McMaster Music Analysis Colloquium



An Active Character in the Design of Ternary Form


Catherine Schwartz


Form: "The shape of a musical composition as defined by all of its pitches, rhythms, dynamics, and timbres. In this sense, there can be no distinction between musical form and specifically musical content . . ."(1)

Noteworthy in this definition is the inclusion of attributes such as dynamics and timbres which emphasize the performative and auditory dimension of analytical considerations. Indeed, form and content of any composition do not exist without their sounds in temporal and physical contexts within which they are performed. And, just as musical experience does not exist outside of cultural frames and interpretive contexts, the above definition might make clearer the view that the structures and forms which content helps to delineate are themselves defined by the very same forms and structures. In this way, there truly is no distinction between them.

Because this is the first formal music analysis paper I have written, I wish to clearly acknowledge that the forms, content, and structures I identify in my analysis do not exist outside the social forums in which we create, perceive, perform, analyse, define, and discuss them. Among the identifying features of the forum in which I present this discussion is a common understanding among its immediate participants (the other members of my class) of the tradition of music analysis. This is a tradition which, I believe, continues to contribute in important ways to our aesthetic experiences of both notated and performed music in the Western classical music tradition. It is my hope then, that my discussion of form and content through the use of music-analytical tools will articulate at some level ways in which I understand my auditory experiences of the music to inform my formal analysis and, of course, the corollary.

"Sich Üben im Lieben" is a da capo aria from J.S. Bach's Wedding Cantata. Formally structured by three sections of which the last is a recapitulation of the first, this aria can be labeled as a ternary-form piece. The musical features such as tonal areas, melodic shape, and instrumentation, to name merely a few, which work to unify the piece and yet differentiate between sections A and B are many. In particular, I am concerned with assumptions in standard ternary-form discussions that some musical material is not considered integral to the form of a piece: namely, what the Appendix, paragraph 18, terms "add-ons".

We must understand that the prelude and postlude, the introduction and coda, like the ritornello, play no part in the form itself; that is the form could stand without them. (Appendix, paragraph 18)

Since I agree with the New Harvard Dictionary that form and content are inextricably connected, my paper seeks to challenge this understanding of "add-ons" with regards to the role of the ritornello in understanding the ternary form of "Sich Üben im Lieben".

In his discussion of the development of the da capo aria, Charles Rosen articulates his disregard of the ritornello in determining even the form of the A section of an aria. He writes,

. . . section A becomes a binary form with two clear phrases, called for by a repetition of the opening quatrain of the text:

                            Ritornello             Solo             Ritornello             Solo             Ritornello
A section                                            A1                                             A2
alone                         I                       I>V                    V                      I>I                     I

As he continues, the agenda of this analysis becomes clear: "Disregarding the ritornelli, this is a slow-movement sonata form (. . .)"(2)

I do not wish to dispute Rosen's observations about the developing relationship between formal aspects of the da capo aria and sonata form, nor do I argue that the ritornello never acts in a primarily "intro" or "inter" function. However, by examining the different roles of the ritornello in "Sich Üben im Lieben," I intend to demonstrate that divisions between the ritornello and formal aspects of the piece can be as misleading as divisions between the piece "itself" and the larger cultural context through which it occurs. So, while normally it is understood that the voice line alone delineates the formal structure of a piece, to my ear, the ritornello does not act as a series of "pillars" around which the form constructs itself, nor simply as a unifying structural device as it is often called.(3)  Rather, the ritornello is an integral and active "character" in the negotiation of structure, content, and rhetoric, all of which are central to how we negotiate an understanding and experience of form.

Any discussion of ternary from is contingent on the dual principles of unification which mark A and B sections as part of the same piece, as well as contrasting features which distinguish A from B and mark the final return of A (A-B-A) as a move back home. In order to examine the integral contribution of the ritornello to the ternary form in "Sich Üben im Lieben," I am concerned not only with its role as a "unifying device" but also ways it acts out altered states of form, content, and functionality in different contexts throughout the piece. This paper will therefore explore the ritornello as unity and difference in its capacities as 1) an identifiable character; 2) a harmonic and melodic structural character; 3) an "affektive" character; and 4) a poetic character.

