a new appendix to Ragtime and Sonata Forms, some practical approaches to ragtime/sonata synthesis

Over the years I have written about my fondness for ragtime and my interest in synthesizing ragtime with 18^th century galant approaches to large-scale form.  I wrote Ragtime and Sonata Forms back in 2020.  Since that time I have read a fairly big pile of books discussing musical time and space; concepts of form; and scholarly reconsiderations and renegotiations of what theorists would call sonata forms.  It has become clearer among scholars over the last two centuries that the whole idea of “sonata form” and “sonata forms” was and is a post hoc description of a very flexible set of practices that were never as schematic or rigid as 19^th century theorists made them out to be. 

 

As Leonard Meyer put it, in the eighteenth century sonata movements were conceived of and written as a flexible set of scripts, but in the nineteenth century they became works written according to plans.  I was telling a friend of mine that this distinction between a sonata as a “script” and a sonata as a “plan” has profound implications.  You can go “off script” any time you want and get back on script.  If you change significant parts of your plan then you may just need a whole new plan.  I enjoy Haydn more than Mozart or Beethoven and I have a fondness for thinking of sonata forms (since the term and the heuristic is still useful and unavoidable now that we’ve had it for two centuries), so I have been thinking of ways to play with sonata forms in a way that treats sonata forms as flexible scripts.  I have also attempted to cross reference approaches to the “punctuation” of musical themes in ways that allow for establishing temporal-spatial correspondences between ragtime and sonata forms. 

 

Ironically I have been helped not just by reading music theory treatises by Jason Yust, Yoel Greenberg, James Hepokoski and L. Poundie Burstein, I have particularly benefited from reading Charles Twombly on John of Damascus; Thomas Torrance on Trinitarian dogmatics; Roger Scruton on the philosophy of music; and then Ferdia Stone-Davis and David Brown’s respective critiques of Scruton.  In other words I dove into theology and philosophy.  Torrance argued that we need to drop the Newtonian conception of time and space in light of patristic witness that time and space are not absolute but are relationally contingent and part of the created order. 

 

So when Jason Yust proposes that we think of musical time and space overlapping in multiple dimensions that just strikes me as a conceptual transposition of the kinds of distinctions that, when Thomas Aquinas reflected on the Trinity, were explicable as filiation, procession and spiration in intra-Trinitarian relationships.  In other words, Christian theology going back to the patristic era and medieval theologies from Aquinas and John of Damascus dispensed with a conception of time and space as some absolute “receptacle”.  L Poundie Burstein’s advice that we avoid getting too rigid in thinking about musical blocks of space and time as fixed “containers”.  Christian theologians agreed with that concern centuries ago and Thomas Torrance was overt in advising us to drop that conception of time and space in the last century.  

 

But I am not writing all this merely to leave things in the realm of speculative philosophy and theology. There are practical applications of such an abstract theological and philosophical rejection of the penchant for Romantic era music theorists to fix sonata forms down into plans while bloviating about how bad convention is.  The irony that I personally savor is that back in the 19^th century writers like Richard Wagner claimed that art could do for religion what religion was no longer able to do itself because of the implausibility of the esoteric or incomprehensible dogmas.   I, obviously, disagree.  I think that if we draw from the doctrine of perichoresis in discussions of the immanent Trinity and the hypostatic union of Christ’s human and divine natures we have a concept of time and space at hand that lets us disregard a Newtonian concept of time and space as either absolute or as rigid containers, and there’s no reason we have to feel constrained by Romanticist concepts of organicism that generally have gotten applied in racist and white supremacist ways (these two are not the same but here is not the place where I want to get into that set of distinctions).

 

So, at the risk of relying on musical works composed by … someone I know … I have drafted a couple of charts that I hope can illustrate in some practical ways how ragtime can be synthesized with sonata forms.  I have drawn on ideas from William Caplin and also from Hepokoski and Darcy’s Elements of Sonata Theory.  Because “ragging the classics” is part of the tradition of ragtime it helps give us a precedent, a precedent where we know the boundaries between the popular style and the canonized works was always permeable from the birth of ragtime.  This means that taking public domain works from the field of classical music and ragging them up is easily done. 

 

So I will proceed, at this point briefly, to a couple of examples of ragtime-sonata fusions.

