Chaucer's Works (ed. Skeat) Vol. I/Notes III

I may remark here that the metre is sometimes difficult to follow; chiefly owing to the fact that the line sometimes begins with an accented syllable, just as, in Milton's L'Allegro, we meet with lines like 'Zéphyr, with Aurora playing.' The accented syllables are sometimes indistinctly marked, and hence arises a difficulty in immediately detecting the right flow of a line. A clear instance of a line beginning with an accented syllable is seen in l. 23—'Slép', and thús meláncolýë.'

1. The opening lines of this poem were subsequently copied (in 1384) by Froissart, in his Paradis d'Amour—

Chaucer frequently makes words like have (l. 1), live (l. 2), especially in the present indicative, mere monosyllables. As examples of the fully sounded final e, we may notice the dative light-e (l. 1), the dative (or adverbial) night-e (l. 2), the infinitive slep-e (3), the adverb ylich-e (9), the dative mind-e (15), &c. On the other hand, hav-e is dissyllabic in l. 24. The e is elided before a following vowel in defaute (5), trouthe (6), falle (13), wite (16), &c. We may also notice that com'th is a monosyllable (7), whereas trewely (33) has three syllables, though in l. 35 it makes but two. It is clear that Chaucer chose to make some words of variable length; and he does this to a much greater extent in the present poem and in the House of Fame than in more finished productions, such as the Canterbury Tales. But it must be observed, on the other hand, that the number of these variable words is limited; in a far larger number of words, the number of syllables never varies at all, except by regular elision before a vowel.

14. The reading For sorwful ymaginacioun (in F., Tn., Th.) cannot be right. Lange proposes to omit For, which hardly helps us. It is clearly sorwful that is wrong. I propose to replace it by sory. Koch remarks that sorwful has only two syllables (l. 85); but the line only admits of one, or of one and a very light syllable.

15. Observe how frequently, in this poem and in the House of Fame, Chaucer concludes a sentence with the former of two lines of a couplet. Other examples occur at ll. 29, 43, 51, 59, 67, 75, 79, 87, 89; i. e. at least ten times in the course of the first hundred lines. The same arrangement occasionally occurs in the existing translation of the Romaunt of the Rose, but with such less frequency as, in itself, to form a presumption against Chaucer's having written the whole of it.

Similar examples in Milton, though he was an admirer of Chaucer, are remarkably rare; compare, however, Comus, 97, 101, 127, 133, 137. The metrical effect of this pause is very good.

23. The texts read this. Ten Brink suggests thus (Ch. Sprache, § 320); which I adopt.

31. What me is, what is the matter with me. Me is here in the dative case. This throws some light on the common use of me in Shakespeare in such cases as 'Heat me these irons hot,' K. John, iv. 1. 1; &c.

31-96. These lines are omitted in the Tanner MS. 346; also in MS. Bodley 638 (which even omits ll. 24-30). In the Fairfax MS. they are added in a much later hand. Consequently, Thynne's edition is here our only satisfactory authority; though the late copy in the Fairfax MS. is worth consulting.

32. Aske, may ask; subjunctive mood.

33. Trewely is here three syllables, which is the normal form; cf. Prologue, 761; Kn. Ta. A 1267. In l. 35, the second e is hardly sounded.

36. We must here read 'hold-e,' without elision of final e, which is preserved by the cæsura.

37. 'The most obvious interpretation of these lines seems to be that they contain the confession of a hopeless passion, which has lasted for eight years—a confession which certainly seems to come more appropriately and more naturally from an unmarried than a married man. 'For eight years,'—he says—'I have loved, and loved in vain—and yet my cure is never the nearer. There is but one physician that can heal me—but all that is ended and done with. Let us pass on into fresh fields; what cannot be obtained must needs be left'; Ward, Life of Chaucer, p. 53. Dr. Furnivall supposes that the relentless fair one was the one to whom his Complaint unto Pite was addressed; and chronology would require that Chaucer fell in love with her in 1361. There is no proof that Chaucer was married before 1374, though he may have been married not long after his first passion was 'done.'

43. 'It is good to regard our first subject'; and therefore to return to it. This first subject was his sleeplessness.

45. Til now late follows I sat upryght, as regards construction. The reading Now of late, in some printed editions, is no better.

48. This 'Romaunce' turns out to have been a copy of Ovid's Metamorphoses, a book of which Chaucer was so fond that he calls it his 'own book'; Ho. of Fame, 712. Probably he really had a copy of his own, as he constantly quotes it. Private libraries were very small indeed.

49. Dryve away, pass away; the usual phrase. Cf. 'And dryuen forth the longe day'; P. Plowman, B. prol. 224.

56. 'As long as men should love the law of nature,' i. e. should continue to be swayed by the natural promptings of passion; in other words, for ever. Certainly, Ovid's book has lasted well. In l. 57, such thinges means 'such love-stories.'

62. 'Alcyone, or Halcyone: A daughter of Æolus and Enarete or Ægiale. She was married to Ceyx, and lived so happy with him, that they were presumptuous enough to call each other Zeus and Hera, for which Zeus metamorphosed them into birds, alkuōn (a king-fisher) and kēūks (a greedy sea-bird, Liddell and Scott; a kind of sea-gull; Apollod. i. 7. § 3, &c.; Hygin. Fab. 65). Hyginus relates that Ceyx perished in a shipwreck, that Alcyone for grief threw herself into the sea, and that the gods, out of compassion, changed the two into birds. It was fabled that, during the seven days before, and as many after the shortest day of the year, while the bird alkuōn was breeding, there always prevailed calms at sea. An embellished form of the story is given by Ovid, Met. xi. 410, &c.; compare Virgil, Georg. i. 399.'—Smith's Dictionary. Hence the expression 'halcyon days'; see Holland's Pliny, b. x. c. 32, quoted in my Etym. Dict. s. v. Halcyon.

M. Sandras asserts that the history of Ceyx and Alcyone is borrowed from the Dit de la Fontaine Amoureuse, by Machault, whereas it is evident that Chaucer took care to consult his favourite Ovid, though he also copied several expressions from Machault's poem. Consult Max Lange, as well as Furnivall's Trial Forewords to Chaucer's Minor Poems, p. 43. Surely, Chaucer himself may be permitted to know; his description of the book, viz. in ll. 57-59, applies to Ovid, rather than to Machault's Poems. But the fact is that we have further evidence; Chaucer himself, elsewhere, plainly names Ovid as his authority. See Cant. Tales, Group B, l. 53 (as printed in vol. v.), where he says—

It is true that Chaucer here mentions Ovid's Heroides rather than the Metamorphoses; but that is only because he goes on to speak of other stories, which he took from the Heroides; see the whole context. It is plain that he wishes us to know that he took the present story chiefly from Ovid; yet there are some expressions which he owes to Machault, as will be shown below. It is worth notice, that the whole story is also in Gower's Confessio Amantis, bk. iv. (ed. Pauli, ii. 100); where it is plainly copied from Ovid throughout.

Ten Brink (Studien, p. 10) points out one very clear indication of Chaucer's having consulted Ovid. In l. 68, he uses the expression to tellen shortly, and then proceeds to allude to the shipwreck of Ceyx, which is told in Ovid at great length (Met. xi. 472-572). Of this shipwreck Machault says never a word; he merely says that Ceyx died in the sea.

There is a chapter De Alcione in Vincent of Beauvais, Speculum Naturale, bk. xvi. c. 26; made up from Ambrosius, Aristotle, Pliny (bk. 10), and the Liber de Natura Rerum.

66. Instead of quoting Ovid, I shall quote from Golding's translation of his Metamorphoses, as being more interesting to the English reader. (The whole story is also told by Dryden, whose version is easily accessible.) As the tale is told at great length, I quote only a few of the lines that most closely correspond to Chaucer. Compare—

See further in the note to l. 136.

67. Koch would read wolde for wol; I adopt his suggestion.

76. Alcyone (in the MSS.) was introduced as a gloss.

78. Come (dissyllabic) is meant to be in the pt. t. subjunctive.

80. Of the restoration of this line, I should have had some reason to be proud; but I find that Ten Brink (who seems to miss nothing) has anticipated me; see his Chaucers Sprache, §§ 48, 329. We have here, as our guides, only the edition of Thynne (1532), and the late insertion in MS. Fairfax 16. Both of these read—'Anon her herte began to yerne'; whereas it of course ought to be—'Anon her herte gan to erme.' The substitution of began for gan arose from forgetting that herte (A.S. heorte) is dissyllabic in Chaucer, in countless places. The substitution of yerne for erme arose from the fact that the old word ermen, to grieve, was supplanted by earn, to desire, to grieve, in the sixteenth century, and afterwards by the form yearn. This I have already shewn at such length in my note to the Pardoner's Prologue (Cant. Ta. C. 312), in my edition of the Man of Lawes Tale, pp. 39, 142, and yet again in my Etym. Dict., s. v. Yearn (2), that it is needless to repeat it all over again. Chaucer was quite incapable of such a mere assonance as that of terme with yerne; in fact, it is precisely the word terme that is rimed with erme in his Pardoner's Prologue. Mr. Cromie's index shews that, in the Cant. Tales, the rime erme, terme, occurs only once, and there is no third word riming with either. There is, however, a rime of conferme with ferme, Troil. ii. 1525, and with afferme in the same, 1588. There is, in Chaucer, no sixth riming word in -erme at all, and none in either -irme or -yrme.

