Although stenographic symbols for standard decorative formulae existed in early 17th-century France, notably in Vallet's lute publication Secretum musarum (1615–16), they were not codified in any systematic fashion. The absence of a notational convention was commented on by Bacilly (H1668, p.135):The majority of these ornaments are never printed in the music, either because they cannot accurately be reduced to print owing to a lack of appropriate musical symbols, or because it may be thought that a superabundance of markings might hinder and obscure the clarity of an air and thus result in confusion.The remedy came soon after, with the flowering of the harpsichord and viol repertories. Ornamental clichés proliferated in manuscripts of harpsichord (Louis Couperin) and viol (Dubuisson) music, and the ‘Table des agréments’ in Chambonnières' Pièces de clavecin (1670) spawned the inclusion of similar tables in the prefaces of most instrumental publications in France. The most comprehensive and fully developed table was by d'Anglebert (Pièces de clavecin, 1689; see fig.2), whose far-reaching influence was felt in Germany when the table was copied by J.S. Bach between 1709 and 1714. These tables maintained a high degree of consistency in the interpretation of named ornaments into the 19th century; the most important – cadence, port de voix, coulé and accent – were retained by Cartier (H1798, 3/c1803) for the violin, albeit with some alteration to their meaning. Indeed, ornamental nomenclature varied not just from one instrument to another but within an instrumental tradition. Such baffling diversity in marking ornaments was bemoaned by Montéclair and many others. Bérard attempted to rationalize French agréments by inventing new symbols, incorporating Greek and Hebrew letters, thus adding to the confusion.
Many composers, however, were content to use a very limited number of symbols, especially in 17th-century vocal parts, preferring to leave the option of more creative embellishment to the performer. Rameau, in his Code de musique pratique (1760), mentioned the need for variety in the execution of ornaments to prevent their becoming ‘insipid'. Often the choice of symbols was governed by tradition, and different sets of symbols were used in the same piece – for example in Rameau's Pièces de clavecin en concerts (1741), where the wavy line in the clavecin part and the cross in the basse de viole part both indicate a tremblement (ex.43). Loulié differentiated between the sign of a tremblement ‘dans les pièces’ and ‘dans la basse continue’, warning that ‘what I am going to say about the comma for les pièces must be understood for the small cross in the basse continue’. The ornaments in his Eléments ou principes de musique (H1696) were replaced in the Amsterdam edition of 1698; it appears that substitutions were necessary because many characters of type were unavailable in resetting the treatise. For example, the symbol for a trill (‘+’, used consistently in the original edition) is replaced by a ‘t’, even where the text continues to speak of a ‘cross’.
Agréments lie at the heart of what the French considered to be proper execution (propreté) and taste (goût). Bacilly observed: ‘Without any doubt a piece of music can be beautiful but at the same time unpleasant. This is usually a result of the omission of the necessary ornaments’ (H1668, p.135). Corrette wrote that ‘a song without any ornament is like an unpolished diamond’ (H1758, chap.15). In the choice of ornaments, Saint Lambert (H1702, p.42) advised that ‘good taste is the only law that one can follow’. Yet contemporary writers were seemingly unable clearly to define what le bon goût actually was. Hotteterre (H1707, §9.23) wrote that ‘one can scarcely give more certain rules for [the] distribution [of ornaments] … it is taste and practical experience, rather than theory, which can teach their appropriate use’. Others recommended study with a master or critical observation of one at work. Clearly, the use of tasteful ornaments was an integral part not only of a performer's artistry but also of his or her technique (Bacilly, H1668, p.89):
as can be observed among the majority of ill-trained singers, there are certain vocal qualities which will never sound satisfactory in themselves, no matter how well handled in performance. For example, some of these faults are singing through the nose, bad breath support, bad cadences and accents or plaintes, use of inappropriate ornaments at the end of an air or the incorrect placement of ports de voix, making passages with the tongue.
