The 'sonata da camera' before Corelli: a renewed search.
At least among the duos and trios, abstract sonatas dominated the printed repertory; only after 1665 did volumes of dances begin to appear as frequently. However, since the wording of title-pages offers only an incomplete guide to the uses of this music, we must often turn to musical content and style, and to norms of scoring and instrumentation, in order to identify the likely functions of each volume or piece. Fortunately, it is quite a straightforward task to divide the pre-Corellian volumes into 'abstract' and 'dance' categories. Some volumes did contain both dances and abstract sonatas, but in most decades before 1670 these amounted to fewer than one-fifth of the published volumes of instrumental music.(4)
Although one can associate likely sacred or secular contexts with much of the earlier seventeenth-century repertory, by the 1660s printed instrumental music had begun to strain hard against the stylistic boundaries imposed by a particular social function. Elements of the abstract sonata and of the dance are found juxtaposed in single works, as long and sometimes virtuoso sonatas incorporate explicit dance movements, and sets of elaborate dance movements include free sections labelled only by tempo designations. The sonatas of Guerrieri (1673e) and Arresti (1665e) are interesting as examples of this trend towards merging the sonata and dance elements. In Arresti's sonatas, one finds triple-metre solo sections called 'arie', and two of Guerrieri's sonatas, 'La Tita, a violino solo' and 'La Benedetta, a 2 violini', conclude with a corrente (so labelled). Several other sonatas in Guerrieri's volume (Nos. 2, 4, 13 & 17) include an arietta, usually as the final movement. Moreover, during the same period, dances were increasingly provided with tempo markings, sometimes two in the course of one movement (e.g. 'Largo' and 'Allegro' in the fourth corrente of Giovanni Battista Vitali's Op. 4, 1668e), suggesting that they were now less suited to a ballroom than to a room filled with listeners. It was at this point, as the music itself began to offer a less and less reliable guide to social function, that the specific label 'sonata da camera' began to appear more frequently, with the stylistic implications familiar to us from Corelli's printed collections.
Beyond its generic use to designate a piece to be played rather than sung, the term 'sonata' had meant one of three things to Italian composers before the mid century: a set of variations over a tune or ground bass (both Salamone Rossi and Buonamente used the term in this way); a short piece in binary form (again, there are examples in the works of Rossi and Buonamente, as well as of Marini); or a work similar in many respects to the ensemble canzona, often with forces reduced to two or three players (see examples by Cima in 1610d, Corradini in 1624a and Cecchino in 1628e). Indeed, the terms 'sonata' and 'canzone' seem to have been favoured by composers trained respectively as virtuoso performers and as church organists, although the works themselves are often indistinguishable in stylistic terms.(5)
Of the various types of sonata, only the variation and binary ones appear to have had primarily a secular function, although the latter could have served as well to introduce sacred concerti performed in church. Rossi includes both binary and variation sonatas in his third and fourth books (1613k; 1622b); and Buonamente incorporates these 'secular' types of sonata in Books 4 and 7 (1626b; 1637d), published in three partbooks, the unfigured bass partbook calling for a basso di viola (basso di viola o da brazzo in 1637d). In contrast, only canzona-style sonatas are found in Buonamente's sixth book (1636), which has both figured (basso continuo) and unfigured bass partbooks. Since Books 4 and 7 (and also Book 5) contain sonatas, sinfonias and a variety of dances, whereas Book 6 contains only sonatas and canzonas, it seems reasonable to regard the binary and variation sonatas as intended primarily for performance in a secular context appropriate for dances.
After the mid century, the canzona-style sonata was the most common type, having developed from a piece in several sections into a work made up of several well-defined movements.(6) Such sonatas in volumes by Vitali, Bononcini, Cazzati and others provide obvious printed models for Corelli's Op. 1 Sonate of 1681. These composers were employed, and had their works published, in the north of Italy; but Roman composers, whose works survive primarily in manuscript, must also have influenced Corelli's sonata style.(7) Born in Fusignano on 17 February 1653, Corelli probably first studied with the local clergy; he received further instruction in nearby Lugo and Faenza, and finally in Bologna, where he went in 1666. He was admitted to the Accademia Filarmonica there in 1670, suggesting that he was already considered a competent composer. He was in Rome certainly by 1675, and perhaps as early as 1671. But if Roman and Bolognese models for Corelli's abstract sonatas are easily identified, Italian sources for his chamber sonatas have been somewhat harder to trace.
If we take the twelve sonatas of his Op. 2 as a model, Corelli's sonata da camera is a set of two or three dances, which more often than not follow an introductory movement:(8)