Both in the present passage and in the Pardoner's Prologue the verb to erme is used with the same sb., viz. herte; which clinches the matter. By way of example, compare 'The bysschop weop for &apos;ermyng&apos;; King Alisaunder, ed. Weber, l. 1525.

86, 87. L. 86 is too short. In l. 87 I delete alas after him, which makes the line a whole foot too long, and is not required. Koch ingeniously suggests, for l. 86: 'That hadde, alas! this noble wyf.' This transference of alas mends both lines at once.

91. Wher, short for whether (very common).

93. Avowe is all one word, though its component parts were often written apart. Thus, in P. Plowman, B. v. 457, we find And made avowe, where the other texts have a-vou, a-vowe; see Avow in the New E. Dict. See my note to Cant. Tales, Group C, 695.

97. Here the gap in the MSS. ceases, and we again have their authority for the text. For Had we should, perhaps, read Hadde.

105. Doubtless, we ought to read:—'Ne coude she.'

106. This phrase is not uncommon. 'And on knes she sat adoun'; Lay le Freine, l. 159; in Weber's Met. Romances, i. 363. Cf. 'This Troilus ful sone on knees him sette'; Troilus, iii. 953.

107. Weep (not wepte) is Chaucer's word; see Cant. Tales, B 606, 1052, 3852, E 545, F 496, G 371.

120. For knowe (as in F. Tn. Th.) read knowen, to avoid hiatus.

126. 'And she, exhausted with weeping and watching.' Gower (Confes. Amantis, ed. Pauli, i. 160) speaks of a ship that is forstormed and forblowe, i. e. excessively driven about by storm and wind.

130. Or read: 'That madë her to slepe sone'; without elision of e in made (Koch).

136. Go bet, go quickly, hasten, lit. go better, i. e. faster. See note to Group C, 667. Cf. Go now faste, l. 152.

Morpheus is dissyllabic, i. e. Morph'ús; cf. Mórph'us in l. 167. I here add another illustration from Golding's Ovid, fol. 139:—

139. The first accent falls on Sey; the e in halfe seems to be suppressed.

154. His wey. Chaucer substitutes a male messenger for Iris; see ll. 134, 155, 180-2.

155. Imitated from Machault's Dit de la Fontaine:—

See Ten Brink, Studien, p. 200; Furnivall, Trial Forewords, p. 44.

It is worth notice that the visit of Iris to Somnus is also fully described by Statius, Theb. x. 81-136; but Chaucer does not seem to have copied him.

158, 159. Two bad lines in the MSS. Both can be mended by changing nought into nothing, as suggested by Ten Brink, Chaucers Sprache, § 299.

160. See a very similar passage in Spenser, F. Q. i. 1. 39, 40, 41, 42, 43. And cf. Ho. of Fame, 70.

167. Eclympasteyre. 'I hold this to be a name of Chaucer's own invention. In Ovid occurs a son of Morpheus who has two different names: "Hunc Icelon superi, mortale Phobetora vulgus Nominat;" Met. xi. 640. Phobetora may have been altered into Pastora: Icelonpastora (the two names linked together) would give Eclympasteyre.'—Ten Brink, Studien, p. 11, as quoted in Furnivall's Trial Forewords, p. 116. At any rate, we may feel sure that Eclym- is precisely Ovid's Icelon. And perhaps Phobetora comes nearer to -pasteyre than does Phantasos, the name of another son of Morpheus, whom Ovid mentions immediately below. Gower (ed. Pauli, ii. 103) calls them Ithecus and Panthasas; and the fact that he here actually turns Icelon into Ithecus is a striking example of the strange corruption of proper names in medieval times. Prof. Hales suggests that Eclympasteyre represents Icelon plastora, where plastora is the acc. of Gk. , i. e. moulder or modeller, a suitable epithet for a god of dreams; compare the expressions used by Ovid in ll. 626 and 634 of this passage. Icelon is the acc. of Gk. , or, like, resembling. For my own part, I would rather take the form plastera, acc. of, a form actually given by Liddell and Scott, and also nearer to the form in Chaucer. Perhaps Chaucer had seen a MS. of Ovid in which Icelon was explained by plastora or plastera, written beside or over it as a gloss, or by way of explanation. This would explain the whole matter. Mr. Fleay thinks the original reading was Morpheus, Ecelon, Phantastere; but this is impossible, because Morpheus had but one heir (l. 168).

Froissart has the word Enclimpostair as the name of a son of the god of sleep, in his poem called Paradis d'Amour. But as he is merely copying this precise passage, it does not at all help us.

For the remarks by Prof. Hales, see the Athenæum, 1882, i. 444; for those by Mr. Fleay, see the same, p. 568. Other suggestions have been made, but are not worth recording.

173. To envye; to be read as Tenvý-e. The phrase is merely an adaptation of the F. à l'envi, or of the vb. envier. Cotgrave gives: &apos;à l'envy l'vn de l'autre, one to despight the other, or in emulation one of the other'; also &apos;envier (au ieu), to vie.' Hence E. vie; see Vie in my Etym. Dict. It is etymologically connected with Lat. inuitare, not with Lat. inuidia. See l. 406, below.

175. Read slepe, as in ll. 169, 177; A.S. slǽpon, pt. t. pl.

Upright, i. e. on their backs; see The Babees Book, p. 245.

181. Who is, i. e. who is it that.

183. Awaketh is here repeated in the plural form.

184. Oon ye, one eye. This is from Machault, who has: 'ouvri l'un de ses yeux.' Ovid has the pl. oculos.

185. Cast is the pp., as pointed out by Ten Brink, who corrects the line; Chaucers Sprache, § 320.

192. Abrayd, and not abrayde, is the right form; for it is a strong verb (A. S. ábregdan, pt. t. ábrægd). So also in the Ho. of Fame, 110 However, brayde (as if weak) also occurs; Ho. of Fame, 1678.

195. Dreynt-e is here used as an adj., with the weak declension in -e. So also in Cant. Tales, B 69. Cf. also Ho. of Fame, 1783.

199. Fet-e is dat. pl.; see l. 400, and Cant. Ta., B 1104.

206. The word look must be supplied. MS. B. even omits herte; which would give—'But good-e swet-e, [look] that ye'; where good-e and swet-e are vocatives.

213. I adopt Ten Brink's suggestion (Chaucers Sprache, § 300), viz. to change allas into A. Lange omits quod she; but see l. 215.

218. My first matere, my first subject; i. e. sleeplessness, as in l. 43.

219. Whérfor seems to be accented on the former syllable. MS. B. inserts you after told; perhaps it is not wanted. If it is, it had better come before told rather than after it.

222. I had be, I should have been. Deed and dolven, dead and buried; as in Cursor Mundi, 5494. Chaucer's dolven and deed is odd.

244. I ne roghte who, to be read In' roght-e who; i. e. I should not care who; see note to Compl. to Pite, 105. Roghte is subjunctive.

247. His lyve, during his life.

248. The readings are here onwarde, Th. F.; here onward, Tn.; here on warde, B. I do not think here onward can be meant, nor yet hereon-ward; I know of no examples of such meaningless expressions. I read here on warde, and explain it: 'I will give him the very best gift that he ever expected (to get) in his life; and (I will give it) here, in his custody, even now, as soon as possible,' &c. Ward = custody, occurs in the dat. warde in William of Palerne, 376—'How that child from here warde was went for evermore.'

250. Here Chaucer again takes a hint from Machault's Dit de la Fontaine, where we find the poet promising the god a hat and a soft bed of gerfalcon's feathers. See Ten Brink, Studien, p. 204.

See also Our English Home, p. 106.

255. Reynes, i. e. Rennes, in Brittany; spelt Raynes in the Paston Letters, ed. Gairdner, iii. 358. Linen is still made there; and by 'clothe of Reynes' some kind of linen, rather than of woollen cloth, is meant. It is here to be used for pillow-cases. It was also used for sheets. 'Your shetes shall be of clothe of Rayne&apos;; Squyr of Lowe Degre, l. 842 (in Ritson, Met. Rom. iii. 180). 'A peyre schetes of Reynes, with the heued shete [head-sheet] of the same'; Earliest Eng. Wills, ed. Furnivall, p. 4, l. 16. 'A towaile of Raynes'; Babees Book, p. 130, l. 213; and see note on p. 208 of the same. 'It [the head-sheet] was more frequently made of the fine white linen of Reynes'; Our Eng. Home, p. 109. 'Hede-shetes of Rennes' are noticed among the effects of Hen. V; see Rot. Parl. iv. p. 228; footnote on the same page. Skelton mentions rochets 'of fyne Raynes'; Colin Clout, 316. The mention of this feather-bed may have been suggested to Machault by Ovid's line about the couch of Morpheus (Metam. xi. 611)—'Plumeus, unicolor, pullo velamine tectus.'

264. We must delete quene; it is only an explanatory gloss.

279. 'To be well able to interpret my dream.'

282. The modern construction is—'The dream of King Pharaoh.' See this idiom explained in my note to the Prioresses Tale, Group F, l. 209. Cf. Gen. xli. 25.

284. As to Macrobius, see note to the Parl. of Foules, 31. And cf. Ho. of Fame, 513-7. We must never forget how frequent are Chaucer's imitations of Le Roman de la Rose. Here, for example, he is thinking of ll. 7-10 of that poem:—

After Macrobeus understand coude (from l. 283), which governs the infin. arede in l. 289.

