the "stylized" dance in italian sonatas of the late baroque
MICHAEL TALBOT University of Liverpool
Izvleček: Raznolikost značajev plesnih stavkov v italijanskih poznobaročnih sonatah se pripisuje »stilizaciji« oz. »idealizaciji «, vendar soobstoj zelo podobnih stavkov z ali brez plesnih naslovov v tem istem repertoarju, ki ga ponazarjajo Albinonijeve in Vivaldijeve sonate, kaže, da ti naslovi služijo kot »parateksti« v smislu definicije literarnega teoretika Gérarda Genettea: kot verbalni pripomočki, ki bralca vodijo k boljšemu poznavanju umetnine.
Ključne besede: Antonio Vivaldi, Tomaso Albinoni, Gérard Genette, paratekst, plesni naslovi.
Abstract: The variability of character of the dance movements in Italian sonatas of the late Baroque is conventionally ascribed to their "stylization" or "idealization ", but the coexistence of very similar movements both with and without dance titles within this repertory, exemplified by sonatas of Albinoni and Vivaldi, suggests that these titles also serve as "paratexts " as defined by the literary theorist Gérard Genette: verbal supplements that guide the reader towards better appreciation of the artwork. Keywords: Antonio Vivaldi, Tomaso Albinoni, Gérard Genette, paratext, dance titles.
Consider the three allemandas and three correnti shown in Example 1. They come from six works included in a manuscript set of twelve violin sonatas by Antonio Vivaldi (the so-called "Manchester" Sonatas) compiled c. 1726 for presentation to Cardinal Pietro Ottoboni, the music-loving Roman prelate known for his patronage of Corelli and Handel, among many others. What must strike anyone with the least familiarity with eighteenth-century dance music is their heterogeneity. It would be hard to deduce from them without reference to external models which were the more typical, which the less typical, specimens of their respective type. Beyond the facts that an allemanda is in quadruple metre, a corrente in triple metre, and that neither dance is slow, it would be difficult to pin down any particular character. Nor does any of the incipits seem especially "danceable".
The standard explanation of this variability, encountered, for example, in William S. Newman's classic study of the Baroque sonata,1 is that the original dances have become "idealized" or "stylized". That is, in accordance with their newly acquired function as movements within a sonata played for enjoyment rather than as an accompaniment to actual dancing, a process of refinement and diversification has been permitted to take place, with the result that what were once the indispensable trademarks of each dance have now become
1 William S. Newman, The Sonata in the Baroque Era, Chapel Hill, The University of North Carolina Press, 1959.
Example 1
Specimens of allemandas and correnti in Vivaldi's "Manchester" Sonatas (c. 1726)
mere options among others. Thus the allemanda need not begin with a short upbeat giving the dancer advance warning of his or her first step, and the corrente may dispense with the pas de courante, the traditional pre-cadential hemiola. A dance may be either faster or slower than would be possible in the ballroom, and there is no longer any need to observe symmetry of phrase structure. The situation is exactly comparable with that of Schubert's waltzes, Chopin's mazurkas or Smetana's polkas.
In purely practical terms, such stylization was a boon to composers. First, it ensured that when many examples of a given dance appeared in close proximity to one another - the "Manchester" Sonatas contain no fewer than eight allemandas and thirteen correnti - each movement could be given individual characteristics. Second, it made the border between dance movements and "abstract" movements more porous: material first appearing in a dance movement could later be recycled in an abstract context.2 The Allemanda and Corrente of Vivaldi's sonata RV 757, illustrated in Example 1, became employed as the principal episodic material in the fast outer movements of a violin concerto,
2 The term "abstract" is perhaps not a particularly happy one for the description of movements lacking a title and identified only by a tempo marking, but since it is so widely used it will be retained here.
RV 329,3 while the Allemanda of RV 3 served similarly for the first movement of a flute concerto, RV 101/437.
However, this porosity operated in both directions. It became possible for a movement not originally conceived as a dance, if compatible in form (binary), tempo (moderate to quick) and scale (small to medium), to be given a dance title when the circumstances encouraged this. Such appears to be the case with the Sarabanda of the eleventh "Manchester" Sonata, which deviates from the norm in two ways. First, it is not cast in the standard binary form with two repeated sections.4 Second, it is in the relative minor key, C minor, not the tonic, E flat major.5 It is a virtual certainty that this movement was originally conceived as abstract and acquired its dance title only in order to become more congruent with the rest of the set.