1 Preludio, Allemanda, Corrente, Gavotta (D)
2 Allemanda, Corrente, Giga (d)
3 Preludio, Allemanda, Adagio, Allemanda (C)
4 Preludio, Allemanda, Grave-Adagio, Giga (e)
5 Preludio, Allemanda, Sarabanda, Gavotta (B flat)
6 Allemanda, Corrente, Giga (g)
7 Preludio, Allemanda, Corrente, Giga (F)
8 Preludio, Allemanda, Sarabanda, Gavotta (b)
9 Allemanda, Sarabanda, Giga (f sharp)
10 Preludio, Allemanda, Sarabanda, Corrente (E)
11 Preludio, Allemanda, Giga (E flat)
12 Ciaccona (G).
Given the dearth of obvious Italian precedents for these sonatas, at least one scholar has looked elsewhere in order to explain their 'sudden' appearance in 1685. John Daverio argues not only that dance suites were much more common in Austro-German and English sources than in those from Italy but also that Georg Muffat may have served as a primary conduit through which Corelli caught on to the idea of grouping particular dances to form a chamber sonata.(9) The statistics Daverio presents are impressive: in Italian publications he found only 30 examples of rudimentary suites (i.e., preludial movements followed by dances clearly specified as belonging together), but ten times that number (300) in 27 contemporary Austrian and German sources (1650-1700). England, too, made a contribution, with fantasia-suites in which an 'imitative fantasia usually served to introduce a group of two or three stylized dances';(10) between 1625 and 1665 composers of such suites included Lawes, Locke, Coprario, Gibbons and Simpson. It is thus undeniable that the suite of dances for instrumental ensemble was well established outside Italy. But groups of dances that belong together were not quite so unknown to Corelli's Italian predecessors as the sources would seem to suggest. Indeed, I demonstrate below that printed Italian sources, published for the most part in northern Italy, especially in Bologna, provided models not only for Corelli's abstract sonatas but also for his sonate da camera.
SUITES IDENTIFIED IN THE SOURCES
In search of an Italian context for Corelli's sonate da camera, we might look first at suites that are explicitly identified as such in the sources. The relatively few examples follow various models, illustrated in Table I. Their composers include Buonamente, [TABULAR DATA FOR TABLE I OMITTED] Marini, Cazzati, Rosenmuller and Guerrieri. Antonio Brunelli's early contribution was simply an arrangement for two instruments and continuo of the vocal piece 'Di quel nudo pargoletto', consisting of a thematically related Ballo, Gagliarda and Corrente.(11) This seems to be a ballo in the sense of a multi-movement, quasi-dramatic presentation; for many individual pieces in Brunelli's print, the table of contents names the dancers who participated in the performance. The Gagliarda and Corrente are set in both vocal and instrumental versions.
Still well before the mid century, Buonamente left ten suites in two volumes (1629a; 1637d), each consisting of a sinfonia followed by two or more dances: a gagliarda and 'sua corrente' in the earlier volume, and a brando, gagliarda and corrente in the later one, where the connection is made explicit by a statement printed between the sonatas and the sinfonias and dances: 'Ogni sinfonia ha il suo Brando, Gagliarda e Corrente'.(12) One difference between the two volumes is in the character of the sinfonias: in 1637d they are longer and all but one incorporate a change of metre. Both title-pages give the printer Alessandro Vincenti responsibility for collecting the volume's contents, and he signed both dedications; Buonamente himself signed the dedication of at least two other volumes (Il quarto libro, 1626d, and the collection of sonatas and canzonas published by Vincenti in 1636). But in the 1629a dedication, the publisher claims status as a 'virtuoso Ladro', a thief who 'copied these sonatas in secret' out of admiration for the composer and who now wished to repay the debt by bringing them into print. But even if Vincenti was entirely responsible for publishing these dances, it seems unreasonable to give him all the credit for the organization of the volume into suites, since his other collections of dance music show no such devotion to this new concept. Buonamente probably intended the grouping of dances as suites, and he may thus deserve some recognition as an innovator. Or perhaps the collection simply reflects Vincenti's adoption in these two volumes of a habit that may have been common among performers, who may regularly have created suites from collections in which the dances were grouped by type.
Between 1620 and 1660, Marini and Cazzati employed two other models for suites. One involves a set of variations followed by one or two dances. Cazzati's grouping of Aria-Ballo dell'Aria-Sua corrente in his Op. 22 (1660a) is a good example.(13) Marini had included similar suites in two earlier volumes (1620g; 1655a). In 1620g, there is a 'Romanesca per violino solo e basso se piace', with four variations over the romanesca bass, followed by a Gagliarda and Corrente also built on the romanesca pattern. ('Basso se piace' here means that a melodic bass instrument may join the chordal basso continuo if desired.) Both dances are in binary form, and the Gagliarda is composed of two complete binary units, labelled 'Prima' and 'Seconda parte'. A suite in Op. 22 (1655a) labelled 'Balletto' has a comparable structure, in which all five parts (the last is labelled 'Corrente') are thematically related. The second model used by these same two composers consists of five or more movements, such as the Entrata-Balletto-Gagliarda-Corrente-Retirata from Marini's Op. 22 (1655a).(14) Cazzati had used a similar format in 1651d: Entrata-Balletto-Trecia del balletto-Gagliarda-Corrente.
In the 1670s, Guerrieri drew on two established models in a collection that consists primarily of abstract sonatas (1673e). The title of this volume promises Sonate di violini a 220.127.116.11. per chiesa e anco aggionta per camera. Although several of the sonate per chiesa may have been performed not only in church but also in the chamber, perhaps with theorbo and basso di viola substituting for organ ('La Benedetta' and 'La Rosciana' carry the instruction 'Qual si puo anco sonare senza Organo pero con Teorba, e Basso di Viola'), the 'additions for the chamber' are presumably the three sonatas found at the end of the volume: 'Baletto Primo per Camera', 'Corrente al Baletto', 'Sarabanda'; 'Baletto Secondo', 'Corente al Baletto', 'Sarabanda'; and 'Partita Sopra Ruggiero' with a 'Corente al Ruggiero'. In these three pieces, Guerrieri adopted Buonamente's approach in combining a duple-metre dance with two in triple metre;(15) but like Cazzati and Marini, he also provided a set of variations followed by its corrente.
