¹ Cf. the passages from Giraut's poems put together by Kolsen, Guiraut von Bornelh (Berlin, 1894), pp. 41–43.

² Cf. Kolsen, p. 43.

³ See Journal des Savants (1937), pp. 197–198.

⁴ Provenzalisches Supplementwörterbuch, vii, 670.

⁵ Lexique Roman, iv, 236.

⁶ Prov. Suppl. Wb., v, 261, No. 3.

⁷ Cf. “Zum Text der Lieder des Giraut de Bornelh,” Bibl. dell' Archivum Romanicum, serie i, vi (1938), 3, where I gave this explanation of paragraphs 3–5.

⁸ Observations sur l'art lyrique de Giraut de Borneil (Amsterdam, 1938) = Mededeelingen der Koninklijke Nederlandsche Akademie van Wetenschappen, Afd. Letterkunde, Nieuwe Reeks, Deel 1, No. 1, p. 93.

⁹ Giraut de Bornelh, whose rectitudo we should apply not only to the moral, but also to the metrical field, ought not to be charged with a stanzaic enjambment, unless it serves a poetical purpose. Besides the above-mentioned case, Salverda de Grave (l.c.) has found five others in Giraut's poems. Of these I can recognize only the case of poem No. 33. But here we have to do with two tornadas which are not infrequently connected with one another syntactically (see also Giraut's poems Nos. 24 and 46, which Salverda de Grave does not mention), not to speak of the fact that the second tornada of No. 33 is offered only by one of 15 manuscripts. All the other cases seem to me highly questionable. Because of the fundamental importance of this matter for our problem, it is worth while, not to say necessary, to examine them. Here they are:

1. No. 8, between stanzas v and vi. Stanza vi is in only one of fourteen manuscripts. So the other thirteen scribes must have thought it superfluous and stanza v complete without it. Jeanroy (Journal des Savants, 1937, p. 198) declared it apocryphal. Even if it belonged to the poem, there is no necessity whatsoever to think it syntactically dependent on stanza v. The Tan by which it is introduced does not connect it with stanza v (Salverda de Grave); it refers to the following per que which, as in so many cases, replaces a simple consecutive que after tan.

2. No. 18, from stanza v to stanza vi. The latter begins with Que, which Salverda de Grave rightly renders by “Car,” putting even a semicolon before this conjunction. Car always introduces a main clause. Here again the syntactical construction does not admit a stanzaic enjambment.

3. No. 30, from stanza v to stanza vi. This case is similar to case two. Stanza vi begins with followed by a main clause, and Salverda de Grave translates (l.c., p. 33) the Provençal text accordingly. Here again both stanzas keep their syntactical independence.

4. No. 5: (a) From stanza iii to stanza iv. The explanation which I gave to the end of stanza iii (see Zum Text etc., p. 3) excludes any idea of a stanzaic enjambment. The Que introducing stanza iv is again “for” (French car). (b) From stanza iv to v. This case is less simple because the text is far from clear. The poem is one of Giraut's works in which he seems to talk to his alter ego or an imaginary interlocutor. The poet's thought is proceeding by jumps, and it is often hard to find the borderlines between the sayings of the two partners. The end of stanza iv and stanza v run thus:

The first two lines are clear : the poet's mysterious interlocutor warns him that he will rouse the ill-will of malevolent people (the famous lauzenjadors) by praising his lady too much. What follows is less easy to understand. Kolsen's translation is not very helpful. So it seems an ingenious solution to substitute a comma for the question mark at the end of stanza iv and see in the first two lines of stanza v the completion of a conditional sentence with quem graziria as its main clause. This is what Jeanroy and Salverda de Grave did. The former translated: “De quoi me saurait-elle gré, si je ne me faisais... ?” (see Kolsen's edition, ii, 23); the latter: “Et comment me serait-elle reconnaissante, si je ne me faisais des ennemis... ?” (Observations etc., p. 93). Indeed, the newly-won conditional sentence is, from the syntactical viewpoint, perfect. But does it lead us to a real understanding of the passage? If we reduce the content of the conditional sentence to the form of a simple statement, it says that only by making himself enemies will the lover obtain his lady's gratitude. I doubt that the troubadour wanted to say that. Moreover, how does this fit lines 31–32 in which the poet declares that there will always be malevolent tatlers anyhow? The following explanation tries to take into account all these doubts and do without the en-jambment introduced by the two above-named scholars. Lines 26–27 are supposed to be spoken by Giraut's partner: “You will certainly have many enemies, because you undertake to praise her too much.” The poet answers with a question: “And what (else) should she thank me for?” Here we have to supply: “if not for my praising her.” It is not his making himself enemies, it is his praising her that will win him his lady's gratitude. The whole passage centers around this question of whether or not he should commend his beloved. No fear of enemies will keep him from doing so. Therefore he comes back to the lauzenjadors in the beginning of stanza v. In an unfinished conditional sentence uttered with the intonation of a question, he asks: “And, even if I really did not like to make myself enemies?” Again we supply: “What use would it be to stop praising my lady?” Then the actual text goes on, giving the reason for this argument: “For some envious, foolish babblers you will always find.” And the poet finishes by saying that he is not afraid of them: “Though I am not arrogant, but even if Ogier (the famous epic hero) blamed her to me, I would fight with him about it.” I hope that the foregoing discussion has made it clear that in none of his poems has Giraut made himself guilty of the metrical negligence which is a stanzaic enjambment.