The aria begins with the exposition of the ritornello. As soon becomes evident, it acts as an introduction. But, an introduction to what? Is it introducing the vocal entry yet to come or is it introducing "itself" as a key player in the piece? I suggest, the latter. The ritornello in "Sich Üben im Lieben" is not a precursor to the main musical content of the vocal line, but in its capacities as melodic content, tonal identity, and accompanimental character, to name a few of its roles, it establishes itself as the beginning of the piece, not as a foreword to it. (4) I begin then by examining my interpretation of the harmonic and melodic structural identity of the opening ritornello.

Figure 1. Opening Ritornello, mm 1-22.


Figure 2. Analysis of Harmonic and Melodic Structure of Opening Ritornello.

In figure 2, the open notes d-a-d mark the overall tonal movement from I-V-I, as well as an interpretation of the fundamental melodic goals within the opening ritornello. I suggest, through the use of stemmed notes, which notes constitute the main melodic movement of the ritornello. This movement begins with an ascent from d-e-f-sharp and leads into a descent beginning on the low a in m.8. This a moves down to g-f-sharp-e with an upward registral shift in m.17, to a repeat of the g-f-sharp-e descent in mm.18-20 to the final descent on the high d in mm.21-22 in the register in which the ritornello begins. In this interpretation the e in measure 20 is implied or carried over from m.19. I have not included the c in m.7 as part of this fundamental movement because I hear it in the context of the scalar descent from the high a to the low a between measures 6 and 8. Likewise, the c in m.17 parallels that in m.7 by functioning within a registral shift. It is interesting to observe that the first c sounds during the descent between a and a leading us to g in m.9, while the c in m.17 sounds during the ascent between e in mm.8-9 and e in m.17 bringing us again to the g, this time in m. 18.

In the first section of the ritornello (mm.1-8 ending with a half-cadence), hereafter referred to as "rit.a," I hear an inner voice which is indicated on the graph by the unstemmed notes in the movement from f-sharp-g-a-(e). This is known as an illusory progression in that the g, as the 7th of V7, resolves to the f-sharp, both immediately in m.4 and on a larger scale in m.5, and yet it also works within a larger over-arching pattern (to be explained momentarily) in its move to the a. Like the c, the e in m.7 might just be considered part of the a scalar passage. Furthermore, the first three notes of this progression have the particular function of reaching-over the main melodic notes: the f-sharp over the d, the g over the f-sharp, and the a over the g to come in mm. 9 (the function and experience of this high a within the ritornello has room for more discussion, and I will return to this point momentarily). I suggest that the low a which sounds in each measure of rit.a works primarily to fill out the D major tonality of the passage and in anticipation of the half-cadence in m.8. We will return in a later section of this paper to the "role" of the inner voice melody in the ritornello.

For now though, I present Figure 3 as a representation of the ritornello's main melodic motion as it appears in the opening ritornello of the piece.

Figure 3. A Reading of the Opening Ritornello's Main Melodic Motion.

However, I suggest that the analysis I have proposed begs further questions. In Figure 4, I present an analysis suggested by Dr. Renwick in the course of on-going discussions about this piece.

Figure 4. Dr. Renwick's Reading of the Opening Ritornello's Main Melodic Motion.

Admittedly, it looks very similar to Figure 3. Nonetheless, there are two fundamental differences: the role of the high a and the "middle" descending section of a-g-f-sharp-e in the lower register. Both of these matters go hand-in-hand. In Dr. Renwick's proposed reading, the main melodic line is in the upper register while "the lower octave motion a down to e that occupies the middle section of the ritornello is a 'motion to an inner voice'--that is the scale segment a-e unfolding and prolonging the harmony V"(5). Thus the melodic motion of the high register is the fundamental melodic line. The question of whether the descent in the lower register is a secondary voice or part of the main voice is, I think, integral to the broader question I am asking about the ritornello's changing and multiple identities.

In my reading of the opening ritornello as a "closed" entity I would suggest that the lower descending passage is indeed part of the main melody, at least as much as the upper-register descending passage which closes the section. First of all, my ear "tells" me this. Though the high a in m.6 is visually climactic in the score, as mentioned earlier, it functions aurally as a reaching-over note and descends somewhat dramatically to the low a. The low a to my ear thus becomes the new melodic point of departure. The descent then from g-f-sharp-e through mm.9-16 comes out of this a and acts as a melodic counter-statement equal in measure numbers to the antecedent phrase (although the consequential descent is obviously not complete at the end of m.16).

