• 2517
• Articles
• 24 Gen 2017

CHAMBER DUETS

The first ‘modern’ musicologist to examine Steffani’s duets was Friedrich Chrysander, who explored them in connexion with his biography of Han-del. He divided those he knew into three groups: ‘small’ duets in one movement, some with da capo repeat; ‘large’ duets, in which solo movements for each singer (some with recitative) were framed by others for both; ‘medium’ duets in more than one movement but without solos. Since he found the solo movements less melodious and artistic (‘kunstreich’) than the ‘lively’ and ‘fiery’ duets, he regarded the ‘medium’ duets as the finest works: here was the mastery of melody and counterpoint that Handel was to take as a model. If Chrysander’s preference was reflected in the first scholarly edition of Steffani’s duets, their contrapuntal nature was adopted by Eugen Schmitz as ‘the hallmark of the form [the Italian chamber duet] in its highest perfection’: Through Steffani’s work, the genre became associated with the notion of a display of superior contrapuntal refinements. Admittedly, it must be emphasized at once that this polyphonic nature does not constitute the norm throughout the history of the chamber duet but rather that it stands out exclusively as a determinant of style only in the case of iso-lated, specially gifted composers of whom Steffani himself is the most important. For this reason, however, it is nevertheless the hallmark of the form in its highest perfection and as such is of utmost significance in the history of the development of the genre. The presence of counterpoint is one of four criteria used by Schmitz to distinguish between various kinds of duet. The others concern the nature of the text (whether it is a dialogue), vocal scoring (whether the work includes solo movements), and instrumental scoring (whether it calls for instruments other than continuo). The question of text is fundamental. When the poem is a dialogue, the singers represent different characters, whether named or not, and engage in conversation: the text is a miniature drama. Even if the characters share some of the words or express similar sentiments, their music inevitably differs. This is not the case when the text is a lyrical monologue. A poem of this kind could have been set for one voice, and occasionally was: some of Steffani’s duets are based on words that had been set as solos by other composers. A deci-sion to score a setting of such verse for two singers has no basis in the poem. The musical result is simply a textural expansion of a solo, and this means that the duet is a medium rather than a form. Most duets before 1650 were madrigals or arias for two voices and continuo only. The madrigal survived for much longer in duet than in solo settings, and the aria a due flourished as both a serious genre and a light-weight canzonet. The first publication to include the word ‘duetti’ in its title was Barbara Strozzi’s Cantate, ariette e duetti, op. 2 (Venice, 1651); printed duets appeared later as ‘duetti’or ‘cantate’. Dialogue settings were comparatively rare, but their presence in Maurizio Cazzati’s Duetti per camera, op. 66 (Bologna, 1677), suggests that even such ‘theatrical’ works could be regarded as ‘chamber’ music. Schmitz’s belief that counterpoint was typical of the chamber style is confirmed by printed ‘duetti’, most of which use imitation and are described as ‘da’ or ‘per camera’. In his search for the sources of Steffani’s style, however, he was continually disappointed, even by Roman composers (though since most of his sources were prints, he could not do justice to Luigi Rossi, Carissimi, Cesti, or Stradella). The most pertinent Venetian was Giovanni Legrenzi, but even his Idee armoniche estese, op. 13 (1678), preferred short antiphonal motives to more sustained contrapuntal material. Bolognese duets, too, were largely ‘theatrical’. If Cazzati began with imitation, the parts were soon combined in parallel motion. The same ap-plied to Giovanni Maria Bononcini and to the Duetti da camera, op. 8 (Bologna, 1691), of his son. A marked interest in the contrapuntal duet was displayed by a minority of composers, of whom the most prominent were Carlo Donato Cossoni, Francesco Petrobelli, and Giovanni Carlo Maria Clari. Insofar as his works eschewed solo movements and obbligato instruments, Clari surpassed even Steffani in his love of the vocal duet. The second half of the seventeenth century saw a number of developments in counterpoint that are relevant to Steffani’s duets. In his Musico pratico (1673) Giovanni Maria Bononcini became the first music theorist to accord the tonal answer equality of status with the real. Johann Adam Reincken proposed a new definition of double fugue (‘contrafuga’): the themes should be contrasted and ‘not continually treated together, but also separately, one after the other, then at times together and against each other, which shows much greater mastery’. He also introduced the concept of stretto into writings on fugue, but it was Bononcini who suggested that close entries be avoided at the beginning and reserved for later on, when listeners would be more familiar with the material and better able to appreciate textural complexity and compositional skill. This advice relates to the ‘a risposta’ technique, ‘whereby the two concertante violin parts [in a trio sonata] begin, not in imitation at the fifth, but by alternation at the unison, with each part presenting the material solo over the bass before turning to more consequential imitation and fugato (which latter then has the effect of stretto)’. Examples of this are found in Bononcini’s early sonatas but appear to be outweighed by such procedures as the counter exposition, with parts interchanged, sequential episodes based on motivic fragmentation, canon, and passages based on pedals or the circle of fifths. The trio sonata was the principal arena in which these contrapuntal developments were applied. The earliest book consisting entirely of such pieces was Legrenzi’s Sonate… libro primo (Venice, 1655). Trio sonatas outnumber every other kind of sonata in the output of Cazzati, Legrenzi, G. B. and T. A. Vitali, Corelli and Torelli, but are balanced by solo sonatas in Albinoni and outweighed by them in Vivaldi. The trio thus marks a midway point between the ensemble and the solo sonata, and the peak of its development ‘coincided with the advent of Corelli and the Stradivari violin’. The upper parts of a trio sonata are nearly always treated as equals: they may employ parallel motion and elaborate passages, but the texture is dominated by