The Journal of Seventeenth-Century Music

Volume 22 (2016) No. 1

Published 2017

Bilingualism, Compositional Process, and the Quest for Meaning: The Dialogue between Italian and French Music in the Ballet de la Raillerie (1659)

Michael Klaper*

Abstract

In the 1650s the French court ballet underwent a sort of Italianization. An example is the Ballet de la Raillerie (1659), whose Intermedio/Intermède is a famous dispute between personifications of French and Italian music. A comparison of the two versions found in the livret, Italian and French, and the version found in the extant scores leads to new insights into the genesis of the dialogue, and shows that its meaning was altered significantly during the process of (re)composition. Though not a manifesto with far-reaching consequences, the dialogue can nevertheless be interpreted as a witness to changing trends in the self-construction of French musical identity.

1. Introduction

2. The phenomenon of bilingualism and translation in French ballet, ca. 1650–1660

3. The Ballet de la Raillerie: Remarks about its structure and the state of the sources

4. The dialogue from La Raillerie: Observations on its textual transmission

5. Comparisons between Italian and French music up to around 1660, and the interpretation and significance of the dialogue from La Raillerie

6. Conclusions

Acknowledgements

Examples

Figures

Tables

Video Example

References

1. Introduction

1.1 One of the numerous ballets performed at the court of the French king Louis XIV during the early part of his reign, the Ballet de la Raillerie (“Ballet of Mockery”), dating from 1659, contains a dialogue sung by the personifications of Italian and French music that has been well known to music historians for over a hundred years. The dialogue was published in 1912 by Paul-Marie Masson, in an article entitled “Italian and French Music: The First Querelle,” and also by Henry Prunières a year later, in his ground-breaking study of the history of Italian opera in France during the seventeenth century.^[1] That the piece has become so famous since then is quite understandable: though not (as suggested by Masson) the very first of the many seventeenth- and eighteenth-century querelles about the respective merits of French and Italian music, it was the first to bring this quarrel onstage. It is thus not surprising that it is often analyzed (or at least mentioned) in the scholarly literature concerning the relationship between these two musical idioms in the early modern period. Yet these commentaries do not agree: whereas some scholars interpret the dialogue as a sort of proclamation of “goûts réunis” avant la lettre,^[2] others conclude that it pronounces the triumph of French music over Italian.^[3] Sometimes the dialogue is seen as a manifesto announcing a programmatic change for the future of Italian music in France,^[4] sometimes it is cited instead as an explanation or summary of the differences underlying French and Italian music around 1660,^[5] and sometimes it is presented as an element of a presumed appeasement politics directed to the enemies of Italian art at the French court.^[6]

1.2 As I will show in the following discussion, this bewildering situation can best be confronted by a look at the transmission history of the dialogue, in particular at the textual differences between its surviving sources. That is, by following strictly philological methods, new insights can be gained into the history of the piece, which is apparently neither straightforward nor simple. Furthermore, I will widen the context for understanding the dialogue, by discussing the topic of bilingualism and the question of translations—or better: of versions of texts in different languages—in the livrets of French ballets performed during the 1650s. The phenomenon of translation in the ballets has never been studied, to my knowledge, and it will prove to be a useful tool not only for the analysis of sung texts, but also in the quest for meaning and thus for historical understanding. Finally, I will situate the dialogue within the main outlines of the seventeenth-century quarrel about French and Italian music, and show that its different surviving versions reflect distinct positions in the quarrel.

2. The phenomenon of bilingualism and translation in French ballet, ca. 1650–1660

2.1 In the 1650s traditional French ballet underwent a sort of Italianization.^[7] This was due partly to the reception of Italian opera at the French court starting in the 1640s, and partly to the presence of Italian artists (singers, instrumentalists, and so on), who—together with their French colleagues—participated in performances that mixed the French and Italian languages and styles. The Italian vocal numbers inserted in these ballets were normally composed by the Italian-born Jean-Baptiste Lully (Giovanni Battista Lulli), who had been appointed “compositeur de la musique instrumentale” to the young Louis XIV in 1653.^[8] For Lully, ballet composition was a kind of experimental field in which he could explore more complicated dramatic structures than just dances.^[9] As a result, what had been a reality in French concert life for many years—the coexistence of French and Italian vocal music, and the performance of the latter by French singers—was now increasingly brought to the stage.^[10]