Example 1. Guitar Sonata No. 5: II. Ragged and Sor(e)

https://www.youtube.com/watch?v=Hhe-ok2oJSI

This sonata form is an adaptation of one of Sor’s Op. 29 etude No. 10, the one in E flat major.  That it is a sonata movement has gotten an explanation elsewhere. This first version of the work does not have a repeating exposition. There is another performance in which the exposition is repeated.  To hear the movement in context and with all scored repeats go here (the central sonata movement is 02:47 to 07:10)

https://youtu.be/5m5w0DFGbkc

 

In the analysis of sonata forms it has been a commonplace to say that repeating expositions are optional and that an analysis of a sonata doesn’t require taking repetitions into account.  Jason Yust has made some cogent arguments against such bromides and they are based on the observation that rhythm is rhythm, and repeats stipulate repetitions that are structurally significant.  That’s the version if you are not rushing out to read Organized Time.  The more relevant translation of Yust’s observation with respect to ragtime is that the structural repeats within rags are paramount. If you want to adapt ragtime vocabulary into anything approaching a set of sonata forms you should use a “bottom up” approach somewhat like Yoel Greenberg’s, or you should make a point of having your structural blocks of musical time and space correspond to both a “sonata” concept of time and a ragtime form.  Greenberg’s observation becomes salient here, noting that in the 18^th century grand binary form was just that, grand binary form, whereas in the 19^th century theorists began to regard sonata forms as giant ternary forms of an especially abstract kind.

 

What that means practically is that your best shot at a ragtime-sonata fusion  is to regard the Joplin rag (AABBACCDD) as a springboard for an approach to form in which the AABB section, taken as a whole, corresponds to the exposition and the CCDD section corresponds to the recapitulation.  The compromise between ragtime and a sonata form is the central A reprise section becomes your development and you will use you’re A and B strains as the basis for your C and D sections—this would be the case regardless of how you script out your sonatas.

 

In this work the exposition is either repeated or not.  When repeated Themes 1 and 2 follow back to back.  When the exposition is not repeated the A and B equivalents retain their order in the exposition and recapitulation and somewhat in the development.  Hepokoski and Darcy have used an idea called “rotation” to explain that the initial presentation of themes in a sonata indicates how they are likely to appear throughout the rest of the sonata movement.  This paradigm is pretty easy to apply to a rag.  There is a precedent in James Sylvester Scott’s rags to bring his B strains back as closing themes. 

 

Now I realize I have not actually discussed this guitar sonata movement in the video.  Don’t worry, I have a chart.  It may be easier to share the performances and then share the chart of the visual timespace of the music (the clunky term comes from George Rochberg, for you theorists who want to know where I got that term). 

 

Example 2. Guitar Sonata No. 6

https://youtu.be/awkeh0HVpO8

Sonatas generally are expected to have modulating and non-modulating transitions.  How would you incorporate those kinds of transitions into a ragtime sonata?  One practical approach is to be strict about the AA section of your rag, your Theme 1.  Any modulating transition you compose will necessarily eat up space that would be devoted to your B strain.  That means you can’t afford to use the old V-I alternating jams that Joplin loved to use because those tend to reinforce your tonic rather than move away from it. But you might find it difficult to just jump straight to a dominant key theme with no transition (we’ll get to whether and how that can be done in a later example). 

 

What you might find useful (needful) is a dependent transition that builds upon Theme 1 materials that leads to your second theme.  That’s what happens in this sonata.  The B strain (i.e. Theme 2) gets only one presentation in the exposition.  If this seems to compromise the internal repetitions that make ragtime what it is remember I mentioned that James Sylvester Scott reprised his B strains at the end of rags whereas, by contrast, Joplin was strict about his CCDD pattern.  If we drop Joplin’s reliance on four-strained rags then not repeating the B strain in the expositional half of a ragtime sonata is not a long-term problem because that B strain/Theme 2 is going to come back and get played twice in the recapitulation.  It might go without saying but you better love your ragtime strains in this approach to musical time and space with form because you’re going to be playing/hearing them a lot in this kind of work. 

 

If the AABB and CCDD sections of your rag have now become exposition and recapitulation zones then what happens to the central A reprise space?  That becomes a development and since many a development starts with Theme 1 derived materials you’ll still get the effect of an A strain reprise but now it launches into a larger than average space that is a sonata development.  You might want to confine your development zone to the length of more or less one ragtime strain for practical reasons.  You want to respect the constraints of the genre and it helps to know that many a Haydn exposition barely lasts longer than two minutes if you’re not observing all the repeat signs. 

