The Classical Concerto First-Movement Cadenza: Origins, Growth, Facilitating Factors and the Eventual Decline^*

“Just as unity is required for a well-ordered whole, so also is variety necessary if the attention of the listener is to be held. Therefore as much of the unexpected and the surprising as can possibly be added should be used in the cadenzas.”

Daniel Gottlob Türk, 1789

Introduction

[0.1] Descriptions of first-movement form in late-eighteenth-century concertos, from Heinrich Christoph Koch ([1793] 1983, 211) and Augustus Frederic Kollmann (1799, 28) through Carl Czerny ([1830] 1848, 159–60) and Ebenezer Prout (1895, 207) to William Caplin (1998, 243) and Mark Evan Bonds (2017, 217), all single out the cadenza as a prominent structural element. Considered by many to be a hallmark of first-movement form in Classical concertos, the cadenza is discussed here with a focus on its origins and evolution in eighteenth-century concerto first movements, suggesting possible reasons for its rise and rapid decline. In this article I offer a fresh perspective and new empirically based observations on the cadenza as a musical phenomenon and its role in shaping eighteenth-century concerto first-movement form in general, grounded in an extensive analysis of a large corpus of works from the period. Here I adopt Philip Whitmore’s (1991, ix) approach to the matter, “the so-called biographical model, by which the history of a phenomenon is described in terms of its antecedents, its birth, growth, prime, decline, and death.” Today, with growing awareness of the advantages a diachronic approach offers, Whitmore’s approach almost goes without saying.⁽¹⁾

[0.2] In this article, after determining what a cadenza is, I describe some of its musical characteristics and the conventions that emerged around the 1730s and 1740s and were consolidated by the 1750s. I demonstrate that the conventional cadenza remained a consistent feature in concerto first movements in the mid-to-late eighteenth century, from both musical and aesthetic standpoints. Building on this understanding of convention, I argue that the cadenza is better viewed not as a homogenous musical phenomenon but as a wide spectrum of possibilities. Next, with the consistent aesthetic role of the cadenza in mind, I discuss how the expansion of concerto movements and the consolidation of sonata-form procedures in these movements contributed to the growth of the cadenza and its subsequent changed role, and how the growth of the cadenza reciprocally stimulated structural changes in the movement overall. Lastly, I discuss the rapid disappearance of the Classical cadenza as a result of what composers regarded as performers’ abuse, with a final note on the Romantic cadenza in this regard.

[0.3] Before going any further, two preliminary facts must be noted. First, it is important to understand that cadenzas also appear regularly in the final movements of mid-to-late-eighteenth-century works, and in many second movements in mid-eighteenth-century concertos. There is good reason to explore the history of the cadenza as an independent phenomenon, across all movements, and even across genres beyond the concerto. Yet here, I limit the discussion to concerto first movements alone. Although several points may extend to cadenzas in other movements, issues like the relationship between the soloist’s playing space in the entire movement vs. that enabled by the cadenza, require limiting myself to first movements only. Second, any discussion involving the cadenza requires a measure of caution. With the origins of the cadenza in extemporized performance, there is relatively little notated material at hand. The cadenza evolved in the eighteenth century not through composition but through performance practice. Eighteenth-century writers who describe the cadenza, such as Quantz ([1752] 1985) and C. P. E. Bach ([1753, 1762] 1949), do so from the standpoint of the performer, not the composer. There is only a small sample of notated cadenzas written in the eighteenth century from which to draw conclusions. Thus, any inference about the cadenza in the eighteenth century must be made with some reservation. Nonetheless, I believe that there is ample material to provide significant insights into the evolution of the cadenza in the eighteenth century within concerto first movements.

[0.4] The term cadenza already appears in relation to the concerto in some of Vivaldi’s early works.⁽²⁾ However, as compelling as it may be to draw a direct line from Vivaldi to Mozart with respect to the cadenza, the familiar Classical cadenza originated in conventions that emerged around the 1730s and 1740s, slightly later and separately from Vivaldi’s practice. The unaccompanied solo passages that Vivaldi labeled “cadenza” were not written in the main body of the score, but as postscripts to the principal part that performers could choose to perform as an additional solo episode after the closing ritornello, thus extending the movement. In such cases the movement was to conclude with an additional closing ritornello that Vivaldi also supplied. Philip Whitmore notes that “Vivaldi was writing before [the cadenza] tradition had established itself, and before [the] term had acquired the definite meaning it was to hold for [others]” (1988, 47–50). It appears that by the 1730s there was a clear distinction between the cadenza and other closely related musical concepts such as the capriccio. Pietro Locatelli (1695–1764), for instance, ends the capriccios he incorporated at the end of the last solos in the outer movements of the concertos in his 1733 L’Arte del violino, op. 3, with a cadenza for the performer to extemporize, as seen in Example 1. Thus, he clearly treats the cadenza as a separate musical entity.⁽³⁾ A rare written-out example in Locatelli’s Violin Sonata, op. 6, no. 12, presented in Example 2, conveys a good sense of how he understood cadenzas. Giuseppe Tartini (1692–1770) provides further evidence in his writings and teachings that around the mid-eighteenth century the cadenza was viewed as a separate musical convention. In an essay published in French as Traité des agrémens de la musique in 1770, but probably written around 1750, Tartini discusses cadenzas as a distinct musical entity ([1770] 1956).⁽⁴⁾ As shown in Example 3, Tartini’s cadenzas are short, maintain the same character throughout, and usually have no clear meter. Erwin R. Jacobi and Willis Wager (1961, 219) assert that “Tartini is in no way referring to an ambiguity of the word ‘cadenza’. . . which was only a short improvisation.”⁽⁵⁾

1. Quantz’s “Blueprint”

[1.1] In his influential Versuch einer Anweisung die Flöte traversiere zu spielen ([1752] 1985), Johann Joachim Quantz provides the most detailed eighteenth-century description of the cadenza.⁽⁶⁾ The many similarities between Tartini’s and Quantz’s conceptions of the cadenza suggest that by the mid-eighteenth century, a largely conventional notion of the cadenza had already been consolidated. Once established, it served as the foundation for further development and the emergence of the Classical cadenza. For its clarity and comprehensiveness, this description by Quantz, which fits in well with Tartini’s ideas, can be viewed as a reflection of the common practice in the mid-eighteenth century and the blueprint for cadenzas in the second half of the eighteenth century.⁽⁷⁾

[1.2] Quantz ([1752] 1985, 179) defines the cadenza as “that extempore embellishment created, according to the fancy and pleasure of the performer, by a concertante part at the close of a piece on the penultimate note of the bass, that is, the fifth of the key of the piece.” According to Quantz, cadenzas “must sound as if they have been improvised spontaneously at the moment of playing” and their objective must be “simply to surprise the listener unexpectedly once more at the end of the piece, and to leave behind a special impression” (181). Grasping this aesthetic goal is key to understanding the cadenza’s musical traits as described by Quantz. It also sheds light on the growth and further development of the cadenza in response to the growing procedures associated with sonata form in concerto movements, as I discuss later.

