Between the middle of the 18th century and the beginning of the 20th attitudes towards the role, function and usage of ornaments underwent a radical transformation. An aesthetic in which almost all music involved an element of free ornamentation gradually gave way to one in which, for the most part, composers expected ornaments to be introduced only where specifically marked. At the same time, the number of ornament signs in common use declined. Furthermore, 19th-century composers increasingly expected ornament signs to function as shorthand for precise figurations; they were not content, as many of their 18th-century predecessors were, to leave the realization to the performer.
(i) Appoggiaturas, anticipatory notes and grace notes.
(ii) Trills, turns and related ornaments.
Ornaments, §9: Late 18th century and the 19th
(i) Appoggiaturas, anticipatory notes and grace notes.
Until the early 19th century small notes extra to the value of the bar indicated several quite different things. The meaning of such notation is (and was) often difficult to determine. Late 18th-century and early 19th-century authorities drew attention to the scope for misunderstanding the intended execution of small notes a 2nd above or below the note they precede. These might indicate any of three things: notes taking a substantial portion of the one they precede (hereafter referred to as appoggiaturas); notes tied to the one they precede and executed very quickly on or just before the beat (the term ‘grace note’ is used here with no necessary suggestion of pre-beat performance); or notes tied to and taking time from the one they follow (anticipatory notes), which were common in the late 18th century and throughout the 19th century in the context of trill endings and certain types of portamento.
During the late 18th and early 19th centuries there were particular problems in distinguishing between the appoggiatura, which has an important harmonic function, and the grace note, which, because it is performed so rapidly that neither the preceding nor the following note appears to lose any significant value, has primarily an accentual or ornamental function. Theorists in the second half of the 18th century periodically suggested that small notes indicate the intended value of the appoggiatura. C.P.E. Bach observed in 1753 that ‘people have recently begun to indicate such appoggiaturas according to their true value’; among composers who began to do so during the second half of the century were Gluck (from the time of his Paris operas), Haydn (from about 1762) and Mozart. Many other composers were much more casual, especially Italians, who often did not trouble to indicate appoggiaturas at all in places where the singer or instrumentalist might have been expected to supply them. Confusion over this type of notation remained a serious problem for many at the end of the 18th century. In the fifth edition (1791) of Löhlein's popular Clavier-Schule, for instance, the editor, J.G. Witthauer, having urged composers to indicate the length of appoggiaturas, concluded: ‘How many pieces would then, at least with respect to the appoggiaturas, be less badly performed, and how much trouble would be spared to the beginner!’
Where it was unclear from the notation whether an appoggiatura or a grace note was implied, some theorists, notably Türk, attempted to assist the performer by providing examples of musical contexts indicating grace note treatment. If it was decided that an appoggiatura was intended, and that the given value was not a reliable indication of its intended length, the performer had to determine what value to give it. Many 18th-century writers advanced general guidelines. The assertion that an appoggiatura should normally take half a binary main note and two-thirds of a ternary main note, promulgated in the mid-18th century by, among others, Tartini, Quantz, Leopold Mozart and C.P.E. Bach, was widely repeated by 18th- and 19th-century theorists. Some musicians, including Francesco Galeazzi and Bernhard Romberg, taught that before a ternary note the appoggiatura should take only a third of the value of the main note; others such as Clementi allowed it to take either a third or two-thirds of a dotted note according to context. Many theorists, following Bach, Mozart and Quantz, felt that an appoggiatura before a tied note, or a note followed by a rest, should take the full value of the note before which it stood, though it was admitted that the resolution on to a rest might not always be permitted by the harmony. Indeed many theorists, having articulated their guidelines, cautioned that the length of appoggiaturas, which by their very nature required a rhythmically unconstrained delivery, might often be conditioned by the expression or by the exigencies of the harmony.
By the end of the 18th century Türk and other theorists were arguing that it would be better to incorporate all appoggiaturas into full-size notation, leaving small notes to indicate grace notes. Beethoven's practice illustrates this changing attitude; he very rarely used small notes to indicate appoggiaturas (except in vocal music), reserving them principally for grace notes. Others, however, resisted that approach on the grounds that the notation of appoggiaturas with small notes was the most appropriate way of eliciting the special manner of performance they required, through either accentuation, flexible length or ornamental resolution (or a combination of these). H.C. Koch (‘Vorschlag’, Musikalisches Lexikon, 1802) articulated this clearly when he remarked that the reason for notating appoggiaturas as ornaments
has its origin in the particular and exceptional manner in which the appoggiatura is performed. Namely … one should markedly bring out the appoggiatura itself by means of a particular accent, or sound it with a certain rapid swelling of the strength of the note: and then slur the following melodic main note to it softly or with decreased strength.
A variety of ornamental resolutions is indicated by musicians as different in time and background as Domenico Corri and Baillot (ex.98).
In the case of an appoggiatura on the major or minor 2nd below the main note, some singing tutors, including Corri and Lanza, considered that in contrast to the falling appoggiatura it should be delivered with increasing strength, so that the main note received the greater accent.