imitation between two (or all three) of the parts, contrapuntal procedures, and ‘chains of suspensions and resolutions created by the intertwining upper parts against the bass’. The prevalence of counterpoint led Newman to regard the trio sonata as ‘an ideal meeting point between the older vocal styles and the newer instrumental styles and between the older, stricter polyphony and the new emphasis on accompanied melody’. A similar view can be taken of the vocal chamber duet as a compromise between polyphony and monody and a halfway house between madrigal and cantata. From the point of view of texture it was equivalent to the trio sonata. In Der vollkommene Capellmeister (1739) Mattheson distinguished between French duets, dependant on parallel motion, and Italian duets characterised by counterpoint; he also suggested that the former were particularly effective in church, that contrapuntal textures were especially appropriate to chamber music, and that Steffani’s duets were models of their kind: ‘The duet, or aria for two voices, is arranged in either the Italian or the French manner. We will give a little idea of each type. The French airs à deux tend particularly to use the same or similar counterpoint, that is, where one voice sings the words at the same time as the other, so that either nothing at all or only here and there something dissimilar or concertato like sneaks in. Such duos sound fine, especially in churches: they are principally devotional and comprehensible. The Italian type of duet lacks much of the aforementioned good qualities of piety and clarity, because of its fugal, artificial, and intertwined nature; however, they require a true man, and are a great pleasure to the musically-trained ear, in the chamber as also in the church (and earlier, in Steffani’s time, even in the theatre), if skilful, well-trained singers can be found for them: of these last we now have fewer than of such works. The said Steffani was incomparably outstanding in this type, before all others whom I know, and to this hour deserves to be taken as a model. For such things do not easily become obsolete.’ The emergence of the contrapuntal Italian duet in the late seventeenth century was noticed also by Burney a century later: ‘Near the latter end of the last [seventeenth] century, a species of learned and elaborate Chamber Duets for voices began to be in favour. The first that I have found, of this kind, were composed by John [Giovanni Bononcini], and published at Bologna in 1691. Soon after, those of the admirable Abate Steffani were dispersed in manuscript throughout Europe. These were followed by the duets of Clari, Handel, Marcello, Gasparini, Lotti, Hasse, and Durante.’ As we have seen, Bononcini’s Duetti da camera were neither the first nor the most representative ‘of this kind’. They must have seemed ‘learned and elaborate’ in comparison with the music of Burney’s day, but they are less consistently contrapuntal than those of Steffani and so are the duets of such earlier composers as Stradella and Cesti and most of his rivals in the field. That Steffani’s duets were dispersed after 1691 indeed, after 1702 is attested by the predominance of his revised versions in the sources, but the fact that Oh! che voi direste bene was composed atMunich proves that he was writing chamber duets before 1688. He appeared on the scene at the right moment: if the trio sonata was crystallized by Corelli, the vocal duet was perfected by Steffani, his junior by only a year.
CANON AND CHRONOLOGY
Steffani composed at least eighty-one of the hundred-plus chamber duets ascribed to him in nearly 250 manuscripts around the world. Six of them should have instruments as well as continuo, but the instrumental staves are blank in the source (GBLbl, RM 23. k. 20). Nineteen duets survive in two forms: a revised version dating from the autumn or winter of 1702–3, and an earlier version; most of the latter (as well as the revisions) are authentic, but the earlier versions of Lungi da l’idol mio and M’hai da piangere are unreliable attributions. With three exceptions, his duets must all have been finished by the end of 1702, when he gathered them together in preparation for a new, complete, manuscript collection in thirteen volumes, with six duets to a book. The exceptions are Dolce è per voi soffrire, of which the countess of Egmont asked him for a copy in June 1711; Dolce labbro, amabil bocca, which appears to have been written at Herten in the summer of 1712; and Quando ti stringo, which dates from the same season or possibly the summer of 1713. Despite the dates mentioned here, which relate to his Hanover and Düsseldorf periods, Steffani also wrote chamber duets in Munich. The only documentary evidence of this is Violanta Beatrice’s letter of 25 August 1693, but circumstantial evidence is provided by his settings of words used previously by other composers. The concordances indicate that these texts were in circulation during Steffani’s periods in Munich and Rome. The first piece in van Geertsom’s anthology is ascribed to ‘Marco Aurelli’ (Marcorelli), who worked in Rome in the 1640s and 1650s; of the remainder, all unascribed, two are by Carissimi and two by Luigi Rossi. Steffani’s È spento l’ardore includes a second stanza, which is missing from van Geertsom and from a related,anonymous version of the piece, for solo voice, in Lbl, Add. MS 14336. Cesti’s death in 1669 provides a terminus ante quem for his solo setting of Tu m’aspettasti, and most of his cantatas are associated with patrons in Rome. Lonati was based there from 1668 to 1677, leading Queen Christina’s orchestra from at least 1673; the source of his duet (I-Bc, MS Mart. 6.1, dated 1719) includes music by Kerll (Il mio cor è un passaggiero) and Ercole Bernabei, and most of the other composers named in it spent part of their life in Rome. Cossoni’s Cantate, op. 13, lacks a title page but must have been published between 1671 and 1679, the dates of his opp. 11 and 14 (op. 12 was reprinted in 1675). Given the date of Albergati’s Cantate, op. 6, Steffani may have encountered Dir che giovi in Italy in the spring of 1688, just after leaving Munich. The concordances suggest that, alongside solo cantatas, he composed chamber duets in the 1670s in Munich and Rome. Some of Steffani’s duets are presumably settings of verse by Munich librettists, but since Domenico Gisberti departed in 1675 and Luigi Orlandi arrived only in 1686, the strongest contender is the composer’s bro-ther, Ventura Terzago. Although Steffani set four of the latter’s librettos, however, no lyrical poetry by Terzago is