Parrots and Nightingales

Although it is not uncommon for songs to reprise elements of other songs and to quote parts of lines from them, especially their incipits,¹ verbatim quotation of a whole line or more is rare. In Die Dialektik des Trobars, his exhaustive study of allusions between troubadours, Jörn Gruber identifies only seven examples of which two are the songs examined in this chapter. Bertolome Zorzi’s “Mout fai sobrieira folia” (74.9) cannibalizes Peire Vidal’s “Quant hom es en autrui poder” (364.39) by incorporating the first four lines of each of the seven stanzas of Peire’s canso-sirventes as the last quatrain of each stanza of his own (the texts of both songs are given in Appendix 8). And Jofre de Foixà’s canso “Be m’a lonc temps menat a guiza d’aura” (304.1) quotes the incipit of a well-known troubadour love song as the final line of each of its six stanzas and its tornada. The incipits thus recycled are Arnaut de Maruelh’s “Las grans beutatz e.l fis ensenhamens” (30.16) and “Si.m destreignetz, dompna, vos et Amors” (30.23) (stanzas I and II); Perdigon’s “Ben aio.l mal e.l afan e.l consir” (370.3) and “Ir’ e pezars e dona ses merce” (370.8) (stanzas III and IV); Folquet de Marselha’s “Amors, merce: non mueira tan soven” (155.1) (stanza V); Gaucelm Faidit’s “Mon cor et mi e mes bonas chansos” (167.37) (stanza VI); and Pons de Capdoill’s “Humils e francs e fis soplei ves vos” (375.10) (tornada; for details and the complete text of Jofre’s song, see Appendix 9).

The uniqueness of these two songs, with their extraordinary OULIPO-like construction, leaps out when they are compared with Gruber’s other examples of lyrics containing full-line quotations (the quoted song is the first of each pair):²

Whereas both Bertolome and Jofre quote extensively, these songs merely repeat the incipit from one other song. The lyrics composed by Bertolome and Jofre are both forms of the canso, but the other five quoting songs are from more marginal genres. Most of the songs they quote are relatively marginal, too, or at least are not quoted elsewhere.⁵ By contrast, the troubadours whose lines are redeployed by Bertolome and Jofre are among the best represented in Appendices 1 and 2, and most of the eight songs in question are quoted elsewhere and/or anthologized, some of them several times.⁶ The incipits included in Jofre de Foixà’s song are sufficiently familiar to evoke a spectrum of other contexts: Arnaut’s “Si.m destreignetz,” for example, is quoted in a grammar, a novas, a razo, and the Breviari, and anthologized in CmDcFa; passages from Folquet’s “Amors, merce” appear in a grammar, a novas, and the Breviari, and are excerpted in DcFaHJ (see discussion in Chapter 4). And although Bertolome’s wholesale plundering is unprecedented—no one else systematically excerpts half stanzas, nor pillages precisely half of an existing text, discarding the other half—his choice of Peire Vidal’s “Quant hom es en autrui poder” similarly places him in the mainstream. Five of its seven stanzas are also quoted in the Breviari d’amor, and it is further quoted in the Doctrina d’acort and excerpted in DcFa; in addition to evoking other quotations of this same troubadour, Bertolome’s radical diminution of Peire’s song seems to reference the rigorous scaling down of lyrics in florilegia. It is safe to infer that, in integrating the words of other troubadours verbatim into their own texts, Bertolome and Jofre are knowingly engaging with practices with which they are familiar from elsewhere than within the lyric. In the other five cases identified by Gruber it is much less clear that there is any such reference to the practice of quotation at large.

On the other hand, Gruber’s five other cases present some illuminating similarities with the songs by Bertolome and Jofre. None of the quoting troubadours cites his source (though Bertolome does so indirectly) but simply assumes another’s line(s) as his own, thereby making the fact of quotation unusually challenging to recognize. Because the quoted words are in the first person, their utterance blurs the subject position of the one quoted with that of the one who quotes. The effects of this coalescence will vary, but it is noticeable that most of Gruber’s quoting poets are, like Bertolome Zorzi and Jofre de Foixà, non-Occitan poets redeploying the works of Occitan troubadours who are far better known than themselves, as if to gain a share in their prestige and a stake in their linguistic patrimony. All these examples, that is, illustrate a desire to claim a subject position in the monolangue.