The second reason I "insist" that the lower descent is part of the main melodic line within the complete ritornello that opens the piece is because I am aurally unable to sustain the high register left at m.6 to the high register resumed in m.17. Interestingly though, as Dr. Renwick's analysis points out, all of the harmonic structural evidence works to bolster an interpretation of the high register as supporting the home or tonic area while the low register supports the dominant area, thereby creating high/I - low/V - high/I. Furthermore, with its disjunct intervals and repeated two-bar segments the melodic character of mm.9-16 does change significantly from mm.1-8 (an important point to which I will soon return) perhaps questioning its melodic "integrity". However, I am no more convinced that mm. 17-22, with its modulation of register which moves quickly into a melodic cadential figure, should be designated as "main" melodic material. Indeed, I now even suggest the idea that the final descent in the upper register is the secondary or quasi melodic motion. In fact, representing this idea, Figure 5 most closely illustrates the way I aurally conceived of the ritornello's melodic structure when I first began this analysis.

Figure 5. Another Perspective on the Opening Ritornello's Main Melodic Motion.

Whatever we choose to identify as the main melodic structure of the opening ritornello, most of the melodic and harmonic features that I have discussed remain intact enough throughout the rest of the piece.  Thus its identifiable character ensures that the ritornello acts as a unifying device. However, the opening ritornello's "complete form" in D major, as played by the oboe and supported harmonically by the continuo, establishes a specific melodic contour, key area, and instrumental texture as "home". This home is not simply a structural pillar though. Instead, it is the establishment of musical "content" unique only to the ritornello in its complete form and yet integral to the identities of every other musical element in the aria. In this way, it is home which both defines what is beyond its "self-contained" walls and yet also home which, conversely, is itself defined by what is beyond those same walls. In its content, it is integral to the form. And, as we shall later see, it is the central part of the A section, and thus is central to the A-B-A form of this aria.


So far, I have examined the ritornello's identity in terms of its unified presentation at the beginning of the aria. However, even now we can see that the ritornello has two distinct areas of structure and content within it, as created by registers, melodic contours, harmonic areas, and notational character. What soon becomes apparent is that these differences become crucial to the identity of the ritornello throughout the rest of the piece. Bearing these distinctions in mind, I discuss the ritornello throughout the rest of the paper in two parts: rit.a (mm.1-8) and rit.b (mm.9-22), and continue my discussion of the ritornello's identifiable character as I proceed with the topic of its contribution to formal design through harmonic structure and melodic voice-leading.

After the opening, the soprano voice enters into duet with the oboe's rit.a at m.23. Interestingly, only the first part of the ritornello (rit.a) is ever played simultaneously with the voice, and with the exceptions of the opening and closing "complete" ritornello of section A, it always works with the voice. Rit.b on the other hand only works in its purely instrumental form. This simple distinction becomes fundamental to each of their identities, structural positions, and functions within the aria. Upon examining the vertical integration of the voice's part with the oboe and continuo parts at m.23, it becomes apparent that the line which has constructed the inner voice which I have already identified in the oboe part as f-sharp-g-a-(e) becomes the linear progression that sounds in the vocal part while the oboe line sounds the main melodic notes of rit.a. See figures 6, 7a, and 7b.

Figure 6. First Vocal Entrance, mm. 23-38.

Figure 7a. Main and Inner Voices of Oboe's Rit.a.

Figure 7b. Main and Inner Voices of Oboe's Rit.a in Conjunction with Soprano.

However, the voice does share with the oboe the main melodic structure of rit.a. Alternation of the main melodic notes (d,e,f-sharp, and A) between the voice and oboe means that at least one of the melodic lines (oboe or voice) sounds a main melodic note from rit.a in every measure throughout the phrase.

Figure 7c. Comparison of Melodic Structures of Rit.a Line and First Vocal Entry.