[1.3] To achieve spontaneous-sounding cadenzas to surprise the listener, Quantz makes suggestions about how to treat various musical aspects such as meter, melodic content, character, key, and length in cadenzas. In terms of meter, Quantz writes that “regular meter is seldom observed, and indeed should not be observed, in cadenzas” (185). For Leopold Mozart, this seems to be a key feature of cadenzas; he notes in his Versuch einer gründlichen Violinschule ([1756] 1951) that in a cadenza “one may never bind oneself to strict time,” and repeats elsewhere that in a cadenza “one is not tied to strict time” (188–91).⁽⁸⁾ In terms of melodic content, the cadenza, for Quantz, “should consist of detached ideas rather than a sustained melody,” and it is preferable if melodic figures are “interrupted with fresh figures,” but he warns readers “not [to] introduce too many ideas” ([1752] 1985, 182). He does not explicitly require a clear thematic relationship between the cadenza and the parent movement, but he does suggest that cadenzas should “include a short repetition or imitation of the most pleasing phrases contained in [the piece],” especially since it is not always “immediately possible to invent something new.” It is therefore advisable “to choose one of the most pleasing of the preceding phrases and fashion the cadenza from it” (181–2).⁽⁹⁾ For Quantz, using previously heard ideas to derive the cadenza does not only “make up for any lack of inventiveness”; it also ensures that the cadenza will “confirm the prevailing passion of the piece” (181). This is the only requirement on which Quantz insists: that the character of the cadenza suit that of the movement. He emphasizes this repeatedly, asserting that “cadenzas must stem from the principal sentiment of the piece,” and that the melodic ideas must “conform to the preceding expression of the passions” (181–2). This assertion, shared by Tartini, is echoed in C. P. E. Bach’s Versuch über die wahre Art das Clavier zu spielen ([1753, 1762] 1949), published a year later. There, Bach states that “such elaborations [of fermatas, including cadenzas,] must be related to the affect of a movement, [and] they can be successfully employed only when close attention is paid to a composition’s expressive aim” (144).

[1.4] Harmonically speaking, Quantz instructs readers not to “roam into keys that are too remote, or touch upon keys which have no relationship with the principal one,” and asserts that “a short cadenza must not modulate out of its key at all” ([1752] 1985, 184).⁽¹⁰⁾ In terms of length, as this last comment suggests, Quantz allows a certain range. For example, he instructs singers and wind instrumentalists to create cadenzas that they can perform in a single breath, whereas string players can make them as long as they like (185); elsewhere he explains how to lengthen or shorten the embellishment. Either way, however, Quantz is consistent in his assertion that cadenzas “must be short and fresh,” and that “reasonable brevity. . . is more advantageous than vexing length” (185). Thus, Quantz seems to allow a certain range within the general sense of a short cadenza.

[1.5] In summary, Quantz’s “blueprint” for a cadenza requires first and foremost that the cadenza should maintain and suit the character of the movement, and secondly that it should give the impression of extemporizing by the following means:

1. Brevity
2. Usually, a lack of definite meter
3. Residing in the principal key throughout, possibly only touching upon closely related keys
4. The use of disjunct ideas, but not too many of them

The examples he provides in his Versuch, three of which are presented in Example 4, reflect this stance well.⁽¹¹⁾

[1.6] Examples 5 and 6 present two examples of cadenzas written for concerto first movements: one from the 1740s for a violin concerto by Joseph Ferdinand Timmer (1708–1771) and one from the 1780s by Johann Matthias Sperger (1750–1812) for a double bass concerto by Anton Zimmerman (1741–1781). Quantz’s description accounts for both, as they are in accordance with the guidelines summarized above.⁽¹²⁾ Both are short and lack a definite meter, and we find no material of thematic quality.⁽¹³⁾ The cadenzas also use distinct ideas and repeating figures, either literal (Sperger) or sequential (Timmer), and remain in the tonic key throughout, with the latter avoiding chromatic notes altogether. For example, in the cadenza to Zimmerman’s concerto we see the use of a common 16th-note figure and a repetition followed by a normal scalar motion. In Timmer’s cadenza we find a unit of a 16th note and three 32nd notes repeated four times in a sequential motion, as Quantz advises. This is not to say, necessarily, that composers were unanimous in their ideas about the cadenza, or that they always tried to achieve the same thing, but rather to say that Quantz’s guidelines remain the best we have to explain what their cadenzas have in common. These are largely the same guidelines that Daniel Gottlob Türk, who was not only influenced by Quantz but also referred readers to him explicitly for further information, echoed in his own treatise in 1789, nearly four decades later ([1789] 1982, 297–309).⁽¹⁴⁾ Arguably, then, the general conception of the cadenza remained stable throughout the second half of the eighteenth century, and Quantz’s account therefore remains the best theoretical guide for cadenzas from that period—and even for later cadenzas discussed below (despite some changes).

[1.7] The appropriateness of Quantz’s description for later eighteenth-century cadenza practices is evident in a collection of 64 cadenzas by Luigi Borghi (1745–1806), probably written during the 1780s and published as the composer’s op. 11. This set, titled “Cadences or Solos in All the Major and Minor Keys, Composed for the Improvement & Practice of Amateurs to Whom They Are Dedicated,” was composed primarily for instructional purposes. It is thus likely that its approach is rather conservative, reflecting what was regarded as the “textbook cadenza,” at least in theory. Indeed, as can be seen in the excerpts shown in Example 7, Borghi’s cadenzas follow the Quantzean idiom in terms of their musical attributes, similarly to the other concerto cadenzas from the 1770s and 1780s presented earlier. One noteworthy aspect of Borghi’s collection is the cadenzas’ lengths. Borghi’s cadenzas do not exhibit uniformity in terms of length, but rather a certain range; still, almost all of them are relatively short. Together with the previous examples, this suggests that short cadenzas were still very much prevalent in the late eighteenth century.

2. Reconceptualizing the Classical Cadenza as a Spectrum

[2.1] Another instructive collection of cadenzas from the same time by Tommaso Giordani (1730–1806), published as his op. 33, includes eight cadenzas.⁽¹⁵⁾ The cadenzas in Giordani’s collection, like those previously discussed, also conform to Quantz’s guidelines. As Example 8 shows, Giordani’s cadenzas are written without a definite meter, mainly reside in the tonic key (quickly regressing back to it if drifting to a closely related tonality), and consist of successions of detached ideas. However, the cadenzas in this collection are longer than Borghi’s. One salient feature across all eight of Giordani’s cadenzas is an explicit instruction to change the tempo, making it possible to sustain the impression of extemporaneous play for longer. Similar directions for tempo changes are evident in the cadenza to the first movement of the Keyboard Concerto in C Major by Giuseppe Buccioni (1759–1830), published in 1783 (Example 9). This cadenza exceeds even the longest of Giordani’s cadenzas and the majority of Mozart’s, providing more evidence that cadenzas gradually increased in length through the late eighteenth century and suggesting that Giordani’s cadenzas reflected at least to some extent the current practice. This example, as well as the well-known tempo alterations in Mozart’s cadenzas, suggest that Türk’s “rule” that “the same tempo and meter should not be maintained throughout the cadenza” was followed commonly enough in cadenzas during the late eighteenth century ([1789] 1982, 301). Since short cadenzas lack sufficient space for tempo changes, this feature seems to reflect the prevalence of longer cadenzas.

