the reference is to the eagle which Æschylus calls Διός πτηνὸς κύων (Prom. 1022). That mediæval writers saw no objection to this association of hounds with angels and ministers of grace, may be seen from the fanciful interpretation current in the Middle Ages of Dominicani, as Domini cani. The context of this line in Purity reminds one of the descriptions of the Harrowing of Hell, and it is possible that the idea of the 'houndez of heven' here, and perhaps also that of the commentators on Dante's Veltro, go back to some obscure legendary conception. The use of the phrase here was probably also influenced by the contrasting idea of the 'hound(s) of Hell,' which was common: 'parinne is mony on hungri hund' (Eleven Pains of Hell 244, in Morris's Old Engl. Miscellany, EETS. 49). 972. clatered be cloudes. The phrase is common; cf. Destr. Troy 4626, 5787, 12501; Alex. C. 555; Sege of Jer. 54. 976. Cf. Pist. of Susan 225: 'But 3it we trinet a trot, þat traytour to take.' 983. þat ho nas. 'without becoming'; i. e. she became a statue at once. 1002. nom[e]n. MS. no mon. M.'s attempt to make sense of this line by inserting so before much-pat alle nazt [so] much nuye had no mon in his hert'—solves only half the difficulty, as two subjects (þat and no mon) remain for the relative clause. Even Fischer's further change, the omission of alle, which makes the line parenthetical, does not dispose of the awkwardness of the following line. The syntactical difficulties disappear and the whole passage reads smoothly, if one assumes that the scribe made the simple mistake of writing no mon for nomen (this form occurs 1. 1281); þat, referring back to Abraham, would be the subject of the relative clause, and leyen would be parallel to nomen, the hade which precedes nomen being understood for leyen: 'Abraham, who had had (felt) anxiety all night and lain awake on Lot's account.' The fact that the verb nym is here used of continued rather than momentary action is unusual, but such straining of meaning is not uncommon in this and other alliterative poems; cf. the use of the word in OE., where it means 'hold, have' as well as 'take' (see definitions in Bosworth-Toller, niman II). It may be noted that nummen alliterates with nyes in Pat. 76. Emerson (Publ. Mod. Lang. Ass. 34. 509) has recently suggested this emendation independently. 1013. Cf. Pat. 370: 'Pe verray vengaunce of God schal voyde þis place.' 1015. faure citees. In discussing the poet's indebtedness to Mandeville (see 1022 n.), C. F. Brown points out that the mention of four cities instead of five is the only important variation from Mandeville's account. 'The number four,' he writes, 'is found only in John of Würtzburg (Tobler, Descrip. Ter. Sanct., p. 178) and in the De Situ, an itinerary of the middle of the 12th century (Marquis de Vogué, Les Eglises de la Terre Sainte, p. 416). But there are no other similarities between these accounts and the passage in Cleanness, and it is extremely improbable that our author was acquainted with them. Moreover, we may account for our author's change in the number of cities, without supposing any such dependence. The fifth city in the Mandeville list is Segor (Zoar), though it is stated that it was not destroyed at the time that the others were, but, through the intercession of Lot, was saved for a long time. It is very likely that our author felt that to include Segor among the destroyed cities was a contradiction of God's promise to Lot that it should be spared (Gen. 19: 21-22), and therefore corrected the number to four. If this explanation be accepted, it furnishes another instance of our author's careful observance of the Biblical texť' (Publ. Mod. Lang. Ass. 19, 152). There would be nothing improbable in this suggestion, especially since the poet states plainly that Segor (Zoar) was saved (992). But in the manuscript þer faure has undoubtedly been partly written by a second hand (see Introd., p. ix, for other instances). The r of faure and the flourish after it (expanded e in the text) are altogether unlike the scribe's handwriting, and a thin line below the a may indicate that this letter has been written over a y. I suspect, therefore, that the original reading may have been fyue, which was changed to faure by a later hand for the sake of consistency with 992. If this were true, we should have the same kind of correction that was made in 924 (see note), where em was changed to broper because broper had been used in 772. It is possible, then, that the poet was following Mandeville after all, and gave the number of cities destroyed as five. 1019. smelle. MS., M. synne. The scribe probably miswrote synne because he had just written it in the line before. The poet elsewhere usually alliterates sm only with itself, and not with simple s (Pur. 226, 461, 771, 732, 955; Gaw. 407, 1763, 1789). Cf. 461. 