Drawing on Jason Yust and Drew Nobile to explore rock and ragtime sonata possibilities

Ethan Hein has a new post up about hypermeter and though I’m not foremost among Beatles’ fans I liked his explication of hypermeter in their songs.  In the post he mentions Jason Yust’s Organized Time.  Something Yust mentions early in his book is that though he drew up Schenkerian categories one of his chief criticisms of the Schenkerian tradition is its failure to deal with rhythm and if a theory fails to develop an adequate account of rhythm then, to put it very sharply indeed, how is it really ultimately very useful for an art that is about organizing time and space?  It is here that I interject that there’s good reason Augustine of Hippo decided to start De Musica with the books on rhythm before he ever got around to discussing melody or harmony.  In fact he never moved beyond the treatment of rhythm and never finished De Musica.

In the twenty-first century it’s not that hard for us to consider that rhythm is foundational to music.  Even pitch is ultimately an expression or variation of a rhythm.  When vibrations in the air reach 440kHz it is a pitch we can hear but the existence of infrasound tells us that not all pitches are at frequencies that we can hear.  Some pitches have too few cycles for the human ear to register them as audible sound just as some pitches cycle so quickly dogs can hear them but we can’t.  Pitch can be thought of as an extension of rhythm into the realm that we hear discreet notes.

Yust has a zinger of an observation on page 43 of Organized Time:

Exact repetitions are usually ignored in tonal analysis … Our tendency to ignore repetition on the music is therefore mostly justified. Yet a general understanding of the structural implications of repetition is nonetheless imperative when dealing with repetition that occurs at deeper levels.

If we are analyzing the first movement of a piano sonata from an 18^th century Viennese composer and ignore the repetition of the exposition we don’t lose all sense of what the music “does”. But as Jason Yust and Yoel Greenberg and Leonard Meyer have pointed out over the last half century, if we find that sonata forms exist in which there is a simple exposition and recapitulation with no development, or if there is an observable trajectory from the 18^th to the 19^th century to have sonata forms weighted more toward ternary than binary forms then it would be a mistake to attempt to analyze an 18^th century sonata that has an exposition and a recapitulation with two themes and find it wanting as a sonata because there are no transitions and no development passages.  Too doctrinaire a concept of sonata form(s) leads us to dismissing something as not being a sonata form at all when it is a sonata form. 

Yoel Greenberg’s bottom-up 2 out of 3 criteria approach seems useful.  Does the exposition repeat?  Does the first theme come back at some point in the original key?  Do the secondary or tertiary themes presented somewhere in the first half of the work return but in the same key as the first theme?  If you say “yes” to all of these you’ve fulfilled the criteria of a sonata in Yoel Greenberg’s taxonomy.  Greenberg stressed in How Sonata Forms, though, that you don’t actually need all three of these.  His bottom-up identification checklist is handy because it can remind us that many sonatas have just some of the structural aspects of a “textbook” sonata form.  As many scholars and historians of music have noted the fully repeating exposition got abandoned in performing sonatas in the 19^th century.  The same can be said for repeating the entire development and recapitulation sections in performance.  When the early 19^th century guitarist composer Wenzel Thomas Matiegka included instructions to actually repeat both his expositions and his development/recapitulations with different endings for the latter in one of his Op. 31 Sonate progressives he was “old school”, overtly drawing on conventions in sonata writing that were falling out of style. Matiegka was also fond of wrong-key recapitulations that eventually self-correct so even Greenberg’s idea needs to be used as a flexible heuristic rather than an ironclad rule.

Music theory has a strong tendency to confuse form with forms, or, more precisely, to confuse the study of form with that of composers’ recipes. Given that the theory of form belongs to a branch of music theory originating in practical compositional teaching, the tenacity of this conflation is understandable. For musical form, the speculative tradition has never quite caught up with its practical tradition. When we look to eighteenth-century writers such as Koch … for a theory of form, what we get is largely recipes; a how-to for building coherent forms out of individual themes. Certainly there is much theory of form implicit in  such recipes and how they are explained, and study of recipes, both as described in treatises and manifest in works, is a worthy undertaking in its own right. But recipes are highly contingent upon time period, composer, and genre.  Good recipes rely upon principles of form to produce pieces with a certain amount of predictable formal structure and coherence, but they can hardly exhaust the possibilities of formal coherence made available by those principles.

Yust, Organized Time page 282

One of Yust’s most simple observations was to point out that as European concert music kept being composed there was a tendency for composers to create formal, tonal and melodic closure in pieces that withheld rhythmic/metric closure.  Beethoven was particularly significant in using this strategy.  In contrast to the claim Peter Kwasniewski has made about how the great masters of the great tradition of classical music created a sense of closure Jason Yust pointed out that relentlessly withholding rhythmic/metric closure long after a formal or syntactic sense of closure had been achieved in his sonata forms was one of the most identifiable traits of Beethoven’s heroic and late styles. Now for people who love Beethoven altogether that is what makes his music full of “struggle” and makes it “profound”.  Sometimes I appreciate that but sometimes I can hear how and why a few of Beethoven’s contemporaries heard him as a stentorian bloviating gasbag, too! Haydn knew when to wrap things up and put a bow on things.  Maybe in our era that seems too conventional but he was not seen as too conventional in his own day.

Ethan pointed out that hypermeter is important for a variety of reasons and one reason it is important to have variable and varied phrases is you don’t want to bore your listener.  If things sound too square too often people disengage as listeners.  If you’re playing utility music where the point is that the hearers are supposed to be doing something besides active listening that’s another matter.