2.2 While it seems clear that most (though certainly not all) of the French texts provided for these ballets can be attributed to the court poet Isaac de Benserade—in fact, many of them were included in a posthumous edition of Benserade’s collected works^[11]—the authors of the Italian texts contained in the printed livrets of the ballets are rarely known. At least some of the latter were written by the diplomat and poet Francesco Buti, an intimate of Cardinal Jules Mazarin (another Italian immigrant: Giulio Mazzarini).^[12] Buti had come to France in 1645 in the aftermath of the flight of the Barberini family from Rome after the installment of a new pope (Innocence X). He wrote the libretto for the first opera created especially for the Parisian court, L’Orfeo (premiered in 1647), as well as the librettos for Le nozze di Peleo e di Theti (1654) and Ercole amante (1662), also performed at court theaters in Paris. To him can also be attributed with certainty the Italian passages in the “Ballet du Roy” Amor malato (1657) and the Ballet de l’Impatience (1661). Moreover, it is probable that in the case of Amor malato (as with the opera L’Orfeo) he was responsible for the whole production as a corago (artistic director).^[13] Any other contributions he might have made to the French-Italian ballets of the time remain unidentified. As for other authors, it is possible, as has often been remarked, that Lully himself composed not only the music but even the words of some of the numbers sung in Italian.^[14] In addition, one cannot know which of the Italians residing at the court of Paris and able to write poetry (for instance, Giovanni Bentivoglio) might have contributed Italian texts to the genre.^[15] At any rate, it seems a fair assumption that the bilingualism led in general to a collaboration between French and Italian authors, whatever that actually meant; at the very least, someone must have coordinated their work. The relationship is seldom explicitly stated. The livret of the Ballet de la Raillerie, in particular, bears only an annotation preceding the first Italian number, specifying that the characters “sing the following lines of Italian poetry, in a version made by someone other than the author of the vers du ballet”—i.e., someone other than Isaac de Benserade.^[16]

2.3 At the same time, the presence of Italian texts created a need for French translators (who are likewise normally unidentified). As long as the Italian vocal numbers were rather short and of a comic nature, the Italian language was apparently not regarded as a problem, and French translations were deemed unnecessary or of little importance (see Table 1):^[17] the Italian récit (vocal solo) “Piaccia a vostra Asinità” from the Ballet des Proverbes (1654), which comprises only twelve lines of verse, has no French counterpart; nor does the “Récit crotesque italien” from the Ballet des Bienvenus (1655), for which not even the Italian text is contained in the livret, perhaps because this one was improvised on the spot by Lully.^[18] But even the relatively extensive “Concert italien” (comprising over fifty lines) in the Ballet de Psyché (1656) has no accompanying French version,^[19] though in 1654 it had already been decided otherwise for a longer music-theater piece, the opera Le nozze di Peleo e di Theti: in the printed libretto the text to be sung in Italian (on the left) is accompanied by a French version (on the right) that in all probability was destined not for singing but for silent reading, for it is a paraphrase in verse rather than a line-by-line translation (see Figure 1).^[20]

2.4 Clearly such translations that appear side by side with the Italian texts (like those from later times, such as the English translations of Handel’s Italian librettos),^[21] can fulfill a multitude of functions. The most obvious seems to be that they facilitate a deeper understanding of the Italian passages by a non-Italian audience, but there is much more to consider about them. It should be emphasized that the French versions of the Italian texts under discussion here are written, like the Italian originals, in verse. A slightly later example shows that this cannot be taken for granted: the Italian libretto of the pastoral opera Nicandro e Fileno by the librettist Philippe-Julien Mancini and the composer Paolo Lorenzani, which was performed before Louis XIV in 1681, was published with a French prose version at its side.^[22] The decision between prose and verse, presumably equally available options, was not only a technical one; it also had cultural and literary implications. A translation is thus a literary text in its own right; as such it might try to maintain as many features of the original as possible (in content and structure)—or, more likely, to free itself from the original, and adapt its content to the cultural expectations and customs of the linguistic community for which the translator works. As Albert La France remarks concerning the French prose version of Nicandro e Fileno, “In fact, the French version is a paraphrase of the libretto in a language familiar to the people of the court, rather than a faithful translation of the Italian.”^[23] Indeed, there was a huge contemporaneous discussion in France about the aims of, and apt means for, translations from foreign languages into French.^[24]

2.5 One need only take a quick look at the excerpt from Le nozze di Peleo e di Theti given here (Figure 1) to realize that on a structural level the Italian and French versions are quite different. On the Italian side, the passage that begins and ends with a refrain (“Deh lascia le pene … E come potrò”) is composed of the regular short lines characteristic of arias and duets of certain types, while the remainder is in versi sciolti, the irregular mixed line lengths of recitative. On the French side, both passages have been rendered in vers alexandrins, the regular long lines that were then characteristic of French spoken drama. The French was probably intended for reading, not reciting; it was certainly not suitable for singing to the music of the opera. Indeed, the majority of contemporaneous French renderings of Italian texts seem to have been intended for silent reading.^[25]