However, both Jofre and Bertolome stand apart by their topsy-turvy positioning of quotation. Instead of reusing another poet’s first line as their own, they invert the quoted material from the beginning to the end of a stanza. This affects musical performance, since the quoted lines, which were originally set to opening cadences, are now sung to closing ones. In a culture where incipits and explicits are the keys to defining and recognizing works, transforming one into the other is a subversive gesture. The love of beginnings that informs the inauguration of a self-consciously written culture for the troubadours undergoes a remarkable twist at the hands of Bertolome and Jofre, and may even be countered head on.

A Song of Three Halves: Bertolome Zorzi’s “Mout fai sobrieira folia”

A Venetian merchant and diplomat, Bertolome Zorzi’s known period of activity as a troubadour is circa 1266 to 1273; between 1263 and 1270 he was held prisoner by the Genoese during their war with Venice.⁹ His corpus of eighteen songs, preserved only in AIK, contains several that build on the same form as songs by other composers.¹⁰ A political strain runs through many of them, and stanzas I–IV of “Mout fai sobrieira folia” seem to follow suit since they progressively mount a veiled attack on an unidentified person (see Appendix 8).

In the tornada of his song, Bertolome describes it as playing with revelation and concealment, at which his editor, Emil Levy, complains that “the disguise of its meaning has been only too successful.”¹¹ Line 64 identifies the work as a dimei chant (half song), the formulation referring presumably to the half that Bertolome actually composed himself. But the term clearly draws attention to the song’s other half, the seven half stanzas by Peire Vidal that Bertolome annexes for his own purposes; and beyond that, to its other other half, those parts of Peire’s text that Bertolome has excised. Bertolome’s boast about his gran sciensa (66, great learning) may allude specifically to his knowledge of Peire Vidal’s song, his mastery of which has enabled him to borrow from it only what serves his apparent nonsense, and thus veil the sense that would emerge more plainly if the remainder were recalled to mind.

The sense that this is a literary joke as much as a satirical polemic is built by the first stanza. Peire Vidal’s reputation, fostered by the troubadour himself and cemented by his biographers, was that of a near lunatic. In defying anyone to call Peire mad (1–2), Bertolome is therefore taking on the whole Peire Vidal legend. His endorsement of Peire’s “great natural judgment” (3) plays with the convention of affirming the authority of an author one is about to quote—paradoxically given Peire’s reputation, and also parodically since the only evidence Bertolome offers for the wisdom of quoting Peire is that he does so himself (4–5). Much of the humor of these opening stanzas derives from systematic equivocation between dire meaning “compose” and “repeat.” If it was true, pace the legend, that Peire Vidal showed sense in composing these verses, does Bertolome make it truer by repeating them? The more the text of “Mout fai” uses the word dire—and it does so frequently (see lines 2, 4, 10, 14, 19, 20, 23, 46, and 58)—the less one knows who is speaking (and there is no sign in the manuscript copies that some of these words are quoted).13 This evasion is at its most comic in stanza III which announces that, contrary to what the enemy of the speaker (which speaker?) may have thought, he (who?) knows what is best for him (whom? 24–25) and can (who can?) interpret other people’s behavior (what others? 26–27), in short he (who?) is thoroughly trustworthy (21) and without a shred of deceit (23)! Whereas elsewhere quotation brings about a strange cohabitation of voices, Bertolome’s self-presentation as Peire Vidal constitutes a veritable drag act that draws attention to citationality as the motor of poetic production.¹⁴

The ambiguity of a dire that is always already a redite must also have been inscribed in the musical performance of this song. The melody of “Mout fai sobrieira folia” does not survive, but three manuscripts (GRW) transmit music for “Quant hom es en autrui poder.” The G and W versions are more similar to one another than either is to R,¹⁵ but in all manuscripts the melody concludes on middle C, while line 4, the midpoint of the stanza, perches unresolved on the G above. This note of expectation is overturned in Bertolome’s drag act that transforms Peire Vidal’s opening sallies into conclusions for his own thoughts, quoting selectively so as to eliminate as much concrete meaning as possible from Peire’s text in favor of a shadow play of remainders and an echo of unheard song.

