They Come in Pairs, Too

May 21, 2007

By Kyle Gann

Two-movement form intrigues me, partly because there are so few two-movement pieces. Unlike three- and four-movement form, its paradigm is so far from being done to death that it is impossible to call any two-movement work typical - it still feels like partly unexplored territory. I love the Clementi two-movement sonatas, which come close to the most perfect balance in the genre, the movements differentiated in meter and density, yet weighted just alike; the Op. 33 No. 2 sonata in F major is the most exquisite case (though less well-known than the F# minor). The elephantine example, of course, is Beethoven's Op. 111, which is perhaps second only to the Concord Sonata as a work that hangs over my life. Other instances are easy to ennumerate:

Webern's Opp. 20 through 22
Nielsen's Fifth
Mahler's Eighth
Schoenberg's Chamber Symphony No. 2
Gottschalk's A Night in the Tropics Symphony
Copland's Piano Concerto
Bernstein's Age of Anxiety Symphony (actually six movements in two continuous parts)
the Becker Third Symphony.

(No one knows John J. Becker anymore, but when we dedicated New Music America '82 to Cage, Cage asked us to include a work by Becker because of his historical importance to the midwest. His Third Symphony is aptly regarded as his best work.) The two-part form of Beckett's novel Molloy, with the second half mysteriously parallel to the first, is an incredibly masterful new paradigm of the genre. Other examples that come to mind seem like cheating; Schubert didn't finish his Eighth, Schuman's Third is in two movements but each divides into two sections (like the Berg Violin Concerto), and the two Branca Symphonies, Nos. 8 and 10, are companion pieces, with four movements between them.

I find the simple duality of two movements inspiring: either this or that, black and white, happy and sad, before and after. Three-movement form is so classical, so easy with its matching fast bookends around an aria (or in Ives's refreshingly reversed slow-fast-slow pattern). By contrast, two-movement form defies you to find a balance. The very fact that one movement is first and the other last makes a true balance impossible; one of them will have the last word, and it can't refer back too much to the first movement and still achieve a satisfying diversity. The second movement's finality represents, in effect, the death of the first movement's idea, and it can't come back to life in the cheery third to make you feel better. These elements are not going to synthesize. One philosophy of life is going to win, and there's no escape.

Perhaps that's why it seems problematic. In 1994 I wrote a two-movement sonata for pianist Lois Svard, and it seemed like no performance would go by without someone coming up to me afterward and blithely telling me, "Great piece, Kyle. You oughta write a third movement." The second 15 people who said this to me never realized how close they came to being throttled within an inch of their lives. I had balanced those movements by making them the same length, but - perhaps thinking subliminally of Nielsen's Fifth, which I adore - I built up a great crescendo of energy in the first movement and slowly released it in the second. I never even had a glimmer of an idea for another movement: those two gestures were the piece, from its first conception. So either I failed to make my bipartite emotional arc convincing, or else my listeners were just so conditioned by the classical Oreo cookie with its creme filling in the middle that they were unable to quit waiting for the third shoe to drop. To mix metaphors.

And so, in my piano concerto, which is "about" Katrina's devastation of New Orleans - thus "before" and "after" - I'm taking a tip from Op. 111. The first movement is seven minutes long, the second somewhere between 16 and 20. I imagine that Beethoven (along with Mahler in the Eighth and Becker in the Third) realized that, to make the classical audience quit waiting for a reassuring synthesis, they had to be dragged through a mysterious, disturbing, second-movement landscape that would kill any yearning for a glib rondo. Where Beethoven's first movement is angry and the second transcendently accepting, my first is pure rowdy fun and the second devastatingly sad (before transcendent acceptance, of course). Like the first movement to the Ives Fourth, my first movement is almost little more than introduction, the setup for the devastation. As I see it, to cure the audience of thinking about third-movement symmetry depends on making the second sufficiently dreary, daunting, and disorienting (being only slightly tongue-in-cheek here) to make them forget there had ever been a first movement. And I swear to god that if people come up afterward and tell me it needs a third movement, I'll double the length of the second movement and only use the pitches C and D-flat in the second half. In whole notes. Pianissimo. Muted.

UPDATE 2015: Can't believe I left out the Wolpe String Quartet.