known: none of the texts of his duets is ascribed to him in a musical source. The only poet whose works appear to have been used in chamber duets by Steffani during his Munich years is Brigida Bianchi, an actress and singer in the comédie italienne in Paris. Six of his duets are based on her verses. The words of Ah! che l’ho sempre detto (‘Presagio’), Ho scherzato in verità (‘Testimonianza di fede’), and Torna a dar vita al core (‘Lontananza’) were published in her L’inganno fortunato … con alcune poesie musicali composte in diversi tempi (Paris, 1659) and reprinted in her Rifiuti di Pindo. Poesie (Paris, 1666), along with E così mi compatite and E perchè non m’uccidete (both entitled ‘Occhi crudeli’) and Sia maledetto Amor (‘Si duole d’Amor’). If Steffani took these poems directly from her books, he is most likely to have done so in Paris in 1678–79. His settings were not necessarily composed there and then: the earlier versions of Ah! che l’ho sempre detto, E perchè non m’uccidete, and Torna a dar vita al core, which were later revised, seem immature, but this is not the case with the others. There is far more evidence of his involvement with chamber duets du-ring his Hanover years. An appealing picture of this activity was painted by Hawkins: ‘It was about this time, when a taste for music prevailed so greatly at Hanover, that Steffani composed his duets, which have acquired him such an universal reputation. It is probable that he might apply his studies so much to this species of composition, in compliance with the taste of the ladies of that court; for it is observable that the poetry of them is altogether of the amatory kind, and it appears by little memorandums, in several copies, that many of his duets were composed at the request of divers ladies of distinction, and that some were made for their own private practice, and amusement.’ Who the particular persons were, we are at a loss to discover, as they are distinguished only by initial letters, denoting their quality; except in the instance of the two duets, beginning Inquieto mio Cor, and Che volete, these appearing to have been made for, and sung by her highness the electress of Brandenburg. ‘For the poetry, to which he adapted his music, he was principally obliged to his friends, the Marquis d’Ariberti, Count Palmieri, Sig. Averara, Abbate Guidi, and Abbate Mauro Hortensio; but the assistance he received from these persons was not so considerable as altogether to free him from the necessity of sometimes composing to words, that, to say the least, did not call for the utmost exertion of his genius: and when every other resource failed him, he would sometimes, as an exercise of his fancy, make use of words that had formerly been set by other masters; of these, the chief are the prince of Venosa, and the famous Alessandro Stradella, so celebrated for his singing, and performance on the harp.’ Although Hawkins was wrong to imply that all of Steffani’s duets were written at Hanover and that amatory verse was necessarily associated with female patrons, his statement that these works were composed for the ‘private practice, or amusement’ of ‘the ladies of that court’ is entirely plausible. Before the arrival of Caroline of Ansbach, in 1705, the musical ladies of Hanover were the duchess (later electress) Sophie, to whom Barbara Strozzi had dedicated her Arie, op. 8 (1664); her daughter Sophie Charlotte, who had married the elector of Brandenburg in 1684; Sophie Dorothea of Celle, who had married Georg Ludwig of Hanover (later George I) in 1682; and their daughter, also Sophie Dorothea, born in 1687. The most important of these, for Steffani and his duets, was Sophie Charlotte, who, even after her move to Berlin, spent as much time as possible in Hanover. She wrote the words of his Crudo Amor, morir mi sento (1698); he sent her three duets from Brussels in 1699—Che volete, o crude pene, Inquieto mio cor, and Placidissime catene—and he also composed Io mi parto for her visit to Elector Maximilian IIEmanuel in Brussels in 1700. The words of this last duet were ascribed to ‘Abbate [Francesco Maria] Paglia’, who was active in Rome and Naples from 1688 to 1705 and supplied texts for Scarlatti and Bononcini, including three of Bononcini’s Duetti da camera (1691). When Steffani collected and revised his duets in the autumn and winter of 1702–3, he corresponded with Sophie Charlotte and had her constantly in mind. The poets mentioned by Hawkins were all associated with Steffani during his Hanover years and are credited in musical sources with the words of six duets that are often grouped together in the same order: Francesco Palmieri also wrote the words of the Accademia per musica (Hanover, 1695) for the wedding of Charlotte Felicitas and Rinaldo D’Este, and the libretto of Briseide (Hanover, 1696). The marquis was presumably the Bartolomeo D’Ariberti who served as Johann Wilhelm’s special envoy in Madrid in 1698; no other poetry by him is known. The identity of Abbate Guidi is uncertain: the most likely candidate, perhaps, is Abbate Giuseppe Guidi, Modenese resident in Hanover, but one cannot rule out Anastasio Guidi, who wrote the words of Bononcini’s one-act pastorella Cefalo (Berlin, 1702). Signor Averara was presumably the well known librettist Pietro D’Averara; that his wife Diana Aurelia sang at Hanover in 1695 suggests a date for Non so chi mi piagò that fits well with the other duets in this group. The words of Inquieto mio cor are not ascribed in any complete copy of the group but are attributed in Add. MS 5330 to ‘Sig. Abbate Conti’, presumably the ‘Abate Carlo Conti, Professor’ identified by Chrysander. As Hawkins observed, Steffani also used words that had been ‘set by other masters’. He did not draw on Gesualdo or Stradella, however—though the earlier versions of Lungi da l’idol mio and Vorrei dire are ascribed to Stradella, and the latter’s Chi dirà che nel veleno is in one source attributed to Steffani. Hawkins finally mentioned Ortensio Mauro, the anonymous librettist of Steffani’s Hanover operas. Accor-ding to a footnote, which Mainwaring must have seen, ‘this gentleman wrote also the words for twelve duets which Mr Handel composed for the practice of the late queen [Caroline (died 1737)], who greatly admired this kind of composition’; Mauro also provided texts for eight duets by Carlo Luigi Pietragrua. While the latter are ascribed to the poet in the musical source, those of Handel and Steffani are not: Mauro was relatively unknown in Düsseldorf, where Pietragrua worked, but he was a familiar figure in the Hanover of Steffani and Handel.