Arguably, the voice becomes dominant in these passages since it sounds the main melodic notes (d,e,and f-sharp) of rit.a for one-and-a-half beats compared to the one beat given to each of these notes in the oboe's line. However, any conception of the vocal line's autonomy is not quite so simple. Not only is the oboe's bright timbre competitive with the soprano's voice, but it is evident very quickly that the voice's line is structurally integrated with and even dependent on the ritornello, both in terms of melodic contour and harmonic function. In the passage directly after this first vocal entry, we do not hear the consequential melodic descent and V-I progression of rit.b which resolved rit.a in the opening ritornello. Rather, in mm.30-38, we hear the oboe take on an accompanimental role while the voice takes over the main melodic line. But, this expected vocal dominance is all very temporary. The vocal passage does make a melodic descent but goes past the marked note of d (the note with which the ascent began), takes on a modulatory function rather than harmonic stability, and leaves rit.a melodically and harmonically unresolved. In its movement from D major towards A major it cadences on V/A major and leaves a b as the last note of the voice waiting to be resolved in m.38. Thus, the return of the rit.a passage (even as it is integrated with the voice) at mm.39, mobilizes rit.a into its now clearly established role, one that remains throughout the aria, as resolution to vocal instability. The ritornello at m. 39 immediately takes up the key of a major and resolves the b melodically to an a as the first main melodic note of the ritornello in its new key. Structural resolution is further created as the rit.a cadences on the tonic of the new key, thereby altering the main melodic motion of 1^-2^-3^5^ to 1^-2^-3^-1^.

Figure 8. Harmonic and Melodic Resolution of mm.38.

So, while the combined oboe/voice rit.a passage reinforces the dominance of melodic context and structure of the rit.a we hear in the introduction, expectations are also thwarted at this new site of the rit.a. Indeed rit.a and rit.b no longer act as antecedent and consequent phrases to each other; but still, the ritornello's new defining structural features guide the form of the piece. Interestingly, the four subsequent rit.a events in the aria (after the one we have just discussed but not including the two ritornellos of the repeated A section) consistently act out a "resolving" function through the same structural roles identified in the above example. At the same time, they can be heard as harmonically and melodically "closed" events so that when rit.b does follow them, it does not act as a descending continuation or tonal consequence to rit.a. Instead, rit.b acts as a link between rit.a, now a sort-of melodically and tonally "closed" area, and the vocal solo sections which, accompanied by the oboe rather than interacting melodically with it, act out tonally transient or modulatory passages. Figure 9 presents the different harmonic structural functions and positions of the rit.a, rit.b and the non-ritornello sections beginning at the first sighting of the new tonally closed rit.a (mm.39-46).

Figure 9. The Harmonic Structures of Vocal Solos, Rit.a Passages, and Rit.b Passages.



Harmonic Resolution: 
vocal solo

Reinforces Tonal Stability: 

Section A



Mm.30-38 (D toV/A)

Mm.39-46 (A) 

Mm. 47-60 rit.b (A)

Mm. 61-68 (A toV/D)

Mm. 69-76 (D)

Mm. 77-98 rit.a &b (D)

Section B



Mm. 99-106 (b to V/f#)

Mm. 107-114 (f#)

Mm. 115-128 rit.b (f#)

Mm. 129-136 (f# to V/b)

Mm. 137-144 (b)

Da capo to opening ritornello of Section A (D)

Perhaps what is most striking about this chart is that the vocal solo line, the line which is supposed to determine the formal structure of the work, the line which can supposedly "stand on its own" from the ritornello is, in fact, utterly dependent on the ritornello for coherence of its own musical content.

Indeed the vocal line is not only harmonically integrated with the ritornello but is melodically interdependent with it as well. The example of melodic resolution from the b to the a as identified in mm.38-39 is virtually identical to the second such passage in section A at mm. 68-69. In both of these cases, the vocal line heard at rit.a acts out or contains the same melodic structure and movement as the oboe so that the vocal line, though not independent from the ritornello, at least embodies a certain structural or formal autonomy within the linearity of its own part. For this reason, some theorists might wish to dispute my emphasis on the oboe's melodic role in the ritornello as a fundamental factor in determining the formal design of the aria.  However, I hold my position for two reasons. Firstly, though this is rather interpretive, I am compelled to hear the voice's role in the melodic structure of rit.a as precisely that, a "role" which cannot be or at least is not ever taken out of the context of its relationship with the oboe's role in rit.a. In this way, I see the two lines as inextricably connected.