UPDATE later in 2015: Ethel Smyth Piano Sonata No. 3 in D.


steven yi says: I've long been a fan of Lutoslawski's two-movement symphonies, and always felt satisfied with Schubert's 8th; I'll be curious to listen to the Becker now and certainly this entry of yours has my mind sparked. Thanks very much for sharing your thoughts on this!

Dan Schmidt says: I thought of Lutoslawski as well (his string quartet too). He has a great way of making the whole first movement a big upbeat to the second. His structure is definitely a model worth looking at.

david toub says: Thanks for remembering the Becker, Kyle. My recollection was that Becker's String Quartet was performed at NMA in Chicago, and I dimly remember hearing it on radio (I can actually remember a few measures of it that have haunted me for years--shame there's no recording I know of).

Jun-Dai says: Does a Prelude and Fugue (or Toccata and Fugue, etc.) count as a two-movement form?
Also, Novak's op.35 String Quartet is in two movements; I would probably never have heard it if it weren't included as filler on my CD of Janacek quartets. I bet people told him plenty of times it would be nice with another movement or two.

alex says: is the rite of spring too diverse a work to be considered as a dyad?
KG replies: I'm embarrassed not to have thought of it.

Bob Gilmore says: ...and not forgetting the Janacek Piano Sonata, which is one of my all-time favourite pieces. It originally had a third movement too. Janacek tried to destroy the whole piece, apparently, and succeeded only with the third movement. Even more so than with the Schubert, I can't imagine anything more perfect than this two-movement form.

Maxmusic says: Alexei Stanchinsky's unjustly neglected 2nd Piano Sonata (1914) is also in two movements, a slow, masterly crafted Fugue, and a kind of wild Toccata with uneven rhythms. They perfectly match each other, IMO.
KG replies: Well, it must certainly represent some kind of validation of the rarity of two-movement sonatas, concertos, and symphonies that the first round of omissions brings us to Stanchinsky. Amazon lists only one CD of his music, available only used, and now I'm curious.

David Cavlovic says: It seems like almost everything by J.C. Bach, though, is in two movements, but I've always suspected it was lack of ambition. I jest, of course, but someone did ask J.C. once why he didn't compose more "heady music" like his brothet C.P.E. J.C's response was, "I compose to live, my brother lives to compose".
KG replies: I didn't really think of following the thread back further than Clementi, but C.P.E. has some interesting two-movement sonatas too. Nice quote.

lawrencedillon says: This post is a nice dissertation on the seldom-acknowledged truism that formal balance in music has little to do with duration: five minutes of X doesn't necessarily balance five minutes of Y. I think you are right to go for an extended second movement, but it can be equally effective to have a very brief second movement, if it's the right material (not in the case of your "devastatingly sad" conception). It's a similar situation to the classic da capo form: repeat, repeat in the initial A; no repeats in the A' - and yet the A' rarely feels like it has understayed its welcome, despite being exactly half as long.
KG replies: It does strike me that the faster, lighter, briefer second movement is the more 18th-century solution, which Webern followed in his Symphony. It makes Op. 111 look like an even bolder departure.

andrea says: to continue with the oreo cookie analogy, you could think of the two-movement work like an open-faced sandwich -- also underrated in the culinary world...

Steve Ledbetter says: Another composer who is very much taken with two-movement form is Leon Kirchner, who casts his early Piano Trio. his mid-career Concerto for Violin, Cello, Ten Winds, and Percussion, and his late Music for 12 (to name three pieces the come to mind immediately) in two movements, occasionally with a kind of "recitative" linking them.

Christopher Culver says: Happy to see others thinking of Lutoslawski. I find his Second Symphony his most successful (though that's quite a minority opinion). Boulez has also contributed to the world of two-movement works with sur Incises and Eclat/Multiples which I find very exciting.
But my favourite two-movement work is Per Norgard's Third Symphony, with its first movement an overt statement of the union of the three infinity series, and the second a more varied landscape where the infinity series governs everything but mainly hidden.

Ken Fasano says: When and where do we get to hear this concerto? It sounds very interesting!
KG replies: October 31, Amsterdam, and then other places in Holland in November. Thanks!