CONTENT AND STYLE
The overwhelming majority of Steffani’s duets are concerned with unrequited love, a subject that had been bequeathed by the medieval courtly love tradition and become the main preoccupation of the Italian cantata in general. It is presented in a variety of guises. Nymphs or shepherds of Renaissance Arcadia are named in one-third of the duets, ancient Classical subjects in one-ninth. Turbini tempestosi begins with Dido and Aeneas in Africa, but one turns to Senti, Filli spietata for mention of Carthage. A few duets in-voke such Roman gods as Jupiter, Venus, and Vulcan; others derive imagery from the Greek legends of Tantalus and Ariadne. The texts also allude to geographical locations such as Abydos (whence Leander swam to Hero), the Nile, the sands of Libya, Gibraltar, Vesuvius, the Adriatic, and the Alps. Cupid is sometimes described as the ‘dio di Gnido’, a refe-rence to Praxiteles’s statue of Aphrodite in the temple to the goddess at Knidos in Asia Minor. Allusions such as these, which occur also in the cantatas of Steffani’s contemporaries, afford insight into Seicento Italian culture and inform one’s approach to the repertory. Although the subject is invariably addressed from the lover’s point of view, it is treated in a remarkable variety of ways. In Aure, voi che volate the lover asks when his servitude will end (‘Quando mai avrà mercè / una lunga servitù?’); in Parlo e rido he cheers himself up with the thought that countless women yield to his advances (‘Cento almeno m’han per amante, / ma di tante / una sola nel cor mi sta’). He may feel betrayed (‘Ah! che l’ho sempre detto, / ch’Amor mi tradirà’), but in that case he can try his luck elsewhere (‘Cangia pensier, mio cor,‘ / se la speme t’ingannò’); if he is still unsuccessful he may seek justice at the tribunal of Love (‘Mi voglio far intendere / al tribunal d’Amor’) or renounce love altogether (Quest’è l’ultima per me; Ribellatevi, o pensieri). Già tu parti and Lungi da l’idol mio deal with parting or separation, while Gelosia, che vuoi da me (two distinct texts) and Tienmi ‘l Cor la gelosia are concerned with jealousy the nearest Steffani gets to a moral subject. The pleasures of love are extolled in Quanto care al core voi siete (and many others), freedom from it in È spento l’ardore and Più non amo. Io voglio provar may be humorous, but one searches in vain for a satirical, philosophical, or historical duet. The texts are almost exclusively lyrical in expression. The singers normally combine in a single person (the voice of the poet) and, except in solo movements, sing the same words. Even in solos they do not become separate characters. In Begl’occhi, oh Dio, non più, for example, they represent Chloris’s lover, addressing her jointly in the duet movements and separately in the solos: [Beautiful eyes, oh God, no more, weep no more…. My Chloris, if your heart is burdened with some doubt as to my fidelity, put such thoughts behind you … If your jealousy is part of your love, oh beautiful Chloris, what then will happen? … My Chloris, pray, stop for a while, stop the flow of those two streams that you endlessly shed from your eyes …]. Nothing could illustrate more clearly the concept of the chamber duet as a textural expansion of a work for solo voice. Four of the texts, however, include passages in narrative mode in which the poet sets the scene or comments on the situation. Such passages are far less common or extended in Steffani’s duets than in solo cantatas of the same period. This manner is adopted at the beginning of D’un faggio all’ombra assiso and Turbini tempestosi, but in Saldi marmi the narrative passage occurs later on, providing an explanation for the preceding lyrical section: [“Oh solid marble slabs that cover the naked corpse of my beloved, that day by day confirm my love more deeply in my heart, what have you to say? Must I oppose this new desire with your coldness, or die?” Thus said Phyllis when she turned one day to the dead beauties of her lost love. She had lived for a long time, happy in the love of Philenus …]. The narrative in Steffani’s duets is always in versi sciolti and set as recitative, and as soon as it is over the singer resumes his or her normal role in the lyrical duo. The only duet in which the singers do not behave in this way is Io mi parto, which was written for Sophie Charlotte and Max Emanuel. Here the performers—the electress and elector—act as separate individuals. Though they are neither named nor sharply characterised, they have dif-ferent (if only slightly different) words in the duet movements as well as in the solos. The text of the opening duet would not have been out of place in an opera: Metaphor, simile, and rhetorical questions are joined in his duets by other figures of speech. Exclamations are not uncommon (‘Ribellatevi, o pensieri, / pera il dio d’Amor!’), and one text bursts into impassioned declamation: Mia speranza illanguidita, a che più lusinghi il cor? La mia gioia è già finita, già comincia il mio dolor. Udite, o voi che in amoroso inferno condanna cruda sorte in maschera di vita a eterna morte … [My enfeebled hope, why do you still flatter my heart? My joy has now finished, my grief now begins. Oh hear, you whom cruel fate condemns, under the semblance of life, to eternal death …]. Tengo per infallibile is far from unusual in exploring a paradox, and Vorrei dire ends with a common antithesis: … sospiro e peno in vano, perchè gelo vicino, ardo lontano? (Vorrei dire) [I sigh and suffer in vain, because I freeze when near and burn far away?]. Even if the final line or couplet lacks antithesis, it may be epigrammatic in a madrigalian manner: così, seguendo le fallaci idee di speme lusinghiera, mai non gode quel cor che sempre spera. (Crudo Amor) [Thus, following the false ideas of flattering expectation, the heart that always hopes will never rejoice.] Some duets have similarly epigrammatic statements in their opening lines. Such examples often recur as refrains, setting the tone of the poem as does the motto of an aria or the text of a sermon. Once the subject has been announced, the remainder of the poem explores its implications, using figures of speech for the purpose, and the composition as a whole becomes a rhetorical exercise. The musical settings, also, are rhetorical in structure and style. The principal aim of a piece of Baroque vocal music was to arouse the passions of the listener by conveying and reinforcing the affections of the words being set. Texture and structure played a part in this process, but the invention and selection of material were fundamental. Steffani’s responsiveness to words was praised by his early biographer, the theorist CountGiordano Riccati, who felt that his music ‘vividly expressed the sentiments of the poetry’. Riccati cited the second movement of Saldi marmi to show that the major mode could convey ‘affetti molli’, and he particularly admired Steffani’s setting of the word ‘morire’: he may have been thinking of the end of the movement,where the voices form exquisite suspensions over a pedal in the bass. Such musical imagery, the stock-in-trade of the Italian cantata, was inherited from the Renaissance madri-gal. While some musical images represent abstract concepts, others portray visible objects or phenomena. Some of these visions are symbols themselves: when Steffani depicts the flight of Cupid’s arrow in rapid scales, or the bonds of love