2.6 With regard to the ballets (see Table 1), a reading version of this kind seems to appear for the first time in La Galanterie du temps (dated 1656), which contains both an Italian “Récit” at the beginning (15 lines of verse) and an Italian “Serenata”/“Sérénade” (more than 40 lines) in the last entrée.^[26] From the publication of this mascarade on, until the beginning of the 1660s, the Italian vocal numbers inserted into French ballets were normally accompanied, in the printed livret, by a French counterpart that either (in the case of relatively short numbers) appears directly beneath the Italian text or (in the case of relatively long numbers) is to be found on the opposite page. In other words, it was deemed desirable for the livrets of the ballets to include French poetic analogues for any Italian poems they contained. These contemporary French versions of the Italian texts have not received much attention from literary scholars; they are nearly absent from standard works such as Les Traductions de l’italien en français au XVIIe siècle.^[27] As I will show in the following case study, a closer look at the French versions, as well as a comparison between the poetic and musical transmission of the pieces, sometimes raises intriguing questions about the genesis and performance history of these bilingual ballets.

3. The Ballet de la Raillerie: Remarks about its structure and the state of the sources

3.1 A case in point, as mentioned above, is the Ballet de la Raillerie from 1659, which was the fruit of a collaboration between the French poet Isaac de Benserade (author of the vers du ballet and at least one of the passages sung in French); an unknown Italian poet, perhaps Buti; and Lully (who composed most of the music, with only minor participation by another musician, Jean-Baptiste Boesset).^[28] The Ballet de la Raillerie is one of the ballets with the most numerous and extended passages sung in Italian; in the version given in the livret, it contains three.^[29] The first (pp. 6–9) is an Italian prologue (Figure 2); it is not named as such, but with its three allegorical figures—La Beffa (“Mockery”), La Saviezza (“Wisdom”), and La Pazzia (“Madness”)—it is clearly comparable to an Italian operatic prologue of the time. The second (pp. 16–19), incorporated in the livret as an intermedio, is the famous dialogue for the personifications of Italian and French music. Finally, the ballet concludes (pp. 27–31) with a sort of Italian epilogue (again not indicated as such) in the form of a dispute by several allegorical figures. La Raillerie is thus framed by Italian numbers, a structural scheme that is roughly similar to those of several ballets beginning with La Galanterie du temps (1656) and continuing to the Ballet de l’Impatience (1661).^[30] The ballet part of La Raillerie, on the other hand, is of rather restricted dimensions: it consists of only twelve entrées in a row (six preceding the intermedio and six following it), without further organization into groups.^[31]

3.2 But there are two French vocal numbers, too, namely a duo with instrumental accompaniment within the eighth entrée (p. 21), and the traditional French récit at the very beginning of the ballet (p. 4: see Figure 3).^[32] From these observations it becomes clear that in La Raillerie not only do French and Italian vocal numbers coexist, but in one case they even occupy the same structural position: the opening of the piece is a mixture of the French tradition (a récit, here for an allegory of French Poetry), and the Italian tradition (an opera-like prologue performed by several allegorical figures).^[33] A similar situation can be found at the beginning of the Ballet d’Alcidiane (1658), where an opening récit in Italian is directly followed by one in French—“auec emulation” (“emulating each other”), as the libretto describes it.^[34] In both pieces, thus, a rivalry between the Italian and French music-theatrical styles is staged by opposing not only the two languages, but also the different structural traditions they represent.

3.3 As far as is known today, the Ballet de la Raillerie was performed only four times in a rather short span of time, between February 19 and 23, 1659.^[35] It seems that it was brought to the stage in a form that corresponded broadly to the printed livret. This much can be deduced from the testimony of the eyewitness Jean Loret, who wrote extensively about La Raillerie on the occasion of its first known performance, in his letter of February 22, 1659:^[36] first Loret describes, in perfect concordance with the livret, the ballet proper (including all of the entrées); then he makes (rather general) comments about the music (“symfonie”); and at the end he comments enthusiastically on the singers. Loret’s remarks confirm in general what the livret says regarding the singers involved in the production: both sources mention Hilaire Dupuis, Anne de la Barre, Anna Bergerotti, Claude Le Gros, and Munier Saint-Elme. Only Laurent Hébert, who in the livret is named for the French duo, is absent from Loret’s letter. In any case it is clear that only one of the singers was Italian: the much-acclaimed Anna Bergerotti.^[37] That means that the lion’s share of the ensemble consisted of French singers who for the most part, though, interpreted Italian numbers.

3.4 Since Loret mentions the singer Saint-Elme, who in the livret is named only for the Italian epilogue, it might be assumed that this epilogue was part of the performances. Otherwise one would have to assume that Saint-Elme replaced the singer Hébert in the duo—certainly possible since Loret does not mention Hébert. In the musical transmission, in any case, the epilogue, in contrast to the other vocal numbers of La Raillerie, seems to be missing: it does not survive in any of the Lully sources, which for the final entrée generally transmit only a ritournelle and a dance piece labeled “La Louchie.”^[38] The epilogue is thus one of numerous Italian vocal pieces by Lully that are apparently lost.

