If you would like to participate in discussions, please sign in or register.

In this Discussion

Most Popular This Week

    Due to changing needs and technologies, the SMT Executive Board has decided to retire SMT Discuss (effective Nov. 9, 2021). Posts will be preserved for archival purposes, but new posts and replies are no longer permitted.

    K. 284i

    Imagine blanking out, say, the ‘second theme’ of  Mozart’s Piano Sonata, K. 284, first movement, exposition, an exquisite development of the opening theme, which relationship may even have been the composer’s deliberate narrative.

    Then imagine, instead, blanking out bars 1-8, which we might call Mozart’s main ‘idea’ here. The latter recomposition may seem more radical, but in fact probably makes vastly more overall sense of what we call the exposition.

    The former is embarrassing—anyone schooled in Western art music can hear how gauche and unviable it is. Yet mine, though also a pointless mutilation, and an equally lamentable model for students of Western art music, would nevertheless work quite well as it were ‘grammatically,’ don’t you think?

    Start the sonata at bar 9, and there’s no immediate problem; it would take a while until even as a highly expert listener you smelled a rat, and if you insist on recomposition, then at least mine has the virtue of demonstrating other than in words that to understand the ‘second theme’ as some kind of insertion, or ornamental, is what I believe our scientific colleagues call: an error.

    To convey the harmonic-contrapuntal function of the K. 284i, ‘second theme,’ I recommend Salzer’s 1962 (in English) distinction between prolongation ‘of’ and prolongation ‘to,’ for which there is an elegant graphic notation, forward- and backward-pointing arrows.

    Jonathan Dunsby

    Professor of Music Theory

    Eastman School of Music

    Sign In or Register to comment.


    • 9 Comments sorted by Votes Date Added
    • Naturally I disagree with Jonathan’s assertion. To be sure, anyone schooled in the post-1800 views of sonata form, which tends to view a second theme serves as a vital counterpoise to the main theme,  would no doubt recognize how gauche and unviable and embarrassing  it would be to remove the start of the second theme. 

      But as I have argued, eighteenth-century theorists, not so schooled (i.e., not so indoctrinated) did not view the second theme as a dialectic counterpart to the main theme, and they would not have been so embarrassed or found it so gauche or unviable—at least not from a tonal or design standpoint.  Without this insertion, the exposition makes tonal sense, and according to the theories of the time, it also makes formal sense. Indeed, in a number of movements, the “middle theme” (what we today call the start of the second theme) was omitted during the recapitulation. This omission is a strict violation of the "sonata principle"--but it happens frequently in the sonata movements of the time. For eighteenth century theorists, to regard the start of the second theme as an insertion was no scientific error: it was standard procedure.

      On the other hand, removing the “middle theme” of K. 284,i, does ruin the narrative trajectory of the exposition, much as one would ruin the narrative trajectory of an action film if one removed the comic interludes. Even if these interludes are extraneous to the plot, strictly speaking, they are vital moments for setting the action in relief. Likewise, without its insertion following the Grundabsatz, the exposition of K. 284/i is just non-stop action, with no relief, and for this reason the results indeed sound gauche.

      To convey the harmonic-contrapuntal function of the K. 284i, I would suggest looking at Schenker’s analysis of Beethoven’s Op. 10/2/i, where he reads the V at the start of what he labels as the “2. Ged.” as lying in the middle of a larger motion to V/V, which arrives right before the start of what he labels as the Nachsatz of the second theme.   

      Poundie Burstein


    • Oops, in the above I meant to say "Quintabsatz," not "Grundabsatz."

      Poundie Burstein


    • The choice of piece obviously biases the issue--this first movement is unusual in having such a clear two-part main theme with two initiating gestures. Therefore, in a very basic structural sense (tonal and formal structure) there is an excision (of mm. 1–8) that happens to work. The problem with that excision is rhetorical rather than structural: the main theme is very clearly (to my perception at least) a symphonic gesture. The symphonic drum bass of m. 9 however, would typically need some preparation. It could be as little as a couple tutti chords, a la the Eroica. So maybe the appropriate recomposition is to change the second half of m. 1 so that it is just two strikes on the tonic chord, then cut out m. 2–8. That would work.

      But Poundie is also right: even though it would trivialize the piece, cutting out, say, mm. 22–37, is at least grammatically admisable in an 18th-century style. (You don't actually say what of the second theme to cut, but m. 38 seems to be the only place where you have a distinct enough initiation that it could follow the MC in m. 21). We're not accustomed to this, because the composers we care about usually weren't tossing off sonata forms so lightly as to go directly into cadential material after the MC. But if you look at, say, the overtures of composers lost to the standard repertoire like Jomelli, Galuppi, Pergolesi, Leo, Hasse, etc., I think you will find this sort of thing often enough: MC in the home key followed directly by cadential material in the key of the dominant.