Galen H. Brown says: Jacob ter Veldhuis has a number of two-movement works. The Rainbow concerto definitely has only two, and if I recall correctly the Tallahatchie and Goldrush concertos do as well. Heartbreakers is two movements as well, and I think there are others, but I don't have the CDs in front of me right now.
David Lang's "Orpheus Over and Under" is two movements (and they aren't called "Over" and "Under").
John Adams's "Grand Pianola Music"
Mendelsohn's "Elijah" is obviously in many movements, but it's in two acts, and I suspect there are a number examples of operas and oratorios that follow that model.
It's also perhaps interesting to note that in popular music a lot of concept albums seem to be divided into two parts, perhaps because of the two-sidedness of the LP. These examples might seem like a stretch, but: Pink Floyd's "The Wall" is in two parts-part one ends with the "Goodbye Cruel World" suicide attempt and part two starts with "Hey You". The Who's "Tommy" is divided in half by "Underture." Pink Floyd's "The Final Cut" seems like two parts as well, with part one ending with "Paranoid Eyes" and part two starting with "Get your filthy hands off my desert."
And speaking of popular music, it occurs to me that The Arcade Fire's new album includes a song called "Black Wave/Good Vibrations" which is a sort of two-movement pop song.
KG replies: Points taken; except that the second part of Grand Pianola Music seems very much like a slow movement that segues without pause into a rollicking third movement, and I seem to recall that Adams talks about it as three movements.

Sarah Cahill says: Fourteen comments so far about two-movement form? You really struck a nerve, Kyle!
Alvin Curran's For Cornelius seems like a classic of this form, also. It's perfect in its own way.

Johnny McWilliams says: "[...] second movement and only use the pitches C and D-flat in the second half. In whole notes. Pianissimo. Muted."
Dang... there goes my idea for Opus 24 down the tubes... [crumples manuscript]... Wait a minute -- You didn't score your 2nd mvt for bayan, harmonica and sacbut obbligato, did you? A-ha! Originality again! [retrieves manuscript and begins afresh]

David Cavlovic says: "Mendelsohn's "Elijah" is obviously in many movements, but it's in two acts, and I suspect there are a number examples of operas and oratorios that follow that model."
Without going into a boring and long-winded history lecture, the Oratorio was crafted historically into two parts. During the "intermission" a sermon on the subject of the Oratorio was provided. Then came television...just kidding.
Händel's Messiah deliberately broke with that tradition, an early example of Enlightenment thinking and, as with the later Brahms Deutsches Requiem, an attempt by the composer to put forward a personal Theology in music.

John Shaw says: Regarding the aftermath of Katrina:
Devastatingly sad, yes.
Transcendent acceptance, no thank you.
I'm guessing you're joking about transcendent acceptance -- your saying "of course" about it seems like a joke.
Gallows humor, if so.
And, given the drift of our industrial society and our polity, the gallows may indeed be ours after all, and not only the Gulf Coast-ers.
I'll never forget seeing the footage of black New Orleanseans being turned away at gunpoint as they tried to walk across a bridge out of New Orleans.
The federal government wouldn't let supplies in.
The federal government allowed neighboring local governments to prevent people from leaving.
Preventing people from leaving while preventing supplies from entering constitutes a siege.
A siege is an act of war.
The federal government was waging war against the (mostly poor, mostly black) people of New Orleans -- literally waging war.
Devastatingly sad.
Best wishes with your piece. It's good to keep this story alive.

Matthew says: The reaction to your sonata probably has as much to do with the fact that it ends with a quiet disspation of energy rather than some sort of affirmation of everything that's preceded it (which doesn't necessarily have to be loud). I like similar endings, which means I've gotten many similar audience suggestions. People want their musical old soldiers to die, not just fade away.