Yet the vast majority of originally abstract movements pressed into service as dances were in fact themselves binary-form structures - whether symmetrical or asymmetrical, rounded or non-rounded - and in that respect were indistinguishable from their dance counterparts. The story of abstract binary movements remains to be written in detail. In Italian music we encounter them as early as 1607 in Salomone Rossi's Sinfonie e gagliarde a tre, and for a long time they appear mostly as partners to dances, which, during most of the seventeenth century are not organized in the sources into regular suites. Rossi's sinfonie are, in fact, the distant ancestors of what sonatas of the Corellian age will know as preludii or introduzioni. While it remained a multi-sectional composition comprising a succession of contrasting through-composed episodes (some of which might be reprised), the sonata was inhospitable to binary form, but after the mid-seventeenth century, when the multi-movement sonata took root, the possibility arose of substituting binary form, in one or more movements, for the through-composed form traditionally applied to music based on imitative processes. Because binary form - highly compact and associated with dance music - connoted lightness, such movements were generally placed at or near the end of the sonata, but over time also the first fast movement (which, of course, was often the second movement of the cycle) became eligible for binary form. We see here an incremental process whereby over the time-span 1650-1750 binary form progressively ousted unitary (i.e., through-composed) form in fast movements and made significant inroads in slow ones, too. The process accelerated after 1700 with the selective absorption of features from the sonata da camera in the new-style, multi-purpose kind of sonata typified by the violin sonatas of Corelli's Op. 5 (1700).
3	For an analysis of the reuse of the sonata's material within the concerto, see the author's introduction to Antonio Vivaldi, Suonate a Violino solo e Basso per il Cembalo (le sonate "di Manchester") [facsimile edition], ed. Michael Talbot, Firenze, S.P.E.S., 2004, pp. 9-55: 36-38. Scholarly modern editions of the sonatas also exist: the present writer has edited the full set in a single volume (as Antonio Vivaldi: The Manchester Sonatas, Madison, A-R Editions, 1976), and, in collaboration with Paul Everett, also each sonata singly (Milan, Istituto Italiano Antonio Vivaldi/Ricordi, 1982-1983, as part of the series "Nuova edizione critica delle opere di Antonio Vivaldi").
4	It is arguably in a variety of binary form without repeats, although activity in the bass bridges the point of junction of the two sections.
5	In the traditional chamber sonata constructed on Corellian principles the dance movements are invariably in the key of the work as a whole, the only movements permitted to adopt different keys being internal "abstract" slow movements.
In that sense, many nominally abstract binary-form movements in sonatas composed after 1700 are potentially classifiable as dances, provided that their character falls within the range of stylizations admissible for a given type. Consider the six movements whose openings are shown in Example 2. They come from the Trattenimenti armoniciper camera, Op. 6, of Tomaso Albinoni, published around 1711.6 Applying the rather loose standards of Vivaldi's "Manchester" Sonatas, the composer could, had he been so minded, have named the first three movements illustrated "Allemanda" and the last three "Corrente". The fact is, however, that he chose not to: ostensibly, at any rate, the Trattenimenti musicali remain from start to finish abstract music - even when some of the movements in compound metre scream "Giga" at the player.
Example 2
In considering the reasons why a composer may in the given circumstances choose to adopt, or not to adopt, dance titles (or a compromise formula such as "Tempo di gavotta"), we have to abandon the concept of "stylized dances" and look for an alternative model. I believe that one lies ready to hand in the concept of "paratexts" as formulated by the French literary theorist Gérard Genette.
6 A modern edition by the present writer of the Trattenimenti armonici per camera, which comprise twelve four-movement violin sonatas, appeared in three volumes from European Music Archive, London, 1981 (Sonatas 1-4), 1986 (Sonatas 5-8) and 2001 (Sonatas 9-12).
To explain what a paratext is, one need merely quote the first sentences of the introduction to Genette's book Paratexts: Thresholds of Interpretation.1 The author is writing of literary works, but his definition is perfectly usable also for musical works.