in a chain of suspensions, he adds musical imagery to symbolic words. The representation of an affection or emotion is not always a straight forward matter. Faced with the words ‘che fermezza Amor non ha’ (‘that Love is not constant’: La fortuna su la ruota), a composer might depict constancy or the lack of it. Steffani does both: the note of constancy is held for some bars but eventually gives way. Some affections occur so frequently that a variety of musical images are needed to convey them: the anguish, pain, or grief of such words as ‘affanno’, ‘dolore’, and ‘duolo’ is expressed by broken figuration and chromatic harmony in Tu m’aspettasti; by rapid gorgie in Aure, voi che volate; and in Cangia pensier by parallel thirds and dissonant suspensions over a pedal. Conversely, some musical effects are employed for a variety of emotions or objects: the passage just cited in Cangia pensier resembles Saldi marmi, even though the words have little in common. Pedals were particularly versatile: in Tu m’aspettasti a pedal overlaid with undulating figuration and static harmony creates an image of the sea; in Forma un mare a similar texture conjures up the wind. Such duplication was inevitable because music is limited as a representational art. It is also unimportant: all the musical images cited here are appropriate to their contexts and purposes; only if they had been used for contrasting purposes in a single piece would they have strained credibility. In other words, the effectiveness of an image depends on its context, as well as on its nature or identity.It is not Steffani’s imagery alone that reflects the contemporary stock-in-trade. He also drew on a pool of common harmonic formulae. Six of the movements in his duets begin with the 42 progression found in his sacred works and overtures: after starting on a unison or octave, the voice rises a second over the bass, turning the latter into a suspension under a 4 –2 chord, from which it falls by step. The formula had been used by Bernabeiand probably every Italian composer of the period. Other movements by Steffani begin with a progression involving a 4 –3 suspension in the vocal line over the dominant note in the bass. The voicemay start on the tonic or leap to it from the dominant, in which case the ensuing leap in the bass from tonic to dominant will sound like an answer. Similar openings are found in Corelli’s trio sonatas, though they are often decorated or varied. As Steffani made use of such formulae, it is not surprising that interrelationships can be found between different duets, the movements of single duets, and the points of single movements. The sections of Quando ti stringo are linked by the inversion of its opening point, and there are similarities between the material of La fortuna su la ruota and that of Forma un mare. Comparison of Quanto care and Rio destin indicates that such similarities are not confined to earlier works. Furthermore, the opening of Rio destin is extensively modified for reuse in the third movement similar interrelationships were exploited in the revision of Begl’occhi, oh Dio, non più. In Che volete the beginning of the first and the end of the last movement are pervaded by the suspension and harmonic progression of the opening bars; the material itself is conventional, but when used so persistently and discreetly it becomes an important formal agent and an eloquent expressive device. Bukofzer’s characterization of seventeenth-century bel canto style as ‘the apotheosis of triple metre’ might have been based on Steffani’s duets. In the works for voices and continuo, there are twice as many movements in triple or compound metre as in common time or §. In one movement out of five there is also a change of time. This normally corresponds with a change of poetical metre or some other feature of the text, and often occurs at the last (epigrammatic) line or couplet of a section, which is thus given prominence. In the six works with additional instruments the movements rarely change metre and are more deeply indebted to dance; the vast majority of the duet movements are in triple or compound metre, but the arias are more evenly divided, most being in common time or §. Tempo markings are rare in the sources and more likely to signal a change of speed during a movement than to set the pace at the start. The principal collection of manuscripts (RM 23. k. 1320) includes five indications (presto, allegro, andante, adagio, and largo), none of which appears more than thrice. In general, tempo must be inferred from the character of the melodic writing, the speed of the word setting, and the time signature. An extensive range of signatures is used for triple and compound metres. Although these signatures occur most often in the original versions of duets that Steffani revised (and recall those in his early sacred works), they also appear in some of his later duets, along with such proportional signatures as 36, 96, 1216, and 2416. The almost ubiquitous broken circle (C: tempus imperfectum) indicates that most of them denote compound duple rather than triple time. The commonest signatures are 32 and C34; the more modern (and more accurate) C64 is used only once. The 31, which is comparatively rare, means what it says: the semibreve is perfect, so undotted, and the reader must understand alteration and dots of division. This is not so in 31, where the semibreve is imperfect, though here, too, the semiminim is void and coloration is used for hemiolas. None of these signatures implies a speed. The opening of Tengo per infallibile, for example, which is in C.31, is evidently brisk: it abounds in hemiolas and syncopation, the harmony is simple, and the bass often leaps a fourth or a fifth.The beginning of Torna a dar vita al core, however, seems rather sedate: the words suggest a slow tempo, and so do the leap on ‘lontano’, the change of harmonic direction and the Neapolitan chord that follows it, and the linear character of the bass. A similar range of speeds is required by movements in31,32, and, especially, 34. The signatures 38, 68, and 128, however, always denote a fast tempo. These were relatively recent in origin and initially associated with vocal music, where the tails and beams of quavers and semiquavers clarified the underlay. Though found in cantatas by Cesti and Legrenzi, they were still ‘rather remarkable’, according to Klenz, in instrumental music of the 1670s; Fedeli’s use of 128 has been described, with reference to instrumental music in Venice, as progressive for 1685. The signature 38 is used in several Steffani duets, including La fortuna su la ruota and No, no, no, non voglio se devo amare, while 68 is found in Oh! che voi direste bene and 128 in Parlo e rido duets which, as we have seen, were probably composed during his Munich years. If they were, he evidently embraced both ancient and modern time signatures from a relatively early age. Given the influence of French dance on his operas, it is not surprising that Steffani used the same metres in his chamber duets. The earliest datable movement that is labelled as such in a duet—though it is probably not the first of its kind—is apparently ‘Che giova lontano’ in Dir che giovi (ca. 1688[?]). Marked ‘sarabanda’, this aria is typical of the French style in its clearly articulated and carefully balanced phrases, its harmonic direction, and general rhythmic uniformity between voice and continuo; but