On the other hand, I did reach early the conviction that ragtime and sonata forms ought to be perfectly suited to each other. My composition teacher averred that form followed function and since ragtime was dance music, primarily, its form was not necessarily amenable to the functions of sonata form. Another professor with whom I shared a fondness for ragtime held a different view, that certainly ragtime and sonata form should, in principle, be a workable fusion of idioms. It'd just be a matter of figuring out how.
The axiom that form follows function can be flipped around, that functions can be assessed by a proper understanding of form. If I were to jettison any residual understanding of sonata forms predicated on 19th century theory or, to put it more polemically, 20th century givens as to what 19th century pedagogy on sonata form at the dawn of the century (either!) were supposed to be, I might find that it would be easier to assess the possible ways to have sonata and ragtime synthesized if I did not assume a class stratification between "high" (sonata form) and "low" ragtime as dance music).
So ... one of my lifelong projects has been, among others, to explore how a composer who loves sonatas and fugues as procedural developmental approaches and also loves ragtime might arrive at a synthesis of the stylistic idioms of ragtime with what we'd rather broadly call sonata form(s). To paraphrase Iannis Xenakis, the most ambitious forms of interstellar travel technology can provide for us may not carry us so far as liberation from our mental shackles could.
One of the gentle warnings Leonard B. Meyer had about musical analysis is that we should be careful not to forget the distinction between musical analysis as a matter of form and musical analysis as an observation of process. That these two aspects of musical analysis often overlap does not mean we won't run into trouble if we misdiagnose early on which of these two aspects of analysis will be most relevant. The kind of modular architectonics that are useful for delineating themes and transitions in a sonata form may be somewhat useless in evaluating the contrapuntal processes guiding a fugue. While linear expansion and development in a fugue "might" give us insights applicable to a sonata form the expectation of a continuously developing thematic economy may run aground in explicitly dance-derived forms, so the way we'd analyze a Bach fugue will be of little help in assessing a ragtime by Joseph Lamb.
Which is not to say that study of counterpoint and advanced forms won't be beneficial in finding relatively new ways to approach ragtime as a style. Being thoroughly steeped in both the history and literature of the sonata as well as the history of ragtime can open up possibilities for the fusion of both idioms that may remain, as yet, under-explored.
Of course to explore those possibilities we may need to establish the basic observations that may be made about these idioms. Rather than attempt to discuss ragtime in terms more suitable for the unfamiliar, I'm simply going to briefly go over the basic outline of the "Joplin" rag as it's commonly described.
AA BB A CC DD
Give or take an introduction, any transitions, and a coda, that's the basic formal outline of a ragtime.
We know that not every theme will be repeated and there are cases in which a theme that appears earlier may appear again in a modified form. For instance, the B strain in James Scott's Modesty Rag is recapitulated at the end where the "D" strain would be and it's slightly recomposed. This example alone should satisfy for the skae of this essay that recapitulating the B group as the final strain in a ragtime already opens up possibilities for a sonata form cast in a ragtime style. For that matter, where the return of the A strain would typically be could very easily be replaced with what in more traditional understandings of sonata form would be the development section. The syntactic climax of the return of the A material in a ragtime may be spatially displaced to a point later in a ragtime than we might expect it to conventionally appear but the syntactic climax of the reprise would still take place. If we bear in mind a simple proposal, that the A or B sections do not have to refer to explicit, literal repeats but may stand in for transitional material, then what we'd call the BB section or B section of a ragtime can be the space in which the second (or also third) thematic strains can be introduced with some suitable transitional materials.
This would be a way of synthesizing ragtime and sonata form in cases where we have sixteen-measure ragtime strains. Of course this would not be the only way a sonata form in a ragtime style could be composed. In fact if we were to make use of material more characteristic of early (rather than late) 19th century sonata materials we might find themes are more apt to be as short as eight measures. If our thematic strains are 12 measures or less immediate and literal repetition might be ill-advised and it might also be less effective, given the fact that so very often thematic differentiation is a weakness in ragtime compared to other styles, then any structural repetitions we might plan to introduce that could bridge the conceptual space between a ragtime and a sonata form would have to lean more toward a conventional understanding of a sonata form than a ragtime.