Secondly, the voice's melodic interdependency on rit.a is pronounced in a particularly unique way in section B. At a cursory glance, just the opposite might seem true. In section B, the voice maintains its own part throughout and when rit.a arrives in the oboe's part, the voice sings a melodic line completely different than that of rit.a and thus unique to the role it has carried out in the formal design thus far. Indeed, Figure 10 reveals the independent character of the vocal melody from the rit.a part in mm. 106-114 and we will discuss the role of this independent character of the vocal line in section B later in the essay.

Figure 10. Dependence and Independence of Vocal Melodies in section B.

Notice the varied contours as well as the contrasting range between the low vocal melody and the high oboe melody before these melodies and ranges begin to converge and cadence together at the end of the phrase. But, what I am most curious about is the fact that the rit.a line in the oboe part acts as the melodic resolution to the preceding vocal phrase. In mm.106-107, visible in Figure 10, notice how the e-sharp of the vocal line as part of the V chord of F-sharp minor, finds resolution not in the vocal line but in the oboe's rit.a. The same thing happens in the move from a-sharp of the voice to b of the oboe in mm.136 and 137 at an identical structural point in the composition. So, the vocal line is not as melodically unrelated to the ritornello as it would initially appear. In both section A and section B, unstable modulatory vocal lines find melodic and harmonic resolution in the rit.a.

Once again, I would like to address the structural role of rit.b. While I have argued above that rit.a becomes a sort of "closed" entity as it functions away from its role in the full, purely instrumental context of the ritornello, rit.b cannot be passed off strictly as an interlude which makes no contributions to musical content of the A and B sections. Rit.b is no more simply a structural pillar or unifying device than is rit.a. In fact, rit.b is not "self-contained" melodically or harmonically, and though in its "independent" appearances (once each in sections A and B), it does reiterate the tonal stability achieved by rit.a, its melodic and harmonic direction comes directly out of the rit.a passages. I recall your attention to the melodic structures of the ritornellos identified in Figures 3 and 5.

Melodic Structure of Full Instrumental Ritornello: d-e-f#-a-g-f#-e-d (mm.1-22)

The basic ascending/descending melodic contour (1^-2^-3^-5^-4^-3^-2^-1^) of this full ritornello is maintained even when rit.b resolves to I rather than V. The only melodic-structural difference in these more "sectionalized" appearances of the two parts of the ritornello is the fourth note. The contour instead becomes: 1^-2^-3^-(1^)-4^-3^-2^-1^.

Melodic Structure of "Sectionalized" Ritornello: a-b-c#-(a)-d-c-b-a (mm.39-60)

Melodic Structure of "Sectionalized" Ritornello: f#-g#-a-(f#)-b-a-g#-f# (mm.107-128)

I have bracketed the "1" pitch because, though it works to harmonically close the rit.a, it does not inhibit the melodic fluidity between the two ritornello sections which is achieved largely by the forward-moving bass passages that do not stop to cadence with the rit.a. See Figures 11a and 11b. Notice also in Figure 11a that the a on which rit.a cadences is immediately picked up by rit.b as an anacrusis.

Figure 11. Transitions Between Rit.a and Rit.b.
11a. Mm. 46-47

11b. Mm.113-115

In this discussion of how the vocal line and the different parts of the ritornello function harmonically and melodically within sections A and B, I hope to have suggested how integral the voice's "content" is to the very structurally unifying device from which it is thought to be independent: the ritornello. The parts are so interconnected, one can not easily distinguish form from content. Without this understanding of how the ritornello's structures work in conjunction with the voice, I do not think we can begin to identify how the vocal line contributes to the delineation of form.


In the midst of writing an essay about the integral relationship between form and content, I also fall into the trap of making arbitrary distinctions between them. Several of the matters which I now identify as "textural and affektive content" are indeed interconnected with the melodic and harmonic structural roles of the ritornello identified in the section above. Also, once again, my discussion of the ritornello is inevitably contingent on musical elements that surround and thereby make up the very content that I seek to delineate.

While the preceding discussion has emphasized consistencies in the ritornello's part throughout the aria, there is no doubt that the "complete" ritornello (mm.1-21) belongs primarily and is most integral to section A. The opening ritornello establishes a "happy" and "light" mood through playful melodic contours created by successive eighth notes, as well as the rhythmic syncopation and forward-motion in the "reach-over" notes. As well, the oboe's instrumentation and major key help to create a bright timbral sonority. To a large degree the rest of section A acts to reinforce this cheerful affekt, by interacting tonally, texturally, and motivically with the ritornello.