[2.2] Further evidence pointing to the prevalence of long cadenzas in late-eighteenth-century practice lies in contemporary critical discourse. Composers and theorists often criticized performers for cadenzas of excessive length. C. F. D. Schubart claims in 1784 that “a long cadenza is a state within a state. . . and is always detrimental to the impression made by the movement as a whole” (Kolneder 1986, 339–40). In 1789, Türk criticizes “the executant [who] goes to excess” in embellished cadenzas and asserts that “cadenzas should not be too long” ([1789] 1982, 301). In 1791, Johann George Tromlitz—writing about performing on wind instruments—is even more blunt, stating that with respect to cadenzas, “the freedom [given] is abused. . . [they are] often made as long as the entire movement. . . so far away from the subject that when the ritornello returns one has to recollect whether or not it belongs to this movement. . . when such a monstrosity has been heard, there still comes a contemptible goat’s trill at the end, and then follows bravo! bravissimo!” ([1791] 1991, 260).⁽¹⁶⁾

[2.3] In summary, we find that cadenzas in the late eighteenth century varied greatly in length. As a result, their content and function differed significantly. Whereas some cadenzas remained mere embellishments of the penultimate dominant harmony at the cadence, others constituted independent segments, becoming, in Koch’s words, a “free fantasy or a capriccio,” thus losing sight of their traditional cadence-embellishing role (Mirka 2005, 297).⁽¹⁷⁾

[2.4] It may be tempting to distinguish between two types of cadenzas in the late eighteenth century: the short cadence-embellishing cadenza and the longer fantasy-like cadenza. Yet there are possible pitfalls in such a clear-cut distinction. First, where do we draw the line? Examining side-by-side the two cadenzas in Example 10, one by Borghi and one by Giordani, makes it simple to classify them as one or the other. In other cases, however, it is more difficult. Consider the two cadenzas in Example 11. They are taken from the very same collections as the cadenzas in the previous example, but there is little if anything to differentiate them. Any decision based on length, especially with respect to in-between cases, is only arbitrary, just like a dichotomous division of people to either short or tall. To complicate matters even more, there are too many variables to consider, such as the parent movements’ tempo, meter, or even style (we know that Bach’s and Mozart’s allegros are not necessarily the same). I already suggested that longer cadenzas are more likely to include tempo alterations, yet this, too, cannot be considered a differentiating criterion. Josef Antonín Štěpán (1726–1797), for instance, published in 1783 a collection that includes 21 fermas, which most likely denote cadenzas. These fermas, as illustrated in Example 12, are mostly brief, yet contain several indications for tempo alterations.

[2.5] Similarly, we find tempo alterations in a cadenza to the first movement of one of Antonio Rosetti’s (ca. 1750–1792) oboe concertos composed ca. 1780, presented in Example 13. Separating the fantasy-like cadenzas from the traditional cadence-embellishment ones also disregards the close linkage between them. They share the same musical language and attributes, and they can all be accounted for using Quantz’s (or Türk’s) guidelines. Again, this does not mean that everyone shared these ideas about the cadenza, but Quantz’s guidelines are still the best we have to explain them. Even Quantz’s plea for short cadenzas is not necessarily undermined by the longer ones found in the 1770s and 1780s. Concerto movements in these two decades were substantially longer than those in the late 1740s and early 1750s when Quantz was writing his treatise. Terms like “short” or “not too long,” used by Quantz and Türk, respectively, to describe good cadenzas, are relative; the length of cadenzas was presumably perceived not by absolute length but relative to the movement as a whole. Thus, whether shorter or longer, cadenzas in eighteenth-century concertos can be viewed as manifestations of the same musical phenomenon. Even with respect to Mozart’s cadenzas, Joseph P. Swain argues that

the first observation we can make about Mozart’s cadenzas is that they follow the guidelines of our theorists [Quantz, C. P. E. Bach, Türk] quite strictly. That they always retain the spirit of the parent work. . . . They never seem too long, and indeed, by today’s standards they are quite short. They are full of thematic references but these are never organized into a continuous fantasy, but rather move from one to another quite abruptly. . . . Most of all, Mozart strictly observes the prohibition against distant modulation. . . [and the cadenza] remains firmly rooted in the tonic key. (Swain 1988, 35)

[2.6] Still, the disparities in length and function between certain cadenzas at the extremes cannot be overlooked. Therefore, I suggest conceptualizing the Classical cadenza in terms of a spectrum of increasing length, with short, cadence-embellishing cadenzas at one end and the quasi-independent, elaborate cadenzas at the other, as plotted in Example 14. The idea of a spectrum expresses the wide range of possibilities, but also the stylistic and syntactic connections between the more conservative shorter cadenzas and the fantasia-like elongated ones, as they share the same musical attributes. Moreover, the idea of a “Classical cadenza spectrum” depicts short embellishments as a viable, non-obsolete option, as they were still very common in late-eighteenth-century practice.⁽¹⁸⁾ Tromlitz, for example, points out that longer cadenzas are more frequent “on stringed and keyboard instruments,” and still dictates, like Quantz forty years before, that “the wind player[s]. . . are limited by the length of their breath,” and allows in some cases two or three breaths, whereas “a string player can make them as long as he likes, if he is rich enough in inventiveness” ([1791] 1991, 260). On this account, perhaps lengthy cadenzas can be seen as one feature that differentiates keyboard and string concertos from concertos for other instruments around the turn of the century. The labels on the spectrum—“short embellishment” and “long, fantasia-like cadenza”—reflect the conspicuous differences between very short and very elaborate cadenzas, thus congruent with Danuta Mirka’s notion that “the growth in size, made possible by the richer keyboard texture [in the case of keyboard concertos], brought about a qualitative leap in the evolution of the cadenza, in effect turning into a relatively autonomous fantasia or capriccio performed near the end of a concerto movement” (2005, 297).

3. Longer Cadenzas: Facilitating Factors and Formal Influences

[3.1] In the late eighteenth century, there were three main factors promoting the growth of cadenzas: the growth of parent concerto movements, the shift of weight from the last solo section to the first, and the growing standardization of sonata-form procedures in concertos. As to how cadenzas grew in size as a response to the longer concerto movements, the story seems straightforward: they grew in proportion to the size of the overall movement. This claim is supported by the musical attributes shared by cadenzas of supposedly different types and lengths. Longer cadenzas began to appear with some regularity in the 1770s. The best indication of this is in C. P. E. Bach’s cadenzas. In his Wq 120 compilation, which dates from no earlier than 1778, Bach left us 80 cadenzas for his keyboard concertos and sonatas—sometimes more than one cadenza for a single concerto movement, such as the five cadenzas designated for the first movement of his Keyboard Concerto in G major, Wq 34. Of all his concerto first-movement cadenzas, the only long, elaborate one is for his last solo concerto, Keyboard Concerto in D major, Wq 45, written in 1778. This cadenza and representative examples of the other cadenzas by the composer that fall into the category of “short embellishments,” are presented in Examples 15 and 16.

[3.2] As I will demonstrate shortly, performers felt more compelled to prolong their cadenzas in response to general changes and processes in concerto movements during the 1770s and 1780s. Yoel Greenberg and I have found that from the 1720s through the 1760s, the last solo was consistently the longest section of ritornello movements (Maliniak and Greenberg 2022, 241). This pattern changed during the 1770s, when a slight shift in weight to the first solo section is evident. There are a few possible explanations for the connection between the slightly shortened last solo and the prolonged cadenza. Perhaps performers who were accustomed to longer solo passages to set up the closing ritornello made up for the reduced space by playing longer cadenzas. Alternatively, composers may have deliberately shortened the last solo to allow soloists more space for elaboration, or this may have been a response to performers taking more liberty with the cadenza. Either way, the longer cadenza seems to counterbalance the allocation of larger portions to the first solo at the expense of the last.

