(ii) Ornamentation in vocal music

(iv) Instrumental music.

Tartini reiterated the standard rules for the length of ‘long’ appoggiaturas – that they should fall on the beat and occupy half the value of the main note, or two-thirds of the value where they are attached to a dotted note. But he allowed for a de-emphasized appoggiatura where the context (a passage descending in 3rds) required it to be short and passing in character. In a case such as ex.28 the grace notes should be as short as possible and the accent should fall on the main notes. He acknowledged the possibility of rising appoggiaturas, but was uncomfortable about using them except in combination with other grace notes that provided an acceptable resolution of the dissonance (ex.29).

For Tartini the turn (and the inverted or lower turn which, like rising appoggiaturas, he found less useful) involved taking as little length and emphasis away from the main note as possible. This produces a rather different-sounding ornament from the evenly measured note groupings described in French and German sources. (Although Tartini did not specify that the three initial notes of the turn should come before the beat, it is difficult to follow his instructions without that being the effect.)

Tartini's coverage suggests that Italian practice in the application of these graces did not always conform to the rules enunciated elsewhere in Europe. Neumann argued from a large number of music examples that it is impossible to be dogmatic about the length and placing of small ornaments (i.e. whether they occur on or before the beat and how much, if any, of the value of the principal note they account for).

Tartini was not alone in discussing vibrato as an ornament alongside the four standard graces considered above. Agricola, in his commentary on Tosi, dealt with vibrato at the end of a chapter on trills, mordents and turns. These and other writers emphasized the importance of adjusting the speed and intensity of vibrato to its context. Geminiani included his famous description of the close shake (see §6) at the end of his discussion of ornaments in A Treatise of Good Taste in the Art of Musick (G1749), which then became example XVIII of The Art of Playing on the Violin (G1751). His encouragement to use this device ‘as often as possible’ needs to be read alongside his advice that the plain shake ‘may be made upon any Note’ and that the appoggiatura ‘will always have a pleasing Effect, and … may be added to any Note’. (Amusingly, Bernhard, Ic1649, included among his ornaments fermo, i.e. without vibrato, noting that it could ‘be regarded as a refinement mainly because the tremulo is a defect’.)

There was general agreement throughout Europe that the attitude towards slow movements – as melodic outlines needing embellishment – was one of the most distinctive and fascinating aspects of the Italian style in the late Baroque period (see Improvisation, §3(iv)). The vogue as it emerged in the later 17th century resembles a resurgence of diminution practices; but whereas diminution typically involved an arithmetical division of long notes into clearly articulated short notes, the most characteristic gestures of late 17th- or early 18th-century embellishment were slurred and asymmetrical, or at least not conspicuously measured, as if unpremeditated and driven by a spontaneous outpouring of emotion. The most famous – and controversial – model for this was the set of graces for the adagios purportedly by Corelli which Roger included in his 1710 edition of the op.5 sonatas. These were not the first such examples for this set of sonatas, and they were far from being the last (though most subsequent sets of graces concentrated on the slow movements from the second part, the sonate da camera). The most systematic instruction in embellishing italianate adagios came from Quantz who, stressing the need for an understanding of harmony, progressed from a demonstration of extempore variation of simple intervals to a consideration of entire movements.

This development parallels the expectations for free embellishment in da capo arias, though the decoration of instrumental movements was not necessarily reserved for repeats. In fact, of all the slow movements in the first part of Corelli's op.5, only the opening binary Adagio of Sonata no.5 in G minor makes any provision for repeats. A related development, which corresponds quite closely to vocal practice, is the use of florid decoration at cadence points. According to Quantz, cadenzas had become popular with the Italians in the early 18th century and

were subsequently imitated by the Germans and others who devoted themselves to singing and playing in the Italian style. … Perhaps the surest account which can be given of the origin of cadenzas is that several years before the end of the previous century, and in the first ten years of the present one, the close of a concertante part was made with a little passage over a moving bass, to which a good shake was attached; between 1710 and 1716, or thereabouts, the cadenzas customary at present, in which the bass must pause, became the mode.

Tartini distinguished between ‘natural cadences’ in which the decoration does not hold up the movement of the bass line, and ‘artificial cadences’ in which the bass line pauses on the dominant while the melodic instrument indulges in a cadenza above it. Both types of Cadenza conclude with a dominant trill (normally on the 5th of the chord) leading on to the tonic.

All forms of free ornamentation were, strictly speaking, the preserve of solo players, though there is evidence (mostly in the form of injunctions against doing it) that some orchestral players could not always repress the instinct to embellish.