286. Métt-e occupies the second foot in the line. Koch proposes him for he; but it is needless; see Cant. Tales, B 3930. In l. 288, read fortúned.

288. This line, found in Thynne only, is perhaps not genuine, but interpolated. Perhaps Whiche is better than Swiche.

292. Cf. Rom. de la Rose, 45-47:—

And again, cf. ll. 295, &c. with the same, ll. 67-74. See pp. 95, 96.

301. Read songen, not songe, to avoid the hiatus.

304. Chaucer uses som as a singular in such cases as the present. A clear case occurs in &apos;Som in his bed'; Kn. Tale, 2173. (C. T. A 3031.) Hence song is the sing. verb.

309. Entunes, tunes. Cf. entuned, pp.; C. T. Prol. 123.

310. Tewnes, Tunis; vaguely put for some distant and wealthy town; see ll. 1061-4, below. Its name was probably suggested by the preceding word entunes, which required a rime. Gower mentions Kaire (Cairo) just as vaguely:—

The sense is—'that certainly, even to gain Tunis, I would not have (done other) than heard them sing.' Lange thinks these lines corrupt; but I believe the idiom is correct.

323. As stained glass windows were then rare and expensive, it is worth while observing that these gorgeous windows were not real ones, but only seen in a dream. This passage is imitated in the late poem called the Court of Love, st. 33, where we are told that 'The temple shone with windows al of glasse,' and that in the glass were portrayed the stories of Dido and Annelida. These windows, it may be observed, were equally imaginary.

328. The caesural pause comes after Ector, which might allow the intrusion of the word of before king. But Mr. Sweet omits of, and I follow him. The words of king are again inserted before Lamedon in l. 329, being caught from l. 328 above.

Lamedon is Laomedon, father of King Priam of Troy. Ector is Chaucer's spelling of Hector; Man of Lawes Tale, B 198. He here cites the usual examples of love-stories, such as those of Medea and Jason, and Paris and Helen. Lavyne is Lavinia, the second wife of Æneas; Vergil, Æn. bk. vii; Rom. Rose, 21087; cf. Ho. of Fame, 458. Observe his pronunciation of Médea, as in Ho. of Fame, 401; Cant. Ta., B 72.

332. 'There is reason to believe that Chaucer copied these imageries from the romance of Guigemar, one of the Lays of Marie de France; in which the walls of a chamber are painted with Venus and the Art of Love from Ovid. Perhaps Chaucer might not look further than the temples of Boccaccio's Theseid for these ornaments'; Warton, Hist. E. Poetry, 1871, iii. 63. Cf. Rom. of the Rose, ll. 139-146; see p. 99.

333. Bothe text and glose, i. e. both in the principal panels and in the margin. He likens the walls to the page of a book, in which the glose, or commentary, was often written in the margin. Mr. Sweet inserts with before text, and changes And into Of in the next line; I do not think the former change is necessary, but I adopt the latter.

334. It had all sorts of scenes from the Romance of the Rose on it. Chaucer again mentions this Romance by name in his Merchant's Tale; C. T., E 2032; and he tells us that he himself translated it; Prol. to Legend, 329. The celebrated Roman de la Rose was begun by Guillaume de Lorris, who wrote ll. 1-4070, and completed about forty years afterwards (in a very different and much more satirical style) by Jean de Meung (or Meun), surnamed (like his father) Clopinel, i. e. the Cripple, who wrote ll. 4071-22074; it was finished about the year 1305. The story is that of a young man who succeeded in plucking a rose in a walled garden, after overcoming extraordinary difficulties; allegorically, it means that he succeeded in obtaining the object of his love. See further above, pp. 16-19.

The E. version is invariably called the Romaunt of the Rose, and we find the title Rommant de la Rose in the original, l. 20082; cf. our romant-ic. But Burguy explains that romant is a false form, due to confusion with words rightly ending in -ant. The right O. F. form is romans, originally an adverb; from the phrase parler romans, i. e. loqui Romanice. In the Six-text edition of the Cant. Tales, E 2032, four MSS. have romance, one has romans, and one romauns.

For examples of walls or ceilings being painted with various subjects, see Warton's Hist. of E. Poetry, ed. Hazlitt, ii. 131, 275; iii. 63.

340. The first accent is on Blew, not on bright. Cf. Rom. de la Rose, 124, 125 (see p. 98, above):—

343. Ne in is to be read as Nin; we find it written nin in the Squieres Tale, F 35. See l. 694.

347. Whether is to be read as Wher; it is often so spelt.

348. The line, as it stands in the authorities, viz. 'And I herde goyng, bothe vp and doune'—cannot be right. Mr. Sweet omits bothe, which throws the accent upon I, and reduces herde to herd&apos; (unaccented!). To remedy this, I also omit And. Perhaps speke (better speken) is an infinitive in l. 350, but it may also be the pt. t. plural (A. S. sprǽcon); and it is more convenient to take it so.

352. Upon lengthe, after a great length of course, after a long run.

M. Sandras points out some very slight resemblances between this passage and some lines in a French poem in the Collection Mouchet, vol. ii. fol. 106; see the passage cited in Furnivall's Trial Forewords to the Minor Poems, p. 51. Most likely Chaucer wrote independently of this French poem, as even M. Sandras seems inclined to admit.

353. Embosed, embossed. This is a technical term, used in various senses, for which see the New Eng. Dict. Here it means 'so far plunged into the thicket'; from O. F. bos (F. bois), a wood. In later authors, it came to mean 'driven to extremity, like a hunted animal'; then 'exhausted by running,' and lastly, 'foaming at the mouth,' as a result of exhaustion.

362. A relay was a fresh set of dogs; see Relay in my Etym. Dict.

A lymere was a dog held in a liam, lime, or leash, to be let loose when required; from O.F. liem (F. lien, Lat. ligamen), a leash. In the Book of St. Alban's, fol. e 4, we are told that the beasts which should be 'reride with the lymer,' i. e. roused and pursued by the dog so called, are 'the hert and the bucke and the boore.'

365. Oon, ladde, i. e. one who led. This omission of the relative is common.

368. 'The emperor Octovien' is the emperor seen by Chaucer in his dream. In l. 1314, he is called this king, by whom Edward III. is plainly intended. He was 'a favourite character of Carolingian legend, and pleasantly revived under this aspect by the modern romanticist Ludwig Tieck—probably [here] a flattering allegory for the King'; Ward's Life of Chaucer, p. 69. The English romance of Octouian Imperator is to be found in Weber's Metrical Romances, iii. 157; it extends to 1962 lines. He was an emperor of Rome, and married Floraunce, daughter of Dagabers [Dagobert], king of France. The adventures of Floraunce somewhat resemble those of Constance in the Man of Lawes Tale. 'The Romance of the Emperor Octavian' was also edited by Halliwell for the Percy Society, in 1844. The name originally referred to the emperor Augustus.

370. The exclamation 'A goddes halfe' was pronounced like 'A god's half'; see l. 758. See note to l. 544.

374. Fil to doon, fell to do, i. e. was fitting to do.

375. Fot-hoot, foot-hot, immediately; see my note to Man of Lawes Tale, B 438.

376. Moot, notes upon a horn, here used as a plural. See Glossary. 'How shall we blowe whan ye han sen the hert? I shal blowe after one mote, ij motes [i. e. 3 motes in all]; and if myn howndes come not hastily to me as I wolde, I shall blowe iiij. motes&apos;; Venery de Twety, in Reliquiæ Antiquæ, i. 152.

Cf. a passage in the Chace du Cerf, quoted from the Collection Mouchet, i. 166, in Furnivall's Trial Forewords, p. 51 (though Chaucer probably wrote his account quite independently of it):—

379. Rechased, headed back. Men were posted at certain places, to keep the hart within certain bounds. See next note.

386. A forloyn, a recall (as I suppose; for it was blown when the hounds were all a long way off their object of pursuit). It is thus explained in the Book of St. Alban's, fol. f I:—

The 'chace of the forloyne' is explained (very obscurely) in the Venery de Twety; see Reliquiæ Antiquæ, i. 152. But the following passage from the same gives some light upon rechased: 'Another chace ther is whan a man hath set up archerys and greyhoundes, and the best be founde, and passe out the boundys, and myne houndes after; then shall y blowe on this maner a mote, and aftirward the rechace upon my houndys that be past the boundys.'

387. Go, gone. The sense is—'I had gone (away having) walked from my tree.' The idiom is curious. My tree, the tree at which I had been posted. Chaucer dreamt that he was one of the men posted to watch which way the hart went, and to keep the bounds.

396. The final e in fled-de is not elided, owing to the pause after it. See note to l. 685.

398. Wente, path. Chaucer often rimes words that are pronounced alike, if their meanings be different. See ll. 439, 440; and cf. ll. 627-630. The very same pair of rimes occurs again in the Ho. of Fame, 181, 182; and in Troil. ii. 62, 813; iii. 785, v. 603, 1192.

402. Read—For both-e Flor-a, &c. The -a in Flora comes at the cæsural pause; cf. ll. 413, 414. Once more, this is from Le Roman de la Rose, ll. 8449-51:—

Cf. also ll. 5962-5:—

405. The first accent is on For; not happily.

408. 'To have more flowers than the heaven (has stars, so as even to rival) seven such planets as there are in the sky.' Rather involved, and probably all suggested by the necessity for a rime to heven. See l. 824. Moreover, it is copied from Le Roman de la Rose, 8465-8:—

410-412. From Le Roman de la Rose, 55-58 (see p. 95, above):—

419. Imitated from Le Roman de la Rose, 1373-1391; in particular:—

Chaucer has treated a toise as if it were equal to two feet; it was really about six. In his own translation of the Romaunt, l. 1393, he translates toise by fadome. See p. 151 (above).