[3.3] The reduction in length of the last solo may have resulted in its assumption of a recapitulatory function, which may in turn have triggered the adoption of longer cadenzas. In other words, the growing tendency toward sonata-form procedures in concertos perhaps facilitated freer, fantasia-like cadenzas. During the 1750s and 1760s, the last solos consisted of long passages of fresh material; thus, long cadenzas were not necessary to leave an impression on the audience, as the shorter traditional embellishments satisfied this aesthetic need. However, as concerto movements grew substantially in size and last solos became more predictable, longer cadenzas became more appropriate, and in some cases, they may have been needed to attract audiences. In the 1770s, with the solo sections becoming the carriers of formal logic in concerto movements and gradually taking on the sonata-form logic of exposition, development, and recapitulation, the treatment of the last solos became stricter. The phrase-oriented melodies associated with sonata form precluded the abundance of passagework that might have been found in mid-century; hence, a designated area for passagework was required. The more rigid approach to the construction of last solos in the 1770s may have influenced the extension of cadenzas around that time, considering the aesthetic goal of the cadenza as asserted by Quantz and echoed by Türk and Tromlitz more than thirty years later: to surprise the listener “once more at the end of the piece” ([1752] 1985, 180). Tromlitz requires that “cadenzas should be short, and when possible, also surprising; even if they are practiced they should be performed as though they were only now appearing for the first time. Diversity and alternation of the figures is very important, so that they do not sound too much in the meter, for this would be against the usual character of a cadenza” ([1791] 1991, 262). By Tromlitz’s account, audiences expected to hear something special at the cadenza and often judged the entire performance accordingly; without it the “whole performance will be worthless.”⁽¹⁹⁾ The aesthetic aim of the cadenza and its importance to the entire performance can best be summarized by Türk’s quote at the beginning of this article: “Just as unity is required for a well-ordered whole, so also is variety necessary if the attention of the listener is to be held. Therefore, as much of the unexpected and the surprising as can possibly be added should be used in the cadenzas” ([1789] 1982, 300). Whereas the structural logic of sonata form is responsible for the unity, the cadenza serves as the element of surprise. Thus, it is quite plausible that the growing tendency toward sonata form in late-eighteenth-century concertos contributed to the growth of the cadenza.

[3.4] Just as the changes in the parent concerto movements generated prolonged cadenzas, prolonged cadenzas reciprocally elicited structural adjustments in late-eighteenth-century concertos. Due to its extemporaneous and non-binding nature, the cadenza is not embedded in the score in the vast majority of eighteenth-century concertos. Therefore, it is usually treated as an extrastructural element, as it probably should be. Though the cadenza itself lies outside the concerto clock, it seems to have influenced “proper” formal elements surrounding it. Here, I specifically focus on the modifications that the cadenza evoked in the passage leading up to it.

[3.5] Traditionally, the cadenza served to embellish the final cadence of the last solo. It is therefore not surprising that in most concertos with a cadenza from the 1740s and 1750s, the fermata indicating a cadenza is located at the end of the last solo passage, leading to it directly without any orchestral interruption.⁽²⁰⁾ In a minority of cases, the orchestra lets the soloist “take a breath” before the cadenza, giving the performer a brief moment, of only a measure or two, to come up with an idea. If the cadenza were just the end of the long solo passage that preceded it, a long cadenza would be unnecessary, especially around that time, when most last solos consisted mainly of new, loose material. In the 1760s, when cadenzas became far more common in concertos, they were placed adjacent to the preceding solo passage in 40% of the concertos examined from that period, and 38% followed a one- or two-measure tutti interjection.⁽²¹⁾ Again, a short embellishment would be the logical choice, avoiding a solo overload. This trend of approaching the cadenza either directly or via a brief one- or two-measure orchestral interruption remained customary in the 1770s, appearing, as summarized in Example 17, in 42% of examined works exhibiting a cadenza space.⁽²²⁾ Taken together with the still-prevalent short cadenzas, the two procedures—leading directly to a cadenza and playing a short one—seem to go hand in hand. With the increased length of the last solo section and the growing tendency to employ longer cadenzas, it is understandable that tutti interjections that allow the soloist to “take a breath” before the cadenza also slightly grew in length. Indeed, tutti interjections of three to four measures in length were found in 34% of 1770s cases. In the 1780s, when longer orchestral preparations become more customary and increase in length, we still find that 25% of concerto movements featuring a cadenza follow the traditional process of placing it at the end of the solo, complementing the still-widespread short cadenza.⁽²³⁾

[3.6] Nevertheless, we must bear in mind that in the 1770s, and even more so in the 1780s, the orchestral preparation for the cadenza sometimes emphasized a different function, that of an independent musical event. In 13% of cases examined from the 1770s, we find more substantial orchestral passages of seven to nine measures preparing the cadenza. In the 1780s, 28% consist of a long orchestral preparation between seven and twelve measures. These passages lead to long and elaborate cadenzas and highlight these cadenzas as structural modules, justifying Daniel N. Leeson and Robert D. Levin’s (1976, 96) recognition of such segments in Mozart’s concertos as short but separate sections labeled “ritornello to cadenza.”⁽²⁴⁾ These passages necessarily suggest longer cadenzas, which require a longer buildup. A brief cadenza of the short, cadence-embellishing type would be only too awkward there. Still, the clear majority of orchestral cadenza preparations from the 1780s are between three and six measures long, offering a range of options for the performer, including the brief cadenza. In the context of the much longer last solos of the 1780s (compared to their 1760s counterparts), three or four measures may be perceived as the soloist taking a brief “breath,” thus justifying a short cadence embellishment, but they can also set up much longer cadenzas, as in Buccioni’s concerto (previously shown in Example 9), where the lengthy cadenza follows only three bars of orchestral preparation. An opposite example can be found in a handwritten short cadenza to the first movement of Leopold Kozeluch’s (1747–1818) Clarinet Concerto no. 2 in E-flat Major (ca. 1780s), shown in Example 18, that follows a considerable six-measure orchestral preparation. Written cadenzas recovered from various sources to Johann Samuel Schröter’s (1753–1788) op. 3 concertos, including a few by Mozart, exemplify the loose relationship between the length of a cadenza and that of its orchestral preparation. Schröter sets up the cadenza in the first movement of both the third and fourth concertos in the collection with four orchestral bars. As Example 19 shows, whereas Mozart found it appropriate to prepare relatively lengthier (metered) cadenzas, other unknown performers settled for less. These examples also attest to the prevalence of shorter “traditional” cadenzas alongside the longer “relatively autonomous” ones, though these cadenzas by Mozart are far shorter compared to some of his later more elaborate ones. The bottom line seems to be that, generally speaking, the growing tendency toward more elaborate cadenzas forced composers to employ a short orchestral statement to end the last solo and to allow performers, often the composers themselves, a lot of flexibility.

[3.7] From a theoretical-analytical standpoint, these findings call for a rethinking of Caplin’s description of Classical concerto form with respect to the cadenza. According to Caplin, the sixth and final principal section of the concerto is “a closing ritornello for orchestra (usually interrupted by a solo cadenza) that completes the structural frame” (1998, 243).⁽²⁵⁾ I propose an amendment. In the majority of cases from the 1780s we find no substantial orchestral passage that would indicate that the cadenza is an “interruption” of the final ritornello. Furthermore, even in instances where the orchestral preparation is substantial—and particularly when it is brief—omitting the cadenza usually reveals that the preceding orchestral material does not convincingly function as the onset of the closing ritornello. Omitting this passage and initiating the closing ritornello with the supposed closing note of the cadenza would make much more musical sense. This would fit Koch’s description, according to which “with the caesura tone of this so-called cadenza, which always ends with a formal cadence, the last ritornello begins” ([1793] 1983, 213). Since in most cases the cadenza, whether a mere embellishment or an elaborate fantasia-like one, is performed in a cadential context, it should be viewed as the close of the preceding solo, not an interruption of the final ritornello. The ambiguous cases are those in which there is a long orchestral preparation, but these are not the majority, and moreover they set up the cadenza as a separate event that leads to the closing ritornello rather than interrupting it. Thus, I suggest viewing the cadenza as part of the last solo rather than part of the closing ritornello when there is a short orchestral preparation, and even more so when there is none. In cases of a long orchestral preparation, I lean toward Leeson and Levin’s notion of a “ritornello to cadenza,” (1976, 96) but I would suggest avoiding the term “ritornello” due to its structural context. Instead, when there is a long orchestral preparation, which usually sets up an elaborate cadenza, I propose simply calling this entire part what it is: an “orchestral preparation and cadenza.”