“Mout fai sobrieira folia” both retains and overturns the love of beginnings as it transforms beginnings into endings and as techniques associated with the schoolroom are travestied in a kind of comic writing that is rarely found in Occitan lyric.

Jofre de Foixà’s “Be m’a lonc temps menat a guiza d’aura” and the Force of Song

The Monk (monge) of Foissan, as Jofre de Foixà is called in chansonnier rubrics, was a Catalan whose dates of known activity range from 1267 to 1295, a little later than Bertolome Zorzi’s but overlapping with them. Jofre seems to have been what we would now call a troubleshooter: he was entrusted with a series of delicate assignments, such as reforming the management of religious houses, which he apparently carried out with exceptional skill.¹⁶ First a Franciscan friar, then a Benedictine monk, Jofre’s services were not confined to these orders. His Regles de trobar, a treatise in the Vidal tradition written for the Catalan court in Sicily between 1289 and 1295, contains snippets of unattributed verse that may be by him. In addition, he is assigned four cansos whose date is unknown but which are assumed to be youthful works composed in Catalonia prior to the Regles; they are transmitted solely in C and R.17 Jofre’s “Be m’a lonc temps menat a guiza d’aura” is remarkable for the way it condenses many songs within its compass, most obviously because of its inclusion of the incipits from seven golden age troubadour songs, but also because it reprises Northern French and Latin poetry. The result is that, like Bertolome’s “Mout fai,” it echoes densely with the Occitan lyric, while also engaging comically with Latin schoolroom practices of composition and quotation.

Jofre’s contribution to the Occitan grammatical tradition sheds interesting light on his treatment of incipits. We have seen how quoting a first line to stand for (some or all of) the rest of a song typifies the scholastic turn taken by courtly poetry in Catalonia and Italy, especially the tendency in grammars and in the organization of chansonniers to grammaticalize troubadour poetry (Chapters 1 and 3). Already there are more instances of contextualizing incipits in the CL text of the Razos de trobar than in the earlier BH redaction. Beyond the Vidal tradition, grammarians’ recourse to incipits only increases. In the two poetic treatises from Ripoll, first lines stand in for all the songs referenced, a usage also adopted by Dante in De vulgari eloquentia; in his Mirall de trobar the early fourteenth-century Catalan Berenguer d’Anoia systematically introduces his examples by quoting the incipit of the song from which they are taken (see Appendix 13). Given the establishment of this norm, it is surprising that in his Regles de trobar Jofre should quote only one first line, that of Bernart de Ventadorn’s “Era.m cosselhatz, senhor” (70.6), using it moreover not in lieu of a title but as a self-contained example illustrating the form senhor.¹⁸ This, in conjunction with the extreme infrequency of verbatim quotation of full lines within lyrics elsewhere, makes the presence of seven incipits in “Be m’a lonc temps menat a guiza d’aura” utterly unexpected. It is as though Jofre de Foixà took stock of what everyone else was doing—at least within the Occitan tradition—and then did the exact opposite. Is “Be m’a lonc temps” actively mocking and transforming the scholastic usage of the incipit, which the later treatise merely resists?

Gilles’s skill in quotation is well displayed in this first stanza. The singer hopes that song can lighten his mood where nature fails to do so, but his distinction between art and the world falters on the delicious ambiguity of the closing line:

Se per mon chant me deüsse aligier

de l’ire grant ke j’ai en mon couraige,

mestier m’avroit, car a moi leecier

ni mi valt riens, ne ne mi rasuaige

fuelle ne flors, chans d’oisiax per boscaige;

plus seux iriés quant plux oi coentoier

la douce voix dou roisignor savaige.

(“Se per mon chant,” 1–7)²⁰

If, from the great distress in my heart, I could lighten my mood by my singing, that would be useful, for nothing cheers or consoles me, neither leaf nor flower nor birdsong in the groves; I just feel worse the more I hear merrily singing “the sweet voice of the wild nightingale.”