429. According to the Book of St. Albans, fol. e 4, the buck was called a fawne in his first year, a preket in the second, a sowrell in the third, a sowre in the fourth, a bucke of the fyrst hede in the fifth, and a bucke (simply) in the sixth year. Also a roo is the female of the roobucke.

435. Argus is put for Algus, the old French name for the inventor of the Arabic numerals; it occurs in l. 16373 of the Roman de la Rose, which mentions him in company with Euclid and Ptolemy—

This name was obviously confused with that of the hundred-eyed Argus.

This name Algus was evolved out of the O.F. algorisme, which, as Dr. Murray says, is a French adaptation 'from the Arab. al-Khowārazmī, the native of Khwārazm (Khiva), surname of the Arab mathematician Abu Ja'far Mohammed Ben Musa, who flourished early in the 9th century, and through the translation of whose work on Algebra, the Arabic numerals became generally known in Europe. Cf. Euclid = plane geometry.' He was truly 'a noble countour,' to whom we all owe a debt of gratitude. That Algus was sometimes called Argus, also appears from the Roman de la Rose, ll. 12994, &c., which is clearly the very passage which Chaucer here copies:—

Here o means 'with'; so that Chaucer has copied the very phrase 'with his figures ten.' But still more curiously, Jean de Meun here rimes nombre, pres. sing. indic., with nombre, sb.; and Chaucer rimes noumbre, infin., with noumbre, sb. likewise. Countour in l. 435 means 'arithmetician'; in the next line it means an abacus or counting-board, for assisting arithmetical operations.

437. His figures ten; the ten Arabic numerals, i. e. from 1 to 9, and the cipher 0.

438. Al ken, all kin, i. e. mankind, all men. This substitution of ken for kin (A.S. cyn) seems to have been due to the exigencies of rime, as Chaucer uses kin elsewhere. However, Gower has the same form—'And of what ken that she was come'; Conf. Am. b. viii; ed. Pauli, iii. 332. So also in Will. of Palerne, 722—'Miself knowe ich nou&#x21D;t mi ken&apos;; and five times at least in the Ayenbite of Inwyt, as it is a Kentish form. It was, doubtless, a permissible variant.

442. The strong accent on me is very forced.

445. A man in blak; John of Gaunt, in mourning for the loss of his wife Blaunche. Imitated by Lydgate, in his Complaint of the Black Knight, l. 130, and by Spenser, in his Daphnaida:—

452. Wel-faring-e; four syllables.

455. John of Gaunt, born in June, 1340, was 29 years old in 1369. I do not know why a poet is never to make a mistake; nor why critics should lay down such a singular law. But if we are to lay the error on the scribes, Mr. Brock's suggestion is excellent. He remarks that nine and twenty was usually written xxviiij.; and if the v were omitted, it would appear as .xxiiij., i. e. four and twenty. The existing MSS. write 'foure and twenty' at length; but such is not the usual practice of earlier scribes. It may also be added that .xxiiij. was at that time always read as four and twenty, never as twenty-four; so that no ambiguity could arise as to the mode of reading it. See Richard the Redeless, iii. 260.

There is a precisely similiar confusion in Cant. Ta. Group B, l. 5, where eightetethe is denoted by 'xviijthe' in the Hengwrt MS., whilst the Harl. MS. omits the v, and reads threttenthe, and again the Ellesmere MS. inserts an x, and gives us eighte and twentithe. The presumption is, that Chaucer knew his patron's age, and that we ought to read nine for four; but even if he inadvertently wrote four, there is no crime in it.

475. The knight's lay falls into two stanzas, one of five, and one of six lines, as marked. In order to make them more alike, Thynne inserted an additional line—And thus in sorowe lefte me alone—after l. 479. This additional line is numbered 480 in the editions; so I omit l. 480 in the numbering. The line is probably spurious. It is not grammatical; grammar would require that has (not is, as in l. 479) should be understood before the pp. left; or if we take left-e as a past tense, then the line will not scan. But it is also unmetrical, as the arrangement of lines should be the same as in ll. 481-6, if the two stanzas are to be made alike. Chaucer says the lay consisted of 'ten verses or twelve' in l. 463, which is a sufficiently close description of a lay of eleven lines. Had he said twelve without any mention of ten, the case would have been different.

479. Lange proposes: 'Is deed, and is fro me agoon.' F. Tn. Th. agree as to the reading given; I see nothing against it.

481. If we must needs complete the line, we must read 'Allas! o deth!' inserting o; or 'Allas! the deth,' inserting the. The latter is proposed by Ten Brink, Sprache, &c. § 346.

490. Pure, very; cf. 'pure fettres,' Kn. Tale, A 1279. And see l. 583, below.

491. Cf. 'Why does my blood thus muster to my heart?' Meas. for Meas. ii. 4. 20.

501. The MSS. have seet, sat, a false form for sat (A.S. sæt); due to the plural form seet-e or sēt-e (A.S. sǽt-on). We certainly find seet for sat in the Kn. Tale, A 2075. Read sete, as the pt. t. subj. (A.S. sǣte); and fete as dative pl. form, as in Cant. Ta. B 1104.

510. Made, i. e. they made; idiomatic.

521. Ne I, nor I; to be read N'I; cf. note to l. 343.

526. 'Yes; the amends is (are) easily made.'

532. Me acqueynte = m'acqueynt-e, acquaint myself.

544. By our Lord, to be read as by'r Lord. Cf. by'r lakin, Temp. iii. 3. 1. So again, in ll. 651, 690, 1042.

547. Me thinketh (= me think'th), it seems to me.

550. Wis, certainly: 'As certainly (as I hope that) God may help me.' So in Nonne Prestes Tale, 587 (B 4598); and cf. Kn. Tale, 1928 (B 2786); Squ. Ta. F 469, &c. And see l. 683, below.

556. Paraventure, pronounced as Paraunter; Thynne so has it.

Compare this passage with the long dialogue between Troilus and Pandarus, in the latter part of the first book of Troilus.

568. Alluding to Ovid's Remedia Amoris. Accent remédies on the second syllable.

569. The story of Orpheus is in Ovid's Metamorphoses, bk. x. The allusion is to the harp of Orpheus, at the sound of which the tortured had rest. Cf. Ho. of Fame, 1202:—

570. Cf. Ho. of Fame, 919; Rom. Rose, 21633. Dædalus represents the mechanician. No mechanical contrivances can help the mourner.

572. Cf.

Hippocrates and Galen are meant; see note to Cant. Tales, C 306.

579. Y-worthe, (who am) become; pp. of worthen.

582. 'For all good fortune and I are foes,' lit. angry (with each other). Hence wroth-e is a plural form.

589. S and C were so constantly interchanged before e that Sesiphus could be written Cesiphus; and C and T were so often mistaken that Cesiphus easily became Tesiphus, the form in the Tanner MS. Further, initial T was sometimes replaced by Th; and this would give the Thesiphus of MS. F.

Sesiphus, i. e. Sisyphus, is of course intended; it was in the author's mind in connection with the story of Orpheus just above; see note to l. 569. In the Roman de la Rose, we have the usual allusions to Yxion (l. 19479), Tentalus, i. e. Tantalus (l. 19482), Ticius, i. e. Tityus (l. 19506), and Sisifus (l. 19499).

But whilst I thus hold that Chaucer probably wrote Sesiphus, I have no doubt that he really meant Tityus, as is shewn by the expression lyth, i. e. lies extended. See Troil. i. 786, where Bell's edition has Siciphus, but the Campsall MS. has Ticyus; whilst in ed. 1532 we find Tesiphus.

599. With this string of contrarieties compare the Eng. version of the Roman de la Rose, 4706-4753. See p. 212, above.

614. Abaved, confounded, disconcerted. See Glossary.

618. Imitated from the Roman de la Rose, from l. 6644 onwards—

Jean de Meun goes on to say that Charles of Anjou killed Manfred, king of Sicily, in the first battle with him [A.D. 1266]—

He next speaks of Conradin, whose death was likewise caused by Charles in 1268, so that these two (Manfred and Conradin) lost all their pieces at chess—

And further, of the inventor of chess (l. 6715)—

He talks of the queen being taken (at chess), l. 6735—

He cannot recount all Fortune's tricks (l. 6879)—

629. Cf. 'whited sepulchres'; Matt. xxiii. 27; Rom. de la Rose, 8946.

630. The MSS. and Thynne have floures, flourys. This gives no sense; we must therefore read flour is. For a similar rime see that of nones, noon is, in the Prologue, 523, 524. Strictly, grammar requires ben rather than is; but when two nominatives express much the same sense, the singular verb may be used, as in Lenvoy to Bukton, 6. The sense is—'her chief glory and her prime vigour is (i. e. consists in) lying.'

634. The parallel passage is one in the Remède de Fortune, by G. de Machault:—

See Furnivall's Trial Forewords, p. 47; and compare the remarkable and elaborate description of Fortune in the Anticlaudian of Alanus de Insulis (Distinctio 8, cap. I), in Wright's Anglo-Latin Satirists, vol. ii. pp. 399, 400.