[3.8] Paying attention to the preparation for the cadenza and acknowledging that both short and long cadenzas were common in the 1770s and 1780s can have immediate implications for present-day practice. In cases where the solo section proceeds directly to the adjacent cadenza, it is advisable to consider using a shorter cadential embellishment. Also, when playing concertos from the 1760s and 1770s, performers should be aware that brief cadenzas are perhaps more idiomatic and better suited. In contrast, long orchestral passages setting up the cadenza call for a more elaborate display.

[3.9] In summary, the tendency by some composers and performers toward longer cadenzas sparked, at least to some extent, the growth of the orchestral passage leading up to them, directly influencing the structural plan as a whole. This may have occurred through a back-and-forth process: to allow for slightly longer cadenzas or to separate them from the essential composed part, composers slightly prolonged the inserted orchestral preparation for the cadenza. In turn, such additions allowed for even longer cadenzas, which required prolonging their orchestral introductions further, and this again led to more expansion of the cadenza. Eventually, at a certain point, some cadenzas were completely detached from their cadence-embellishing role. In conclusion, in a more general sense, the longer cadenza grew as a result of the musical terrain of the 1760s and 1770s and reciprocally elicited some changes in the arrangement of concerto movements later to accommodate them.

4. The Decline of the Classical Cadenza and the Dawn of the Romantic Cadenza

[4.1] In the 1790s and the first few years of the 1800s, the cadenza was on the decline. Whereas 77% of concerto movements I examined from the 1770s include a space for a cadenza, the same is true for only 53% in the 1790s and 38% in concertos examined from the years 1800–1805.⁽²⁶⁾ The cadenza’s fall from favor is well reflected in a comparison of Jan Ladislav Dussek’s (1760–1812) concerto output from the 1780s and early 1790s with that from the latter part of the 1790s; whereas he employed cadenzas regularly in the 1780s and early 1790s, in the latter part of the 1790s he no longer allocated space for a cadenza. Another interesting example comes from the four oboe concertos composed in 1794 by the English composer and oboist William Ling (1775–1831). The earliest of the four includes a cadenza set up by six orchestral measures. However, the next three concertos from later that year have no cadenza. In one of them, as shown in Example 20, the composer was thinking of a cadenza and even jotted some of it down, only to cross it out (along with the lead-up to it) and resume the solo.

[4.2] I believe that three factors contributed to the rapid decline of the Classical cadenza. The first, already noted, was composers’ mistrust of performers’ handling of the cadenza space. Second, from the performers’ perspective, the cadenza was a possible source of criticism. Being extempore or extempore-like, the cadenza may sometimes fail; although the performance of the written music may not go as planned either, playing a cadenza (especially a long one) adds to the risk. Thus, the cadenza can sometimes expose performers, often the composers themselves, to unwanted and perhaps unnecessary criticism. At times the cadenza may in fact overshadow the piece and become the center of attention. As Türk expressed it, “It is not seldom that a concerto or the like seems to be played merely because of its cadenzas.. . . The good impression left on the listener by the composition. . . [can] for the most part. . . [be] cadenzaed out of him” ([1789] 1982, 298, italics original). Or, in the words of Kollmann, “Nothing can more torment a musical ear, or more spoil the effect of a concerto, or more expose an author or player, than a bad fancy cadence [cadenza]” (1799, 23). When the cadenza makes a good impression there is much to gain, but if it goes wrong, there is even more to lose. Leopold Mozart, for example, once reported to his son on the performance of one of his violin concertos, informing him that the violinist, though playing “excellently” overall, missed some notes occasionally, and “nearly came to grief in a cadenza” (Anderson 1997, 297). On a different occasion, Leopold wrote to his son about a performer who “did not play at all badly, [although] his cadenzas were detestably Pinzgerish” (that is, in the manner of a violinist named Pinzger, of whom the Mozarts were far from fans) (545).⁽²⁷⁾ If these letters are any example of musicians’ discourse among themselves, much of the performance is judged by the cadenzas.

[4.3] The third factor possibly contributing to the decline of the Classical cadenza after Mozart is the longer, more brilliant writing for the solo, which may have made the cadenza excessive or even redundant at times. Whitmore sees this as the primary cause for the decay of cadenzas in piano concertos. He states that

one factor which led to the decline of the cadenza was the rise of the so-called virtuoso concerto, a type used increasingly by travelling virtuosi such as Steibelt, Field, and Hummel. Conceived primarily as display pieces, these concertos usually had somewhat perfunctory tuttis, the focus of attention being the spectacular virtuosity of the solo sections. (1991, 59)

This is not a new idea. In 1830, after cadenzas had mostly disappeared, Czerny wrote:

In the more ancient Concertos (those of Mozart, Beethoven &c) it was here [in the final part of the movement] customary to add an energetic tutti, which, after a few bars, made a pause on the dominant seventh, and the player was left to extemporize a grand cadence [cadenza], after which the orchestra performed a short conclusion. But in modern Concertos, which are written as brilliant as possible, we omit this cadence, and end immediately with, or after, the concluding passages of the pianist, when the orchestra performs a few powerful chords. ([1830] 1848, 159–60)⁽²⁸⁾

This correlation between more brilliant writing for the solo and the omission of the cadenza is already evident in some virtuoso concertos from the 1770s, when composers were negotiating the need for a cadenza at the tail of a long, brilliant solo. Joseph Bologne, Chevalier de Saint-Georges (1745–1799), for example, found no need for a cadenza in his concertos after very elaborate virtuosic last solos.⁽²⁹⁾ The same can hardly be said about Tommaso Giordani, who, as we already saw, valued and was very much occupied with cadenzas. Nevertheless, in his op. 14, he decided against a cadenza in two out of the six piano concertos, all including elaborate, demanding, at times even wild solo passages. Fast-forward to the 1790s, with solos increasing in technical display and brilliant play (taking Johann Baptist Cramer’s concertos as an example), and there was no need to impress the listener any further.

[4.4] While composers and performers were shifting away from the cadenza, Beethoven was still using it as an important structural and expressive vehicle. In Beethoven’s hands, however, the cadenza became something entirely different in scope and function so that it could no longer be regarded as a type of Classical cadenza.⁽³⁰⁾ Beethoven, as a virtuoso pianist famous for his improvisations, favored very long cadenzas. These were long, unaccompanied fantasias that were by no means embellishments of the cadential event. I have already mentioned the problem of possible excess in having a very long cadenza following an elaborate solo section. Beethoven understood that; he solved the problem in his Violin Concerto in D major, op. 61 (1806), and his Third and Fourth Piano Concertos, opp. 37 and 58 (1800; 1806), by moving the cadenza slightly back, detaching it from its cadential function. Though he maintained the customary approach to the cadenza, with orchestral preparation and a stop on a six-four chord over $\hat{5}$ in the bass, his handling of the exit from the cadenza marks a break from tradition. Instead of ending the cadenza with a bold perfect authentic cadence that signals the entrance of the closing tutti, the cadenza subtly leads back in to the movement proper with the gradual entrance of the orchestra.⁽³¹⁾ Thus, for Beethoven, the six-four chord over $\hat{5}$ in this context was not part of the cadential event, but a signal for a cadenza.⁽³²⁾ Furthermore, Beethoven explicitly undermines the cadenza’s cadential role by not fulfilling the listener’s expectation that the long trill will be resolved over the dominant chord back to the tonic; instead he fades it out and, as mentioned, slowly brings back the orchestra.⁽³³⁾ This insertion of a free, unaccompanied fantasy, not necessarily at the cadence, may have contributed greatly to our casual use of the term cadenza nowadays.

