1, 3. chant,verchant: Jeanroy prints chan, verchan.
3. flor: Jeanroy prints flors, but flor here is necessary for the internal rhyme of each first and third line, first noticed by Pillet. In a this rhyme is violated at 3. flors, 38. enqer, 43. salvaire, and 45. genzer; see notes to these lines. In his preface to the songbook which is represented by a, Bernart Amoros says that he emended his exemplar in order to conform to correct language; see MS. Florence, Riccardiana 2814, p. 2. flors, salvaire, and genzer are probably his doing, and show an overriding concern for grammar in the face of prosodic constraints. For another example of flor nominative singular, cf. Marcabru, Bel m’es quan son li fruich, 37.
6. Dejeanne suggested correcting joi to lei.
7. no·m lau is De Lollis’ emendation of non lau.
11. For the dual subjects Jovenz and Faig see 1.15.
12. Jeanroy lists luecs as a’s reading, but the MS reads luec.
13. Jeanroy lists amiztatz as a’s reading, but the MS reads amistatz; he emends to amistat in his text.
17. drudejan is probably either a verbal noun or a third person plural from the otherwise unattested *drudejar; see Adams, Word Formation in Provençal, pp. 357-60.
18. guizardos: Jeanroy emends this to guizardo, taking it strictly as the object of ditz in the next line.
19. reprocher: Jeanroy emends this to reprovier.
20. qi glazi fai: De Lollis emends this to q’a glazi fer.
27. moiller e marit is Dejeanne’s correction; the MS reading is shown to be incorrect by tug tres: moillerat and marit both denote the husband.
29. creman is Dejeanne’s emendation, for the rhyme; cf. Marcabru, Hueymais dey esser alegrans, 31: estas putas ardens, cremans.
30, 32. Jeanroy reverses the order of these lines.
31. Dejeanne suggested emending desliau to e truan, but failed to see that the n of deslian had probably been mistaken for u.
34. [ja] is Jeanroy’s conjecture; he also emends no·m clam to no clam.
38. n’encor lo cor tristan: The cruces encor and tristan have raised much debate. A reading in -or, with close o, is needed for the internal rhyme; Pillet’s correction encor (encorre, “to incur, contract”) has been adopted here. It seems unlikely, despite the view of Appel, Kolsen and Lejeune, that tristan refers to the medieval hero, simply because it would not make sense in this context: there is no reason why the fact that the lady is false should cause the poet to have “the heart of Tristan”: Tristan was certainly never wounded by Iseult’s falseness. Jeanroy prints n’enqer and tristan.
41. f[ez]es is Pillet’s correction.
43. Salvador is Pillet’s correction; the subject case for the vocative is not consistently used in Old Prov., and Salvador is necessary here for the internal rhyme. Jeanroy prints Salvaire.
45. genzor is De Lollis’ correction; both grammar and rhyme require genzor, which Jeanroy prints.
48. m’en ira[i]: The MS reading s’en ira conqes, which Jeanroy prints, cannot stand, as s’ refers back to a feminine antecedent, requiring conqiza. Dejeanne suggested con q’es “as she is,’’ but the result was obscure. Because of the traditional alba situation here, m’en ira[i] is offered to preserve grammar and logical consistency.
|