 

Now when we get to the recapitulation we don’t have a repeating A strain.  This is a point at which a Theme 1/A strain derived non-modulating transition does the work of fleshing out the CC space of a ragtime sonata.  Since in a strict Joplin rag the A strain that returns is never played more than once most of the time it’s natural for the non-modulating transition to combine with the returned A to establish the CC zone of the rag and also be the first part of a sonata recapitulation.  When the B strain/Theme 2 returns it is going to get repeated just like a DD section in one of Joplin’s rags would be. 

 

Now here’s how the musical timespace would map out.

This approach has the advantage of being amenable to translation into ritornello form and, thus, to a mini-concerto format.  This extra modification allows the ragtime-sonata synthesis to become a duet for piano and guitar where the ritornello effect of the concerto is preserved and the structural repeats of ragtime are also respected.

 

Example 3. Sonata for piano and guitar (adaptation of the preceding solo guitar sonata)

https://youtu.be/ai315vMBuCk

 

And this is how that maps out in the proverbial musical timespace.

 

Example 4. Guitar Sonata No. 7: III. and IV. (includes a score)

https://youtu.be/We9F3iKhT7Q

The entire sonata from which these movements are excerpted can be heard here:

https://youtu.be/xd_QGDigJds

A valuable observation made by Hepokoski and Darcy in Elements of Sonata Theory is to distinguish within the expositional zone whether or not themes have an S function or a C function, whether the theme is a second theme in a secondary key or a closing theme that establishes a tonal/rhythmic/melodic resolution for the exposition.   What this means in practical terms is that it is often more accurate to describe sonatas as having three themes rather than two.  Can a ragtime-sonata fusion be written that takes into account that three-themed sonata expositions and recapitulations happen?  Yes.

 

The first theme still corresponds to the A strain of a Joplin rag. The internal repeat is still an inviolable pattern (unless you are ragging a classic as was done with the Sor etude). The modulating transition is still probably necessary.  In this case Theme 2 and Theme 3 in the secondary key need to fulfill the function of the BB space and the yet also provide some basis for the DD/Theme 2 recapitulation zone to come.  In this particular piece what the composer decided to do was create Themes 2 and 3 that have internal sequence and flow one into the other.  Theme 2 is 32 measures long so while it doesn’t repeat is sequentially presents thematic material in a way that fulfills the “punctuation” or sentential equivalent of the B space in a rag.  Theme 3 is also 32 measures long.  The two themes together are larger than Theme 1 even when repeated. The structural potential in this is something that we’ll keep off to the side for when the recapitulation comes.

 

The development here is short and goes by quickly.  Theme 1 returns in the way we would expect it to in a sonata form.  The modulating transition is replaced by a non-modulating transition. Theme 2 does not come back in the same way it was presented in the exposition.  It has been rewritten entirely.  Theme 2 is really a traditional shape-note hymn. The ragtime form it took in the exposition was a disguise.  When it appears in the recapitulation in its original form it is also played in bottleneck technique after the style of Blind Willie Johnson.  An outworking of Hepokoski and Darcy’s idea of “rotation” is being used here.  The idea is that if you observe the rotational order of the themes strictly in the exposition and recapitulation you have set up an expectation in the audience for Themes 1, 2 and 3 to show up in a strict order.  As a compensatory freedom, Theme 2 can be rewritten in a completely different style and this will not violate the listener’s expectation that Theme 2 comes after Theme 1.  Similarly, Theme 3 can be dramatically revised into a different style (a 1970s R&B/rock sound) and so long as it retains the identifiable melodic content of Theme 3 an audience will still know they are hearing a version of Theme 3.  This also allowed for a Haydnesque propensity to go to town developing materials in the exposition zone that did not show up in the development section.

 

But the crucial element where respecting the structure of ragtime goes is this—Themes 1, 2 and 3 are presented in ways where, no matter how far afield the variations on Theme 2 go, the variations respect the strict internal repeat paradigm in ragtime.  I.e. AABB is still observed and CCDD is still observed.  The variations on Theme 2 in the recapitulation zone are internally consistent to their own space.  The return of Theme 3 as a coda allows Theme 2 to fill out, at some length, the CC section of a rag and the traditional practice of a secondary development in the recapitulation zone.

 

And how would that map out?  Like this.