When, in late 18th- and early 19th-century scores, an appoggiatura on the 2nd above or below is found before a pair of notes with a strong–weak metrical stress which are of equal length and on the same pitch, it seems clear that the appoggiatura was meant to take the whole length of the note before which it stood. This practice can be found at least as early as the 1760s and as late as the 1820s, but apparently it was not discussed at the time by theorists. In a letter of 1768, however, Haydn specifically stated that in such cases the realization of ex.99a should be as in ex.99b, not as in ex.99c. This treatment can also be found in Corri's realization of J.C. Bach's ‘Nel partir bell’idol mio' in A Select Collection. It was clearly intended, too, in Schubert's operas, as indicated by comparison of the vocal part with the orchestral parts (ex.100). Interestingly, Schubert consistently gave the appoggiatura half the value of its intended realization. A similar usage is found in Weber (perhaps deriving from his lessons with Michael Haydn), for instance in Der Freischütz.
This notation, confined largely to German composers, raises the broader question of how such pairs of notes on the same pitch should be treated when they have no indication for an appoggiatura. Crutchfield (J1989) has argued persuasively that an appoggiatura of some kind is appropriate almost always in recitative and often in arias. The practice was so well known that Italian composers in particular rarely troubled to notate appoggiaturas in such circumstances, and if a composer wanted the music sung as notated he would have to specify it, as Verdi did in Rigoletto (no.13). The preservation of this tradition among 19th-century artists is demonstrated by early recordings. Charles Santley (1834–1922), for instance, added appoggiaturas, as well as other ornamentation, in both recitative and aria in his recording of Mozart's ‘Non più andrai’.
By the second quarter of the 19th century the use of small notes to indicate appoggiaturas of the above types was fast disappearing. Where single small notes were still employed they were intended to be performed very rapidly as grace notes on or just before the beat. A sign of changing practice in this respect is Philip Corri's treatment of the matter in his L'anima di musica (1810), where, reversing Türk's approach, he instructed readers to assume that small notes represented grace notes except in a limited number of circumstances, of which he gave examples. In later treatises discussion of appoggiaturas was largely intended as an aid to the performance of older music, which formed an increasingly large proportion of the contemporary repertory. By the middle of the 19th century the now customary notation of grace notes was widespread.
General rules for an appropriate manner of grace note performance in any given period are impossible to formulate. Practice varied from time to time, place to place and individual to individual. The matter is also complicated by wider questions of historical performing practice in respect of tempo rubato, rhythmic freedom in general and, particularly, the practice in keyboard music of playing the left hand before the right. For such reasons a simplistic rule of on or before the beat, grafted on to an otherwise ‘modern’ style of playing, is essentially meaningless. During the Classical period German authorities generally taught that in most if not all circumstances grace notes should be performed on the beat (i.e. against the bass note that pertained to the main note before which the grace note stood). Milchmeyer's Die wahre Art das Pianoforte zu spielen (1797) was among the few late 18th-century German sources to recommend a pre-beat conception of grace notes as the norm. Pre-beat performance, especially in the context of tierces coulées, was commonly associated with what Türk called the ‘French style or the so called Lombard Taste’. Leopold Mozart recognized the possibility that ex.101a could imply pre-beat performance, but considered that the composer would specify this more clearly by writing it out as in ex.101b. Löhlein's explanation of the similar figure in ex.102a, in his Anweisung zum Violinspielen, as indicating anticipatory notes (ex.102b) was ‘corrected’ to ex.102c in Reichardt's 1797 edition of the treatise. There was always a degree of ambiguity in such circumstances.
Throughout the 19th century German writers continued predominantly to instruct that grace notes should be performed on the beat, as did many theorists of other nationalities. Although there was no unanimity among late 18th-century and 19th-century musicians as to whether the grace note or the main note should receive the greater accent, the majority – with the notable exception of Hugo Riemann – seem to have favoured the latter conception. In particular instances there was always the possibility of disagreement. Edward Dannreuther, for example, considered that the small note in bar 3 of Schubert's A Moment musical op.94 ex.103a was ‘meant for a Nachschlag’, and illustrated it as in ex.103b; while Riemann in his annotated edition of the work indicated an accented performance on the beat (ex.103c). Among musicians on whom French influence was strongest, however, a pre-beat conception not only of grace notes but also of ornaments of two or more notes, which in the German tradition were still widely regarded as occurring on the beat, seems to have been the norm. In 1840 Fétis and Moscheles observed in the Méthode des méthodes pour piano:
Acciaccaturas, slides and groups of two or three notes are placed immediately before the principal note. In the old school it was understood that they should share in the time of the principal note, but they are now to be played quickly and lightly before the time of the large note.
It was not, though, merely a question of nationalities. The German violinist and pedagogue Andreas Moser, writing at the beginning of the 20th century, supported a pre-beat conception in many cases and considered that preference for a pre-beat or on-beat conception of grace notes was largely determined by the nature of different instruments. He observed that even at that time ‘there were the most contradictory opinions among practical musicians’ (Violinschule, iii, 28), noting that keyboard players still tended to favour placing grace notes firmly on the beat while the majority of singers and string players anticipated them, and he suggested that this had been the case continuously since the mid-18th century. Although this view was probably shared by Joseph Joachim, whose direct experience went back to the 1830s, documentary evidence suggests that, in theory at least, the French–German split was as strong among violinists as among keyboard players in the mid-19th century: Spohr explicitly required on-beat performance, while Baillot envisaged the performance of grace notes before the beat.
Nevertheless, in practice this theoretical distinction may have mattered little if the grace note was performed quickly and lightly, as the vast majority of writers said it should be. Where concrete evidence for the performance of grace notes exists, such as barrel organs or, at the end of the 19th century, recordings and piano rolls, it is often difficult to determine whether in particular instances a grace note occurs on or before the beat.