636. Chaucer seems to have rewritten the whole passage at a later period:—

Compare also Man of Lawes Tale, B 361, 404. 'The scorpiun is ones cunnes wurm thet haueth neb, ase me seith, sumdel iliche ase wummon, and is neddre bihinden; maketh feir semblaunt and fiketh mit te heaued, and stingeth mid te teile'; Ancren Riwle, p. 206. Vincent of Beauvais, in his Speculum Naturale, bk. xx. c. 160, quotes from the Liber de Naturis Rerum—'Scorpio blandum et quasi virgineum dicitur vultum habere, sed habet in cauda nodosa venenatum aculeum, quo pungit et inficit proximantem.'

642. A translated line; see note to l. 634.

651. Read—''Trow'st thou? by'r lord''; see note to l. 544.

653. Draught is a move at chess; see ll. 682, 685. Thus in Caxton's Game of the Chesse—'the alphyn [bishop] goeth in vj. draughtes al the tablier [board] rounde about.' So in The Tale of Beryn, 1779, 1812. It translates the F. trait; see note to l. 618 (second quotation).

654. &apos;Fers, the piece at chess next to the king, which we and other European nations call the queen; though very improperly, as Hyde has observed. Pherz, or Pherzan, which is the Persian name for the same piece, signifies the King's Chief Counsellor, or General—Hist. Shahilud. [shahi-ludii, chess-play], pp. 88, 89.'—Tyrwhitt's Glossary. Chaucer follows Rom. Rose, where the word appears as fierge, l. 6688, and fierche, l. 6735; see note to l. 618 above. (For another use of fers, see note to l. 723 below.) Godefroy gives the O. F. spellings fierce, fierche, fierge, firge, and quotes two lines, which give the O. F. names of all the pieces at chess:—

Caxton calls them kyng, quene, alphyn, knyght, rook, pawn. Richardson's Pers. Dict. p. 1080, gives the Pers. name of the queen as farzī or farzīn, and explains farsīn by 'the queen at chess, a learned man'; compare Tyrwhitt's remark above. In fact, the orig. Skt. name for this piece was mantrí, i. e. the adviser or counsellor. He also gives the Pers. farz, learned; farz or firz, the queen at chess. I suppose it is a mere chance that the somewhat similar Arab. faras means 'a horse, and the knight at chess'; Richardson (as above). Oddly enough, the latter word has also some connection with Chaucer, as it is the Arabic name of the 'wedge' of an astrolabe; see Chaucer's Astrolabe, Part i. § 14 (footnote), in vol. iii.

655. When a chess-player, by an oversight, loses his queen for nothing, he may, in general, as well as give up the game. Beryn was 'in hevy plyghte,' when he only lost a rook for nothing; Tale of Beryn, 1812.

660. The word the before mid must of course be omitted. The lines are to be scanned thus:—

The rime is a feminine one. Lines 660 and 661 are copied from the Rom. Rose; see note to l. 618, above. To be checkmated by an 'errant' pawn in the very middle of the board is a most ignominious way of losing the game. Cf. check-mate in Troil. ii. 754.

663. Athalus; see note to l. 618, above. Jean de Meun follows John of Salisbury (bishop of Chartres, died 1180) in attributing the invention of chess to Attalus. 'Attalus Asiaticus, si Gentilium creditur historiis, hanc ludendi lasciuiam dicitur inuenisse ab exercitio numerorum, paululum deflexa materia;' Joan. Saresburiensis Policraticus, lib. i. c. 5. Warton (Hist. E. Poet. 1871, iii. 91) says the person meant is Attalus Philometor, king of Pergamus; who is mentioned by Pliny, Nat. Hist. xviii. 3, xxviii. 2. It is needless to explain here how chess was developed out of the old Indian game for four persons called chaturanga, i. e. consisting of four members or parts (Benfey's Skt. Dict. p. 6). I must refer the reader to Forbes's History of Chess, or the article on Chess in the English Cyclopædia. See also the E. version of the Gesta Romanorum, ed. Herrtage, p. 70; A. Neckam, De Naturis Rerum, ed. Wright, p. 324; and Sir F. Madden's article in the Archæologia, xxiv. 203.

666. Ieupardyes, hazards, critical positions, problems; see note on Cant. Tales, Group G, 743.

667. Pithagores, put for Pythagoras; for the rime. Pythagoras of Samos, born about B.C. 570, considered that all things were founded upon numerical relations; various discoveries in mathematics, music, and astronomy, were attributed to him.

682. 'I would have made the same move'; i. e. had I had the power, I would have taken her fers from her, just as she took mine.

684. She, i. e. Fortune; so in Thynne. The MSS. have He, i. e. God, which can hardly be meant.

685. The cæsural pause preserves e in draughte from elision. It rimes with caughte (l. 682). Similar examples of 'hiatus' are not common: Ten Brink (Sprache, § 270) instances Cant. Tales, Group C, 599, 772 (Pard. Tale).

694. Ne in is to be read as nin (twice); see note to l. 343.

700. 'There lies in reckoning (i. e. is debited to me in the account), as regards sorrow, for no amount at all.' In his account with Sorrow he is owed nothing, having received payment in full. There is no real difficulty here.

705. 'I have nothing'; for (1) Sorrow has paid in full, and so owes me nothing; (2) I have no gladness left; (3) I have lost my true wealth; (4) and I have no pleasure.

708. 'What is past is not yet to come.'

709. Tantale, Tantalus. He has already referred to Sisyphus; see note to l. 589. In the Roman de la Rose, we find Yxion, l. 19479; Tentalus, l. 19482; and Sisifus, l. 19499; as I have already remarked.

717. Again from the Rom. de la Rose, l. 5869—

Chaucer's three strees (i. e. straws) is Jean de Meun's prune.

723. By the ferses twelve I understand all the pieces except the king, which could not be taken. The guess in Bell's Chaucer says 'all the pieces except the pawns'; but as a player only has seven pieces beside the pawns and king, we must then say that the knight exaggerates. My own reckoning is thus: pawns, eight; queen, bishop, rook, knight, four; total, twelve. The fact that each player has two of three of these, viz. of the bishop, rook, and knight, arose from the conversion of chaturaṅga, in which each of four persons had a king, bishop, knight, rook [to keep to modern names] and four pawns, into chess, in which each of two persons had two kings (afterwards king and queen), two bishops, knights, and rooks, and eight pawns. The bishop, knight, and rook, were thus duplicated, and so count but one apiece, which makes three (sorts of) pieces; and the queen is a fourth, for the king cannot be taken. The case of the pawns was different, for each pawn had an individuality of its own, no two being made alike (except in inferior sets). Caxton's Game of the Chesse shews this clearly; he describes each of the eight pawns separately, and gives a different figure to each. According to him, the pawns were (beginning from the King's Rook's Pawn) the Labourer, Smyth, Clerke (or Notary), Marchaunt, Physicien, Tauerner, Garde, and Ribauld. They denoted 'all sorts and conditions of men'; and this is why our common saying of 'tinker, tailor, soldier, sailor, gentleman, apothecary, ploughboy, thief' enumerates eight conditions.

As the word fers originally meant counsellor or monitor of the king, it could be applied to any of the pieces. There was a special reason for its application to each of the pawns; for a pawn, on arriving at its last square, could not be exchanged (as now) for any piece at pleasure, but only for a queen, i. e. the fers par excellence. For, as Caxton says again, 'he [the pawn] may not goo on neyther side till he hath been in the fardest ligne of theschequer, & that he hath taken the nature of the draughtes of the quene, & than he is a fiers, and than may he goo on al sides cornerwyse fro poynt to poynt onely as the quene'; &c.

726. These stock examples all come together in the Rom. de la Rose; viz. Jason and Medee, at l. 13433; Philis and Demophon, at l. 13415; &apos;Dido, roine de Cartage,' at l. 13379. The story of Echo and Narcissus is told fully, in an earlier passage (see ll. 1469-1545 of the English version, at p. 154); also that of 'Dalida' and 'Sanson' in a later passage, at l. 16879. See also the Legends of Dido, Medea, and Phillis in the Legend of Good Women; and the story of Sampson in the Monkes Tale, B 3205:—

779. M. Sandras points out the resemblance to a passage in G. de Machault's Remède de Fortune:—

The rime of table and able settles the point. Mr. Brock points out a parallel passage in Boethius, which Chaucer thus translates:—'the soule hadde ben naked of it-self, as a mirour or a clene parchemin.... Right as we ben wont som tyme by a swifte pointel to ficchen lettres emprented in the smothenesse or in the pleinnesse of the table of wex, or in parchemin that ne hath no figure ne note in it'; bk. v. met. 4. But I doubt if Chaucer knew much of Boethius in 1369; and in the present passage he clearly refers to a prepared white surface, not to a tablet of wax. 'Youth and white paper take any impression'; Ray's Proverbs.

791. An allusion to the old proverb which is given in Hending in the form—'Whose yong lerneth, olt [old] he ne leseth'; Hending's Prov. l. 45. Kemble gives the medieval Latin—'Quod puer adsuescit, leviter dimittere nescit'; Gartner, Dicteria, p. 24 b. Cf. Horace, Epist. i. 2. 69; also Rom. de la Rose, 13094.

799. John of Gaunt married Blaunche at the age of nineteen.

805. Imitated from Machault's Dit du Vergier and Fontaine Amoureuse.

These are, no doubt, the lines to which Tyrwhitt refers in his remarks on the present passage in a note to the last paragraph of the Persones Tale. Observe also how closely the fifth line of the latter passage answers to l. 812.