The withholding of closure is a concept I’ve been thinking about lately and there are a variety of ways of withholding closure in writing a piece of music.  Yust has done a formidable job of explaining how in classical pieces rhythmic closure can happen at multiple levels and that a closure at one level can be counteracted by a lack of closure at other levels. 

But in terms of a theory of form I’ve been thinking about what makes a development distinct from an exposition or a recapitulation.  I am also thinking about ways to synthesize the vocabulary of rock, pop, ragtime and other American vernacular styles with the syntactic processes attributed to 18^th and 19^th century European concert music.  It is at this point I’m going to pivot to Drew Nobile’s Form as Harmony in Rock Music.

Early in his book Nobile introduces the concept of the harmonic circuit. I won’t attempt to get detailed on what the term means and how he uses it across an entire book but he proposes that when a verse and chorus or a verse, pre-chorus and chorus or a chorus and verse have run their course and cycle back on themselves a functional circuit has been traversed. He also proposes that once a song has had two full functional circuits it can end. 

With respect to a sonata movement drawing upon rock materials I propose that theme 1 and theme 2 could be as simple as functional circuit 1 and functional circuit 2.  Perhaps theme 1 is a 12 or 24 measure blues idea and theme 2 is a ragtime strain or a verse, pre-chorus and chorus module.  There may or may not even need to be a transition.  Given the flexibility of blues tonality there might just be a first group and a second group back to back.  The first group could have the traits of a verse-chorus unit and the second group or theme could function like a bridge that leads back to the first group.  That could give you a Type 1 sonata form in Hepokoski and Darcy’s Elements of Sonata Theory, a sonata with an exposition and a recapitulation but no development.

One of Jason Yust’s most striking and practical observations was how in 18^th century music secondary themes do not have to start on the tonic of their expositional key (Yust, page 81). Secondary themes can start on the V rather than the I of the new key.  Something could be said for the IV being as suitable.  Now why do I mention this?  Well, because IV-I vamps are common in rock and this means that if you have a sonata based on rock ideas you could start in A minor or A blues for Theme 1 and your Theme 2 could abruptly start on an F to C shuttle (Nobile’s term, I think) as the announcement of your Theme 2. 

If you were writing a ragtime and wanted to create a Type 1 sonata using traditional ragtime materials you could start with an A minor idea for you’re `A’ strain and then your `B’ strain (Theme 2) could be a thoroughly cliché V-I V-I vamp of the kind strewn all over Joplin’s rags and the new theme could be in C major.  If you wanted to create a Type 1 sonata using the modules of rag then after you bring back you’re A minor `A’ strain/Theme 1 you could immediately follow it up with the `B’ strain as your Theme 2 transposed down into A minor.  The closing strain could be new material or derived from Theme 1 or whatever but the point is that by this time you have created everything you need for a ragtime Type 1 sonata form that respects the modular construction of ragtime while taking up the syntactic parameters of a sonata form type outlined in the work of Elements of Sonata Theory.

But what if you wanted to create a sonata using rock or ragtime ideas and have a development section/  What principles could be provided for that?  My own guess, for what it’s worth, is to draw on Nobile’s concept of functional circuit to clarify a pattern that could be observed in 18^th century sonatas of the Type 3 variety.  It’s a commonplace in scholarship on sonata forms that in the development ideas from the exposition are developed in all sorts of ways but it’s rare for entire themes to be presented.  There may be ideas developed in this or that order. Hepokoski and Darcy have a case for how and why the concept of rotation informs a development. If an exposition has themes 1 and 2 than in the development we can at least guess that themes 1 and 2 could get developed in that order before they return in the same order in the recapitulation. 

Where rock comes into this, if an exposition could be said to be two fully delineated functional circuits (of whatever types); and if a recapitulation could be said to be two functional circuits that fulfill the syntactic goals of a sonata where themes 1 and 2 return in the same key; then a development could be a section which there are two spaces equivalent to the functional circuits of the exposition and recapitulation but which are in some way disrupted.  This would be where you would want to not let your thematic groups run their full course. 

So what techniques could you use to disrupt the completion of your two functional circuits? If your first theme is a 12-bar blues then you don’t let that blues run its complete course in the development.  Something, anything, that disrupts that formal closure can be introduced.  Perhaps an idea can be sequenced so as to prevent the blues theme from resolving because the sequence leads into the next functional circuit.  Or perhaps Theme 1 can be altered by having Theme 2 materials appear where we would expect resolution of Theme 1’s functional circuit.  Both of these are pretty simple options.  Maybe the ideas from Theme 1 show up upside down. Maybe Theme 2 is played backwards.  If you were playing with a ragtime-based sonata you might run the full length of your `A’ strain with all four phrases but you might traverse a new key region in every phrase.  At the risk of suggesting examples from the work of … someone … you can go here, here and here. (Update 5-27-2024 also here)

Whatever the case may be the functional circuits of the development are altered in such a way that they continuously move toward the return of the first theme in a way that would work both as a recapitulation in a sonata movement and as the shift from the bridge to the chorus or verse in a rock song.  Alternately, the reprise of the big `A’ strain in a ragtime becomes the double return in a sonata form and you adjust everything about the nature of the recipe accordingly.

For the purposes of discussing musical time as both time and space I hope to have established that, as a matter of pure or mere theory, there isn’t any reason you couldn’t create sonata forms using American rock and pop materials due to some purported constraints “inherent” to the material. 

These are not my most organized thoughts but I wanted to set them down anyway.