3.5 The complicated source situation for the early Lully ballets is well known: for the most part their music is transmitted only in copies that were made after the composer’s death—therefore, some decades after the performances—and that vary from each other to different degrees.^[39] For this reason, a tentative reconstruction of the performed versions of these ballets depends above all on information gained from contemporaneous livrets and accounts of the events.^[40] But sometimes, as is the case with La Raillerie, such a reconstruction seems exceptionally challenging because too many elements remain obscure. Nonetheless, a detailed comparison between the sources can prove useful, insofar as it might allow us to gain insight into compositional or redactional processes, and hence to uncover shifts of meaning. This is the case regarding the dialogue for the personifications of Italian and French music. Three different versions exist (see Table 2) and they relate to each other in a fascinating way: an Italian livret version, a French livret version, and a score version.

4. The dialogue from La Raillerie: Observations on its textual transmission

4.1 In the livret, which would have been printed before the performances of the ballet,^[41] the texts of the Italian prologue and epilogue are loosely paraphrased in the French versions (but in octosyllables and vers libres, not alexandrins: see Figure 2 for the prologue). A similar loose parity characterizes the Italian and French texts of the dialogue, called “Intermedio de la Musica Francese, è la Musica Italiana” and “Intermède de la Musique Françoise, & de la Musique Italienne” respectively (Figure 4). Rather than privileging either language, I shall henceforth call the interlocutors “Italian Music” and “French Music.”

4.2 An English translation of the Italian version of the dialogue may be seen in Example 1.^[42] The dialogue pronounces ideas and concepts that around the middle of the seventeenth century had become topoi in the literature about the respective characteristics of the two musical styles:^[43] Italian music is characterized as strange, exaggerated, and overladen with ornaments; French music is described as melancholic, monotonous, and restrained in the expression of affects. All this is rather conventional for its time, as is also the differentiation between a style of composing and one of singing (a manner of performance). Interestingly, at the end the struggle gives way to the perspective of a possible combination of the most valuable aspects of both styles, so that Italian and French Music sing harmoniously together.

4.3 At some places the French text is more explicit than the Italian one (see Table 2, middle column): for example, in the first passage for French Music, where the Italian version has “Your manner is sometimes strange” (“Strana è ben tal’ hor tua vena”), the French version reads: “The excessive liberties that you take in your songs sometimes make them extravagant” (“Le trop de liberté que tu prends dans tes chants / Les rend par fois extrauagans”). The description of the deficits of Italian music in the eyes of the French (too much license and extravagance) is herewith much clearer. Similarly, the intended reconciliation between the two styles, the combination of an Italian manner of composing with a French manner of singing,^[44] comes across more clearly in the French version: whereas in the Italian version, French Music merely says, “Well, let us admit our respective merits, mine in composing and yours in singing” (“Deh cediam l’vn l’altra il vanto, / Io in comporre,^[45] e tù nel canto”), she goes further in the French version: “I will compose like you if you want so sing like me” (“Ie composeray comme toy, / Si tu veux chanter comme moy”). But in general, the content of the two versions of the dialogue, as Rose Pruiksma has already noted,^[46] is more or less the same. This is not to deny the obvious differences between them. Still, one cannot find here the kind of intentional reworking of an existing text that can be seen, for instance, in the French version of the Italian libretto of Cavalli’s Ercole amante.^[47]

4.4 Nonetheless, it is not quite clear why there should be two complete versions of this dialogue, one in Italian, and the other in French. Pruiksma thought it “possible that this entire dialogue was meant to be sung in Italian.”^[48] This seems unconvincing, though, since the personification of French music singing in Italian would have made a strange effect. It can only have been intended as a mixed Italian-French dialogue from the outset, with the passages for Italian Music sung in Italian (by Signora Bergerotti), and the passages for French Music sung in French (by Mademoiselle de la Barre). The existence of the French versions of the passages meant to be sung in Italian may of course be explained by the desire to make them accessible to the French audience. But what is the reason, then, for the Italian versions of the passages meant to be sung in French? There is a plausible explanation for that: in all likelihood, the dialogue was originally conceived either by an Italian author in Italian, or by a French author in French, and was afterwards translated into the other language in its entirety.

4.5 There is one observation that might speak in favor of arguing that the Italian text came first: the Italian version of the dialogue is homogeneous from a structural point of view (Table 2, left-hand column).^[49] It consists exclusively of ottonari (eight-syllable lines), i.e., structured lines of verse that are especially apt for a musical setting in Italian aria style, half of which are destined for the figure of Italian Music, and the other half for French Music. At the beginning there are six couplets for the two Musics in alternation: a pair of lines for Italian Music answered by a pair for French Music, and so on. These consist entirely of versi piani (lines in which the final stress falls on the penultimate syllable), with paired rhymes: a a b b c c d d e e f f. There follows a quatrain for Italian Music, and another for French Music, in which—probably for the sake of variety—new rhyme schemes are introduced, along with two versi tronchi (lines in which the final stress falls on the last syllable):^[50] g h(vt) g h(vt) i k k i. The concluding passage, the one where Italian and French Music sing together, is again essentially a quatrain, but it is extended to five lines by the repetition of the first line as a refrain (l m m l L).^[51] All in all, the division of the material into passages of various lengths for Italian Music and French Music respectively is exactly evenhanded.