One other model for the suite was found in the brando (branle), a functional dance-cycle used at the Este court in Modena and found in collections by Bononcini, Colombi, Vitali and Uccellini. These brando-suites incorporate the actual sequence of dances of the French branle as described by Arbeau in his Orchesographie (1588): a multi-sectional brando followed by its gavotta and one or two correnti. Not only did Bononcini work in Modena, where French influence at the Este court was extremely strong, but he had also studied with Uccellini, who had left Modena for Parma by 1665. Such brando-suites are usually set apart from other dances in a volume, are often for four rather than two or three instruments, and may dispense with the chordal continuo (unfigured and marked 'violone' in 1667d; marked 'tacet' in 1669g).
While it is dances intended for the chamber that are of primary interest as models for Corelli, it is not always possible to separate those for listening from those meant to accompany actual dancing. However, some do carry explicit designations. For example, Bononcini's Op. 2 is entitled Sonate da camera e da ballo a 18.104.22.168. (1667d). The dances are not separated according to function in the table of contents, but they are clearly labelled within each partbook by means of headings at the start of major sections (shown below as they appear in the first-violin partbook):(16)
1-8 Allemana, Sarabande, Corente, e Gighe a Violino solo Da Camera 9-14 Allemana, Corente, e Balletti, a due violini Da Camera 15-26 Corente a 3. due violini, e Violone Da Camera 27-29 Brando, e Corente' in stil francese a 3 Da Ballo 30-36 Brando, e Corente in stil francese a quatro Da Ballo
Other contemporary volumes that specify 'da ballo' or 'da camera' for at least some dances include Vitati's Op. 3 (1667f) and Colombi's Op. 1 (1668a). The former includes a series of ballettos, each paired with a corrente alla francese per ballare; after several dances whose function is unspecified, there are four ballettos labelled 'per Camera', followed by correnti, a sarabanda and two sinfonias that carry no special designations. In Colombi's Op. 1, a sinfonia and several correnti are labelled 'da Camera'; presumably the other dances (brando alla francese, corrente alla francese) are for dancing.
With the exception of Buonamente's Book 7, the suites described thus far represent only a small portion of the dance content of the volumes in which they are found; most of the dances here are arranged not in suites but in groups, together with others of the same type. In Buonamente's fifth book (1629a), for instance, the two suites are followed by eight sinfonias, nine correnti and fourteen arie. But in 1667, the German immigrant Johann Rosenmuller, who had lived in Venice since 1658, devoted an entire volume to suites.(17) Here a sinfonia (numbered) is followed by 'la sua alemanda' (sic), corrente, ballo and sarabanda; in two out of eleven suites, an intrata is inserted after the corrente. Despite their publication in Venice, these suites are similar to those found in German and Austrian sources, and they seem to have had little impact on native Italian composers.
The examples noted above make it clear that suites of dances - whether da camera or da ballo - can indeed be found in Italian prints. However, there are far stronger Italian precedents for Corelli's sonate da camera among groups of dances not overtly identified as a unit and associated more by common tonality and by their position in the volume than by explicit labels. Wording on title-pages or in tables of contents that might identify groups of dances is, admittedly, rare; however, when adjacent dances of unlike type share the same tonic, the possibility that they formed a set in the eyes of the seventeenth-century musician is strong, as is the likelihood that the order of the dances in the volume may appear rather haphazard to the modern scholar perhaps relying primarily on lists of contents in Sartori's catalogue. And even outside Italy, where the suite was well established, groups of dances were not always given a collective label.
Between 1607 (the date of Salamone Rossi's first book) and 1685 (when Corelli's Op. 2 appeared), the arranging of the dances in a volume mainly by type, so that all the ballettos precede all the correnti, increasingly gave way to other methods of organization. Even in the earliest of these collections, one finds some dances arranged in pairs or short groups, in addition to the few in explicitly labelled suites.(18) But especially among composers active in the 1660s and '70s, grouping unlike dances by key was becoming a standard practice, reflected most clearly in the volumes printed by Giacomo Monti in Bologna. Indeed, the published volumes in which groups of dances began to predominate appeared precisely during the years Corelli spent in Bologna; he would surely have known both their composers and the music, and he may even have heard discussions of the merits of one or another approach to the organization of dances in printed collections. Once in Rome, he would have been predisposed to arrange his dances in similar groups, regardless of his personal contact with music and musicians there. One can hardly deny the possibility of a strong Roman influence on Corelli's compositional choices, but the printed corpus from Bologna should not be disregarded.
Although there is little Italian precedent for the use of the term 'sonata da camera' as a label for groups of dances such as those in Corelli's Op. 2, one might argue convincingly that only the name was changed in 1685.(19) In what follows, I offer such an argument by examining the contents of relevant sevententh-century dance collections, especially those printed between 1666 and 1685.