      --Jason Yust


    • An addendum to Jason's analysis: If mm. 1-8 of K. 284/i were excised, then the movement would begin with what Gjerdingen calls a Quiescenza and what Hepokoski and Darcy call a "circular" 8-b7-6-n7-8 pattern. Each source says that, in music from around 1750-1800, this design is more commonly used as a post-PAC tonic extension than as a beginning, but one can nonetheless find examples, even many examples, in which this kind of music initiates a movement.

      -Scott Murphy

      Professor, Music Theory

      University of Kansas

    • Poundie is surely right. I would only add that  there is a chromatic voice exchange within the tonic in bars 1-33, which is very common. See Eric Wen's analysis of K. 550/II in the Schachter Festchrift, and his unpublsihed conference analysis of Haydn's Sym. 99/I.

      The notion of the secon theme as parenthesis is common in Tovey; see Kimball, G. Cook. “The Second Theme in Sonata Form as Insertion,” The Music Review 52/4 (November 1991): 279-293.

    • In a rare moment of what may be seen as critical hubris, I did call the second theme “an exquisite development of the opening theme, which relationship may even have been the composer’s deliberate narrative,” and with all the good will in the world I find it hard to co‑conceptualize such a creative gesture as also an insertion in the 18th‑century sense, or in Tovey’s.

      I appreciate the fascinating points in this thread—not least if I may say so Scott’s generous stylistic validation of my own rather implausible excision, and also Poundie’s analogy with Schenker on Beethoven’s Op. 10/2/1, though when Schenker writes that “the first theme…and antecedent of the second theme are elevated to the status of an organic whole by the force of the melodic progression and arpeggiations,” not to mention the whole rhetorical context of Schenker’s synthesis here, I’m not sure I then entirely get the point or that Schenker would either, since delay (in this case delay of level C's initial move to VI, let alone the ensuing V of V) never did mean structural excisability, or we'd lose all our recaps too!

      But to put my general point more crudely, is it at least possible that J.G. Portmann just didn’t get it, in this case, K. 284i; failed to understand musically the intricate relationship between P and S (or whatever you want to call them) and defaulted to fashionable pseudo‑linguistic theory?

      That kind of thing does happen all the time, then as now. 

      Jonathan Dunsby

      Professor of Music Theory

      Eastman School of Music

    • Since the nineteenth century there has been a tradition of regarding the second theme of a sonata-form movement as forming a vital part of the dialectic that frames the movement. Remove the second theme, and the resulting opposition that seems to serve as the very essence of the movement is gone, and thus the movement’s structure would be ruined.  Furthermore, since the second theme of the exposition often launches the prolongation of the tonic of the new-key area, omitting the start of the second theme would cause the tonal structure of the exposition to collapse as well—though again, without the dialectic opposition of themes, the whole dramatic point of the movement is already made nonsensical.

      Musicians of the eighteenth-century, however, did not seem to share the outlook described above. When I first saw Portmann’s analysis, I was excited to see a type of proto-Schenkerian analysis of a middleground—but I was disappointed that his reading seemed to me to be just plain wrong. How could he have missed the start of the second theme? Didn't he realize its vital importance? Likewise, when I first read Galeazzi, I was surprised that he said that the "characteristic passage," which he also called the passo di mezzo, could be omitted. How could you omit the start of the second theme! Wouldn’t that make hash of the structure?  And why did Galeazzi call the second theme a "middle passage"—what is it in the middle of?

      As I have become more acquainted with music of the third quarter of the eighteenth century, however, the attitude of Portmann and Galeazzi started to make more sense.  In the manner that Galeazzi noted was possible, there are many expositions from this time that plunge from the departure from the main key straight into the periodo di cadenza, without any intervening "characteristic passage." Furthermore, there are many movements in which the "characteristic passage" from the exposition is excised during the recapitulation. However odd this might seem from the standpoint of sonata-form of the late eighteenth century onward, this was standard operating procedure in music of the Galant era, as Jason Yust rightly observes. This practice also accords with comments by Riepel and Koch, who discussed analogous passages as  insertions or parenthetical passages.

      With this in mind, I reconsidered Portmann’s analysis of Mozart’s K. 284/i. It has long seemed to be dogma in Schenkerian studies that a standard second theme in a major-key exposition should normally prolong V—though, as I suggested in my previous post, Schenker was far more flexible in this regard that many of his followers.  Might the tonal structure of the second theme of Mozart’s K. 284/i be fruitfully understood in a manner somewhat similar to Schenker’s reading of Beethoven’s Op. 10/2/i—that is, might the local tonic at the start of the second theme (which in K. 284/i is only implied on the surface of the music) be understood to lead to the deeper-level V/V that is so strongly pronounced in mm. 34–37? In other words, in the manner described by eighteenth-century theorists, could the start of the second theme in K. 284/i be regarded as a type of interpolation in the midst of the dramatic flow?

