [kn]owe<note>Bx.5.480: <hi>knowe</hi>: The reading of alpha, <hi>Ax</hi> and <hi>Cx</hi> (from where Cr derives it). Beta has <hi>owe</hi>.</note></l>
<l> But for þi mykel mercy · mitigacioun I biseche</l>
<l> Ne dampne me nouȝte at domesday · for þat I did so ille</l>
</lg>
<lb/>
<lg>
<l> ¶ What bifel of þis feloun · I can nouȝte faire schewe</l>
Bx.5.484KD.5.472
<l> Wel I wote he wepte faste · water with boþe his eyen</l>
<l> <note>Bx.5.485: The line is omitted by alpha, but it is in <hi>AC</hi>.</note> And knowleched his gult · to cryst ȝete eftsones</l>
<l> Þat <foreign>penitencia</foreign> his pyke · he shulde polsche newe</l>
<l> And lepe with hym ouer londe · al his lyf-tyme</l>
Bx.5.488KD.5.476
<l> For he had leyne bi <foreign>latro</foreign> · luciferes aunte</l>
</lg>
<lb/>
<lg>
<l> ¶ And þanne had repentaunce reuthe · and redde hem alle to knele</l>
<l> For I shal biseche for al synful · owre saueoure of grace</l>
<l> To amende vs of owre mysdedes · and do mercy to vs alle</l>
</lg>
<lb/>
<lg>
Bx.5.492KD.5.480
<l> ¶ Now god quod he þat of þi goodnesse · gonne þe worlde make</l>
<l> And of nauȝte madest auȝte · and man moste liche to<note>Bx.5.493: <hi>to</hi>: LR and WHmCO, but not in MCrGF or most <hi>C</hi> mss. (RK.7.123). <hi>Bx</hi>.5.489-528 are not in <hi>A</hi>.</note> þi-selue</l>
<l> And sithen suffredest [hym]<note>Bx.5.494: <hi>hym</hi>: Beta reads <hi>for</hi>, but alpha is supported by <hi>Cx</hi>.</note> to synne · a sikenesse to vs alle</l>
<l> And al for þe best as I bileue · what-euere þe boke telleth</l>
Bx.5.496KD.5.483α
<l> <foreign>O felix culpa o necessarium peccatum ade<note>Bx.5.496: <hi><foreign>peccatum ade</foreign></hi>: The word-order <hi>ade peccatum</hi> in MGO is also recorded in four <hi>C</hi> mss.</note> &c</foreign></l>
</lg>
<lb/>
<lg>
<l> [¶]<note>Bx.5.497: <hi></hi>: The paraph has support from WHmCR.</note> For þourgh þat synne þi sone · sent was to þis<note>Bx.5.497: <hi>þis</hi>: Better supported for <hi>Bx</hi> than <hi>þe</hi> in MCrG. F omits, as does <hi>Cx</hi>.</note> erthe</l>
<l> And bicam man of a mayde · mankynde to saue</l>
<l> And madest þi-self with þi sone · and vs synful yliche</l>
Bx.5.500KD.5.486α
<l> <foreign>Faciamus hominem ad ymaginem et similitudinem nostram</foreign></l>
<l> <foreign>Et alibi qui manet in caritate · in deo manet & deus in eo</foreign></l>
</lg>
<lb/>
<lg>
<l> ¶ And sith with þi self sone · in owre sute deydest</l>
<l> On gode fryday for mannes sake · at ful tyme of þe daye</l>
Bx.5.504KD.5.489
<l> Þere þi-self ne þi sone · no sorwe in deth feledest</l>
<l> But in owre secte was þe<note>Bx.5.505: <hi>þe</hi>: Alpha has <hi>þat</hi>. Without a parallel in <hi>C</hi>, we follow copy-text.</note> sorwe · and þi sone it ladde</l>
<l> <foreign>Captiuam duxit captiuitatem</foreign></l>
</lg>
<lb/>
<lg>
<l> ¶ Þe sonne for sorwe þer-of · les syȝte<note>Bx.5.507: <hi>syȝte</hi>: So LHmCG and alpha, and secure for <hi>Bx</hi>. M (corrected) and CrWO read <hi>light</hi>; if this is an error picked up from the following line, as the alliterative pattern suggests, it nevertheless became <hi>Cx</hi>.</note> for a tyme</l>
Bx.5.508KD.5.492
<l> Aboute mydday whan moste liȝte is · and mele-tyme of seintes</l>
<l> Feddest [þo]<note>Bx.5.509: <hi>þo</hi>: Alpha's reading is supported both by sense ("at that time fed ...") and by <hi>Cx</hi>.</note> with þi fresche blode · owre forfadres in derknesse</l>
<l> <foreign>Populus qui ambulabat in tenebris · vidit lucem magnam</foreign></l>
<l> <note>Bx.5.511: WHm and R have a paraph.</note>And thorw þe liȝte þat lepe oute of þe · lucifer [it]<note>Bx.5.511: <hi>it</hi>: R's reading of the b-verse is supported by <hi>Cx</hi>, which revises the a-verse to avoid the syntactic difficulty (what does <hi>it</hi> refer to?), which is smoothed in different ways by beta and F.</note> blent</l>
Bx.5.512KD.5.495
<l> And blewe alle þi blissed [þennes]<note>Bx.5.512: <hi>þennes</hi>: Alpha's adverb is included in the revised line in <hi>C</hi>.</note> · in-to þe blisse of paradise</l>
</lg>
<lb/>
<lg>
<l> ¶ Þe thrydde daye after · þow ȝedest in owre sute</l>