823. Is, which is; as usual. I propose this reading. That of the MSS. is very bad, viz. 'Than any other planete in heven.'

824. 'The seven stars' generally mean the planets; but, as the sun and moon and planets have just been mentioned, the reference may be to the well-known seven stars in Ursa Major commonly called Charles's Wain. In later English, the seven stars sometimes mean the Pleiades; see Pleiade in Cotgrave's French Dictionary, and G. Douglas, ed. Small, i. 69. 23, iii. 147. 15. The phrase is, in fact, ambiguous; see note to P. Plowman, C. xviii. 98.

831. Referring to Christ and His twelve apostles.

835-7. Resembles Le Roman de la Rose, 1689-91 (see p. 164)—

840. Koch proposes to omit maner, and read—'No counseyl, but at hir loke.' It is more likely that counseyl has slipped in, as a gloss upon reed, and was afterwards substituted for it.

849. Carole, dance round, accompanying the dance with a song. The word occurs in the Rom. de la Rose several times; thus at l. 747, we have:—

Cf. Chaucer's version, ll. 759, 810; also 744. Dante uses the pl. carole (Parad. xxiv. 16) to express swift circular movements; and Cary quotes a comment upon it to the effect that &apos;carolæ dicuntur tripudium quoddam quod fit saliendo, ut Napolitani faciunt et dicunt.' He also quotes the expression 'grans danses et grans karolles&apos; from Froissart, ed. 1559, vol. i. cap. 219. That it meant singing as well as dancing appears from the Rom. de la Rose, l. 731.

858. Chaucer gives Virginia golden hair; Doct. Tale, C 38. Compare the whole description of the maiden in the E. version of the Rom. of the Rose, ll. 539-561 (p. 116, above).

861. Of good mochel, of an excellent size; mochel = size, occurs in P. Plowman, B. xvi. 182. Scan the line—

894. 'In reasonable cases, that involve responsibility.'

908. Somewhat similar are ll. 9-18 of the Doctoures Tale.

916. Scan by reading—They n' shóld' hav' foúnd-e, &c.

917. A wikked signe, a sign, or mark, of wickedness.

919. Imitated from Machault's Remède de Fortune (see Trial Forewords, p. 48):—

Line 922 is taken from this word for word.

927-8. 'Nor that scorned less, nor that could better heal,' &c.

943. Canel-boon, collar-bone; lit. channel-bone, i. e. bone with a channel behind it. See Three Metrical Romances (Camden Soc.), p. 19; Gloss. to Babees Book, ed. Furnivall; and the Percy Folio MS., i. 387. I put and for or; the sense requires a conjunction.

948. Here Whyte, representing the lady's name, is plainly a translation of Blaunche. The insertion of whyte in l. 905, in the existing authorities, is surely a blunder, and I therefore have omitted it. It anticipates the climax of the description, besides ruining the scansion of the line.

950. There is here some resemblance to some lines in G. Machault's Remède de Fortune (see Trial Forewords, p. 49):—

957. For hippes, Bell prints lippes; a comic reading.

958. This reading means—'I knew in her no other defect'; which, as no defect has been mentioned, seems inconsistent. Perhaps we should read no maner lak, i. e. no 'sort of defect in her (to cause) that all her limbs should not be proportionate.'

964. A common illustration. See Rom. de la Rose, 7448; Alexander and Dindimus, ll. 233-5. Duke Francesco Maria had, for one of his badges, a lighted candle by which others are lighted; with the motto Non degener addam, i. e. I will give without loss; see Mrs. Palliser's Historic Devices, p. 263. And cf. Cant. Ta. D 333-5.

973. The accents seem to fall on She and have, the e in wold-e being elided. Otherwise, read: She wóld-e háv' be.

982. Liddell and Scott explain Gk. as 'the fabulous Egyptian bird phœnix, first in Hesiod, Fragment 50. 4; then in Herodotus, ii. 73.' Vincent of Beauvais, Speculum Naturale, bk. 16. c. 74, refers us to Isidore, Ambrosius (lib. 5), Solinus, Pliny (lib. 10), and Liber de Naturis Rerum; see Solinus, Polyhistor. c. 33. 11; A. Neckam, De Naturis Rerum, c. 34. Philip de Thaun describes it in his Bestiaire, l. 1089; see Popular Treatises on Science, ed. Wright, p. 113. 'The Phœnix of Arabia passes all others. Howbeit, I cannot tell what to make of him; and first of all, whether it be a tale or no, that there is neuer but one of them in all the world, and the same not commonly seen'; Holland, tr. of Pliny, bk. 10. c. 2.

Scan: Th' soléyn | feníx | of A | rabye ||. Cf. 'Com la fenix souleine est au sejour En Arabie': Gower, Balade 35.

987. Chaucer refers to Esther again; e.g. in his Merchant's Tale (E 1371, 1744); Leg. of G. Women, prol. 250; and in the Tale of Melibee (B 2291).

997. Cf. Vergil, Æn. i. 630: 'Haud ignara mali.'

1021. In balaunce, i. e. in a state of suspense. F. en balance; Rom. de la Rose, 13871, 16770.

1024. This sending of lovers on expeditions, by way of proving them, was in accordance with the manners of the time. Gower explains the whole matter, in his Conf. Amant, lib. 4 (ed. Pauli, ii. 56):—

Chaucer's Knight (in the Prologue) sought for renown in Pruce, Alisaundre, and Turkye.

There is a similar passage in Le Rom. de la Rose, 18499-18526. The first part of Machault's Dit du Lion (doubtless the Book of the Lion of which Chaucer's translation is now lost) is likewise taken up with the account of lovers who undertook feats, in order that the news of their deeds might reach their ladies. Among the places to which they used to go are mentioned Alexandres, Alemaigne, Osteriche, Behaigne, Honguerie, Danemarche, Prusse, Poulaine, Cracoe, Tartarie, &c. Some even went 'jusqu'à l'Arbre sec, Ou li oisel pendent au bec.' This alludes to the famous Arbre sec or Dry Tree, to reach which was a feat indeed; see Yule's edition of Marco Polo, i. 119; Maundeville, ed. Halliwell, p. 68; Mätzner, Sprachproben, ii. 185.

As a specimen of the modes of expression then prevalent, Warton draws attention to a passage in Froissart, c. 81, where Sir Walter Manny prefaces a gallant charge upon the enemy with the words—'May I never be embraced by my mistress and dear friend, if I enter castle or fortress before I have unhorsed one of these gallopers.'

1028. Go hoodles, travel without even the protection of a hood; by way of bravado. Warton, Hist. Eng. Poet. § 18 (ed. Hazlitt, iii. 4), says of a society called the Fraternity of the Penitents of Love—'Their object was to prove the excess of their love, by shewing with an invincible fortitude and consistency of conduct ... that they could bear extremes of heat and cold.... It was a crime to wear fur on a day of the most piercing cold; or to appear with a hood, cloak, gloves or muff.' See the long account of this in the Knight de la Tour Landry, ed. Wright, p. 169; and cf. The Squyer of Low Degree, 171-200.

What is meant by the drye se (dry sea) is disputed; but it matters little, for the general idea is clear. Mr. Brae, in the Appendix to his edition of Chaucer's Astrolabe (p. 101), has a long note on the present passage. Relying on the above quotation from Warton, he supposes hoodless to have reference to a practice of going unprotected in winter, and says that 'dry sea' may refer to any frozen sea. But it may equally refer to going unprotected in summer, in which case he offers us an alternative suggestion, that 'any arid sandy desert might be metaphorically called a dry sea.' The latter is almost a sufficient explanation; but if we must be particular, Mr. Brae has yet more to tell us. He says that, at p. 1044 (Basle edition) of Sebastian Munster's Cosmographie, there is a description of a large lake which was dry in summer. 'It is said that there is a lake near the city of Labac, adjoining the plain of Zircknitz [Czirknitz], which in winter-time becomes of great extent.... But in summer the water drains away, the fish expire, the bed of the lake is ploughed up, corn grows to maturity, and, after the harvest is over, the waters return, &c. The Augspourg merchants have assured me of this, and it has been since confirmed to me by Vergier, the bishop of Cappodistria' [Capo d'Istria]. The lake still exists, and is no fable. It is the variable lake of Czirknitz, which sometimes covers sixty-three square miles, and is sometimes dry. It is situate in the province of Krain, or Carniola; Labac is the modern Laybach or Laibach, N.E. of Trieste. See the articles Krain, Czirknitz in the Engl. Cyclopædia, and the account of the lake in The Student, Sept. 1869.

That Chaucer really referred to this very lake becomes almost certain, if we are to accept Mr. Brae's explanation of the next line. See the next note.