INSTRUMENTAL MUSIC IN PRINTED SOURCES
In seventeenth-century Italy, volumes of instrumental music were typically arranged according to considerations of scoring, function or genre. Often one finds all the solo works at the outset, followed by duos, then trios and works for larger ensembles. Thus Marini presented works for two, three and four players in separate sections of his Op. 8 (1626m). If not scoring, then function may be the primary determinant, so that one finds sonatas of different types (e.g. multi-movement 'church' versus one-movement 'chamber' works) and dances for chamber and/or the ballroom in separate sections. Legrenzi's Op. 4 (Venice, 1656) contains, according to the title-page, Sonate da chiesa, da camera, correnti, balletti, alemane, e sarabande a tre, doi violini e violone . . . Indeed, the contents seem to appear in precisely that order: six multi-movement sonatas (probably da chiesa), six one-movement sonatas (apparently da camera), six correnti, six ballettos and then three sarabanda-allemanda pairs. In Colombi's first book (1668a), a group of eight sinfonie da camera is followed by two brando-suites a4, each followed by a single corrente or paired corrente da camera and aria; at the end of the volume come two abstract sonatas a2, 3 or 4 se piace which fall into the usual multi-sectional format. And in Bononcini's first published collection (1666c), abstract sonatas are followed by a section devoted to 'Brando Arie Corrente . . . Per Camera'.
Many volumes, however, included only dances, all scored for one size of ensemble; in that case, dance-type often served as the ordering principle. Vitali's first published collection (1666a) presents ballettos and correnti - twelve of the former followed by twelve of the latter - not paired in any obvious way:(20)
Balletto 1 (d), Balletto 2 (c), Balletto 3 (d), Balletto 4 (C), Balletto 5 (a), Balletto 6 (B flat), Balletto 7 (A), Balletto 8 (F), Balletto 9 (D), Balletto 10 (c), Balletto 11 (D), Balletto 12 (E flat)
Corrente 1 (d), Corrente 2 (c), Corrente 3 (D), Corrente 4 (g), Corrente 5 (C), Corrente 6 (B flat), Corrente 7 (e), Corrente 8 (c), Corrente 9 (a), Corrente 10 (D), Corrente 11 (E flat), Corrente 12 (A)
Of the composers most active in the realm of instrumental music, Uccellini stands out for his devotion to dance- or movement-type as a primary method of organization. In five collections first published between 1639 and 1660, he kept strictly to this approach, habitually including within a single volume one section devoted to sonatas, and others devoted to sinfonias, arie or correnti, the last often separated into French and Italian varieties. Only in his Op. 9 (1667g) did he include two mixed groups of dances clearly identified as suites by the label 'brandi alla francese per ballare' (each brando consists of sections marked 'Allegro e presto', 'Gaii', 'Amener' and 'Gavotta'):
1 Sinfonia 1-2 a violino solo 2 Sinfonia 3 a violino, & violone 3 Sinfonia 4-7 a due violini 4 Sinfonia 8-9 a tre violini 5 Sinfonia 10-12 a due violini, & violone 6 Sinfonia 13-25 a quatro 7 Primo Brando alla francese per ballare 8 Secondo Brando alla francese per ballare 9 Corrente 1-6 alla francese per ballare 10 Introdutione de Balli al italiana 11 Ballo 1-15 al italiana 12 Canon Comunis est via, a due violini 13 Secondo Canon a due violini
In their earliest collections, Bononcini and Vitali followed Uccellini in arranging most or all dances by type, but both were shortly to abandon that approach in favour of pairs and longer groups of unlike dances in the same key. This change is reflected in many volumes published between 1667 and 1692 (see Table II).(21)
PAIRS AND GROUPS OF DANCES
As Table II demonstrates, several mid-century composers included a few paired dances in collections otherwise organized by type: thus Legrenzi's sonatas (multi- and single-movement), correnti and ballettos are followed by three sarabanda-allemanda pairs (Op. 4, 1656d). By 1670, pairs were becoming a more prominent feature of dance collections. Pizzoni's Op. 1 (1669c) presents twelve ballettos, each followed by one other dance (corrente, giga or 'sarrabanda' (sic)); the paired dances have the same tonality. Pietro Degli Antoni made the pairings explicit in his Op. 3 (1671c), where each dance is followed by 'its' partner, from the initial 'Balletto primo, sua corrente' to the final 'Giga, sua Sarabanda'.(22) Several additional volumes by Bononcini, Guerrieri and Vitali arrange dances in pairs or groups; other volumes, while they include more exceptions to this rule, adopt pairs and suites as their primary means of organization. Thus Cazzati's Op. 22 and Op. 30 (1660a; 1662) were almost entirely presented in balletto-corrente pairs, with a few suites or independent pieces to open or close each volume.