Le Gallois (H1680, pp.76–8) chastised ‘flashy players’ who exhibited poor taste in embellishing:
Their cadences are often played too rapidly, and as a result quite crudely, having been produced with too much energy … one hears nothing but a perpetual cadence, which prevents one from hearing the basic melody clearly. They continually, habitually add passages, especially from one note to its octave, which Chambonnières used to call ‘rattling’ [chaudronnier].
But he had special praise for Chambonnières' ability to improvise new embellishments (ibid., 70):
Each time he played a piece he added new delights with ports de voix, passages and different ornaments, including double cadences. In a word, he so varied them with all these different adornments that he was always able to disclose some new beauty in them.
François Couperin had less confidence in his readers' taste; in the preface to the third book of his Piéces de clavecin (1722), he declared that his ornaments and music were indissolubly linked:
I am always surprised, after the pains I have taken to indicate the agrémens which suit my pieces, and of which I have given separately a quite intelligible explanation in a particular méthode known as L'art de toucher le clavecin, to hear people who have learnt them without following my instructions. This negligence is unpardonable, the more so since it is not at all an arbitrary matter to introduce such ornaments as one wishes. I declare that my pieces must be executed as I have marked them, and that they will never make much of an impression on persons of true taste unless everything I have indicated is observed to the letter, without adding or subtracting anything.
Ornaments served a practical purpose: to provide shape and character to the melody.To sing or play proprement is to execute French melody with the ornaments that suit it. This melody, being nothing by the mere force of the sounds, and not having by the same any character, only receives it [character] by the affective contours that one gives it in executing it. These contours, taught by the masters of goût du chant, make up what one calls the agrémens of French song. (Rousseau, H1768, under ‘Proprement’, p.396)The most important ornaments, such as those with long appoggiaturas, provide melodic dissonance, thereby fulfilling a crucial harmonic role. It was this concept of ‘dissonance as decoration’ which Brossard, in his Dictionaire des termes (H1701, 2/1703), called the supposition and equated with ornement du chant. Masson (H1694, 2/1699, p.59) wrote that dissonance was required ‘for giving beauty to the melody, by adding a note which functions as an ornament … [and] for connecting the intervals, i.e. for rendering the melody smoother and sweeter’. Such ornaments acquire their expressiveness by being sounded against some present or implied consonant chord of resolution while simultaneously being used to ‘suppose’ or substitute for the displaced consonant. The ubiquity of such ornaments, used to enhance almost all cadential points, testifies to the growing notion of harmony, rather than melody, as the essential determinant of musical structure.
The playing of ornaments, whether supplied by the composer or provided by the performer, was never the sole prerogative of the melodist. Though continuo players often encounter bass-line embellishments in the form of divisions, bass-line ornament signs are less common. In the trio sonatas of François Couperin, however, the basse d'archet part surpasses the basse chiffrée in ornamentation, presumably because the latter was prepared first, so that the additional ornaments in the former were a result of afterthought and refinement.
The use of embellishment in orchestral playing was generally censured, although Muffat sanctioned ornaments other than diminutions in his introduction to book 1 of Florilegium (I1695). Lully's obsession with orchestral discipline and uniformity was mentioned by Sénecé (H1688, p.299), Montéclair (H1736, pp.86–7) and Le Cerf de la Viéville (H1705, p.227):[Lully's] instrumentalists did not take it upon themselves to ornament their parts. He would not have allowed them to do this any more than he allowed it with his singers. He did not think it was right when they imagined they knew more than he did and added graces to their parts. When this happened, he grew angry and quickly set them straight. It is a true story that more than once in his life he broke a violin across the back of a musician who was not playing it the way he wanted.As Bourdelot (H1715, pp.297–9) observed:
it is hard for a harpsichord, a viol and a theorbo (to say nothing of the string and wind instruments) all to hit upon just the same ornaments at the same time. One plays one figure, another plays a different one, and the result is such total cacophony that the composer no longer recognizes his own work, which seems completely deformed.
By mid-century, limited use of ornaments seems to have become accepted practice (Rousseau, H1768, p.200, under ‘Ensemble’).