VIVALDI: L'Estro Armonico, Op . 3

L'Estro Armonico is a title that defies translation; neither The Harmonic Fancy nor The Musical Flush suggests quite the right combination of genius and fantasy that prompted Estienne Roger, the shrewdest of the 18th-century publishers, to issue this set of twelve concertos in 1711. Prior to this, Vivaldi's only printed works had been two sets of sonatas published in Venice: twelve trio sonatas (Opus 1) in 1705 and twelve solo sonatas (Opus 2) in 1709, both stemming from his activities as violin teacher at the Conservatorio dell' Ospedale della Pietà.

This orphanage was one of the four famous institutions in Venice that offered a musical training to young girls, and Vivaldi directed concerts on Sundays and feast-days which very soon acquired a reputation with visitors from far outside the city. 'The transcendant music is that of the hospitals. There are four of them, all consisting of bastard girls or orphans, and those whom their parents are in no position to bring up. They are brought up at the expense of the State, and they are trained only to excel in music. They sing like angels, and play the violin, the flute, the organ, the oboe, the cello, the bassoon; in short, there is no instrument so large that it could frighten them ... I swear to you that there is nothing more pleasing than to see a pretty young nun in a white habit, with a spray of pomegranate blossom behind her ear, conducting the orchestra and beating time with all the grace and precision imaginable' (Charles de Brosses, Lettres familières sur l'Italie).

It was the publication of Opus 3, however, that made Vivaldi's reputation in Europe, and the collection was soon reprinted by John Walsh, in London (Vivaldi' s most Celebrated Concertos) and by Le Clerce Cadet in Paris, with the title somewhat bizarrely rendered as Les Troharnonico. Quantz heard them for the first time in Pirna, near Dresden, in 1714:'as musical pieces of a kind that was then entirely new, they made no small impression on me. I was eager to accumulate a good number of them, and Vivaldi's splendid ritornelli served as good models for me in later days.'

Of the ten keyboard transcriptions that Bach made from Vivaldi's concertos, six are taken from Opus 3, but significantly, not from the printed versions. It is clear, however, that at least some of these works had already been in circulation in manuscript versions for several years; six of them are still extant in widely separated collections (three in Dresden, and one each in Vienna, Naples and Schwerin). Since there would have been little point or profit in making a copy of a work that was in print, we can deduce that these works were known and widely distributed perhaps even as early as 1700. The collection that Vivaldi put together for Roger was designed to impress by its diversity, both of style and scoring. By drawing on more than ten years' experiments with concerto form he could include examples in the old, Corellian manner, like the two polysectional concertos, IV and VII, or modelled on the concertos for 'due violini che concertano soli' of Torelli in his Opus 8, or, in the most advanced manner, introducing a lyricism and drama that had previously belonged to the opera house, in those solo concertos where Vivaldi clearly saw himself as the protagonist.

The collection was not put together chronologically, but in a complex arrangement designed to show a maximum of variety if the set is played as a whole (a scheme that seems to have involved Vivaldi in a bit of a last-minute rewriting). The concertos are arranged in four groups of three, each containing a solo, double and quadruple concerto. In addition Vivaldi employs a pairwise arrangement by keys, each concerto in the major being followed by one in the minor, with the exception of the final pair, where this system is reversed in order to end the entire set in the major. This seems to have been a typically Italian arrangement; Albinoni (Opp. 7 and 9) and Torelli (Opp. 5 and 8) employed the same pairwise system.

The rewriting appears to have been necessary to give Vivaldi enough concertos with four soloists to satisfy his complex arrangement, and it is most obvious in VII, which shows every sign of having originated as a concerto with the normal concertino of two violins and continuo; the final Allegro, in fact, retains the original grouping (although the continuo line now has to become 'violoncello obligato'), and elsewhere the duality of the solo parts has been inflated by simple extension and repetition. In several other concertos it is possible to deduce the existence of an original continuo line, and the often redundant 'violoncello obligato' seems only to hide the slightly uneasy compromise that Vivaldi had to make in order to issue the set with a single continuo part.