its wide leaps, ornamentation, Neapolitan harmony, and imitation in the bass are more characteristic of the Italian manner: the mould is French, but the filling is Italian, a mixture typical of Steffani in the 1680s and 1690s. Movements labelled ‘menüet’ appear only in Mia speranza illanguidita and S’intimi guerra a la crudel beltà, but the crotchetminim pattern typical of the dance recurs in other works, including La fortuna su la ruota. The signature and halfbar anacrucis of the gavotte occur in five arias, of which ‘Che rimedio è ’l star lontano’ (D’un faggio all’ombra assiso) is the simplest and the only one labelled in the sources. The others make greater use of repetition, some of it sequential, and of counterpoint between voice and bass. ‘Chi sarà ch’hoggi mi scampi’ (Luci belle, non tanta fretta) is exceptional in lacking a double barline halfway; it is bisected by a dominant cadence, but the first section is irregular (18.5 bars) and, uncharacteristically, longer than the second: the aria is in the style of a gavotte but not in its normal form. The signature also appears, together with a crotchet upbeat, in three movements, of which two are labelled ‘bourée’ (sic). It would be hard toestablish when such dances were introduced into Italian vocal music, because most cantatasources are undated, but since the gavotte was first published in Italian instrumental music in 1666 and the gigue in 1669, Steffani was probably one of the earliest composers to employ them in arias and duets.As in metre, so in key there is a pronounced difference between his duets with continuo and those with additional instruments. While theformer are mostly in the minor mode, the latter prefer the major. Furthermore, while the continuo duets in major keys are predominantly on the sharp side (only six are in F or B), half of those with instruments are in flat keys. The instrumental duets are so few in number that it is dangerousto generalise, but the emphasis on the major mode is consistent with their more dancelike metres and contributes to their lighter character. The range of keys in Steffani’s output extends from B to A major and from F to E minor. None of his works is in E major, but several movements have a cadence in B or even F. Similarly, although there are no movements in B minor, Troppo cruda è la mia sorte cadences in D major and touches on G.While most of his key signatures conform with those of the tonal system, that of A major is short of a sharp sign and those of B major and of G, C, and F minor each lack a flat. There are two exceptions: Tengo per infallibile is in G Mixolydian (no sharp), and Labri belli, dite un pò in D Dorian (no flat). As Steffani’s time signatures preserve features of the mensural system, so his key signatures retain vestiges of the modal. All his duets, without exception, end in the key in which they begin, and most include at least one movement that either begins in one key and ends in another or lies entirely outside the tonic. The ‘modulating’ movements occur mainly in works without solos and contribute to a larger design. The central and final movements of Che volete, for example, are tonally complementary, the former ending and the latter beginning in the relative major key. The solo movements that modulate are recitatives with cavata, related to their context in similar ways. Since they promote continuity, modulating movements are particularly common in throughcomposed works. The movements in keys other than the tonic are all solos: in other words, all the ‘nonmodulating’ duet movements are in the tonic key. Since the latter occur often at the beginning and end of a work, they provide the tonal framework for the internal movements, which may be solos in contrasting keys. The range of such keys is narrow— relative major in minor-key works; relative minor, dominant, or subdominant in major—but the distribution of duet and solo movements makes for balance overall. A wider range of keys is used within movements. Among the commonest modulations in major-key movements is that to the dominant of the dominant. Oh! che voi direste bene goes one step further, to the dominant of the dominant of the dominant, while Luci belle modulates exceptionally to the relative minor of the dominant. Che sarà di quel pensiero and Tu m’aspettasti visit the relative minor of the subdominant, but the only modulations to the subdominant itself are found in the Mixolydian Tengo per infallibile. In minor-key movements the normal excursions are to the relative major, the dominant, and the relative major of the dominant. The modulations to the subdominant in Dir che giovi and the relative major of the subdominant in Troppo cruda are exceptional: as in the major-key movements and other Italian music of the period (e.g., Corelli’s concertos), excursions to the subdominant are rare. Although Steffani’s bass lines are virtually unfigured, his harmonic intentions in duet movements are clear from the three notated parts. If he uses the 6– 4 chord, he prepares the fourth and resolves the triad on to 5–3, often via 5– 4. The dominant seventh is not shunned: the seventh is normally a passing note or a suspension, but it occasionally ‘passes’ so slowly as to seem consonant. The Neapolitan sixth occurred in Dir che giovi; other chromatic chords are found in such duets as Troppo cruda and Occhi, perchè piangete? which make a special feature of chromaticism: the opening of Troppo cruda includes both a diminished seventh and, arguably, an augmented chord. The harmony of Steffani’s recitative differs in some ways from that of his arias and duets. The most obvious differences are the widespread use of bass pedals (with diverse chords above them); the resolution of a last inversion of a dominant seventh on to a root position tonic; and the practice, seen in his solo cantatas (discussed earlier in this chapter), of starting a phrase on a chord that is unrelated to the chord that precedes it. His harmonic vocabulary is enriched by frequent suspensions. Dissonance and resolution season the harmony, highlight the counterpoint, generate rhythmic momentum, and impart a sense of direction. Even an early duet such as La fortuna su la ruota employs all the normal dissonances—fourth, major and minor seventh, and major and minor ninth—as well as a chain of suspensions over an extended sequence that would be exceptional in his later duets. E perchè non m’uccidete, for example, includes a passage with dissonance on nearly every strong beat, but the harmony is not sequential and some of the dissonances are only fleetingly resolved. His duets are remarkable also for their frequent suspensions in the bass, a distinctive feature that attracted Riccati and epitomizes the contrapuntal role of the continuo. Sevenths in the bass in No, no, no, non voglio se devo amare spring from the use of imitation; fourths and ninths are less common but can also be found. Approaching a cadence (which may include parallel seconds alla Corelli), Steffani sometimes uses an unprepared melodic seventh or a 4 –3 chord, both distinctive features of Seicento music. The seventh can be prepared by the continuo player ,but the 4 –3 chord is more difficult to explain. In the earlier version of E perchè non m’uccidete the fourth and third also appear as the sixth and fifth over a different bass note in the previous bar, suggesting that the 4 –3 chord should be viewed, anachronistically