1029. Carrenare. Mr. Brae suggests that the reference is to the 'gulf of the Carnaro or Quarnaro in the Adriatic,' to which Dante alludes in the Inferno, ix. 113, as being noted for its perils. Cary's translation runs thus:—

It is called in Black's Atlas the Channel of Quarnerolo, and is the gulf which separates Istria from Croatia. The head of the gulf runs up towards the province of Carniola, and approaches within forty miles (at the outside) of the lake of Czirknitz (see note above). I suppose that Quarnaro may be connected with Carn-iola and the Carn-ic Alps, but popular etymology interpreted it to mean 'charnel-house,' from its evil reputation. This appears from the quotations cited by Mr. Brae; he says that the Abbé Fortis quotes a Paduan writer, Palladio Negro, as saying—'E regione Istriæ, sinu Palatico, quem nautæ carnarium vocitant'; and again, Sebastian Munster, in his Cosmographie, p. 1044 (Basle edition) quotes a description by Vergier, Bishop of Capo d'Istria—'par deça le gouffre enragé lequel on appelle vulgairement Carnarie, d'autantque le plus souvent on le voit agité de tempestes horribles; et là s'engloutissent beaucoup de navires et se perdent plusieurs hommes.' In other words, the true name Quarnaro or Carnaro was turned by the sailors into Carnario, which means in Italian 'the shambles'; see Florio's Dict., ed. 1598. This Carnario might become Careynaire or Carenare in Chaucer's English, by association with the M.E. careyne or caroigne, carrion. This word is used by Chaucer in the Kn. Tale, 1155 (Six-text, A 2013), where the Ellesmere MS. has careyne, and the Cambridge and Petworth MSS. have careyn.

For myself, I am well satisfied with the above explanation. It is probable, and it suffices; and stories about this dry sea may easily have been spread by Venetian sailors. I may add that Maundeville mentions 'a gravely see' in the land of Prestre John, 'that is alle gravele and sonde, with-outen any drope of watre; and it ebbethe and flowethe in grete wawes, as other sees don': ed. Halliwell, p. 272. This curious passage was pointed out by Prof. Hales, in a letter in the Academy, Jan. 28, 1882, p. 65.

We certainly ought to reject the explanation given with great assurance in the Saturday Review, July, 1870, p. 143, col. 1, that the allusion is to the chain of mountains called the Carena or Charenal, a continuation of the Atlas Mountains in Africa. The writer says—'Leonardo Dati (A.D. 1470), speaking of Africa, mentions a chain of mountains in continuation of the Atlas, 300 miles long, "commonly called Charenal." In the fine chart of Africa by Juan de la Coxa (1500), this chain is made to stretch as far as Egypt, and bears the name of Carena. La Salle, who was born in 1398, lays down the same chain, which corresponds, says Santarem (Histoire de la Cosmographie, iii. 456), to the of Ptolemy. These allusions place it beyond doubt [?] that the drie see of Chaucer was the Great Sahara, the return from whence [sic] homewards would be by the chain of the Atlas or [sic] Carena.' On the writer's own shewing, the Carena was not the Atlas, but a chain stretching thence towards Egypt; not an obvious way of returning home! Whereas, if the 'dry sea' were the lake of Czirknitz, the obvious way of getting away from it would be to take ship in the neighbouring gulf of Quarnaro. And how could Chaucer come to hear of this remote chain of mountains?

1034. 'But why do I tell you my story?' I. e. let me go on with it, and tell you the result.

1037. Again imitated from Machault's Remède de Fortune:—

Line 1039 is closely translated. See Furnivall's Trial Forewords, p. 48.

1040. I here substitute lisse for goddesse, as in the authorities. The blunder is obvious; goddesse clogs the line with an extra syllable, and gives a false rime such as Chaucer never makes. He rimes blisse with kisse, lisse, misse, and wisse. Thus in the Frankelein's Tale, F 1237—

Lisse is alleviation, solace, comfort; and l. 1040 as emended, fairly corresponds to Machault's 'C'est ce qui me soustient en vie,' i. e. it is she who sustains my life. The word goddesse was probably substituted for lisse, because the latter was obsolescent.

1041. I change hoolly hirs into hirs hoolly, and omit the following and. In the next line we have—By'r lord; as before (ll. 544, 651, 690).

1047. Leve (i. e. believe) is here much stronger than trowe, which merely expresses general assent.

1050. Read—'And to | behold | e th'alder | fayrest | e.' After beholde comes the cæsural pause, so that the final e in beholde does not count. Koch proposes to omit alder-. But how came it there?

1057. The spelling Alcipiades occurs in the Roman de la Rose, 8981, where he is mentioned as a type of beauty—'qui de biauté avoit adès'—on the authority of 'Boece.' The ultimate reference is to Boethius, Cons. Phil. b. iii. pr. 8. l. 32—'the body of Alcibiades that was ful fayr.'

1058. Hercules is also mentioned in Le Rom. de la Rose, 9223, 9240. See also Ho. Fame, 1413.

1060. Koch proposes to omit al; I would rather omit the. But we may read al th.'

1061. See note to l. 310.

1067. He, i. e. Achilles himself; see next note.

1069. Antilegius, a corruption of Antilochus; and again, Antilochus is a mistake for Archilochus, owing to the usual medieval confusion in the forms of proper names. For the story, see next note.

1070. Dares Frigius, i. e. Dares Phrygius, or Dares of Phrygia. Chaucer again refers to him near the end of Troilus, and in Ho. Fame, 1467 (on which see the note). The works of Dares and Dictys are probably spurious. The reference is really to the very singular, yet popular, medieval version of the story of the Trojan war which was written by Guido of Colonna, and is entitled 'Historia destructionis Troie, per iudicem Guidonem de Columpna Messaniensem.' Guido's work was derived from the Roman de Troie, written by Benoit de Sainte-Maure; of which romance there is a late edition by M. Joly. In Mr. Panton's introduction to his edition of the Gest Historiale of the Destruction of Troy (Early Eng. Text Society), p. ix., we read—'From the exhaustive reasonings and proofs of Mons. Joly as to the person and age and country of his author, it is sufficiently manifest that the Roman du Troie appeared between the years 1175 and 1185. The translation, or version, of the Roman by Guido de Colonna was finished, as he tells us at the end of his Historia Trioana, in 1287. From one or other, or both, of these works, the various Histories, Chronicles, Romances, Gestes, and Plays of The Destruction of Troy, The Prowess and Death of Hector, The Treason of the Greeks, &c., were translated, adapted, or amplified, in almost every language of Europe.'

The fact is, that the western nations of Europe claimed connexion, through Æneas and his followers, with the Trojans, and repudiated Homer as favouring the Greeks. They therefore rewrote the story of the Trojan war after a manner of their own; and, in order to give it authority, pretended that it was derived from two authors named Dares Phrygius (or Dares of Phrygia) and Dictys Cretensis (or Dictys of Crete). Dares and Dictys were real names, as they were cited in the time of Ælian (A.D. 230); and it was said that Dares was a Trojan who was killed by Ulysses. See further in Mr. Panton's introduction, as above; Morley's English Writers, vi. 118; and Warton, Hist. Eng. Poetry, ed. Hazlitt, ii. 127 (sect. 3). But Warton does not seem to have known that Guido mainly followed Benoit de Sainte-Maure.

The story about the death of Achilles is taken, accordingly, not from Homer but from Guido de Colonna and his predecessor Benoit. It may be found in the alliterative Geste Hystoriale, above referred to (ed. Panton and Donaldson, p. 342); or in Lydgate's Siege of Troye, bk. iv. c. 32. Hecuba invites Achilles and Archilochus to meet her in the temple of Apollo. When they arrive, they are attacked by Paris and a band of men and soon killed, though Achilles first slays seven of his foes with his own hand.

Here 'the kyng' is Achilles, and 'the knyght' is Archilochus. It may be added that Achilles was lured to the temple by the expectation that he would there meet Polyxena, and be wedded to her; as Chaucer says in the next line. Polyxena was a daughter of Priam and Hecuba; she is alluded to in Shakespeare's Troilus, iii. 3. 208. According to Ovid, Metam. xiii. 448, she was sacrificed on the tomb of Achilles.

Lydgate employs the forms Archylogus and Anthylogus.

1071. I supply hir; Koch would supply queen. I do not find that she was a queen.

1075. Trewely is properly (though not always) trisyllabic. It was inserted after nay, because nede and gabbe were thought to be monosyllables. Even so, the 'amended' line is bad. It is all right if trewly be omitted; and I omit it accordingly.

1081. Penelope is accented on the first e and on o, as in French. Chaucer copies this form from the Roman de la Rose, l. 8694, as appears from his coupling it with Lucrece, whilst at the same time he borrows a pair of rimes. The French has:—

In the same passage, the story of Lucretia is told in full, on the authority of Livy, as here. The French has: 'ce dit Titus Livius'; l. 8654. In the prologue to the Legend of Good Women, Chaucer alludes again to Penelope (l. 252), Lucrece of Rome (l. 257), and Polixene (l. 258); and he gives the Legend of Lucrece in full. He again alludes to Lucrece and Penelope in the lines preceding the Man of Lawes Prologue (B 63, 75); and in the Frankelein's Tale (F 1405, 1443).

1085. This seems to mean—'she (Blaunche) was as good (as they), and (there was) nothing like (her), though their stories are authentic (enough).' But the expression 'nothing lyke' is extremely awkward, and seems wrong. Nothing also means 'not at all'; but this does not help us. In l. 1086, stories should perhaps be storie; then her storie would be the story of Lucrece; cf. l. 1087.

1087. 'Any way, she (Blaunche) was as true as she (Lucrece).'

1089, 1090. Read seyë, subjunctive, and seyë, gerund. Cf. knewë, subj., 1133.

Yong is properly monosyllabic. Read—'I was right yong, the sooth to sey.' In. l. 1095, yong-e is the definite form.

1096. Accent besette (= besett') on the prefix. Else, we must read Without&apos; and besettë. We should expect Without-e, as in 1100. Without is rare; but see IV. 17.