The concertos were issued in eight partbooks (four violins, two violas, violoncello, and 'violone e cembalo'). The frequent contention that this implies antiphonal performance is contradicted by the works themselves; the only antiphony that is required is between concertino and tutti (as in VII), or between individual soloists in I, IV and X, for example, in too mercurial a style to allow spatial separation. There is no antiphonal writing at all for the two violas. The simple answer is the correct one: eight part-books is the minimum that can contain the variegated scoring that Vivaldi demands, with a maximum of four solo violins, a cello part which occasionally diverges from the continue, and two violas which occasionally divide. The assumption by Vivaldi, of course, was that each part would be played by a single player, and the parts were so organized that in the solo concertos this would give a tutti of three violins; in the double concertos each soloist would be doubled in the ritornelli; where four soloists are used together the indications of 'tutti' and 'solo' were, as usual, navigational aids only. It is the same assumption that Bach made for the performance of the Brandenburg Concertos, and Brandenburg III is probably the nearest German derivative of Vivaldi's scoring for multiple string soloists. In the present recording, the first to use solo strings throughout, the players are laid out with first and second violins to the left, and third and fourth violins to the right of a central continuo section, except in the concertos 'con due violini obligati' where the two soloists are placed antiphonally.

The large part that improvisation played in 18th-century concerto performances - particularly in the 'gracing' of the solo parts and the realisation of the continuo - leaves the modern player with many problems. The little evidence we have of Vivaldi's practice suggests that where he required elaborate soloistic decoration, he would indicate it (as in the lavish slow movements of V, VI and IX), or use the convention of a corona ( ) to indicate a momentary halt for a cadenza. Elsewhere, Quantz's advice seems most apt: 'One ought to avoid varying the lyrical ideas of which one does not easily tire, and, likewise, the brilliant passages which have a sufficiently pleasing melody themselves. One should only vary such ideas as make no great impression'. And, most pertinently for the slow movement of XI, he says: 'A siciliano ought to be played very simply, with scarcely any trills and not too slowly. One should not employ many other ornaments here except for a few appogiaturas, because it is an imitation of a Sicilian shepherds' dance'.

As regards the improvised continuo element, two fragments of a realisation written out by Vivaldi himself confirm the view that the Italians regarded the continuo as a background support to the harmony, rather than a foreground rival to the soloist.

It is illuminating to see how L' EstroArmonico (which, after Corelli's Opus IV, were the most popular set of concertos of the whole century) provided the rules from which such writers as Quantz, Marcello and Mattheson judged and advised other composers. Quantz, in his Versuch einer Anweisung die flöte Traversiere zu spielen, writes that what was looked for was a 'magnificent opening ritornello, with all the parts well elaborated' (II, IV or VIII for example), or possibly 'the unison, performed in a lofty and majestic manner, with a fire and vigour such as is not given to the notes of another sort of melody' (V, or the Largo of I). 'A skilful mixture of the imitations in the concerted parts' is best displayed with four soloists in the first movement, a contrast between the ritornello and the lyrical solo is recommended; Vivaldi's solutions for the Largo movements include a persistent unison tutti (as in I), sustained harmony (VI), gently repeated chords (V, XI) and a frequent lightening of the texture by omitting the bass line completely. The last movement 'should differ from the first both in the nature and metre... as humorous and gay as the first is serious'; one interesting alternative to the bravura finale is provided by the last movement of VII, a minuet in the French ballet style, which would traditionally end a suite of court dances (compare the final minuet and trios of Brandenburg I). In this context, it now appears that the Italians were more conscious of the stile alla Francese than had been assumed in the past, and several passages in Vivaldi's concertos call for interpretation in the French manner.

Those concertos from Opus 3 which surprised 18th-century commentators need little note, since they have the same impact on us: the theatrical opening of II (remarkably similar to the start of 'Winter' in The Four Seasons), or the extraordinary passage of scoring for the four soloists in the Larghetto of X, where four distinct methods of arpeggiation are specified simultaneously. Of the whole set, it was the Eleventh Concerto which excited most comment and imitation. The drama of its opening could never be repeated, but the following fugue subject tempted many to the sincerest form of flattery. It was even remarked on that Vivaldi, 'being of a volatile disposition (having too much mercury in his constitution)', should have shown such contrapuntal skill. Dr William Hayes, Professor of Music at Oxford, further declared that 'in the eleventh of his first twelve concertos, Opus 3, he has given us a specimen of his capacity in solid composition... in the others he ues himself upon a certain brilliance of fancy and execution, in which he excelled all who went before him, and in which even Geminiani has not thought him unworthy to be imitated. But in the above concerto is a fugue, the principal subjects of which are well invented, well maintained, the whole properly diversified with masterly contrivances, and the harmony full and complete' (from Remarks on Mr Avison's Essay on Musical Expression).


Title Page of L'Estro Armonico

Antonio Vivaldi