perhaps, as a second inversion of a secondary seventh. It appears in major and minor keys, and the dissonance is often protracted and stressed, implying that the sonority was pleasurable. Cadences in the earlier duets are rarely approached via a dominant pedal, a standard device in such later works as Che volete (where Steffani also displays ingenuity in avoiding repeated V–i progressions, occasionally by flattening the leading note). ‘Imperfect’ progressions occur in most duets, but the plagal cadence is not a feature of his style. The continuo promotes unity of texture and continuity of structure. In duet movements, where contrapuntal interest is borne mainly by the voices, the bass mostly reinforces the harmonic rhythm, although it may also echo or anticipate vocal material, especially in a link between sections or a coda at the end. In solo arias and cavatas, however, the continuo plays a more contrapuntal role, imitating the vocal line or supplying an ostinato. Its role is slightly different in movements for bass voice, which, as Legrenzi and Gaffi acknowledged, poses a particular problem. Steffani’s writing for bass or baritone, like that of his Italian contemporaries, is characterized by rapid scales, wide leaps, and a desire to treat the voice as equal to the others in range and agility. Such material is less suitable for imitation by the continuo, which therefore adopts a simpler, supporting role, often presenting an outline of the vocal part. The singer usually joins the continuo at cadences and rarely descends below its line. Ostinato basses, common in Steffani’s Munich operas but less so in those for Hanover or Düsseldorf, are found in only five solo movements of his chamber duets. This might suggest that most of his duets are relatively late, but it is more likely to mean that ostinato basses, possibly because of their insistence on a single affection, were more appropriate to the theatre than the chamber—an idea supported by their comparative rarity in the cantatas of Luigi Rossi, Carissimi, Savioni, Gasparini, and Handel. Steffani’s ostinatos were regarded by Riemann as a high point in the history of the technique and a probable model for Purcell. They display various kinds of treatment, ranging from literal repetition of a melody several bars long to constant reiteration of a motive of as few as four notes. One of the longer examples in Steffani’s duets is the two-bar bass of ‘Sospirar per la bellezza’ in Cangia pensier, which is heard twelve times and in three keys; the first four notes link adjacent statements in different keys and furnish the basis for two ‘free’ bars (23–25) toward the end. Above this framework the voice moves relatively freely, bridging some of the cadences and coinciding with others. The identity and distribution of the vocal material, together with the cadences, determine the overall form. Greater variety of treatment is found in three solos where, as in cantatas by Rossi and Savioni, the ostinato is confined to a section of a movement. In ‘Struggerò con calde stille’ (Labri belli) the ostinato is also taken from the vocal part, a practice favoured by Stradella. The aria ‘Che sarà, mal gradita fedeltà?’ (Il mio seno) is a tour de force of ‘quasi ostinato’ construction. After the opening eight bars, which are ‘free’, the bass is entirely composed of two motives of four quavers each .The motives are grouped in varying quantities into phrases of varying length, which are punctuated by rests of varying duration. Both motives are stated in all five closely related keys, and no two phrases are identical in structure. The first phrase avoids motive (b), but this appears up to three times in each subsequent phrase; although this motive has cadential force, it occurs in the middle as well as at the end of a phrase, and toward the end of the movement its final quaver is frequently omitted, possibly to illustrate the word ‘muor’, to which the voice often supplies the missing note. The vocal line incorporates motive (a), but most of its material is free; though some of its phrases are repeated and rather predictable, its form (abbccdeff) is largely independent of the bass, which is wholly unpredictable. It is not far from this to a running bass in continuous quavers with no ostinato at all. The scoring of Steffani’s duets involves six combinations of voices. Many duets reveal his preference for voices pitched a fifth or twelfth apart (soprano with alto, alto with tenor, and soprano with bass) and his interest in those an octave apart (soprano with tenor, alto with bass). Less than one duet in seven is scored for equal voices (two sopranos). It would be wrong to consider soprano-tenor an equal-voice combination: soprano and tenor may have been interchangeable in early-seventeenth century duets and later solo cantatas, but the counterpoint of Steffani’s duets discourages substitution. His preference for unequal scoring contrasts strongly with that of the earlier chamber duet and contemporary trio sonata. The latter is normally scored for two equal instruments: sonatas with ‘unequal’ melody instruments are rare in the late Baroque period, and the most notable examples—Bach’s sonatas for organ, or for viola da gamba and harpsichord—dispense with continuo realization. Steffani’s duets for soprano and alto, soprano and tenor, or alto and tenor could likewise be performed without chordal accompaniment, and sometimes possibly were, though probably not for preference, in the seventeenth and eighteenth centuries. Early Seicento vocal duets tend to be scored either for high voice and bass or for two equal voices: all but two of Monteverdi’s duets are for a pair of sopranos or tenors, only three of Carissimi’s are not for two sopranos, and the same combination predominates in ‘D’autori romani’ (I-Bc, MSS Q. 44 –50), an extensive collection of music from mid-seventeenth century Rome. There is greater diversity in succeeding generations: fewer than half of Cesti’s duets are scored for equal voices, while for Stradella the proportion is less than one in three. These developments are reflected in Steffani’s duets and were probably encouraged by the growing interest in counterpoint in the second half of the century. Of the works by him for which early dates may be suggested (on the basis of concordances), over half are scored for soprano and bass while another is for two sopranos; by contrast, of the eight remaining duets that can be approximately dated, three quarters are for soprano and alto. Although these later duets were probably written for particular singers, the fact remains that nearly one-third of Steffani’s soprano-alto duets are relatively late compositions. Since equal voices generally excel in similar parts of the range, their use tends to favour parallel motion; counterpoint is not excluded, but it is more difficult for the composer to maintain the independence and distinctness of two equal parts. Scoring for unequal voices does not guarantee a contrapuntal texture, but it facilitates imitation at the fifth, octave, or twelfth and creates tonal space for manoeuvre: it provides opportunities for counterpoint, and Steffani exploits them to the full.The texture of his duets derives largely from his use of imitative and invertible counterpoint. The subject is normally announced by the upper voice, almost invariably a soprano. The pitch of the answer depends partly on the identity of the other voice. Alto