[4.5] Beethoven encouraged his students, such as Ferdinand Ries (1784–1838), to compose their own cadenzas to be performed in his concertos.⁽³⁴⁾ This suggests that it was customary for performers to prepare their cadenzas in advance, but for lack of evidence, it is nearly impossible to generalize solely from Beethoven’s cadenzas regarding how performers approached the subject in the first decade of the nineteenth century. By his Fifth Piano Concerto, op. 73 (1810), however, Beethoven waived the performer’s choice of the cadenza altogether, dictating in the score of the first movement what was to be played and signaling the end of the ad libitum cadenza tradition in new works. Later in this movement, just at the point where the cadenza would traditionally appear, he explicitly instructed performers not to play a cadenza and to move directly to the end.⁽³⁵⁾ Furthermore, Beethoven was probably among the last to incorporate cadenzas in his works on a regular basis, as cadenzas were generally considered outdated and redundant by younger composers and performers, who seem to have preferred a freer, less formal approach. Of Ferdinand Ries’s eight piano concertos, for example, only two have cadenzas, all written into the score as obligatory long passages inserted in the middle of the last solo section with no cadential function.⁽³⁶⁾ Johann Nepomuk Hummel (1778–1837), who famously wrote cadenzas for Mozart’s concertos, incorporated none in his own. Similarly, other virtuosos such as Friedrich Wilhelm Kalkbrenner (1785–1849) and Ignaz Moscheles (1794–1870) would freely incorporate fantasia-like embellishments in various places throughout their works, as shown in Example 21.

[4.6] When the famous younger Romantics such as Liszt, Mendelssohn, and Schumann came onto the scene, they used the term “cadenzas” freely, potentially referring to any unaccompanied solo passage. Liszt, for instance, places a “cadenza” only ten bars into the first movement of his Piano Concerto no. 1, S. 124 (1856). Schumann is more conservative in his Piano Concerto in A minor, op. 54 (1845), but he uses his cadenza similarly to Beethoven’s, as a lead-in to the coda. In his famous Violin Concerto op. 64, as noted, Mendelssohn uses a cadenza as a lead-in to the recapitulation. These composers may have employed cadenzas in homage to the concertos of their predecessors Mozart and Beethoven, whom they revered, but in terms of its structural function, there was nothing left of the Classical cadenza. All that was left from the traditional cadenza is the cadenza as a topic, already employed in eighteenth-century works outside the realm of the concerto.⁽³⁷⁾

Conclusion

[5.1] Recalling Koch’s comment that the free fantasia near the end of the first concerto movement “is mistakenly called cadenza” ([1793] 1983, 211) may suggest that he is probably uninformed or simply unconcerned with the origins and traditions of concerto cadenzas. His words do, however, encapsulate the short history of the Classical cadenza. Though the term cadenza could mean various things in the early eighteenth century—including unaccompanied, virtuosic add-in passages like those found in the writings of Vivaldi and his disciples—by the 1750s the term was considerably restricted to a short embellishment of the penultimate dominant harmony at the end of the last solo. The primary aesthetic purpose of the cadenza in concerto movements was to sound extemporaneous, and to surprise listeners one last time before the close of the movement. As last solos, which were once the most exciting and elaborate passages in concertos, became more rigidly organized along the lines of the sonata-form recapitulation from the 1760s through the 1780s, performers felt compelled to introduce longer and more elaborate passages at the cadenza to draw listeners’ attention and leave one final impression. Nevertheless, during that period the short embellishments remained in common use, and they represent standard late-eighteenth-century practice just as well as Mozart’s familiar longer embellishments. The tendency toward longer cadenzas, which probably peaked in the 1770s and 1780s, corresponds to structural changes in the movement proper, primarily calling for longer orchestral preparations. At a certain point, cadenzas became long and perhaps even independent fantasias that no longer fulfilled their traditional structural role of mere cadential embellishments, leading late-eighteenth-century writers such as Koch to confidently distinguish between the cadence and the cadenza. Beethoven’s cadenzas—which cannot be regarded as manifestations of the Classical cadenza, but were rather precursors of the Romantic cadenza—clearly demonstrate this. Around the turn of the eighteenth century, when Koch was writing his comments, the Classical cadenza was already in its twilight. It was a short-lived phenomenon that declined during the 1790s and the first decade of the 1800s, being used less and less, as the more brilliant writing for the solo instrument in concertos made it redundant.

Omer Maliniak
96 Poleg
Rosh Haayin 4862296
Israel
omermaliniak@gmail.com

Works Cited

Agawu, V. Kofi. 1991. Playing with Signs: A Semiotic Interpretation of Classic Music. Princeton University Press. https://doi.org/10.1515/9781400861835.

Anderson, Emily. 1997. The Letters of Mozart and his Family: Chronologically Arranged, Translated, and Edited with an Introduction, Notes and Indexes. Edited by Stanley Sadie and Fiona Smart. Palgrave Macmillan.

Bach, Carl Philipp Emanuel. (1753, 1762) 1949. Essay on the True Art of Playing Keyboard Instruments. Translated by William John Mitchell. W. W. Norton.

Bandy, Dorian. 2022. “Beethoven’s Rhetoric of Embellishment.” 19th Century Music 46 (2): 125–62. https://doi.org/10.1525/ncm.2022.46.2.125.

Bonds, Mark E. 2017. Listen to This. 4th ed. Pearson.

Caplin, William E. 1998. Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven. Oxford University Press. https://doi.org/10.1093/oso/9780195104806.001.0001.

Czerny, Carl. (1830) 1848. School of Practical Composition; or, Complete Treatise on the Composition of All Kinds of Music. . . op. 600. Translated by John Bishop. Robert Cocks & Co.

Feldman, Martha. 2007. Opera and Sovereignty: Transforming Myths in Eighteenth-Century Italy. University of Chicago Press. https://doi.org/10.7208/chicago/9780226044545.001.0001.

Greenberg, Yoel. 2022. How Sonata Forms: A Bottom-Up Approach to Musical Form. Oxford University Press. https://doi.org/10.1093/oso/9780197526286.001.0001.

Hill, John Walter. 2014. Joseph Riepel’s Theory of Metric and Tonal Order, Phrase and Form: A Translation of His Anfangsgründe zur musicalischen Setzkunst, Chapters 1 and 2 (1752/54, 1755). Pendragon.

Jacobi, Erwin R., and Willis Wager. 1961. “G. F. Nicolai’s Manuscript of Tartini’s ‘Regole per ben suonar il Violino.’” The Musical Quarterly 47 (2): 207–23. https://doi.org/10.1093/mq/XLVII.2.207.

Koch, Heinrich Christoph. (1793) 1983. Introductory Essay on Composition: The Mechanical Rules of Melody. Sections 3 and 4. Translated by Nancy K. Baker. Yale University Press.

Kollmann, Augustus Frederic Christopher. 1799. An Essay on Practical Musical Composition: According to the Nature of That Science and the Principles of the Greatest Musical Authors. London: printed for the author.

Kolneder, Walter. 1986. “The Solo Concerto.” In New Oxford History of Music. Vol. VI, Concert Music (1630–1750), ed. Gerald Abraham, 302–76. Oxford University Press.

Leeson, Daniel N., and Robert D. Levin. 1976. “On the Authenticity of K. Anh. C 14.01 (297b), a Symphonia Concertante for Four Winds and Orchestra.” Bärenreiter. 70–96.

Maliniak, Omer, and Yoel Greenberg. 2022. “Follow the Solo: The Formal Evolution of the Concerto in the Eighteenth-Century.” Music Theory Spectrum 44 (2): 231–59. https://doi.org/10.1093/mts/mtac009.

Mirka, Danuta. 2005. “The Cadence of Mozart’s Cadenzas.” Journal of Musicology 22 (2): 292–325. https://doi.org/10.1525/jm.2005.22.2.292.