1108. Yit, still. Sit, sittteth; pres, tense.

1113. I. e. you are like one who confesses, but does not repent.

1118. Achitofel, Ahitophel; see 2 Sam. xvii.

1119. According to the Historia Troiana of Guido (see note to l. 1070) it was Antenor (also written Anthenor) who took away the Palladium and sent it to Ulysses, thus betraying Troy. See the Geste Hystoriale, p. 379; or see the extract from Caxton in my Specimens of English from 1394 to 1579, p. 89. Or see Chaucer's Troilus, bk. iv. l. 204.

1121. Genelon; also Genilon, as in the Monkes Tale, B 3579. He is mentioned again in the Nonne Preestes Tale, B 4417 (C. T. 15233), and in the Shipmannes Tale, B 1384 (C. T. 13124), where he is called 'Geniloun of France.' Tyrwhitt's note on Genelon in his Glossary is as follows: 'One of Charlemaigne's officers, who, by his treachery, was the cause of the defeat at Roncevaux, the death of Roland, &c., for which he was torn to pieces by horses. This at least is the account of the author who calls himself Archbishop Turpin, and of the Romancers who followed him; upon whose credit the name of Genelon or Ganelon was for several centuries a synonymous expression for the worst of traitors.' See the Chanson de Roland, ed. Gautier; Dante, Inf. xxxii. 122, where he is called Ganellone; and Wheeler's Noted Names of Fiction. Cf. also the Roman de la Rose, l. 7902-4:—

1123. Rowland and Olivere, the two most celebrated of Charlemagne's Twelve Peers of France; see Roland in Wheeler's Noted Names of Fiction, and Ellis's Specimens of Early Eng. Metrical Romances, especially the account of the Romance of Sir Otuel.

1126. I supply right. We find right tho in C. T. 6398, 8420 (D 816, E 544).

1133. Knew-e, might know; subjunctive mood. See note to l. 1089.

1137. Accent thou. This and the next line are repeated, nearly, from ll. 743, 744. See also ll. 1305-6.

1139. I here insert the word sir, as in most of the other places where the poet addresses the stranger.

1152-3. Cf. Rom. de la Rose, 2006-7:—

1159. For this, B. has thus. Neither this nor thus seems wanted; I therefore pay no regard to them.

The squire Dorigen, in the Frankelein's Tale, consoled himself in the same way (F 947):—

1162. Tubal; an error for Jubal; see Gen. iv. 21. But the error is Chaucer's own, and is common. See Higden's Polychronicon, lib. iii. c. 11, ed. Lumby, iii. 202; Higden cites the following from Isidorus, lib. ii. c. 24:—'Quamvis Tubal de stirpe Cayn ante diluvium legatur fuisse musicæ inventor, ... tamen apud Græcos Pythagoras legitur ex malleorum sonitu et chordarum extensione musicam reperisse.' In Genesis, it is Jubal who 'was the father of all such as handle the harp and organ'; and Tubal-cain who was 'an instructor of every artificer in brass and iron.' The notion of the discovery of music by the former from the observation of the sounds struck upon the anvil of the latter is borrowed from the usual fable about Pythagoras. This fable is also given by Higden, who copies it from Macrobius. It will be found in the Commentary by Macrobius on the Somnium Scipionis, lib. ii. c. 1; and is to the effect that Pythagoras, observing some smiths at work, found that the tones struck upon their anvils varied according to the weights of the hammers used by them; and, by weighing these hammers, he discovered the relations to each other of the various notes in the gamut. The story is open to the objection that the facts are not so; the sound varies according to variations in the anvil or the thing struck, not according to the variation in the striking implement. However, Pythagoras is further said to have made experiments with stretched strings of varying length; which would have given him right results. See Mrs. Somerville's Connection of the Physical Sciences, sect. 16 and 17.

1169. Aurora. The note in Tyrwhitt's Glossary, s. v. Aurora, runs thus:—'The title of a Latin metrical version of several parts of the Bible by Petrus de Riga, Canon of Rheims, in the twelfth century. Leyser, in his ''Hist. Poet. Med. Ævi'', pp. 692-736, has given large extracts from this work, and among others the passage which Chaucer seems to have had in his eye (p. 728):—

Warton speaks of 'Petrus de Riga, canon of Rheims, whose Aurora, or the History of the Bible allegorised, in Latin verses ... was never printed entire.'—Hist. E. Poet. 1871, iii. 136.

1175. A song in six lines; compare the eleven-line song above, at l. 475. Lines 1175-6 rime with lines 1179-80.

1198. Koch scans: Ánd | bounté | withoút' | mercý ||. This is no better than the reading in the text.

1200. 'With (tones of) sorrow and by compulsion, yet as though I never ought to have done so.' Perhaps read wolde, wished (to do).

1206. Dismal. In this particular passage the phrase in the dismal means 'on an unlucky day,' with reference to an etymology which connected dismal with the Latin dies malus. Though we cannot derive dismal immediately from the Lat. dies malus, it is now known that there was an Anglo-French phrase dis mal (= Lat. dies mali, plural); whence the M. E. phrase in the dismal, 'in the evil days,' or (more loosely), 'on an evil day.' When the exact sense was lost, the suffix -al seemed to be adjectival, and the word dismal became at last an adjective. The A.F. form dismal, explained as les mal jours (evil days), was discovered by M. Paul Meyer in a Glasgow MS. (marked Q. 9. 13, fol. 100, back), in a poem dated 1256; which settles the question. Dr. Chance notes that Chaucer probably took dis-mal to be derived from O.F. dis mal, i. e. 'ten evils'; see l. 1207.

We can now see the connexion with the next line. The whole sentence means: 'I think it must have been in the evil days (i. e. on an unlucky day), such as were the days of the ten plagues of Egypt'; and the allusion is clearly to the so-called dies Ægyptiaci, or unlucky days; and woundes is merely a rather too literal translation of Lat. plaga, which we generally translate by plague. In Vincent of Beauvais, Speculum Naturale, lib. xv. c. 83, we find:—'In quolibet mense sunt duo dies, qui dicuntur Ægyptiaci, quorum unus est a principio mensis, alter a fine.' He goes on to shew how they are calculated, and says that, in January, the Egyptian days are the 1st, and the 7th from the end, i. e. the 25th; and he expressly refers the name Ægyptiaci to the plagues of Egypt, which (as some said) took place on Egyptian days; for it was asserted that there were minor plagues besides the ten. See also Brand's Pop. Antiquities, ed. Ellis, from which I extract the following. Barnabe Googe thus translates the remarks of Naogeorgus on this subject [of days]:—

'The Christian faith is violated when, so like a pagan and apostate, any man doth observe those days which are called Ægyptiaci,' &c.—Melton's Astrologaster, p. 56; in Brand, ii. 47. 'If his Journey began unawares on the dismal day, he feares a mischiefe'; Bp. Hall, Characters of Virtues and Vices; in Brand, ii. 48. 'Alle that take hede to dysmal dayes, or use nyce observaunces in the newe moone,' &c.; Dialogue of Dives and Pauper (1493); in Brand, i. 9. 'A dismol day'; Tale of Beryn, 650. Compare also the following:—

In the Pistil of Swete Susan (Laing's Anc. Pop. Poetry of Scotland), l. 305, Daniel reproves one of the elders in these terms:—

In Langtoft's Chronicle, l. 477 (in Wright's Polit. Songs, p. 303), John Baliol is attacked in some derisive verses, which conclude with:—'Rede him at ride in the dismale&apos;; i. e. advise him to ride on an unlucky day. Cf. The Academy, Nov. 28, 1891, p. 482; &c.

The consequence of 'proposing' on an unlucky day was a refusal; see l. 1243.

1208. A priest who missed words in chanting a service was called an overskipper; see my note to P. Plowman, C. xiv. 123.

1219. Similarly, Troilus was reduced to saying—

1234. 'Unless I am dreaming,' i. e. unintentionally.

1246. Cassandra. The prophetic lamentation of Cassandra over the impending fate of Troy is given in the alliterative Geste Hystoriale (E. E. T. S.), p. 88, and in Lydgate's Siege of Troye, bk. ii. c. 12, from Guido de Colonna; cf. Vergil, Æn. ii. 246.

1248. Chaucer treats Ilion as if it were different from Troye; cf. Nonne Prestes Tale, B 4546 (C. T. 15362). He merely follows Guido de Colonna and others, who made Ilion the name of the citadel of Troy; see further in note to Ho. of Fame, l. 158.

1288. M. Sandras (Étude sur Chaucer, p. 95) says this is from Machault's Jugement du Bon Roi de Behaigne—

1305-6. Repeated from ll. 743, 744. Cf. ll. 1137-8.

1309. Imitated in Spenser's Daphnaida, 184. The Duchess Blaunche died Sept. 12, 1369. The third great pestilence lasted from July to September in that year.

1314. King, i. e. Edward III; see note to l. 368.

1318. Possibly the long castel here meant is Windsor Castle; this seems likely when we remember that it was in Windsor Castle that Edward III. instituted the order of the Garter, April 23, 1349; and that he often resided there. A riche hil in the next line appears to have no special significance. The suggestion, in Bell's Chaucer, that it refers to Richmond (which, after all, is not Windsor) is quite out of the question, because that town was then called Sheen, and did not receive the name of Richmond till the reign of Henry VII., who renamed it after Richmond in Yorkshire, whence his own title of Earl of Richmond had been derived.

1322. Belle, i. e. bell of a clock, which rang out the hour. This bell, half heard in the dream, seems to be meant to be real. If so, it struck midnight; and Chaucer's chamber must have been within reach of its sound.