Bononcini and Vitali are especially prominent among the composers whose dances are arranged in pairs and longer groups. In five publications that appeared between 1671 and 1678 (Op. 4, 5, 7, 9 & 12), Bononcini abandoned grouping dances by type in favour of pairs or groups of three to six pieces. Op. 5 (1671f) offers a clear illustration of this approach. After an introductory sinfonia, we find groups of two to five dances, in the order allemanda-corrente-sarabanda:
1 Sinfonia per introduzione, Allemanda, Corrente, Sarabanda (d) 2 Allemanda, Corrente, Sarabanda (B flat) 3 Allemanda, Corrente, Sarabanda (g) 4 Allemanda, Corrente, Sarabanda, Corrente, Sarabanda (E flat) 5 Corrente, Sarabanda (f) 6 Allemanda, Corrente, Sarabanda (c) 7 Allemanda, Corrente, Sarabanda (B flat) 8 Allemanda, Corrente, Sarabanda (F)
In one case, the allemanda is omitted, and in another an 'extra' corrente and sarabanda are appended; but the method of grouping different dances in suites by key is clear.
Even earlier, by the late 1660s, Vitali and Bononcini had begun to arrange dances less by type than in pairs and groups. Bononcini included several paired dances as well as brando-suites in his Opp. 2 and 3 (1667d; 1669g) while continuing to group some dances by type in those volumes. Colombi (1668a) and Vitali (1667f; 1668e) adopted similar strategies. These transitional volumes reflect in their somewhat haphazard ordering of dances the trend away from grouping by type towards grouping in pairs or suites. Vitali's Op. 4 may serve to illustrate this approach:
1 Balletto 1 (b) 2 Balletto 2 (B flat) 3 Balletto 3 (E flat) 4 Balletto 4, Giga (D) 5 Allemanda 1, Giga (d) 6 Allemanda 2, Zoppa (g) 7 Corrente 1, Giga (b) 8 Corrente 2, Giga, Sarabanda (D) 9 Corrente 3, Giga, Sarabanda (e) 10 Corrente 4, Sarabanda (B flat) 11 Corrente 5 (f) 12 Balletto, Giga (D) 13 Giga (d) 14 Giga (a)
In this collection, two dances of the same type are rarely placed in adjacent positions, nor are the adjacent pairs confined to two types of dance; rather, adjacent dances of various types share a tonality, forming somewhat varied groups of two or three different dances. The 'groups' are not labelled as such in the print, and the existence of several 'extra' dances - the ballettos at the outset and the gighe at the end - is confusing if one is looking for clear-cut suites. Nonetheless, these collections seem to reflect in their apparent disorder a transitional period during which composers and publishers were moving from one method of organization to the other and were not fully committed to either.
By the early 1680s, Vitali (in Opp. 8, 11 & 12), like Bononcini, had begun to arrange entire volumes in pairs or suites. Op. 8 (1683e) is devoted mainly to balletto-corrente pairs, preceded by a suite of three movements (Balletto, Corrente, Giga) and followed by two Capriccios. Op. 11 (1684b) contains nine three-movement dance groups and one pair, followed by a closing three-movement sinfonia:
1 Capriccio, Giga, Gavotta (B flat) 2 Balletto in stile francese, Corrente alla francese, Borea (g) 3 Capriccio, Giga, Gavotta (e) 4 Balletto per camera, Zoppa, Gavotta (c) 5 Capriccio, Balletto, Giga (E flat) 6 Introdutione, Balletto, Borea (C) 7 Balletto, Giga, Gavotta (A) 8 Capriccio, Giga (D) 9 Introdutione, Giga, Sarabanda (F) 10 Introdutione, Balletto, Giga (a) 11 Sinfonia (D)
Op. 12 (1685j), which appeared in the same year as Corelli's Op. 2,(23) consists of nine suites, the last two of which are brando-suites. The other seven consist of three to six movements, always beginning with a balletto:(24)
1 Balletto, Giga, Minuet, Borea (D)
2 Balletto, Giga, Gavotta, Minuet (G)
3 Balletto, Giga, Minuet, Borea (B flat)
4 Balletto, Giga, Minuet, Borea (g)
5 Balletto, Giga, Borea, Sarabanda, Gavotta, Minuet (F)
6 Balletto, Giga, Borea, Minuet (C)
7 Balletto, Borea, Minuet (G)
8 Brando primo-Gaij-Amener-Gavotta,(*) Corrente, Gagliarda figurata, Gavotta, Minuet (B flat)
9 Brando secondo-Gaij-Amener-Gavotta (F)(*)
* The individual sections of the brando are not listed separately in the table of contents
Vitali's posthumously published Op. 14 is similarly organized.
Other composers, too, began to group dances in pairs or suites rather than by type (see Table II, above); volumes by Pandolfi (1669i) and Colombi (1674c) are especially tightly organized.(25) However, a few volumes of the time present a somewhat more confusing picture. The collections of Prattichista (1666b), Cazzati (1669a) and Orazio Polaroli (1673d) exhibit little internal evidence of larger groupings. Yet given the evidence of grouping by key in the majority of volumes published after 1666, it is difficult to believe that conventional groupings were never adopted in performances drawing on such seemingly random collections. Performers may have created groups even from sets of dances in different keys. One such suite is explicitly identified in Marini's Op. 22 (1655a): the group, labelled 'Balletto 2', consists of an Entrata (E), Balletto (G), Gagliarda (a), Corrente (a) and Retirata (D).(26)
CORELLI'S OP. 2
Our picture of the organization of dance volumes is now well enough developed for a closer look at Corelli's first set of chamber sonatas to be needed. Here, the sonatas, eight of which begin with a slow prelude, are made up of two or, more often, three dance movements; at the end of the volume stands a ciaccona (Largo-Allegro), and in addition there are three sections identified only by tempo designations. The specific content of the groups varies; indeed the only constant is the presence of at least one allemanda in each sonata (similar to the balletto in Vitali's Op. 12), but its tempo may be fast or slow (and the choice does not depend on the presence or absence of a prelude). Corelli's Op. 4 (1694a) is somewhat more standardized. All groups begin with preludes (the closes of three of which are harmonically open) and are made up of one, two or three dance movements. Six of the twelve groups include transitional slow movements, much shorter than those in Op. 2. The allemanda is no longer a necessary constituent, but all groups include either an allemanda or a corrente, and two include both. All the allemandas are fast. In both collections, neither the number of dance movements nor their character is standardized; moreover, slow and fast dances do not necessarily alternate.