Mozart, Leopold. (1756) 1951. A Treatise on the Fundamental Principles of Violin Playing. Translated by Editha Knocker. 2nd ed. Oxford University Press.

Prout, Ebenezer. 1895. Applied Forms. Augener.

Quantz, Johann Joachim. (1752) 1985. On Playing the Flute. Translated by Edward R. Reilly. Faber and Faber.

Swain, Joseph P. 1988. “Form and Function of the Classical Cadenza.” The Journal of Musicology 6 (1): 27–59. https://doi.org/10.2307/763668.

Talbot, Michael. 2005. “The Italian Concerto in the Late Seventeenth and Early Eighteenth Centuries.” In The Cambridge Companion to the Concerto, ed. Simon P. Keefe, 35–52. Cambridge University Press. https://doi.org/10.1017/CCOL9780521834834.005.

Tartini, Giuseppe. (1770) 1956. Treatise on the Ornaments of Music. Edited by Sol Babitz. Carl Fischer.

Tosi, Pier Francesco. (1723) 1757. Anleitung zur Singkunst. George Ludwig Winter.

—————. (1723) 1743. Observations on the Florid Song; or Sentiments on the Ancient and Modern Singers. Translated by Johann Ernst Galliard. J. Wilcox.

Tromlitz, Johann George. (1791) 1991. The Virtuoso Flute-Player. Translated and edited by Ardal Powell. Cambridge University Press.

Türk, Daniel G. (1789) 1982. School of Clavier Playing; or Instructions in Playing the Clavier for Teachers & Students. Translated by Raymond H. Haggh. University of Nebraska Press.

Whitmore, Philip. 1991. Unpremeditated Art: The Cadenza in the Classical Keyboard Concerto. Clarendon Press.

—————. 1988. “Toward an Understanding of the Capriccio.” Journal of the Royal Musical Association 113 (1): 45–56. https://doi.org/10.1093/jrma/113.1.47.

Zaslaw, Neal. 2008. “One More Time: Mozart and his Cadenzas.” The Century of Bach and Mozart: Perspectives on Histography, Composition, Theory, and Performance, ed. Sean Gallagher and Tomas Forrest Kelly. Harvard University Press.

Footnotes

* This research was supported by the Israel Science Foundation (grant no. 242/22), the Fulbright Israel Postdoctoral Fellowship, and the Rothschild Postdoctoral Fellowship. The choice of words in the title “The origins, the growth, and the eventual decline” is taken from the preface (p. vii) to Philip Whitmore’s Unpremeditated Art: The Cadenza in the Classical Keyboard Concerto (1991) in awe of his work, approach, and observations on the cadenza in Classical concertos. I would like to thank Alexander Rehding for all of his support and useful suggestions.
Return to text

1. Yoel Greenberg (2022), for instance, adopted a diachronic perspective to examine a large corpus of binary movements, spanning from the late seventeenth century to the second half of the eighteenth. Using statistical analyses, he was able to demonstrate how changes in musical conventions and the interaction between them over time produced new conventions. The statistics presented here were obtained in a similar fashion.
Return to text

2. Whitmore (1988, 50) and Talbot (2005, 46) credit Vivaldi with introducing the cadenza in a small minority of his concertos, all probably predating 1730.
Return to text

3. Locatelli makes a distinction between a capriccio and a cadenza: if the capriccio ends with a cadenza, it cannot be equivalent to one. It is far more plausible that after a relentless virtuosic capriccio, the cadenza, if it stands for anything, denotes a short embellishment of the cadence rather than yet another virtuosic display passage, as associated with much later cadenza practice. The same distinction can be found in Louis-Gabriel Guillemain’s (1705–1770) Amusement pour le violon seul, op. 18, from 1755.
Return to text

4. According to Sol Babitz, in his preface to Tartini ([1770] 1956), “the manuscript apparently bore no date, but it must have been written prior to 1750 in order to allow several years for it to circulate and reach Leopold Mozart in time to be incorporated into his book before its writing in 1754.”
Return to text

5. With respect to the cadenza vs. capriccio discussion, Whitmore (1988, 49–52) agrees with Jakobi and Wager’s general claim on Tartini’s differentiation between the cadenza and capriccio, noting that Tartini “recognizes that the two phenomena overlapped more in practice than he was prepared to allow in theory.” This assertion comes in response to Tartini’s comment that “today, every singer or instrumentalist permits himself to draw [the cadenza] out to such a degree, and with such different expressions that it is not reasonable to call it cadence [cadenza]; one must perforce call it caprice [capriccio]” ([1770] 1956, 100). Tartini explicitly distinguishes between a cadenza and a capriccio, which, unlike the cadenza “can be extended as far as one wishes, and can consist of pieces and different sentiments with different time signatures.” Jacobi and Wager (1961, 219) note that for Tartini, the cadenza and capriccio are “two forms that fundamentally had nothing to do with each other.” A manuscript by one of Tartini’s prominent pupils, Jean-Pierre (André-Noël) Pagin (1723–1799), clearly reflects these differences. The manuscript was probably compiled around 1742, while Pagin was studying with Tartini. Pagin’s compilation includes several cadenzas and capriccios, and he explicitly distinguishes between them, reflecting the distinction made by his teacher. See Whitmore 1988, 54–55, for a summary of these and other differences between the capriccio, as handled by Vivaldi, Locatelli, and Tartini, and the early cadenza.
Return to text

6. Many of Quantz’s ideas concerning the cadenza were echoed by Daniel Gottlob Türk in 1789 (1982), as described below.
Return to text

7. Here I discuss Quantz’s ideas about the cadenza in the context of the instrumental cadenza, and specifically in concertos. It is important to note that the following ideas, which also echo Tartini’s ideas, as well as others, are also true for vocal cadenzas (Whitmore 1988, 10). This suggests that the idea of a cadenza around the mid- to late-eighteenth century was similar across genres. Pietro Francesco Tosi, in his 1723 Opinioni de’ cantori antichi e moderni, was the first to discuss these ideas in the realm of vocal cadenzas, and his ideas were echoed in the 1757 translation of his book by one of J.S. Bach’s and Quantz’s pupils, Johann Friedrich Agricola ([1723] 1757). See Feldman 2007, 85–95.
Return to text

8. Tartini implies the same by attaching time signatures to capriccios while setting true cadenzas apart from them (see note 5).
Return to text

9. Around the same time, in 1755, Joseph Riepel made a similar suggestion to those struggling to come up with a cadenza. He advised them to “take the next best measure. . . which will provide. . . an introduction to it” (see Hill 2014, 198).
Return to text

10. By “modulating,” Quantz means what is described today as “tonicizing,” and not moving to a new key established with a cadence, which we now refer to as “modulating.”
Return to text

11. Quantz warns that he does “not pretend that the examples found here [in his Versuch] are complete and finished cadenzas; they are simply models from which [the reader] can, to some extent, learn. . . the characteristics of cadenzas in general.”
Return to text

12. Additional examples are provided in Appendix B. Perhaps the most “textbook-like” examples that follow Quantz, or at least for which Quantz’s description provides the best explanation, are the two cadenzas to the first movements of Dittersdorf’s two concertos for two violins in D major (Kr.155) and C major (Kr.deest) from the mid-1760s. I encourage readers to read Quantz’s guidelines for cadenza for two soloists on pp. 187–192 and apply it to these two cadenzas.
Return to text

13. That there is no requirement for a thematic relationship between the cadenza and its parent movement in the late eighteenth century is also evident, as Neal Zaslaw (2008, 245) points out, in cadenzas written by Mozart before 1779. Zaslaw deems them “simple and generic.”
Return to text

14. Türk can hardly be viewed as outdated when his book was published considering that it was republished in 1792 and 1802.
Return to text