4.6 This is not wholly true, however, of the French version (Table 2, middle column). Although the rhyme scheme at the beginning of the French text is essentially identical to that of the Italian version—i.e., a series of six rhymed couplets, here alternating between feminine and masculine line endings (a’ a’ b  b  c’ c’ d  d  e’ e’ f  f)^[52]—and in what follows is very near to it (g’ h  h  g’ i  k’ i  k’ l’ l’ m  m  l’ m), the closing duo of the French version is longer than that of the Italian version, six different lines of poetry. What is more, the line lengths of the French version as a whole vary between twelve, eight, and ten syllables in an irregular manner; and the textual refrain, which in the Italian version concludes the final duet (and is propitious for a musical setting), is missing here. This could be a reason for arguing that the Italian text was the starting point for the French one (and not vice versa). It would then be fair to assume that the French translator treated the dialogue in a manner like that he adopted in translating the other Italian texts of the livret, namely as if this were a number to be sung in Italian, and he were producing a version intended for reading, not singing. In that case he would not strictly imitate the structural features of the original.

4.7 On the other hand, it is worth emphasizing that the French version also lends itself easily to a musical setting: the free mixture of lines of different length and non-strophic rhyme schemes (vers libres, or vers mêlés) that characterizes the French version of the dialogue gained popularity in French literature around the mid seventeenth century, and vers mêlés were often set to music.^[53] In fact, Lully’s librettist Philippe Quinault would later use them in his tragédies en musique.^[54] Under the circumstances, with only these two texts in the livret to consider, one could propose the opposite argument: that the French version came first, and then an Italian poet “Italianized” it by imposing regularity.

4.8 In any case, there is no evidence that the French text as given in the livret was ever set to music. Instead, the composer was apparently provided with another version of the passages to be sung by French Music. This is shown by a comparison of the livret with the musical scores of the ballet (see Table 2, right-hand column). As no single authoritative source for the music of the early Lully ballets is known, I have taken into consideration five scores, two stemming from the so-called Philidor tradition, two from the Foucault tradition, and one from the Third tradition.^[55] In the passages sung by Italian Music, there are no textual variants whatsoever between the livret and the scores. This is not the case in the passages for French Music: they appear in the scores in yet another version, one that is not totally independent of the French text in the livret but is more structurally congruent with the Italian text of the dialogue. The French passages in the version found in the scores consist entirely of seven-syllable lines—thus, vers impairs: lines with an odd number of syllables, relatively uncommon in classical French poetry. Here French Music’s regular seven-syllable lines are a counterpart to Italian Music’s regular eight-syllable lines. The French lines normally have masculine endings. The only exception to this rule occurs in French Music’s quatrain, where two lines with feminine endings frame two lines with masculine ones. Thus, the introduction of variety in line endings is restricted to the quatrain for each singer: tronco lines for Italian Music, feminine endings for French Music. As a result, the score version restores a perfect structural balance between the parts for French and Italian Music.

4.9 It is possible that the poetry for French Music contained in the livret was the outcome of a misunderstanding (or productive misreading) on the part of the translator—or perhaps Lully requested a modification once he started thinking about a musical setting. Or should we rather assume that the French text was there first, and was then remodeled and translated into Italian? There does exist an additional possibility: namely, that the livret version of the French text (though transmitted earlier) was composed later than the score version, which would mean that the French lines set to music by Lully were later transformed into a “reading version” to be included, for purely literary reasons, in the printed livret. Indeed, as we have seen (Chapter 2 above and Table 1), such French paraphrases of Italian texts were common in the livrets of French ballets at that time. In this case, it would be a matter of paraphrasing a French text for singing, turning it into one for reading.

4.10 In any case, it is the French text of the dialogue that appears as an unstable element in the transmission history: not only is the French text transmitted in two different versions (a livret version and a score version), but the score version itself, with regard to the passages sung in French, is subject to variants, while the Italian text seems to be transmitted uniformly everywhere (see Table 2). Two of the variants in the French text are of an unproblematic nature—“explique” for “exprime” and “touche” for “presse,” both in the final quatrain—since the alternate readings, though not entirely synonymous, do not alter the overall sense of the passage and, moreover, have the same syllable count. The remaining variant, “languissants” for “extravagants” in French Music’s first couplet, is more problematic: the two terms have totally different meanings and different syllable counts. “Languissants” is also used in Italian Music’s second couplet; in all likelihood, its appearance in French Music’s couplet is simply a copying error. The other two variants (“explique” and “touche”) are real alternatives that could point to a troubled history of the French version of the dialogue—one of working and re-working.