First of all, the tonal movement from the stable tonic of D major to the dominant of A major establishes calm and stability while contributing to the sense of propelled motion and direction. This tonal movement also reinforces the melodic structure of the opening ritornello which can be seen as a fundamental movement from d-a-d (see Figures 4-6). Secondly, section A reinforces the oboe itself as fundamental to the aria's sonority through 1) its constant presence which never allows the listener to lose touch with the action of the oboe's line and thus with the ritornello, and 2) its integral role in the melodic and harmonic structures explored in the above section. Thirdly, when the voice enters at m.22, the oboe begins into a playful duet with it, not only by sharing the main melodic motion of the ritornello, as already explained, but by interweaving with the oboe in yet another dimension: through motivic interplay and voice exchange. See Figure 12.

Figure 12. Motivic Interplay and Voice Exchange Between Voice and Oboe.

This interplay between voice and oboe occurs two more times in section A (mm.39-46 and mm.69-76), designating the ritornello with all of its musical properties as an integral contribution to the playful yet stable sonority and rich texture of the section.
In contrast, while the ritornello retains an integral structural role in section B as it had in section A, and in that way contributes to content, the affekt of this section is defined largely through the different role of the ritornello within it. Its altered textural, timbral, tonal, and motivic contributions alter the "affektive" identity of section B which, through its instability and sparseness, speaks a character of freedom and anticipation.
First of all, not only does section B not begin with the ritornello as section A had, it does not even include the oboe in its first entrance relegating the voice to a much more independent position than in section A. So, while in section A the ritornello had established the oboe as a central voice in the piece, the absence of the ritornello marks also the absence of the oboe in section B. For example, this more independent role of the voice affirms itself again in the second vocal solo which is also texturally independent of the oboe, since the oboe is largely excluded from even an accompanimental role in this passage (mm.134-136). Further creating a more unstable and sparse textural character, the first voice/oboe duet in section B at mm.31-37 is, as already discussed in the previous section, characterized by differences between the two parts in terms of register and melodic-contour, as opposed to the tightly woven voice/oboe duets in section A.

In addition to the altered textural and timbral features of section A, the ritornello's diminished involvement in section B becomes associated with a decreased tonal stability. Section A began firmly in D major and thereby established clear expectations, in its tonal function, for its move to the dominant and back home again. In contrast, section B begins immediately with a modulatory passage beginning only momentarily on D's relative minor (B minor) chord at m.99. So, although the section moves to f sharp minor (the dominant of B minor) and closes in b minor, the momentum to get there is not created as much by clear tonal drive as by the tension of tonal instability.

As explained earlier, when rit.a does appear in section B it does create melodic closure to the modulatory vocal line. Nonetheless, the ritornello is now altered in its transformation into the minor mode, and in this new tonal disposition it does not act with the same authority as it does in its more dominant role in the secure context of the major mode of section A. Furthermore, while rit.a does resolve the melodic tensions left hanging by the vocal lines in section B (see discussion about voice-leading in Figure 10), the voice maintains not only melodic independence as discussed above, but moves forward in its harmonic and eventual cadential closure. The absence of the ritornello and its tonal stability in the opening of section B as well as the overall stability of its major mode and clear harmonic direction throughout section B, reconfirms the section A ritornello's presence as happy stability.

Indeed, what I hope this discussion of the ritornello in section B demonstrates is that the power and character of the ritornello embedded in section A is present in the very absence of this same role in section B. Furthermore, the new role the ritornello takes up works to help distinguish B from A, particularly as it functions reflexively to reconfirm the opening ritornello's role within section A. So, when, at the end of section B, the da capo returns us to the same material that began the piece, the ritornello hardly functions as a link or a structural pillar between sections A and B. To be sure, the ritornello now marks the return of A, all the more so because of its differences from B. Simultaneously though, with its new found history in the "life" of section B, it also marks a point of unity and mutual function between the two sections, acknowledging their part in a shared whole.