 

Earlier I mentioned that it’s good to ask whether or not we need transitions across our themes to write ragtime sonata forms.  They are not actually necessary.  Theorists have pointed out that Mozart dispensed with transitions altogether in a variety of his slow sonata movements and other composers, similarly, did without transitions.  That leads us to another way to synthesize ragtime with sonata forms.

 

Example 5. Prelude in C major

https://youtu.be/bvElYZv_rTk

In this case the AA and BB sections are Themes 1 and 2 in the keys of C major and G major.  No separable transitions are involved.  We just get a second ending for Theme 1 that chromatically pivots into the new theme in the dominant key.  This is where blues tonality and ragtime conventions are our friends.  This particular work is openly indebted to themes from guitar sonatas by the 19^th century Bohemian guitarist Wenzel Thomas Matiegka.  Ragging the classics, in other words, is going on here.  Themes 1 and 2 are presented in the tonic and dominant keys like they would be in a sonata form but this is all ragtime.  The development presents Themes 1 and 2 ideas moving through a variety of keys and then leading to the recapitulation.  The return of Theme 1 fulfills the CC equivalent in a Joplin rag and instead of endings 1 and 2 as in the exposition, the returning A strain gets a substantial revision that allows it to fulfill the purpose of the non-modulating transition while still recognizably fulfilling the CC function that we’d expect in a Joplin rag.  Theme 2 returns in C major and here is where dispensing with distinct transitions helps out a composer, in a rag there would rarely be a transition between the C and D strains.    So how does that map out in musical time and space?  Like this.

 

[update 12-30-2023: there is now a corresponding fugue to go with the prelude over here]

Now if we have played this fast and loose with the conventions and scripts of ragtime and sonata forms the natural question that emerges from this is, “Why do we have to restrict ourselves to ragtime materials?” We don’t have to.  Ragtime is a style in which all the canonical works are wonderfully public domain so in that respect it is a goldmine of materials composers can sample directly or emulate.  But if you can develop a variety of syntheses of ragtime and sonata scripts other styles are also available.  That’s how we get to this closing example.

 

Example 6. in memoriam Aretha Franklin, Sonata-prelude in C minor

https://youtu.be/Cu2JjUCVVgw

 

This work is the first movement of a prelude and fugue in C minor and if you’re interested in hearing the prelude and then its fugue you can go over here:

https://youtu.be/crHaaw-bERM

 

L Poundie Burstein pointed out in Journeys Through Galant Expositions that in the eighteenth century composers did not tend to go for high contrasts between first and second themes.  It was more common to have contrasting “punctuation”. A way to translate this concept is to note that first themes might end with half cadences that lead to transitions that established a new key and the entrance of new thematic material, which might or might not actually contrast with the initial theme.  The new material would have a different kind of phrasing.  If a first theme was a binary form the second theme might be only different by being a rounded binary form or having ternary elements.  For this sonata-prelude in C minor the contrast is simple:  Theme 1 is a blues theme and Theme 2 has verse-verse-prechorus-chorus form.  That’s how you can translate Burstein’s idea of contrasting punctuation in galant era expositions into terms easily explicable within songcraft in contemporary popular music.  So this C minor sonata-prelude has a Theme 1 that is a blues and a Theme 2 that has a more complex verse-prechorus-chorus structure.  Conveniently, music theorists have liked to point out that Theme 1 in a sonata is usually tightly constructed while Theme 2 is more loosely constructed (William Caplin).  That’s what happens here in this sonata. 

There is no repeating exposition. There is an exposition, a development and a recapitulation and it is all explicable in terms of what Hepokoski might call a “triple rotational” form.  Themes 1 and 2 follow each other across each of the major structural subdivisions of the movement.  Since recapitulation zones offer room for secondary development and this sonata form has a blues for Theme 1 there is, unsurprisingly, embellishment of the material in the recapitulation zone. Theme 2 is kept pretty strictly but then C minor is hardly the kind of key where guitarists might want to just “cut loose”, especially since Theme 2 in a recapitulation will want to observe what Yoel Greenberg has called the “end rhyme” of recapitulations.  All in all probably pretty straightforward and the map of the musical timespace here doesn’t make reference to ragtime because this sonata is pretty obviously a blues/rock based creation.

 

I had been thinking of writing an appendix like this to Ragtime and Sonata Forms ever since I wrote the original treatise.  The works cited are all part of a slow incremental process of synthesizing ragtime and sonata forms until we get to the last example. Examples 1 through 4 predate 6 and 5 postdates the rest.  The most challenging version of a ragtime-sonata fusion, for this person I know, turned out to be the one that dispensed with all transitions in the exposition and recapitulation.  Obviously, it is possible to write such a ragtime-sonata synthesis. 