4.11 It is difficult to surmise what this might tell us about the genesis of the dialogue. Difficulty in creating the French version may have arisen from the newness of the situation or from lack of time: indeed, as the Gazette tells, La Raillerie was produced in a hurry;^[56] and so far as I know, a mixed Italian-French vocal number (as opposed to adjacent Italian and French numbers) had never been brought onstage before. Be that as it may, one should keep in mind the existence of two versions of the Intermedio/Intermède from La Raillerie (a livret version and a score version) since the differences between them do not end here.

4.12 A translation of the version found in the score may be seen in Example 2.^[57] Superficially it might seem that the score version of the dialogue is not much different from the livret version. A closer examination, though, reveals one big difference: the final duet that is found in the livret is entirely absent from the musical sources. That this is no coincidence is suggested by a detail of the textual transmission: in Table 2 see the last of the six couplets. In the livret version the final reconciliation of the two rivals is prepared here by French Music’s encouragement: “Let us admit our respective merits, mine in composing and yours in singing” (Italian version) or, alternatively, “I will compose like you if you want to sing like me” (French version). In the score version, however, French music declares: “I don’t order anything at all about your tastes, but I want to sing according to mine” (“Je n’ordonne point du tien / Mais je veux chanter au mien”). The perspective of a possible rapprochement has completely disappeared here: both French and Italian music irreconcilably confirm their respective viewpoints. Herewith the message of the dialogue is radically altered: the vision of a peaceful symbiosis between the two musical styles has given way to confrontation and deep-rooted distance.

4.13 At the same time, the reason for the diverse opinions about this dialogue in the secondary literature becomes clear: sometimes reference is made to the livret version with its ending of reconciliation;^[58] sometimes reference is made to the score version with its ending of opposition;^[59] and sometimes the two versions are even intermixed with each other.^[60]

It cannot be determined with certainty when the change at the end of the dialogue occurred: before or during the performances of the ballet in February 1659, or at a later moment. Contemporaneous testimonies do not mention any changes at all;^[61] the known copies of the 1659 livret are uniform in transmitting the long version of the dialogue, including the final duet;^[62] and the scores, entirely uniform in conserving the shortened version, without the final duet, stem from a much later time—the earliest from around 1680 and most from after Lully’s death. In any case, the context for both versions of the dialogue can be approached via the (mainly French) seventeenth-century literature on the respective characteristics of Italian and French music.

5. Comparisons between Italian and French music up to around 1660, and the interpretation and significance of the dialogue from La Raillerie

5.1 The early writings on this topic—that is, commentaries from before 1660, such as those by Marin Mersenne and André Maugars—have some general tendencies in common:^[63] they characterize the Italian musical style as variegated, passionately moving, dramatic, and unencumbered by rules, but also as extravagant; and from a more technical point of view they observe in Italian music the use of chromaticism, numerous repetitions of textual phrases, and a discriminating expression of the affects, but also the fact that this style requires getting used to. The French musical style, on the other hand, is described as graceful, clear, regular, and natural, but also as pedantic; and from a more technical point of view as having a preference for the diatonic genre. In sum, though the French authors from this period assume the superiority of the French musical style over the Italian, they nevertheless normally recognize positive (as well as negative) traits on both sides, and they declare the idea of perfecting the French musical art through the assimilation of certain Italian elements, which should (of course) be moderated by French composers.

5.2 One can easily see how the livret version of the Intermedio/Intermède (with the final duet of reconciliation) fits into this framework: its author was well aware of the traditional outlines of the discussion and took them up with at least a hint of parody (appropriate for the ballet’s general theme of mockery). From this point of view, it is interesting to note that according to the livret version of the dialogue, Italian and French Music would have performed, at the end, a duet together—and that would mean singing simultaneously in their respective languages. By this means, the “goûts réunis” (so to speak) that the text envisages would have been expressed further by a simultaneous mixing of languages in one and the same musical number.