and bass often answer at the fourth or fifth (or eleventh or twelfth) below, but the alto may answer at the unison or octave and the bass at the octave or double octave. Although the tenor frequently answers the soprano at the octave below, entries at the fourth, eleventh, and twelfth are also abundant. In duets for two sopranos, answers at the fourth and fifth, above and below, occur alongside those at the unison. There are also examples of imitation, though not of initial answers, at such intervals as the second and third. If the subject begins and ends in one and the same key, the answer is normally real. In the first section of a movement the subject is invariably in the tonic, but later sections frequently begin in the dominant; in this case the answer may be in either key. Since an answer in the tonic has a cadential effect, it is often withheld until the end—as in the final section of Dolce è per voi soffrire, a masterly piece of tonal pacing. The tonal answer appears comparatively rarely with non modulating subjects, but examples can be found, even in such relatively early duets as La fortuna su la ruota. Modulating subjects are not uncommon, and the modulation is invariably to the dominant; the answer is normally adjusted to remain in that key or return to the tonic, though if pitched at the unison or octave it may repeat the subject’s progression. Some of Steffani’s duets recall the ‘a risposta’ technique, each voice presenting an extended subject separately before combining in counterpoint. The procedure is normally found at the beginning of a movement and elaborated. In Non, no, no, non voglio se devo amare, for example, the head of the answer is detached from the tail and set against the subject in the opening bars—a false answer that camouflages the underlying plan; in Labri belli the answer begins before the subject has finished. A number of techniques are available at the end of the answer. In Labri belli each phrase of the subject is restated by the soprano and immediately imitated by the other voice (bass), creating a series of strettos. The section concludes with a repeat of the second half of this series, transposed down a fifth to end in the tonic. If this kind of stretto is particularly characteristic of ‘a risposta’ sections, repetition and transposition of chunks of material, often accompanied by inversion of the texture, are standard in all types of fugal section. Two further devices are employed in Crudo Amor: the stretto is anticipated by a false entry on the head of the subject ,and in the partial restatement of the stretto the time lag between the voices is halved. Telescoping of entries is not confined to sections employing the ‘a risposta’ technique, but since it tightens the texture it is often reserved for the end. False entries and strettos are among the commonest means by which Steffani enriches his expositions, a sensible measure when only two imitative parts are involved. His practice is illustrated by the duet movement in Tu m’aspettasti. The entries of the opening phrases—(a) ‘Conducetemi verso il porto’ and (b) ‘o nocchieri del mare d’amore’—are shown in Table 9.9. The false entry is the first statement of (a) in the tenor. It is imitated in stretto by the soprano’s second (a), which is immediately answered in closer stretto by the tenor’s. As a result of the false entry, the soprano sings the complete subject twice before the tenor finishes it once. Steffani could haveavoided this by omitting the false entry and inverting the remainder, but he would have sacrificed both the first stretto and the cumulative impact of two strettos in quick succession. The effect of a stretto is enhanced if it creates a new piece of double (or invertible) counterpoint. An example occurs in the earlier version of Ah! che l’ho sempre detto (1b), of which the second movement (‘In due luci tutte ardore’) includes a section composed of two phrases—(a) ‘poi con guardo lusinghiero’ and (b) ‘mi trafisse a morte il core’ (Table 9.10). Bars 46–51 constitute the exposition; bars 50–53 are a stretto in which (b) acts for the first time as a countersubject to (a); bars 53–57 are the same stretto inverted and transposed, and bars 57–61 are a series of strettos, at diminishing intervals, on (b) alone. The only permutation of material that is not employed is (a) in combination with itself. Although double counterpoint is often introduced in the course of a section, as here, it is normally announced at the beginning and constitutes its principal texture. The phrases are usually similar in length, forming a self-contained block of material capable of inversion and transposition. If in these respects it seems rather limited, the texture also affords opportunities for separate strettos on each phrase in turn. The most characteristic examples in Steffani’s duets are set to the final lines of such comparatively late pieces as Crudo Amor and the revised version of Quanto care, where extended sections of intense counterpoint are preceded by brief passages of recitative a due. At the beginning of a movement the texture may be looser and the contrapuntal potential of the material less thoroughly explored. Though standard in the trio sonata, double counterpoint is comparatively rare in the chamber duet, possibly because it entails the simultaneous delivery of different portions of text. When the latter are complementary, antithetical, or epigrammatic, however, double counterpoint is a highly appropriate means of presenting them. Steffani, who used the texture in early as well as late duets, may have been one of the first composers to exploit it in vocal works. He did not, however, employ modulatory episodes or middle entries in related keys. In some duets the idea of a middle entry is ‘replaced’ by an exceptional repeat. In the last movement of Che volete, for example, the opening point returns over threefifths of the way through. The passage in question is almost a literal repeat of the exposition, but because it is widely separated from those bars by other material and appears in a different key (and the interval between the voices has grown by an octave), it has a startling effect comparable to that of a middle or final entry in a fugue. That the same device is used in Placidissime catene, which, like Che volete, dates from the late 1690s, suggests that at this stage of his career Steffani investigated new ways (for him) of extending contrapuntal sections. Although his duets also lack the harmonic freedom that is typical of fugal episodes, they use textures found in such sections, including motivic interplay between the voices; a brief, incisive rhythmic figure in one voice against a long, florid subject in the other; and homophony. The last occurs mostly in short passages of recitative a due and brief introductions to more extended sections of counterpoint; several duet movements begin homophonically, but none approaches the consistent parallel motion of the air à deux, and none is a dance. These textures provide contrast within and between contrapuntal sections of movements. Despite their importance, however, it is the quality of Steffani’s counterpoint and the beauty of his melodic and harmonic expression that make him the greatest exponent of the Italian chamber duet in the late seventeenth or early eighteenth century.

POLYMATH OF THE BAROQUE
Agostino Steffani and His Music
COLIN TIMMS
OXFORD UNIVERSITY PRESS 2003