1022-48. These lines are based on Mandeville's Voyage d'Outre Mer (ed. Warner, p. 50): 'Entour celle mer croist mult dalum et dalketran.. Leawe de cel mere est mult amere et salee; et si la terre estoit moillie de celle eawe, elle naporteroit point de fruit. Et la terre de luy change souent sa colour. Et iette fors .. del eawe vne chose qe homme appelle aspalt, auxi grosses pieces come vne chival, touz les iours et a toutes cousteez. Et est dit Mer Mort pur ceo qelle ne court point; ne homme ne beste qad en luy vie ne purroit morir en cel meer. Et ceo ad estee proue mointefoitz qe homme gettoit dedeinz gentz qi auoient deseruy mort, et demorroient iii. iours ou iiii., mes ils ne poaient morir. . . . Et qi metteroit fer dedeins, il noeroit par dessure; et qi mitteroit vne plume dedeins, elle irroit au founz. Et si croissent arbres delez qi portent pommes tres beles et de bele colour a regarder et toutes maners a semblant, mes qi les brusera ou trenchera parmy, il ne trouera dedeins qe cendres.' 1026. to founs. Mandeville's au founz; cf. Osgood's note on Pearl III. 1048. wyndowande askes. Cf. Mandeville's Travels (Cotton MS., ed. Halliwell, p. 107): 'And there let Julianus Apostata dyggen him up, and let brennen his Bones, Askes in the Wynd.' and let wyndwe the 1057-66. The passage here paraphrased consists of 11. 8021 ff. (ed. Marteau) of the Roman de la Rose, this part being written, as the poet rightly states, by Jean Clopinel, better known as Jean de Meun. It is part of Reason's advice to the Lover about the manner in which he should undertake to set free the imprisoned Bel Acueil, son of Courtesy; and the point of the passage, as our poet sees, is that favor is to be won by observing the character of the lady beloved, and doing that which pleases her best. I quote the more relevant parts of this long discourse: De Bel-Acuel vous prenés garde 1065 ff. This application of Jean de Meun's advice is of course entirely the poet's own; cf. Introd., p. xlii. 1067. The word conforme is borrowed from Roman de la Rose 8024 (see above). 1068. as be perle selven. This comparison of Christ with the pearl reflects the common interpretation of the pearl of great price (Matt. 13. 45, 46) as a symbol of Christ. Among the commentators who gave this explanation were Origen (Migne, Patr. Græc. 13. 856); Ephraem Syrus (Select Works, p. 84, ed. Morris); Augustine, who gives other possible interpretations (Patr. Lat. 35. 1371); Maximus of Turin (Patr. Lat. 57. 528); Bede (Patr. Lat. 92. 69); Walafrid Strabo (Patr. Lat. 114. 133); Radbertus (Patr. Lat. 120.505). For other interpretations, see Osgood's note on Pearl 735. 1075-80. The poet frequently renders homage to the Virgin (cf. Pearl 423 ff., 453 ff.; Gaw. 647 ff.); but nowhere has he written fairer lines in her praise than this series of contrasts describing her joy in the birth of Christ. The belief in the Virgin's painless delivery arose early (cf. Livius, The Blessed Virgin in the Fathers of the First Six Centuries, London, 1893, pp. 204-7); and this is reflected in the innumerable hymns on the Five Joys of the Virgin, which regularly include the birth of Christ as one of the five joys. With 1. 1077 may be compared Birth of Jesus 599-60 (in Horstmann's Altengl. Legenden, 1875): Heo bar a betere burþone pan wymmen now do, Heo hedde elles igroned sore and nou3t ascaped so. So in the Nativity Plays, the miraculous delivery of Mary excites the wonder of the midwives (Chester 528-66; Coventry 203 ff., ed. Hemingway). 1078. seknesse al sounde. Cf. Roman de la Rose 4441-2: C'est langor toute santéive C'est santé toute maladive. 1084. were. This use of the subjunctive to express an unconditional occurrence is noteworthy; we should expect watz; cf. also 209 and Gaw. 143, though in the latter instance were may possibly be an ind. plural due to attraction. Kellner (Engl. Stud. 18. 290) calls attention to a subjunctive in rhyme in Ipomadon (1596), where we should expect the indicative: A messyngere, it semyd, he were, For be his syde a box he bare, and explains this instance and a similar use of were in Guy of Warwick (2798) as due to the necessities of rhyme. It is noteworthy that the two instances in Purity occur at the end of the line, where the weak metrical ending is usual. 1086. The adoration of the ox and the ass, like the singing of the angels (1080 ff.), had become traditional. Both are found in the Pseudo-Matthew, Chap. 14 (Cowper, Apocryphal Gospels, p. 53). 1103-8. Christ's clean cutting of the bread is mentioned in the Towneley Play (No. 28) Thomas of India 264-5: Ihesu, goddis son of heuen at sopere satt betweyn; These lines are spoken by Peter in order to convince the doubting Thomas of Christ's resurrection, and the reference is to the supper at Emmaus, where, according to Luke 24. 35, the disciples recognized Christ in fractione panis. The passage in Purity and that in the Towneley Play probably go back to a common source based on Luke's words. 'particular' 1109. kyryous and clene. The lines on Christ's cutting the bread have, of course, little to do with Christ's purity, and the passage is introduced only by means of a kind of play on words. Clene means at once 'pure,' and, with reference to cutting, 'smooth, sharp, without ragged edges.' So kyryous_in_this_line='skilful' as far as the cutting of the bread is concerned, but in connection with Christ's abhorrence of everything vile. 1118. hym. This use of the dative (accus.) for the nominative is extraordinary. Einenkel quotes examples of various cases where the dative pronoun was substituted for the nominative (Grundr., 2d ed., 1. 1085, § 141 a; 1093, § 144 8), but none is quite like this instance. It may be that we have a case of attraction, the pronoun, |