5.3 Lully’s composition of the dialogue reinforces the parodistic component and contributes to the characterization of the two musical styles by illustrating the ideas with music itself (see Figure 5).^[64] At the beginning of the dialogue, Italian Music and French Music are shown reproaching each other for their respective “deficits.” In the composed version each does this by imitating her enemy in an exaggerated manner. Italian Music, for “Tu formar altro non sai / Che languenti e mesti lai” (“You don’t know how to do anything but laments and sad songs”), sings in the style of her adversary by combining, for the word “lai,” a chain of 7-6 suspensions in the bass with pronounced descending half steps in the voice, and by moving in long note values at the end of the passage (Figure 5, p. 32, third and fourth systems). French Music, on the other hand, for “Et croys tu qu’on ayme mieux / Les longs fredons ennuyeux” (“And do you think that people prefer long boring melismas?”), imitates the sequential coloratura of the Italian style, after a transition to an Italianate triple meter (Figure 5, p. 33, first two systems). This parodistic mimicry comes to an end with French Music’s affirmation that she wants to remain faithful to her own taste and manner of singing (“Mais je veux chanter au mien”): here French Music confronts the “boring” and repetitive melismas she had assigned to Italian Music with her own syncopated and detailed ornaments (Figure 5, p. 33, fourth system); this is also the style in which French Music sings at the very end of the piece as it is transmitted in the score (Figure 5, starting on the last system on p. 34). Italian Music, too, sings in her own style: for “Chi risente un mal di morte / Più che può grida mercé” (“She who feels a pain unto death cries ‘mercy’ as much as she can”) she uses ascending chromaticism and large leaps in changing directions (Figure 5, p. 34, second and third systems). One might say, therefore, that Lully and his poet(s) created a relatively complete picture of traditional assumptions and prejudices about the Italian and French musical styles, or, in other words, that they put them on the level of “metareference” and made them audible in one and the same piece.^[65] Of course one would like to know, as Philippe Beaussant has pointed out, how the piece was performed in its own time:^[66] perhaps with different continuo groups for the two musics, and with an exaggerated interpretation, emphasizing the metareferential aspects—as Rachel Redmond and Anna Reinhold demonstrated in their performance in 2011 under the direction of William Christie (Video 1).

5.4 Nonetheless, in view of the aforementioned commentaries comparing Italian and French music, which encouraged French music to assimilate certain Italianate elements, the score version of the dialogue takes a novel approach: its message of irreconcilable differences between the Italian and French musical styles is apparently entirely new. This message, though, does have a context.

5.5 In April 1659, only two months after the documented performances of the Ballet de la Raillerie, the very first French opera had its premiere: the Pastorale by the librettist Pierre Perrin and the composer Robert Cambert.^[67] Given first in a private setting and repeated for the royal court a month later, this opera was meant to be the exact opposite of Italian opera and to lay the foundations for a French operatic tradition in the authors’ and audience’s native language. As is well known, Perrin commented upon his piece in a manifesto (the famous Lettre) that today is considered the first true poetics of French opera.^[68] This text is in one way radically different from the earlier comparisons of Italian and French music: Perrin straightforwardly condemns the Italian style in every respect in order to place it in stark opposition to French music, the latter with its alleged regularity and naturalness. In Perrin’s view, a synthesis of the two styles is completely unthinkable. Clearly the altered version of the dialogue from La Raillerie transmits a similar message.

5.6 An important background for Perrin’s Lettre is the gradual discrediting of Italianism that can be observed in France during the 1650s, and a corresponding awakening of French national pride.^[69] The fact that both Italian and French numbers were staged many times in the ballets from these years could have been seen by aestheticians as an advantage, since it might have sharpened the awareness of the stylistic differences, and consequently might have helped to define French music more precisely.^[70] But against this background the older ideal of a symbiosis between French and Italian music must certainly have lost its attractiveness. Was the altered version of the dialogue a reaction to this new situation? Obviously, we cannot be sure about that. Indeed, if the changes in the dialogue had already been made in 1659, they would have led to a paradoxical situation: the audience reading one version (and its message) in the livret and hearing another version (and its completely transformed message) onstage.

5.7 It is unlikely that the score version resulted from later seventeenth-century redactional processes, for two reasons: the lost early “master score” of La Raillerie, from which the existing copies stem, must have contained the changed ending since it is common to all traditions;^[71] and Philidor (who would have been aware of the final duet in the livret) evidently hoped to find the missing music somewhere because he left room for it, and even notated the clefs, in his copy from 1690 (Figure 5, p. 35).^[72] It is improbable, therefore, that the elimination of the final duet had anything to do with tendencies to revise the posthumous picture of Lully for French music history (i.e., by styling him as a “real French composer” who would never have considered mixing the Italian and French musical styles). Quite the opposite seems to be true: although Philidor probably did not think of his copies as a “critical edition” of Lully’s music, in the modern sense, he nevertheless generally—and certainly in this case—tried to reconstruct the “original” form of the ballets, so far as possible, by consulting the livrets.^[73]

5.8 It follows that there are two likely possibilities: either the score version of the dialogue (with the missing final duet) is an indication of later performances of La Raillerie for which no descriptions have been found;^[74] or the change had already been made before or during the performances in February 1659 without this having been reflected in the livrets. It can be deduced, thus, that the dialogue of La Raillerie was “in a state of flux.”^[75] For this reason we should be very cautious about drawing any conclusions from the situation; nevertheless, we should always keep in mind the existence of two different versions that must be treated separately. Though we might assume that the reworking of the dialogue was influenced, in one way or another, by contemporary cultural and political circumstances, the main point is that over the course of its transmission history, the dialogue acquired two entirely different meanings.