15. Giordani’s collection, which also includes fourteen preludes, bears no indication of pedagogical purposes on the cover, but a lost collection by the composer from 1777 titled “Cadences for the Use of Young Practitioners” suggests that his op. 33 collection had similar aims.
Return to text

16. Christian Friedrich Daniel Schubart, Ideen zu einer ästhetik der Tonkunst, published in 1806 but written in 1784, translated by Kolneder (1986, 339–40); Joseph P. Swain (1988, 33) translated “excess” in Türk’s comments as “pointless length.” There is no way to know with certainty whether Buccioni’s cadenza and other cadenzas of similar length would be subjected to this “anything but new. . . often heard complaint,” in Türk’s words, or whether they represent the high end of what was considered acceptable at the time. Any presumed conclusion would be somewhat subjective and thus debatable.
Return to text

17. Some of the confusion between early cadenzas and the concept of the cadenza on the one hand and free flowing passages like capriccios and fantasias on the other presumably stems from our familiarity with what the cadenza became around 1800, as addressed later.
Return to text

18. In our current conception of the Classical cadenza, Mozart’s practice and the cadenza practices of Beethoven and his Romantic successors overshadow the shorter, more traditional practice, which was still common in the late eighteenth century. The cadenzas of Beethoven, and the early Romantics are addressed later.
Return to text

19. “The concerto or solo player has the liberty to play, or rather, that he must play if his playing is not to fail to please; and [even] if he frequently performs the most beautiful concerto in the clearest and most correct manner, and makes no such rubbishy fantasy at the end, or only a short one, whether suitable or not, his whole performance will be [considered] worthless. Now because everyone knows this, nobody will quit the stage without having done such a thing” ([1791] 1991, 260).
Return to text

20. Eleven concertos from the 1740s and twelve from the 1750s, out of a corpus of 60 concertos examined from those decades.
Return to text

21. Cadenzas appear in 77% of works examined, based on a sample of 30 concertos from the 1760s. All examined works are listed in Appendix A.
Return to text

22. Based on a sample of 135 concertos from the 1770s.
Return to text

23. Based on a sample of 85 concertos from the 1780s.
Return to text

24. For Leeson and Levin this orchestral passage leading to the cadenza is the sixth of seven parts of Mozart’s general structural plan for a concerto. Composers would also highlight the arrival at the cadenza by means of tonicization, approaching $\hat{5}$ in the bass (and occasionally in other voices) using its leading tone (♯ $\hat{4}$). All of Borghi’s and Giordani’s cadenzas in their collections are approached in this manner, which Mozart employed in 27 of his 31 concertos examined. See Mirka 2005, 300, for more on this.
Return to text

25. Ebenezer Prout (1895, 207) expressed a similar view with reference to Mozart and Beethoven, writing that “it was the custom of Mozart and Beethoven to introduce in the last orchestral tutti of the recapitulation a pause, always on the second inversion of the tonic chord, to allow a brilliant improvisation on the part of the soloist. Such an improvisation was called ‘Cadenza,’ and in it the performer was expected not only to show his technical skill, but his ability in treating material taken from the movement itself.” It is noteworthy that Prout refers to Czerny’s School of Practical Composition, op. 600 ([1830] 1848), as “almost the only theoretical work. . . that gives a detailed account of concerto form”; however, Czerny (159–60) does not place the cadenza in the closing ritornello, which, like Prout, he calls “a short conclusion.”
Return to text

26. Based on a sample of 59 concertos from the 1790s and 32 from 1800 to 1805.
Return to text

27. Mozart once wrote to his father about another violinist who played “ten million times better than Pinzger” (Anderson 1997, 607).
Return to text

28. This source is John Bishop’s English translation of Czerny’s Die Schule der Praktischen Tonsetz Kunst, op. 600, originally published in German in 1830.
Return to text

29. There may be a geographical nuance to this as well, which I have not discussed here. Saint-Georges lived and primarily worked in France. Other French composers, like Dieudonné-Pascal Pieltain (1754–1833), Antoine Hugot (1761–1803), François Devienne (1759–1803), Adrien-Louis Le Bugle (1744–1812), and Jean-Louis Duport (1749–1819), also left cadenzas out of their concertos. The French taste probably had more and more influence as Paris was emerging as the leading musical center in Europe, eventually dethroning Vienna in the music history books. Of course, there were other composers from other nations who also omitted the cadenza and French composers, even leading ones like Simon Leduc (1742–1777), who favored it. As noted, I did not delve into this geographical nuance, so for now this is merely conjecture and anyone is welcome to pursue the matter.
Return to text

30. For more on the divergence of Beethoven’s cadenzas from Mozart’s, see Swain 1988. I am not repeating Swain’s ideas here since they have already been written, and more importantly because Beethoven’s cadenzas and their foreshadowing of the Romantics’ conception of the cadenza are off-topic. However, because our current conception of the cadenza is highly influenced by the Romantics’ notion, it seems appropriate to round off the discussion of the Classical cadenza by making a few brief remarks about the Romantic perspective.
Return to text

31. Another aspect of Beethoven’s different treatment of the cadenza has to do with his view of the closing segment of the movement, in which the soloist takes an active part. This segment can be better read as a coda rather than being reminiscent of the ritornello tradition, which Mozart basically followed. But the large-scale structure in Beethoven’s concertos is a whole different discussion that lies outside the scope of this study.
Return to text

32. This may suggest that by then, following decades of practice, the cadenza was so closely associated with the cadential six-four chord that the six-four chord assumed an additional function that it traditionally did not possess.
Return to text

33. See Beethoven’s cadenza to the first movement of his Piano Concerto No. 4, op. 58, for a clear example.
Return to text

34. See Swain 1988, 45–46.
Return to text

35. For the rhetorical aspects of Beethoven’s treatment of the cadenzas in his fourth and fifth piano concertos and his role in moving away from the improvised cadenza as a part of “long-range compositional planning,” see the discussion in Bandy 2022.
Return to text

36. Opp. 42 (1811) and 132 (1823). We also find a very short, non-mandatory written passage marked “cadenza” only ten bars into the development section in op. 115 (1809). All of Ries’s concertos were published, suggesting they were intended for the use of others and not just for the composer’s own use.
Return to text

37. The idea of a cadenza as a topic is not new; Agawu (1991) uses the idea of the cadenza as a topic when discussing unmarked cadenza-like passages in non-concerto works by Haydn, Mozart, and Beethoven. This also connects to the changed conception of the solo-tutti relationship in Romantic concertos and the dissolution of the Classical concerto form, which are outside the scope of this discussion.
Return to text

Copyright Statement

Copyright © 2025 by the Society for Music Theory. All rights reserved.

[1] Copyrights for individual items published in Music Theory Online (MTO) are held by their authors. Items appearing in MTO may be saved and stored in electronic or paper form, and may be shared among individuals for purposes of scholarly research or discussion, but may not be republished in any form, electronic or print, without prior, written permission from the author(s), and advance notification of the editors of MTO.

[2] Any redistributed form of items published in MTO must include the following information in a form appropriate to the medium in which the items are to appear:

This item appeared in Music Theory Online in [VOLUME #, ISSUE #] on [DAY/MONTH/YEAR]. It was authored by [FULL NAME, EMAIL ADDRESS], with whose written permission it is reprinted here.

[3] Libraries may archive issues of MTO in electronic or paper form for public access so long as each issue is stored in its entirety, and no access fee is charged. Exceptions to these requirements must be approved in writing by the editors of MTO, who will act in accordance with the decisions of the Society for Music Theory.

This document and all portions thereof are protected by U.S. and international copyright laws. Material contained herein may be copied and/or distributed for research purposes only.