For all their status as classical models, Corelli's sonatas seem to vary a great deal. Without the title 'Sonata', we would be able to identify groups only by key and location, much as is the case with Vitali and Bononcini. Vitali's arrangement of dances in his Op. 4 is not quite as tidy as Corelli's: one cannot easily identify the suites from the table of contents, since the evidence for them lies within the volume in terms of shared tonality and printed order. But in his Opp. 11 and 12, Vitali arranged his dances in perfectly coherent suites which lack only the label 'sonata da camera' to make them comparable with those of Corelli. It is these volumes and their forerunners that demonstrate the existence of an Italian printed context for Corelli's Op. 2.
Nevertheless, this context cannot have been the only one within which Corelli operated. Manuscript sources demonstrate that Roman composers, too, were likely to group different dances by key. In particular, the manuscript sources for works by Lelio Colista demonstrate his commitment to an idea of the chamber sonata similar to that held by Corelli.(27) As for the putative role of Georg Muffat, one can only note that a Roman influence on the chamber sonatas in his Armonico tributo (Salzburg, 1682) cannot be ruled out, since he visited the city in the early 1680s: the volume's dedication was signed in Rome on 4 September 1682. Thus, far from Muffat having introduced Corelli to the idea of the sonata da camera, he may himself have adopted the habits of Corelli and his Roman colleagues.
Even in the 1680s, the model developed most clearly by Bononcini and Vitali, and evidently influential on Corelli, does not represent the only method of organization. As late as 1682, Placuzzi's collection included sixteen dances that seem not to fit into any of the pairs and groups otherwise apparent in the volume. And while Colombi (or his publisher) did adopt the label 'sonata' for eleven groups (each an Adagio followed by three dances) in his Op. 5 (1689b), one giga is left at the end of the volume apparently unattached.(28)
One of the confusing aspects of the pre-Corellian sources involves the conventions of numbering the dances, conventions decidedly at odds with our idea of coherent suites. In Vitali's Op. 4 and in several contemporary volumes, the numbering schemes evince less concern with consecutive order within each group, or with the order of groups as a whole, than with how many ballettos or correnti there are in total. Thus dance groups could apparently be composed of, say, Allemanda 1, Corrente 3 and Giga 2 without disturbing the seventeenth-century player. And if performers can be assumed to have used dances arranged by type as the raw material for the suites they invented at will, those accustomed to finding their dances grouped by type may even have rearranged such collections as Vitali's Op. 4 into more personally satisfying suites by treating individual elements as the single and separable entities suggested by the numbering scheme. Indeed, dances from Vitali's Op. 4, as well as six from his Op. 1, turn up without attribution in an English manuscript (Durham Cathedral Library, MS M 179-80) and in a Walsh print (c. 1703; here they are attributed to Marc' Antonio Ziani).(29) In these two sources, the dances are rearranged more firmly into suites, in that each of the opening ballettos from Op. 4 introduces a series of dances in the same key, and three ballettos are paired with three correnti from Op. 1.(30) The manuscript copy serves at the very least as evidence that an English copyist preferred to arrange suites definitively in the score rather than in performance.
The conclusion that the development of the sonata da camera can be observed in Italian printed sources is inescapable. While the term itself is apparently not used in Corelli's sense before 1685, it is abundantly clear that Vitali and his contemporaries left behind many well-organized volumes with dances grouped into suites not very different from those of Corelli except in the matter of the name. In individual cases, the interpretation of the term 'sonata', or of the modifiers 'da camera', 'da chiesa' and 'da ballo', may remain problematic. We simply cannot be sure that before 1685 'sonata' invariably referred to a single piece rather than to a group of separate movements (say allemanda-corrente-sarabanda). One can readily identify groups (as in Rosenmuller 1667h and Buonamente 1637d) without depending on this title, even if the groups vary in length and composition (as do Corelli's). That is, a collective title is not a necessary condition, although it may be a sufficient one, for the identification of dance groups. Moreover, performers and composers may have thought in terms of groups of dances and may even have called such groups 'sonatas' long before that usage was clearly reflected in the printed sources. Certainly, the convention of grouping dances by key was becoming more and more securely entrenched in the 1680s, and the use of the term 'sonata' to designate a particular group of dances simply provides evidence for the final stage of that entrenchment.(31)
1 For a contrasting view, see Mildred Pearl, The Suite in Relation to Baroque Style (unpublished dissertation), New York University, 1957, p. 91. Pearl suggests that the terms 'da camera' and 'da chiesa' were in fact common, citing Merula's 'per chiesa, e camera' (1637a). See also the extended discussion of terminology and function in Peter All-sop, The Italian Trio Sonata, Oxford, 1992, pp. 47-66.