“J’oi coentoier” (I hear merrily singing) announces both a literal referent (the annoying nightingale) and an incipit (from the Chastelain’s apparently equally annoying song); it barely matters which, since Gilles’s mood is so bad that listening to either makes it worse. Given that nightingales epitomize the spontaneity and integrity of the lyric subject, it is especially piquant to find one at the very point in the stanza where a meta-discourse opens out and where the subject position equivocates between Gilles and the Chastelain. In each of the three other stanzas of this version, the same ambiguity is wittily replayed between quoting a famous lyric and simply completing what comes before.

István Frank proposes a possible antecedent for Gilles in Latin versus cum auctoritate, also called chansons glosées (glossed songs), a form probably invented by Walter of Châtillon and common in Latin goliardic or student poetry, which consists of quatrains each concluding with a well-known quotation from a Classical poet or else from the Bible.21 Such verses are humorous distortions of hymns that quote, at the end of each stanza, the incipits of famous earlier hymns, whereas in versus cum auctoritate the quoted line that the rest of the stanza purports to gloss can come from anywhere in the “authority.” Given the insistence on incipits in “Se per mon chant,” Gilles may actually be imitating the liturgical archetype directly. Frank singles out as a likely model the twelfth-century Cistercian hymn “Corusca Sion inclitis / ornatibus, cum gloria,” which includes among its seven quotations two incipits by Venantius Fortunatus plus another by Sedulius Scotus, and is also set to the music of an Ambrosian hymn.²² Similar liturgical compositions abound through the thirteenth and into the fourteenth century; Gruber claims to have identified over a hundred of them.²³

Whereas Bartolome Zorzi’s travesty of Peire Vidal turns on the question “Who is speaking?” I suggest that Gilles’s song is rather leading its audience to ask, “What am I hearing?” The song as a whole offers its listeners the option of hearing its unsung but nevertheless potentially present Latin models, liturgical and/or goliardic. Each stanza extends the possibility of either hearing or not recognizing the trouvère incipit that it quotes. Each incipit, once recognized, allows the audience the choice mentally to hear out, or not to entertain, the rest of the song from which it is taken. And aside from words that may or may not be heard, the song confronts the listener with music heard and unheard given that, alongside Gilles’s own melody (notation for which survives in all copies except CI), the musical settings of the four songs quoted would also be evoked, making “Se per mon chant me deüsse aligier” a kind of virtual motet.

The question “What am I hearing?” is even more appropriate to Jofre de Foixà’s “Be m’a lonc temps menat a guisa d’aura,” which draws on Gilles’s example while also exceeding it at every turn. For example, Jofre follows “Corusca Sion inclitis” more faithfully than Gilles, like it, including seven incipits where the trouvère supplied at most four. The spectacle of the troubadour in his monk’s habit singing a secular love song modeled on a popul ar style of hymn must have been quite delicious to contemporary audiences.

In addition Jofre de Foixà, even more than Gilles de Viés-Maisons, capitalizes on the celebrity of Gace Brulé’s “Tant m’a mené force de signorage.” This song enjoyed an astonishingly wide diffusion. The Swiss Minnesänger Rudolf von Fenis composed a German imitation (though probably not a contrafactum), which Jofre may have known since, somewhat like his own song, it recasts various troubadour songs.24 The same form as “Tant m’a mené force de signorage” is also found in Gaucelm Faidit’s “Pel messatgier que fai tan lonc estatge” (167.46).²⁵ This song in turn has a series of formal equivalents in Occitan including, amusingly, a tenso between Albertet (de Sestaro?) and a certain Monge (16.17 = 303.1) on the question whether the French or the Catalans are nobler, in which the Monk—as though anticipating the interest in the trouvères of the future Monk of Foissan—takes the side of the French.²⁶ Jofre de Foixà’s “Be m’a lonc temps menat a guiza d’aura” is another member of this extensive, multilingual family of nightingale-style recreations.