This basic adjustment to small-scale sonata form to resemble the style of ragtime is actually not the least bit difficult. Rather than an AA BB paradigm as we'd expect in ragtime we could substitute a short repeating exposition that would present as AB AB that could fill the commensurate durational space (George Rochberg's concept for this, called "time-space" is inelegant and bluntly literal, but it's arguably the best possible term we're looking for. After all, for the aim of this essay we ARE looking at corresponding durational time-spaces within sonata form and ragtime for any possibilities of fusion.
So that's to say this, if a ragtime strain is conventionally understood to be sixteen measures long and a conventional ragtime has an AABB procession before A returns, then the number of measures in that durational stretch would be sixty-four measures. This means that if you have a small-scale sonata exposition (something in the zone of forty measures) then you would be able to fulfill the cumulative AABB in a ragtime best with an AB exposition that has an internal structural repeat. A simple but perhaps necessary observation about sonata forms in the 18th century is to note that the repetitions should be considered structural. If you keep this in mind then an exposition with two themes can be mapped out as
exposition repetition development recapitulation
A B A B C A B
Oh, wow, what does that resemble?
verse chorus verse chorus bridge verse chorus
In strictly modular, macro-structural terms, the sonata form with a repeating exposition that's observed before the development and recapitulation can be thought of as remarkably similar in its overall concept to a standard pop song format. There are crucial, substantial differences in thematic developmental economy between a sonata form and a pop song, to be sure, but the differences are not necessarily reducible to a matter of sheer modular form. Rather, we're looking at a continuum of differences in the realm of procedural development and expansion of thematic content. When advocates of high art musical culture dismiss popular music they do not necessarily reject the elegant simplicity of pop music at the level of macro-structural considerations. They're rejecting the rudimentary presentation and repetition of thematic materials without long-form development.
Okay, fine, but who says this isn't possible in vernacular or popular idioms? Just because a majority of composers or musicians don't try out the possibilities of a fugue based on blues or ragtime vamps doesn't mean Henry Martin and Nikolai Kapustin haven't written such fugues. And even if you were to decide their attempts at a fusion of blues or jazz idioms with fugal writing don't "work" this can be a case in which the question is not whether the experiment is worth doing (at any rate I'll say directly I think it is) but why this or that experiment somehow falls short of a successful fusion of diverse musical idioms.
So we've established at a conceptual level the potential for macro-structural overlap between an 18th century sonata form and a 20th century pop song. Given the lack of redundancy we observed at the outset in the AABBACCDD of conventional ragtime don't we run into the problem of a lack of redundancy across themes? Won't we run into a problem in which it seems that the recursive interiority of ragtime strains preclude the possibility of thematic development?
Well, no. The short reason for this is to say that at a melodic and harmonic level the boundaries between a ragtime strain, as a form of late 19th century music, and a sonata form from the same period, are permeable. The long reason for this was my survey of early 19th century guitar sonatas in which I highlighted the ways in which a number of sonata themes by Giuliani and Carulli lent themselves to simple transformation from sonata material to ragtime material. For all of that go over here.
But if you still don't believe me and want a case study for how an early 19th century sonata form for solo guitar could be recomposed into the style of ragtime, here's an example I worked on this last weekend (after the break).
You'll see that this would be a case of an ABABCAB variant of sonata form with ragtime elements. The Sor etude in E flat required very little manipulation in terms of melodic lines or harmonic vocabulary to make the transition from a sonata form to a sonata-ragtime hybrid. A slightly different (in substance) adaptation of Sor's Op. 29, etude 10 in E flat into a ragtime/sonata hybrid can be heard over here starting at 02:45 and ending at 07:09.
Now that approach to hybridizing sonata form and ragtime depends on the brevity of the core themes. We'd be talking about expositions that are essentially only about as long as playing through two ragtime strains, and then repeating that material. What if we wanted to synthesize ragtime and sonata form at a larger scale, something approaching the traditionally longer or larger sense of scale we expect of sonata forms in the art music tradition?