      To help demonstrate this possibility, in my recent SMT-V video article I recomposed Mozart’s exposition by excising this passage. To alter a Mozart work in such a manner inevitably will be heard as a desecration, not as an improvement! The question here is: what is lost by this recomposition? If you feel that excising the passo di mezzo here removes the dialectic opposition that serves as the very essence of this exposition’s structure (as described in the opening paragraph of this post), then you would regard such a recomposition as entirely ruining the structural integrity of the exposition in a manner that is simply not viable.

      But if you feel that the passo di mezzo could be understood as an insertion—a notion that I feel is fully in accordance with eighteenth-century practice and  theory (as described in the third and fourth paragraphs of this post)—then removing this passage does not ruin that underlying tonal structure or design. It does make hash of the drama: the unrelenting action, with no Einschiebsel (insertion) to break things up, becomes monotonous; Mozart would not compose such a tedious exposition. But I maintain that the altered version is structurally viable, and that this understanding accords with the theories and practice of the time.

      Since most listeners no doubt would find that altering a Mozart piece is disturbing, I initially thought of demonstrating this possibility in my video article by instead having the Portmann piece altered by removing its passo di mezzo.  But if you want an example of another piece for which shortening it in such a manner can be clearly imagined and that would not cause so much distress, you might look at the symphony by A.C.F. Kollmann that Kollmann analyzes in his composition treatise, An Essay on Practical Composition, first published in 1799 (a copy of this symphony may be found in the Da Capo reprint of Kollmann's treatise). This symphony of Kollmann’s (which is not on a Mozartean level, to say the least!) is used to demonstrate what we now call sonata form. In this piece, there is a "characteristic passage" in the exposition (mm. 25–32) that Kollmann omits during the recapitulation. In other words, in the piece that he cites as an exemplar of standard formal practice, Kollmann evidently did not consider the start of the second theme to be so essential that it could not be excised without ruining the work’s structure. Indeed, one could easily imagine that the start of the second theme (mm. 25–32) could be removed during the exposition as well, without harming the structure of Kollmann's symphony—in fact, I would argue that removing mm. 25–32 (and perhaps also mm. 1–8)  actually would improve Kollmann’s work by tightening its structure. (One might further argue that Kollmann’s symphony would be even more improved by removing all of its measures—but that’s another story!).

      In answer to Jonathan’s question: is it possible that Kollmann was simply wrong? The answer is: absolutely. As I have argued in a series of publications, an analysis (at least in many of its manifestations, such as Kollmann's) does not present a scientific proof. Analyses are like performances: you propose a way of hearing a piece, which others find either convincing or not.

      Before dismissing Portmann’s reading out of hand, however, I would encourage people to consider whether it is Portmann who fails to understand the crucial structural role of the second theme, or if it is we who misunderstand its role. In this light, I would encourage people to at least consider whether Portmann's reading—even if it seems so out of sync with modern approaches that we initially might regard it as absurd—might nonetheless offer a refreshing and exciting possibility for hearing and performing Mozart’s K. 284/i. That is, in considering Portmann’s reading, I would encourage people to ask themselves: does trying to understand the start of the second theme of Mozart’s piece as an inserted passage, lying in the middle of a larger motion to V/V, seem to be a rewarding way to hear and perform this movement?  If not, well, then you will not be convinced by Portmann's reading (though I would hope you would at least recognize that others might reasonably hear it somewhat differently than you do). On the other hand, If you do feel that Portmann’s reading offers a fruitful and suggestive way for understanding this movement, do realize that your hearing accords with a long tradition of others who are well schooled in  Western art music, including those theorists from the eighteenth century mentioned above.





      Poundie Burstein


    • I defer to nobody in my love of scholarship and admiration for Poundie's, in this case, in particular.

      If one were persuaded that Mozart here is responding to what he perceived as a typical weakness in contemporaneous compositional practice, then perhaps there is some common ground between those who try to follow the musical thinking that a composer inscribes in a masterpiece and those interested in how a work might have turned out otherwise.

      But on the other hand, I doubt that Mozart himself would tell us that he was responding to contemporaneous practice, so much as transcending it. when noticing it at all.

      It is, BTW, focused research into actual musical creativity that has always fascinated me in Schenker's essays and books.

      Jonathan Dunsby

      Professor of Music Theory

      Eastman School of Music

    • Further to my earlier comments, see also the paradigms in Suurpää, Lauri. “Continuous Exposition and Tonal Structure in Three Late Haydn Works,” Music Theory Spectrum 21/2 (Fall 1999): 174-199.