2 See John Daverio, 'In Search of the sonata da camera, before Corelli', Acta musicologica, lvii (1985), 195-214.
3 Ibid., p. 200 n. 25.
4 For a discussion of differences in the scoring of the bass line in church and chamber volumes, see Sandra Mangsen, 'The Trio Sonata in Pre-Corellian Prints: When Does 3 = 4?', Performance Practice Review, iii (1990), 138-64.
5 See Eleanor Selfridge-Field, 'Canzona and Sonata: Some Differences in Social Identity', International Review of the Aesthetics and Sociology of Music, ix (1978), 111-19.
6 See John Daverio, Formal Design and Terminology in the Pre-Corellian 'Sonata' and Related Instrumental Forms in the Printed Sources (unpublished dissertation), Boston University, 1983, where this view of the development of the sonata is subjected to close scrutiny. Daverio's outline aptly describes the instrumental duos and trios also discussed in Sandro Mangsen, Instrumental Duos and Trios in Printed Italian Sources, 1600-1675 (unpublished dissertation), Cornell University, 1989.
7 For a discussion of the relative importance of specific Bolognese and Roman elements in Corelli's works, see Allsop. The Italian Trio Sonata, pp. 227-39.
8 Sonate da camera a tre, doi violini, e violone, o cimbalo, Op. 2, Rome, Bologna & Modena, 1685. Introductory movements are those placed at the beginning of a work and not based on a particular dance type; terms such as 'sonata' 'sinfonia', 'introduzione' and 'preludio' were commonly applied to such movements.
9 Daverio, 'In Search of the sonata da camera', pp. 203-8.
10 Ibid., p. 197.
11 The work is transcribed in Paul Nettl, 'Eine Sing- und Spielsuite von Antonio Brunelli', Zeitschrift fur Musikwissenschaft, ii (1919-20), 385-92.
12 'Each sinfonia has its Brando, Gagliarda and Corrente'. The two volumes are entitled Il quinto libro de varie sonate, sinfonie, gagliarde, corrente, & ariette per sonar . . . (Venice, 1629) and Il settimo libro di sonate, sinfonie, gagliarde, corrente, et brandi . . . (Venice, 1637).
13 The original score is reproduced in Thomas D. Dunn, The Instrumental Music of Biagio Marini (unpublished dissertation), Yale University, 1969, pp. 47-51.
14 Both of Marini's Op. 22 suites are transcribed ibid., pp. 23-43. A Balletto and an Allemanda from 1649a are included in Richard Hudson, The Allemande, the Balletto and the Tanz, Cambridge, 1986, ii, Ex. 107.
15 His balletto-corrente-sarabanda is a format nearly identical to the allemande-courante-sarabande order followed by French lutenists early in the seventeenth century; see David Fuller, 'Suite', The New Grove, xviii. 339-40.
16 The 'Aria Discordia Concots, a 4' and its corrente at the end of the volume are not classified as chamber or ballroom pieces.
17 Sonate da camera cioe sinfonie, alemande, correnti, balletti, sarabande, da suonare con cinque stromenti da arco, et altri, Venice, 1667.
18 In Italy, 'classical' suites (allemanda-corrente-sarabanda) were somewhat more common in volumes for guitar than in those for ensemble or for keyboard. David Fuller ('Suite', p. 340) cites publications by A. M. Bartolotti (1640), Francesco Corbetta (1643) and G. B. Granata (1651) in this regard; Granata was still in Bologna during Corelli's years there.
19 The meaning of 'sonate' on earlier title-pages is rather slippery. Pearl (The Suite in Relation to Baroque Style, p. 91) cites Rosenmuller (she gives 1670, but that is a reprint of 1667h) as the first time 'the specific meaning of a suite of dances is first imposed on the term sonata da camera'. Daverio ('In Search of the sonata da camera', p. 197 n. 11) understands Rosenmuller's 'sonata' as a single piece rather than a group of pieces. Unfortunately, the usage seems destined to remain somewhat unclear.
20 However, there is some thematic similarity to be noted between selected pairs (Balletto 1, Corrente 1; Balletto 2, Corrente 2; Balletto 4, Corrente 5; Balletto 6. Corrente 6; Balletto 12, Corrente 11).
21 While Table II does not constitute a comprehensive list of dance volumes, it is relatively complete for duo and trio publications before 1675.
22 Three balletto-corrente pairs, two balletto-giga pairs and one each of allemanda-corrente, aria-corrente, capriccio-giga and giga-sarabanda.
23 The dedication to Corelli's Op. 2 is dated 9 July 1685; Vitali's dedication is undated.
24 These printed quintets are reduced to trios in the manuscript copy of this volume in Modena, Biblioteca Estense, Mus F1246. For further discussion of differences between manuscript and printed versions of duos and trios, see Sandra Mangsen, 'The Dissemination of Pre-Corellian Duo and Trio Sonatas in Manuscript and Printed Sources: a Preliminary Report', The Dissemination of Music: Studies in the History of Music Publishing, ed. Hans Lenneberg, Philadelphia, forthcoming.