Gilles de Viés-Maisons’s “Se per mon chant” explicitly acknowledges Gace’s influence, since his incipit “Tant m’a mené force de signorage” is quoted at the end of Gilles’s second stanza (in the form “Trop m’ait greveit force de signoraige”). Both French grands chants have stanzas of seven decasyllabic lines arranged in coblas doblas with just two rhymes per stanza pair. Gilles, however, does not retain Gace’s distribution of masculine and feminine rhymes, a'ba'bba'b, instead inverting it to produce the schema ab’ab’b’ab’. By contrast, the Occitan poems—Jofre’s “Be m’a lonc temps menat,” together with Gaucelm Faidit 167.46 and the Albertet-Monge tenso—retain the same versification as Gace. Jofre too signals his debt to Gace, since he alludes to the trouvère’s incipit “Tant m’a mené force de signorage” in his own opening line.²⁷

Assuming that Jofre adopts Gace’s melody as well as his form, audiences who heard “Be m’a lonc temps menat” performed might be quite forcibly reminded of the text of “Tant m’a mené force de signorage.” It is tempting, then, to speculate what the effect might be of hearing Gace’s closing melodic phrase displace the opening cadence of all the famous troubadour songs that Jofre quotes. Notation for “Tant m’a mené force de signorage” is given in trouvère chansonniers KLMNOTVX, among which KLMNOX share essentially the same melody, T resembles the majority group in its first section but develops differently, and V has a third distinct melody.²⁸ All begin with an ABAB structure, which parallels Gace's a'ba'b rhyme scheme, and in the majority of manuscripts (KLMNOX) the final melodic line, E, ends with a recapitulation of the closing notes of the B melody, which also chimes with the return, in the final line, of the b rhyme. If “Be m’a lonc temps menat a guiza d’aura” were sung to this most widely copied version of Gace’s melody, not only would the incipits of the famous songs quoted at the end of each stanza lose their own melody, they would resonate musically with the second and fourth lines of their new context, and their integration to its versification would by the same token be reinforced.

Whatever the musical realization of “Be m’a lonc temps menat a guiza d’aura,” the way it inserts the final incipits does differ interestingly from the model offered by Gilles. The French trouvère scrupulously holds open two alternative readings, in one of which the incipit is allowed to be recognized as such, and thus as evoking the voice of its author, whereas in the other it simply completes the ongoing utterance in the voice of Gilles. The “I hear merrily singing” of the first stanza is followed by “I have heard said and witnessed” (line 13; j’ai oï contier et tesmoignier), “reminds me of” (20; me fait remenbreir), and “consoles me” (27; me fait reconforteir). Jofre’s skill, by contrast, lies in effacing all traces of the rhetoric of quotation from his stanzas so that there is no sense whatever of any gap between the other troubadours’ discourse and his own. (This is also true of its copies, since C and R do not distinguish the quoted verses from the others.) As Frank observes in his commentary on “Be m’a lonc temps menat a guiza d’aura,” the lines preceding each quotation are marked by intricate syntax together with enjambment and rejets, which together enable a line that originally opened another troubadour’s song to fall seamlessly into place at the end of the stanza in question.29

The effect can be extremely witty. In stanza III, in preparation for the last line, Perdigon’s “May the suffering and anguish and longing be rewarded,” the speaker wavers between the drawbacks and advantages of suffering, and jestingly asserts that he would willingly die for his lady if only it would not prevent him ever enjoying her love! But the seamless integration of the quotations also has important consequences for subjectivity. I have elsewhere sketched a theoretical reading of how a subject supposed to know plays out in “Be m’a lonc temps menat,”³⁰ so a brief example of my argument will suffice here: the quotation from Gaucelm Faidit at the end of stanza VI. Even in its original context, the line “My heart and me and my good songs” casts the first person in a somewhat alien and uncanny light. Now that it has been assimilated to a new textual “I” with its own (different?) heart, its own (separate?) “me,” and who knows what songs, the line becomes a complex jeu d’esprit that casts doubt on the very existence of a subject of utterance.