To the books, two in particular. I've found William E. Caplin's treatment of classical form to be quite useful. Seeing as his books deal with themes in their modular elements at different hierarchical levels his scholarship permits a composer to think through the different structural levels at which a sonata form and a ragtime could overlap in terms of two, four, eight, and sixteen measure phrases. Caplin's work is also useful for highlighting tendencies such as the absence of modulating transitions in some slow sonata forms. But while Caplin's book is fantastic for an overview of macro-structural considerations it's not quite as strong on what might be called the syntactics of thematic differentiation. For that sort of thing, and for a consideration of what Rochberg called "time-space" we may benefit by turning elsewhere.
In my case elsewhere does not just so happen to be ideas formulated by James Hepokoski and Warren Darcy in their book Elements of Sonata Theory. Now I can appreciate why some feel that their book doesn't do enough to address the boundless variation in Haydn's approach to what's generically called sonata form. I can appreciate why some feel that the book overemphasized Mozart and Beethoven at the expense of Haydn, who was indisputably the big influence within his own generation. I happen to adore Haydn's music and admire a lot of Beethoven while preferring Clementi and Reicha to Mozart by a longer shot than Mozart fans will think I have any right to.
I've blogged at length about the viability of the Type 2 category of sonata in explicating early 19th century guitar sonatas. But there's something else in the taxonomy of sonata theory by Hepokoski and Darcy I haven't really seen discussed. The thing is that if they outline five sonata types then we should no longer take as given an academic bromide to the effect that sonata form(s) have been obsolete. Maybe the stereotypical high Romantic German sonata form is obsolete. But if two scholars can propose no less than five general "scripts" that can be used to compose sonata forms in 18th century terms those are five scripts we can make use of right now.
When H&D discuss "space" we can think of it as something like what George Rochberg called "time space". For a larger scale form we can regard the B space (BB, since we might as well regard the repetition of the B strain in a ragtime as non-negotiable and structural) as inclusive--it will include not just what we'd call the B strain or Group 2 thematic material that's formally identifiable as such, it would include any modulating transition that gets us there. Caplin's observation about slow-movement sonata forms is here immensely valuable--Caplin noted that in slow movement forms the transition into the second group is often elided into the second group, makes anticipatory use of Theme 2 elements or skips a transition altogether if the movement starts in a minor key and Theme 2 is in the relative major.
That means that if we started with a minor-key ragtime strain then our B material strain could start in the relative major with no preparation. It might recapitulate as minor-key material later on and no transitions would be needed. After all, if we had a minor-key A strain returning in a recapitulation then the B strain could be brought back with no need for a modulating transition. A sonata form of this sort would be so skeletal as to seem exempt from being a sonata at all but if we go with Hepokoski & Darcy's taxonomy of sonata types this kind of ragtime/sonata hybrid could look roughly like this:
AABB (development) A B or ...
AB AB (development) AB or ...
AA BB A BBA
AB AB AB
and all of this could fit into the Type 3 category (versions 1 and 2) or the Type 1 category (version 3).
If you simply subtracted all the development section in the previously presented E flat major ragtime sonata you could get the fourth version.
If you've read this far you know I'm presupposing your familiarity with sonata forms that have an exposition with a number of themes; a development section (which may be optional depending on the type of sonata form we're using); and a recapitulation (which may be an incomplete or otherwise truncated recapitulation depending on other variables I won't discuss here).
All that said, in many cases larger scale sonata forms frequently have three rather than two core thematic groups. Whatever fusion of ragtime with larger scale sonata forms we arrive at will need to be able to account for moderately long expositional spaces.
But first let's take a detour. We need to review some details about ragtime that can be too readily missed by those who aren't already steeped in the style. For instance, there's no obligation to necessarily have four distinguishable strains. For ragtime, we don't "have" to have D material at the end. James Scott could recapitulate and recompose his B strain with a new ending, as he did in the Modesty Rag. So we're going to have that interrupting Trio but Scott presents us with a ragtime that has an A strain followed by a B strain that has an A return and a B return. In other words, the possibilities of composing a sonata exposition even with the essentially modular strains of ragtime is pretty easy to imagine.