25 Colombi had already included several corrente da camera-aria pairs in his Op. 1 (1668a).
26 Pearl (The Suite in Relation to Baroque Style, p. 98) notes that this group 'violates the primary rule of key unity' and 'must be considered an interesting and isolated example of a cul-de-sac of tonal adventure in the early history of the suite'. Daverio ('In Search of the sonata in camera', p. 202) suggests that it derives from a series of pieces 'originally employed in a courtly entertainment'.
27 I am indebted to Peter Allsop and Eleanor McCrickard for this information. Helene Wessely-Kropik lists three sonate da camera by Colista, consisting of two to six movements, in Lelio Colista, ein Romischer Meister vor Corelli: Leben und Umwelt, Vienna, 1961, p. 117. For a general discussion of Roman practice, see Allsop, The Italian Trio Sonata, pp. 192-210.
28 Michael Talbot's very suitable appellation for such isolated and unattached dances is 'stubborn singletons'. See his discussion in 'The Taiheg, the Pira and Other Curiosities of Benedetto Vinaccesi's "Suonate da camera a tre", Op. 1', Music & Letters, lxxv (1994), 344-64.
29 Ziani's Aires or Sonatas in 3 Parts for Two Violins and a Thorow-Bass . . . Opera Prima.
30 Balletto 1, Corrente 1; Balletto 4, Corrente 5; Balletto 9, Corrente 3.
31 See Michael Talbot's apt proposal of a series of stages in the development of the sonata da camera, in 'The Taiheg, the Pira and Other Curiosities of Benedetto Vinaccesi's "Suonate da camera a tre", Op. 1'.
Organization of Dances in Selected Italian Publications, 1607-94
(1) Arranged by type
1613k, Rossi Bk 3; 1626d, Buonamente Bk 4; 1639b, Uccellini Bk 2; 1642a, Uccellini; 1644b, Neri Op. 1; 1645f, Uccellini Op. 4; 1650b, Todeschini Op. 1; 1654a, Cazzati Op. 15; 1660d, Uccellini Op. 7; 1666a, Vitali Op. 1
(2) Arranged primarily by type (with some pairs, suites or singles)
1607c, Rossi Bk 1; 1608h, Rossi Bk 2; 1622b, Rossi Bk 4; 1626m, Marini Op. 8; 1629a, Buonamente Bk 5; 1638c, Selma Bk 1; 1651d, Cazzati; 1655a, Marini Op. 22; 1656d, Legrenzi Op. 4; 1666c, G. M. Bononcini Op. 1; 1667g, Uccellini Op. 9
(3) Arranged by type, in pairs and suites
1667d, G. M. Bononcini Op. 2; 1667f, Vitali Op. 3; 1668a, Colombi Op. 1; 1668e, Vitali Op. 4; 1669g, G. M. Bononcini Op. 3
(4) Arranged in pairs or suites
1637d, Buonamente Bk 7; 1667h, Rosenmuller; 1669c, Pizzoni Op. 1; 1671c, P. Degli Antoni Op. 3; 1671e, G. M. Bononcini Op. 4; 1671f, G. M. Bononcini Op. 5; 1673e, Guerrieri Op. 1; 1675b, G. M. Bononcini Op. 9; 1677a, G. B. Degli Antoni Op. 3; 1677c, Bassani Op. 1; 1677+, G. B. Degli Antoni Op. 4; 1678a, G. M. Bononcini Op. 12; 1678g, A. Grossi Op. 1; 1679a, A. Grossi Op. 2; 1681b, Piazzi Op. 2; 1684a, D. Gabrielli Op. 1; 1685d, G. Bononcini Op. 1; 1685j, Vitali Op. 12; 1686d, Torelli Op. 2; 1687i, Marino Op. 1; 16871, Vinaccesi Op. 1; 1690d, G. B. Degli Antoni Op. 6; 1691h, Legrenzi Op. 16; 1692d, Vitali Op. 14; 1694a, Corelli Op. 4
(5) Arranged primarily in pairs or suites (with some by type or singly)
1655b, Gandini Op. 4; 1660a, Cazzati Op. 22; 1662, Cazzati Op. 30; 1665d, Prioli; 1669i, Pandolfi; 1670b, P. Degli Antoni Op. 1; 1673i, G. M. Bononcini Op. 7; 1674c, Colombi Op. 3; 1682i, Placuzzi Op. 2; 1682e, Albergati Op. 1; 1683e, Vitali Op. 8; 1685a, Corelli Op. 2; 1684b, Vitali Op. 11; 1687+, Torelli Op. 4; 1689g, Colombi Op. 5
(6) Apparently haphazard arrangement
1617c, Marini Op. 1; 1650a, Falconieri Bk 1; 1666b, Prattichista; 1669a, Cazzati Op. 50; 1673d, Polaroli Op. 1
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Italian instrumental ensemble music; Archangelo Corelli|
|Publication:||Music & Letters|
|Date:||Feb 1, 1995|
|Previous Article:||Music and the Quadrivium in early Tudor England.|
|Next Article:||The Paris Opera chorus during the time of Rameau.|