Theorizing quotation from a modern perspective, Mary Orr draws attention to how “quotations invite onward transmission”; there is, she suggests, a vibrancy about quotable expressions that exceeds any context in which they are repeated and ensures that they remain more memorable than it.31 An impulse toward their own reiteration can be discerned in the lines Jofre chooses, since so many of them are quoted elsewhere. In their case, “incipit,” literally “it begins,” could be understood psychoanalytically with the “it” (or “id,” as Freud’s Es is usually translated in English, French opting for the straightforwardly equivalent ça) taken in the psychoanalytic sense of an unconscious drive. “It begins” would then designate a form of repetition compulsion where the drive is not only to repeat, but to begin again and thus potentially to change. “Be m’a lonc temps menat” can be thought of as the plaything of such a drive. Although the song sometimes sounds like a plea for something or a complaint against something, it seeks above all for song to be constantly renewed. In the tornada, for instance, where the singer seemingly humbles himself before his lady, his reprise of Pons de Capdoill’s humbling himself before his lady effectively redirects the act of reverence away from one lady or the other and toward poetry as such—toward the symbolic structure in which alone the subject’s objects, and his supposed knowledge and love of them, can be essayed.

As treated by Jofre de Foixà, this drive constantly to begin again remains resolutely tied to music. The net effect of detaching each quoted incipit from its own melody is to make the song’s symbolic texture extraordinarily polyphonic. Condensed within it are Latin sung traditions associated with both church and tavern, a French grand chant by Gilles de Viés-Maisons, another by Gace Brulé, and a series of formally identical songs in Occitan by Gaucelm Faidit, Albertet and his monkish interlocutor, as well as potentially a Swiss song. All this makes “Be m’a lonc temps menat” the expression not just of an ineluctable drive to keep beginning but also of a drive to keep singing that seems all the more vibrant for having been so long divorced from quotation in the Occitan tradition. In thus calling attention to song, Jofre’s “it begins” reconnect the parrots’ way to that of the nightingales’ way in a playful but also significant renewal of the earlier Occitano-Catalano-Italian practice.

I do not believe there is any mutual influence between the two songs by Bertolome and by Jofre. Both react independently to the well-established practice of troubadour quotation and seek to reinsert what has become a purely textual usage back into the domain of song. In so doing they are turning the dominant mode of quotation on its head. If Bertolome sets out to conceal in full view the stanza openings of a single, well-known song, Jofre’s strategy, with its ludic inversion of that scholastic shibboleth the incipit, draws a series of songs by well-known troubadours into relation with quoting practices from outside Occitania, all of which (like his own song) involve music. Whereas Bertolome does not question the parrots’ way, Jofre approaches it from the perspective of the Chastelain de Couci’s and Gilles de Viés-Maisons’s northern French nightingales. The Monk of Foissan raises the stakes of quotation and expands the virtuous circle more than any other poet to quote from other lyrics, and this is no doubt why Petrarch later chose his song as a model for his own “Lasso me” (see Chapter 11).

Giraut de Bornelh 242.40	Iois e chanz / e solatz /	(canso)
	E cortesia.m platz
Anonymous 461.142a	Jois e chans e solatz	(descort)
	E amors certana
	E cortesia.m platz
Raimon Jordan 404.7	Per solatz e per deport	(canso)
	Mi conort
Guillem de Salaignac 235.2	Per solatz et per deport	(descort)
	Me conort³
Peire Guillem de Luzerna	En aquest gai sonet leuger	(sirventes)
344.3
Anonymous 461.104	En aquest gai son e leugier	(descort)
Cadenet 106.10	Be volgra, s’esser pogues	(crusade song)
Guiraut d’Espanha 244.1a	Ben volgra s’esser pogues	(dansa)⁴
Bernart de Rovenac 66.2	D’un sirventes m’es grans	(sirventes)
	volontatz preza
Luquet Gatelus 290.1a	D’un sirventes m’es grans volontatz preza	(sirventes)

CHAPTER 7

Songs Within Songs

A Song of Three Halves: Bertolome Zorzi’s “Mout fai sobrieira folia”

Jofre de Foixà’s “Be m’a lonc temps menat a guiza d’aura” and the Force of Song