6. Conclusions

6.1 The dialogue between Italian and French music in the Ballet de la Raillerie does not mark an end to the (overwhelmingly French) debates about these two musical idioms in the seventeenth century—it is only part of an enduring quarrel—nor was it a turning point in Lully’s career as composer: there continued to be Italian numbers in French ballets even after the known performances of La Raillerie in 1659 (see Table 1). This was the case even though Italian influence at the French court gradually diminished during the 1660s—i.e., after Mazarin’s death in 1661 (at which time Louis XIV declared that he would rule alone, without a prime minister). Even the idea of bringing the Italian-French querelle onstage had a continuing history. An example of this is the pastoral opera Nicandro e Fileno, mentioned above, which was performed at the royal court in 1681: its spoken prologue and intermedi, performed by members of both the Comédie Italienne and the Comédie Française, were all dedicated to the discussion of the respective merits of Italian and French opera (the latter now a mature art form though it had been in its infancy in 1659).^[76] In the case of Nicandro e Fileno, then, there was a “metatheatrical” frame to the performance of an Italian opera in France—the only Italian opera to have been performed there in a long time.^[77]

6.2 The idea of juxtaposing the Italian and French languages within one and the same musical number is also not completely absent from the French stage after the Ballet de la Raillerie: in Lully’s Carnaval mascarade, LWV 52, staged at the Opéra in Paris in 1675, there is, in fact, a number in which one character (Monsieur de Pourceaugnac) sings in Italian, and two others (comic lawyers) in French.^[78] Though this number is not directly comparable to the dialogue from La Raillerie, and though the juxtaposition of two different languages seems to fulfill other dramaturgical functions in Le Carnaval mascarade, this piece shows at least that the idea of bilingualism for sung texts was not considered all that strange in the seventeenth century. A final example, from the beginning of the eighteenth century, might be cited in this connection: the prologue of Méléagre (1709) by the Italian-born composer Jean-Baptiste (Batistin) Stuck, on a libretto by François-Antoine Jolly, brought onstage a quarrel by allegories of Italy and France about their respective musical merits—a quarrel that is finally resolved by Apollo.^[79] Like the tragédie itself, this prologue is written in French, with one singular exception: a number sung by one of the followers of Italy has an Italian text, and this is set to music (not surprisingly for its time) as a da capo aria.

6.3 These later examples aside, the significance of the dialogue from La Raillerie can be outlined relatively precisely. Surely it was no manifesto with far-reaching consequences, though the idea of bringing the personifications of both Italian and French music onstage in one and the same musical number, discussing (and even imitating) their respective characteristics in an ironical manner, and joining them in a bilingual duet—and all this interpreted by two of the era’s greatest singers from both nations—was highly original. The dialogue may be recognized as a product of the 1650s, when court ballet had playful traits, in contrast with the 1660s, when it became more “absolutist” and “monarchic” in its focus.^[80] Moreover, the dialogue was one of many instances of Lully’s and his librettists’ efforts around the middle of the seventeenth century “to use the medium of art to attest to their own actions: thus, to write and compose pieces that thematicized the framework of art itself.”^[81] It is perhaps no coincidence, then, that the dialogue from La Raillerie forms part of a discussion about Italian and French music and art that must sometimes have been virulent, but that the piece itself, subject to the fluid compositional process of the theater, does not represent a clear point of view in that discussion.

Acknowledgements

This article is an expanded version of a paper presented at the 2014 annual conference of the Society for Seventeenth Century Music (San Antonio, Texas) and the 2014 Biennial International Conference on Baroque Music (Salzburg). I would like to express my thanks to those who have commented upon the paper and made useful suggestions, especially Don Fader, Rebecca Harris-Warrick, Margaret Murata, and Louise Stein, as well as the anonymous readers of an earlier version of the article. Heartfelt thanks to Lois Rosow for her improvements of my English, her marvelous editorial work on my text, and her many thoughts and comments that contributed immensely to the article’s final form.

Examples

Example 1. Ballet de la Raillerie, Italian version of dialogue, from livret

Example 2. Ballet de la Raillerie, Bilingual version of dialogue, from score

Figures

Figure 1. Les Nopces de Pélée et de Thétis: excerpt from bilingual opera libretto

Figure 2. Ballet de la Raillerie: beginning of Italian prologo in livret

Figure 3. Ballet de la Raillerie: French récit and introduction to Italian prologo in livret

Figure 4. Ballet de la Raillerie: Intermedio/Intermède in livret

Figure 5. Ballet de la Raillerie, Intermedio/Intermède in score copied by Philidor

Tables

Table 1. French ballets including passages sung in Italian, ca. 1644–1664

Table 2. The textual versions of the Intermedio/Intermède from Ballet de la Raillerie

Video Example

Video 1. Ballet de la Raillerie, Intermedio/Intermède