Dee Aghilev says: Prokofiev's Symphony No. 2, two movements, one of his least known and best pieces.

casey anderson says: Perhaps this is a mistake on my part, but I almost never compose multi-movement works (favoring instead one through-composed movement), and now that this has been brought up I feel compelled to reinvestigate my thoughts on such formal constructions.
In regards to the many comments on Opus 111, and Lawrence Dillon's point concerning proportionality and balance without symmetrical movement durations: are there many (or any) works that reverse the form of Opus 111 (aside from the aforementioned Webern)? In other words, does anyone know of anything that starts out with a grandiose (or whatever), longer and "weightier" movement, and then pairs it with a shorter movement? Maybe this is more common than I think it is, but nothing is springing to mind right now (however this sounds intriguing to me).
Now that I think about it, it seems like it would be difficult to make the second movement, in such a formal design, not seem like a coda (or some sort of less-important addition), but to still balance it out with the larger first (or to even justify a direct contradiction with the larger-in-whatever-number-of-ways first). Any suggestions of works which are designed in this manner (again, where the "smaller" second movement is not just conclusory, but a constituent part in its own right) would be appreciated.
I would be interested to hear how often anyone else here composes in a more "standardized," multi-movement form, whether it is one that is infrequently used throughout history or not. Like I said, to date I have only written what I consider single movement, through-composed works, however that may just be the way I think about them. Now I am wondering if avoiding multi-movement designs is a mistake.

By the way, I look forward to hearing this work, Kyle.
KG replies: Well, Casey, you're in good company. In the '80s, Morton Feldman proclaimed multimovement form dead. I do feel a little old-fashioned, thinking first movement, second movement, slow movement, fast movement. Most of my pieces are one-movement, and the prestige of a long, one-movement form is not likely to disappear soon. But, given the rare opportunity to work with a large ensemble, the temptation to try out a couple of different experiments with it is hard to resist. It does solve a couple of formal problems, and there's no reason to throw away an old-fashioned idea that intrigues you.
I keep thinking there are some recent two-movement pieces with smaller second movements, but aside from Webern, I can't think what they are.

Scott says: Adding to the list, I'd like to mention Bartok's two Rhapsodies for violin and piano (and later for violin and orchestra), which are both in two movements, and are both stunningly cool pieces of music, suitable for thoughtful listening as well as headbanging. I just played the 2nd rhapsody a couple weeks ago, and it was incredible fun to get inside, with a very convincing 2-movement form.

Jennifer Higdon says: I love 2-movement forms and use them from time to time... my Piano Trio is divided into 2 movements, "Pale Yellow" and "Fiery Red". My Viola Sonata is a 2-movementer. It's a great form to just show contasting styles, without the need to go back to the first movement's energy or even language. Why not do things differently?

Galen H. Brown says: I just realized that while Steve Reich almost never writes two-movement works he does have at least one: "It's Gonna Rain." Hmm. . .

Honi Soit says: Would the "Recitative -> Aria" structure count as a two-movement form?
Granted, by the time you get to, say, Donizetti, the recit doesn't really stand up as a musical unit in its own right. But I feel like in baroque works it often does. Any opera scholars out there want to back me up/shoot me down?

Samuel Vriezen says: And I swear to god that if people come up afterward and tell me it needs a third movement, I'll double the length of the second movement and only use the pitches C and D-flat in the second half. In whole notes. Pianissimo. Muted.
Actually, that would make a fantastic piece. Quite Wandelweiser.
There's a special type of quasi-2-movement form that I've found myself using every now and then, which I think of as "initiation form". In this case, you have a process, and then the same process is done again, only the second time it's somehow much longer and more complex. It feels like between one and two movements; it happens in my sine wave Toccata VI, where I call it one movement, and in the piano piece Within Fourths/Within Fifths, where I call it two movements.
Jurg Frey has a few one-movement pieces which are one movement played twice without changes. There's something not-quite-one-movement-ish about that, I feel.

Eric Bruskin says: Hey, don't forget Beethoven's unjustly maligned Op.54 Piano Sonata, whose smashing second movement toccata was the basis for one of the greatest downtown pieces ever, Peter Gena's "Beethoven in Soho"!
KG replies: I had intentionally slighted Beethoven's early two-movement sonatinas that got inserted into the set with higher opus numbers, but I forgot Op. 54, which *is* a neglected gem.

Alfredo says: Arvo Pärt's "Tabula Rasa" is also in two movements, with a second one more or less twice as long as the first.

kac attac says: Three years later, no one has mentioned Scriabin's 4th Piano Sonata.

Copyright 2007 by Kyle Gann

Return to the PostClassic Blog Archive

Return to the Kyle Gann Home Page

return to the home page