 

UPDATE 1-17-2024

Example 7

Exposition             Development     Recapitulation 

Intro AA B CC      D Intro               A B CC Intro as coda

G minor     B flat   B flat minor       G minor to G major to G minor

Another possibility to consider is shuffling around the strains.  For instance, if you compose an exposition in which Theme 1 and Theme 2 are taken by A and C strains in a rag then the B strain, which tends to have V-I oscillations anyway, can be used as a modulating theme.  The alternate endings of the B strain can take the roles of a modulating and non-modulating transition in the exposition and recapitulation.

The D strain, provided it has a clear relationship to the other strains, can be introduced in the development section.  In this hypothetical example I'm referring to the four strains are all drawing on prime, inversion, retrograde inversion and inversion forms of a single gesture.

https://www.youtube.com/watch?v=qb64PRHd8OI

I have it on good authority the piece has recently been revised into a sonata-prelude and fugue in G minor for solo guitar in open G tuning. 

While at some point I’ll need to expand the bibliography of Ragtime and Sonata Forms to include more recent reading for the holiday week I want to cut myself a teensy bit of slack and link to this recent post that lists the reading materials that were in the background of this appendix post.

 

The Presbyterian minister Kwame Bediako has pointed out that the transformation of Hellenistic philosophical concepts into Christian doctrinal terms took centuries.  He proposed that comparable, or parallel, processes of transformation within African theological contexts might be similarly long and slow.  Similar processes of indigenous Christian transformations of cultural concepts are taking place in Native American Christian theology (I have Daniel Kleven to thank for some references to Randy Woodley and Michael Newnham of Phoenix Preacher for reference to Richard Twiss).  I would venture that what Stoic philosophers meant by perichoresis is not what John of Damascus meant by perichoresis centuries later. 

 

Similar transformations of concepts and applications can happen within music that can parallel slow but seismic conceptual realignments that happened within theology and philosophy.  While Aristotle’s work influenced Aquinas what Aquinas articulated regarding filiation, procession and spiration within the immanent Trinity can be wielded by musicians and music theorists and composers today against what are by now parochial bromides within partisanship for the long 19^th century of the “timeless” classics of the Western canon.  Any claim that blues material has to be "transformed" before it can be used in a "classical" or "serious" piece of music belies the claim that art can do for religion what religion can no longer do for itself. The opposite is the case, the whole point of the Christian teaching of the hypostatic union of Christ's divine and human natures is that Jesus was and is fully human and did not need that human nature "transformed" to be compatible with His divinity. By extension, a pseudo-Christian dogmatics cannot be leveraged to claim that "pop" is some sarx that is incompatible with a divine pneuma of "serious music" and requires "transformation" or even some kind of "resurrection body".  That is a smokescreen used by people who don't believe vernacular musical vocabularies can be used to create "serious music" by way of using large-scale forms.  It's generally bad Christian dogmatics and even when it's not the dogmatics are wielded in bad faith, something that even non-religious readers who think things through will usually see through.  

If art mystics were consistent about their art mysticism they ought to concede that John Coltrane's A Love Supreme is on a level playing field with Beethoven's Ninth Symphony.  People who say otherwise reveal the bad faith of their art religion.  I don't happen to subscribe to art religion myself but I've read enough theologians who have taken up Rudolf Otto's The Idea of the Holy to know that several of them would affirm that Coltrane and Beethoven are on a level playing field in making musical art expressing the aspiration to universal human peace and fellowship.

I have only made passing reference to contemporary African and Native American Christian theology and to Syrian medieval reflections on the Trinity but biblical scholars and theologians don’t need to be told which sources I’ve so glancingly alluded to.  My contention has been that partisans of the Romanticism of the long 19^th century have claimed that art can do for religion what religion can no longer do for itself are, ultimately, not demonstrably competent in explicating religion and it is precisely the most arcane and inscrutable doctrines regarding the immanent and social Trinity and the hypostatic union of Christ’s divine and human natures that provide a springboard for rejecting concepts of musical time and space and concepts of the all-too-frequently racist versions of “organicism” in nineteenth century aesthetic theory that have crippled the possibilities within music theory as a discipline for exploring just how permeable the boundaries between “classical” and “pop” have always been in the past, can be in the present, and should be (again) in the future.