¹⁰ Prov. Inedita, p. 32.

¹¹ Prov. Suppl. Wb., vii, 669 s.v. so No. 9.

¹² Paris, 1929.

¹³ Romania, Liii, 136.

¹⁴ E. Lerch, Satzglieder ohne den Ausdruck irgendeiner logischen Beziehung (in: Haupt-probleme der französischen Sprache, I,36 ff., especially pp. 39 and 43) speaks in such cases of a “logical zero-category.”

¹⁵ Whether ney means “snow” (Appel) or “refusal” from negar (Schultz-Gora) is irrelevant for our purposes; see the controversy between the two scholars, Arch. f. d. Studium d. Neueren Sprachen, CXLV, 104–105 and CXLVI, 248–249 and Gruenenthal, cxlviii, 101–102.

¹⁶ ab ponher d'asperos (literally: “with setting spurs”) is to be understood in a figurative sense: “despite my efforts.”

¹⁷ This is the ab in question.

¹⁸ Meyer: “ce nonobstant.”

¹⁹ Meyer: “Malgré cela.”

²⁰ Meyer: “malgré l'intempérie.”

²¹ The empero clearly indicates that ab tola l'ira means “despite all their grief.”

²² Anglade: “Malgré tout.”

²³ i, 784, lines 14 ff.

²⁴ This is a meaning which has not yet been found for Old Provençal ab so que. This is an amazing fact, because here ab would have its original sense, that of simply adding one fact to another. Maybe this lack of evidence is due to a mere coincidence.

²⁶ Cf. my article, “Old French veaus, seveaus, siviaus; Old Provençal sevals, sivals,” Studies in Philology, XLI, 16 ff., especially p. 37.

²⁶ Que-Sätze anknüpfend an adverbiale Ausdrücke der Versicherung, Beschwörung, Vermutung, Bejahung, Vereinung, an Interjektionen.

²⁷ This German kaum daß bears some resemblance to French à peine si, with which Tobler deals in his Verm. Beitr., v, section 4 under the caption of A peine si elle répondait à son salut. Tobler too points out (p. 18) that sentences with à peine si belong in the same syntactical category as those treated in the above-quoted article of his Verm. Beitr.

²⁸ Attention may be called to the adversative nature of nock dazu (wo), which originally only denotes “in addition to.”

²⁹ It will be remembered that Old Provençal so stands for both French ce and cela. Prov. per so is equal to French pour (par) cela, Prov. per so que equal to French parce que.

³⁰ Zschr.f.rom. Phil., LV, 656–663.

⁸¹ Loc. cit., p. 659.

³² Zschr.f.fz. Spr. & Lit., lii (1929), 100, note 5.

³³ Zschr.f.rom. Phil., liii, 87 ff.; especially pp. 91–92.

³⁴ iii, 253.

⁸⁵ (Paris, 1922), no. i, stanza 2.

³⁶ This translation of covertis is questionable.

³⁷ Edited by Clovis Brunei, Classiques Français du Moyen Age (Paris, 1914).

³⁸ “Valley.”

³⁹ Read: Que farem mays en cesta vaur, lassetas, e que poyrem far... ? Put comma after Lassas (line 1670).

⁴⁰ That of line 1672 is, of course, the interrogative pronoun, while those of lines 1671 and 1674 are the same as we find it in the following passage from Claudel's Annonce faite à Marie, iv, 4: La maison est-elle vide que toutes les portes soient ouvertes?

⁴¹ Behrens-Festschrift (Jena, 1929), p. 181.

⁴⁵ It should be remembered, though, that MS. H has el instead of que.