A literary work consists [...] of a text defined (very minimally) as a [...] sequence of verbal statements that are [...] endowed with significance. But the text is rarely presented in an unadorned state, unreinforced and unaccompanied by a certain number of verbal or other productions, such as an author's name, a title, a preface, illustrations. And although we do not always know whether these productions are to be regarded as belonging to the text, in any case they surround it and extend it, precisely in order to "present" it, in the usual sense of this verb but also in the strongest sense: to "make present", to ensure the text's presence in the world, its "reception" and consumption [...].
Transferring Genette's concept to a musical context, we see that to call a movement an allemanda is to make a decision that affects its reception and consumption. Conversely, not to do so when the possibility exists constitutes a decision of the same order. The vital question to ask at this point is: which factors encourage a composer to provide the dance title, and which to dissuade him from doing so, in a situation where, because of the tolerance introduced to the world of dance music via the process of stylization, either decision is in principle readily acceptable?
I believe that at least part of the answer lies in the social cachet - the "aura", one might say - of dance music in the Italian society of the day. The five dances that, between them, made up the vast bulk of the movements in a post-1680 chamber sonata - the allemanda, corrente, sarabanda, giga and gavotta - were associated intimately with the lifestyle of the nobility.8 Whereas in more recent ages leisure has to observe some boundaries in order not to attract censure as idleness, in Vivaldi's and Albinoni's day it was widely regarded, rather, as an ennobling good of which it was impossible to have too much. An allemanda or corrente thus pays homage to the noble caste. If the recipient - player or listener - is himself or herself a noble, the movement is an affirmation of status. If he or she is not, it offers a vicarious glimpse of a social ideal. However, this identification is bought at a price: an explicit allemanda is not suitable for performance in church on account of its worldli-ness, and perhaps also sits a little uneasily in the setting of a music society (accademia) or bourgeois home.
The "titling strategy" of composers therefore has a lot to do with the anticipated destination of a work or group of works. As we know, Vivaldi's "Manchester" Sonatas were copied out specially for presentation to Cardinal Ottoboni, to whom Corelli had dedicated the chamber sonatas of Op. 4 in 1694, and who would later receive similar published col-
7	Gérard Genette, Paratexts, Thresholds of Interpretation, translated by Jane E. Lewin, Cambridge, Cambridge University Press, 1997, p. 1. In the original French-language edition (Paris, Éditions du Seuil, 1987) the book is entitled simply Seuils [Thresholds].
8	Newman's preference for "court sonata" over "chamber sonata" as a translation of sonata da camera speaks for itself.
lections from Giovanni Battista Somis (Op. 4, 1726) and Giovanni Mossi (Op. 6, 1733). By ensuring that his offering to all appearances conformed strictly to the da camera stereotype (even though the usual boundaries of some dances were on occasion overstepped), Vivaldi celebrated the courtly life of the cardinal in his Roman palace, the Cancelleria. It is a telling fact that in his other violin sonatas preserved in manuscript, several of which are concordant in one or more movements with those of the Manchester set, Vivaldi only rarely opted for dance titles. These were destined for ordinary customers or professional colleagues and therefore had no need to restrict their possible future use by advertising whatever dance characteristics they possessed.