For that matter ... having a ragtime open up with minor key themes was also done a couple of times by Joseph Lamb. Eh, we'll just go with Arpin for now. And then there's Nightingale, another Joseph Lamb rag that has a minor key opening theme. So if you wanted to you could open with minor-key ragtime themes and get all Beethoven by having them recapitulate in parallel major modes. You could even have a formal development section.
The development section, in this large-scale fusion of ragtime with sonata form, would almost by necessity be where in a traditionalist ragtime the A material would be coming back. This means that the syntactic climax of the A material would be, as we've noted earlier, pushed forward. If we were to conceive of a sonata form and a ragtime as a three-dimensional space that could map out the durational time-spaces for thematic regions we could look at it this way:
Now there are a number of things that could be said about the recapitulatory space in a ragtime/sonata fusion. The first is that in a recapitulation region the example of Haydn shows us we could substantially reconceive the style or tone of a theme within a recapitulation area. To give a non-random example, I could take a them that was presented as a ragtime strain in the sonata exposition and recompose it in the style of Texas Gospel blues in the form of a short set of variations of Theme 2 and even have this section played entirely as a slide guitar solo. When Hepokoski & Darcy discuss what they call deformations that are possible in sonata forms they don't mean deformations in some pejorative way. The way I took their meaning they were saying that deformations are not only not deformations in a negative sense they are to some degree "expected" surprises that subvert expectations while satisfying them. One such example could be transforming a theme that was a ragtime strain in an exposition into an aggressive blues slide guitar version of the Theme 2 and Theme 3 materials in the recapitulation zone. It would be important to maintain the linear and fundamental harmonic elements of the thematic materials, obviously, but drastic stylistic mutations in a recapitulation space are normal in a sonata by, say, Haydn.
The reason this could work is because of another concept Hepokoski & Darcy introduced in their Elements of Sonata Theory, the concept of rotation. The simplest way to describe a concept that has tripped up some scholars and reviewers is to say that if A, B and C are presented in the exposition then we should hear an observation procession of A, B and C in the development section as well as the recapitulation section. Or in my ad hoc parlance, Groups 1 through 3 need to be presented in the same order across the major sections of the sonata form. This isn't always the case, certainly not in Chopin's B flat minor piano sonata, for instance, but in the context of a sonata where we plan to drastically change the style or idiom of a theme in a recapitulation then what rotation does is set up the expectation that A, B and C definitely will return but that there is room for surprise as to precisely how or in what form A, B and C will return. If we were to humor an unsuspecting audience it would seem best to save the dramatic mutations of recapitulation region themes to B and C material while leaving the A material in tact. This would, not coincidentally, also be the most plausible way to preserve all those syntactic points of commonality between what we 'd identify as a traditional sonata with a traditional ragtime.
And, of course, if you've already read the chart with the green, yellow and blue boxes you'll have seen all this just minutes ago but I'm writing this on the surmise that not everyone will be able to see the image and that some kind of written explanation of what I'm proposing is featured in addition to a chart that I hope can summarize things in a handy visual way. Conceiving of a sonata form and a ragtime form as a three-dimensional temporal space, like a row of wooden blocks with which you might make something, makes this easier to think through and quite a bit easier to try to describe. It allows me to demonstrate what the points of durational correspondence can be between a sonata form as we've been given it in so many a textbook and ragtime as we may have read it described.
In summary, the possibilities for a fusion of the style of ragtime with sonata form have been latent within the ragtime style since its inception. That any number of scholars and musicians and composers have not taken this conceptual possibility seriously is just a matter of people having not considered ragtime a style able to introduce what some regard as loftier and more substantial intellectual approaches to the musical materials. I consider this a big and terrible mistake but an understandable error for people to fall into if they are immersed in an approach to musicology that treats ragtime as somehow categorically different from 19th century art music in a more "serious" tradition. I believe that one of the core problems in art music repertoire has to do with pedagogy and the pedagogical problem is, as Richard Taruskin has been asserting steadily for a decade or so, a chasm between what he call the repertoire canon and the academic canon. That ragtime is part of the repertoire canon, the canon of music which people will happily part with their time and money to hear, is pretty much beyond dispute. That ragtime is by and large not a particularly serious presence in the academic canon would also seem fairly hard to dispute.