Although I have attempted to clarify in this essay that the vocal line does not work alone in determining form as some writers seem to suggest, I hope my discussions of the ritornello in relationship to the voice does not suggest that the voice's musical line plays no role at all. Likewise, the ritornello also interacts with another important part of the voice's contribution to the aria: the text. Indeed, the poem is an important part of my experience of "Sich Üben im Lieben" and, in my interpretation of both ritornello and textual contributions to the ternary design of the work, I find the ritornello's part interconnected with the enunciation of the textual meaning. Here is a translation of the German original:

To train oneself in love, in lighthearted jests
is better than spring's transitory desire.

Here run waves laughing and watching,
the victorious palms on lips and breasts.(6)

The first two stanzas are the text of section A, and the second two, the text of section B. Normally the different affekts of the A and B sections of a da capo aria help to articulate differences in the meaning of the poetic texts. In my interpretation of this aria, I am able to link the musical character of section A that I have identified in the previous section of this paper with feelings of content and stability which are often felt in loving relationships. Furthermore, the intricate weaving of voice and oboe create, in my ears, a love duet. I interpret the less predictable character of section B which, in particular, deviates from the musical patterns established in section A, as the playful part of love which "runs" freely and "laughs". The voice and oboe, "free" in their relationship can work together melodically and harmonically, without being "tied" down. They are free in love and free to anticipate return to each other.

In so far as the ritornello works integrally with the voice's poetic message, it contributes in yet another way to the formal differentiations between sections A and B. At the same time, the formal contributions made by the relationship between ritornello and poetic message can also be understood at the level of integration between the two distinct sections. The recurring ritornello throughout the aria, its harmonic and melodic structural consistencies, and its particularly vocal presence in the return of section A sound out Love's constancy as opposed to Spring's "transitory desire". For me, the ritornello takes on a distinct poetic character in the formal design of the aria.


I began this essay with both a preface and introduction before I began the "body" of the paper. And, I suggest that the formal design of this "body" could not stand on its own from the introduction any more than the formal design of "Sich Üben im Lieben" could stand without the ritornello. Form and content are integral to each other, and the content of structural pillars such as introductions and conclusions or ritornellos work reflexively within the "bodies" of works to define the very identities of form-defining "content". In this way, they are in inextricably embodied within the forms of their works.

In "Sich Üben im Lieben," the ritornello plays a particularly integral and dominant character. It is vocal in several ways including its role as introduction and conclusion, its contribution to harmonic and melodic structuring, its part in the distinct affekts of sections A and B, as well as in its voice in the creation of poetic meaning and interpretation. Since structure, affektive content, and text all play important roles in the creation of formal design, both in its features which distinguish sections as well as those that provide coherent unity, the ritornello is integral to that form. My point is certainly not that the vocal line is not also fundamental to the formal design of the piece, but rather that its very identity and therefore formal contributions are created through its interactions with other elements of the music. In "Sich Üben im Lieben" the ritornello happens to have an especially dominant role. But, I suggest that in all works with ritornellos, even those in which they are less active throughout the piece, the ritornello's structural role cannot be removed from its role as content. The two live inside of each other.


1. "Form," The New Harvard Dictionary of Music. Ed. Don Randel. Cambridge: Belknap Press of Harvard University Press, 1986.

2. Quotations and graph from Rosen, Charles. Sonata Form. New York: W.W. Norton, 1980, p.30.

3. "Pillar construction" is a term used by Friedrich Blume regarding the role of the ritornello in his book Renaissance and Baroque Music. New York: W.W. Norton, 1967, p.137.

4. In some arias, the opening ritornello seems to be more of a precursor to the main part of the piece yet to come than it is in "Sich Üben". For example, in Bach's aria "Bereite dich Zion" (as analyzed by Megan elsewhere in this journal issue) the vocal entry subsumes the ritornello line by repeating its main melodic passage. Similarly, many Handel arias use a ritornello to introduce the main melodic line of the aria which is taken over by the voice. Interestingly, in the Wedding Cantata, a couple of arias other than "Sich Üben . . ." also incorporate melodic instruments (such as violin and oboe) as dominant parts which are hardly submissive (in "content") to the vocal line. However, as I hope it will become clear, it is not my intention to suggest the ritornello is only a fundamental part of determining form if its melody stands apart from the vocal line.

5. From a note by Dr. Renwick to me on LearnLink, October 28, 1999.

6. Translation from CD jacket of RCA's Bach, with Kathleen Battle and James Levine, New York, 1992.

© Copyright 1999 by Catherine Schwartz.