If we move on to Albinoni's Trattenimenti armonici, we see that here, too, the choice is rational. This is apparently the earliest known collection of sonatas by a composer resident in Italy that intentionally bypassed local printers in favour of the publishing house of Estienne Roger in Amsterdam. Its prime market was therefore transalpine and bourgeois, even though the sonatas were dedicated to a minor Venetian noble, Giovanni Francesco Zeno, who appears to have been a violin pupil of the composer. To have engraved "Alle-manda" or "Giga" over appropriate movements might, certainly, have conjured up pleasant visions - but it might easily also have injected an unwelcome note of levity. Seriousness and total uniformity of layout characterize Albinoni's set. To have labelled the final movements variously "Corrente", "Giga" and "Gavotta" would have disrupted the regularity with which one plain "Allegro" follows another. Albinoni is far from averse, per se, to employing dance titles. In his published music he does so in his Op. 3 (Balletti a tre, 1701, dedicated to Grand Prince Ferdinando of Tuscany) and in his Op. 8 (Balletti e sonate a tre, 1722, dedicated to Count Christian Heinrich von Watzdorf); but in both those instances it is appropriate to allude overtly to a pastime of a distinguished noble patron and (one might add) to adhere to traditional models more closely than Vivaldi usually does.9
To sum up: in the late Baroque period we witness a stylistic confluence between genuine dance movements, whose elasticity has been increased by stylization, and nominally abstract movements identical to them in form. This creates an intermediate zone, a borderland, within which movements may, or may not, be given dance titles by the composer. In such circumstances, dance titles cease to act as reliable generic labels, still less to define function: the titles become, instead, paratexts according to Genette's definition - descriptions that aim to influence the reception of the movement by making an allusion to a certain lifestyle and ambience. When such allusion is perceived as appropriate and advantageous, the titles are retained - or even newly created. Where they are viewed as incongruous, they are not adopted (and perhaps even suppressed where once present, although I can recall few concrete instances of this). In general - and one could illustrate this more widely by reference to collections of violin sonatas by Giuseppe Valentini (his Allettamenti per camera, Op. 8, of 1714) and many others - what goes for one sonata goes for the entire set: uniformity of presentation appears to enjoy a high value in its own right.
9 In the "Manchester" Sonatas there is only one work, RV 12 (Sonata 2), that conforms to the "classic" outline of the sonata da camera by featuring the sequence Preludio-Corrente-Giga-Gavotta, thereby eschewing the internal slow movement (which, when in triple metre, is - often rather unconvincingly - labelled "Sarabanda" but otherwise is left without a dance title).
Looking more widely, I can see that Genette's beautifully simple concept of the paratext has wide potential applicability to musicology. It connects with the idea of the "generic contract" recently popularized by Jeffrey Kallberg,10 but is a more flexible analytical tool and has the virtue of creating parallels with the other arts, visual as well as literary.
»STILIZIRANI« PLESI V ITALIJANSKIH SONATAH POZNEGA BAROKA
Povzetek
Članek dokazuje, zlasti na podlagi Vivaldijevih »manchestrskih« violinskih sonat (ok. 1726) in Albinonijevih Trattenmentiarmoniciper camera, op. 6 (ok. 1712), da močno spremenljive narave stavkov, ki v italijanskih poznobaročnih sonatah nosijo ime določenega plesa, ni mogoče v celoti razložiti s tradicionalnim sklicevanjem na njihovo »stilizacijo« ali »ideali-zacijo«. S privzemom binarne oblike v 17. stoletju tudi za abstraktne stavke, s težnjo torej, ki se je polno razmahnila po letu 1700, se je ustvarilo vmesno področje, v katerem je imel skladatelj na izbiro kar precej možnosti za odločitev, ali bo uporabil plesne naslove ali ne. »Manchestrske« sonate so naslovljene s plesnimi imeni celo v primerih, kjer je bil naslov očitno dodan post factum. Nasprotno pa jih Trattenimenti armonici ne uporabljajo niti takrat, kadar je plesni značaj povsem očiten.
Za razlago tega navidezno protislovnega položaja je potrebno pojem »stiliziranih« plesov dopolniti z drugim konceptom: z uporabo plesnih naslovov kot »paratekstov« (izraz je skoval francoski literarni teoretik Gérard Genette). Paratekst je opis zunaj samega teksta, njegov namen pa je usmerjati recepcijo dela pri porabniku (kar pri glasbi vključuje tako izvajalca kot poslušalca). Zaradi povezovanja »standardnih« plesov (allemanda, corrente, sarabanda, giga, gavotta) z dvorskim življenjem so skladatelji radi ohranjali ali uvajali plesne naslove pri glasbi, namenjeni dvorskemu okolju. Nasprotno pa so bili pri glasbi, namenjeni meščanskemu tržišču ali možni izvedbi v cerkvi, plesni naslovi manj primerni.
Članek se konča z razmislekom, da je Genettov koncept parateksta, ki se očitno navezuje na idejo Jeffreya Kallberga o »splošnem dogovoru«, za preučevanje glasbe široko uporaben.
10 Jeffrey Kallberg, The Rhetoric of Genre: Chopin's Nocturne in G Minor, 19th-century Music 11 (1987-8), pp. 238-261.