So, in our last episode, I wondered about the ways in which we all take for granted that most major instrumental pieces are broken up into distinct (often self-sufficient) movements - distinct movements which, nevertheless, are generally expected to be performed in the context of the whole. In other words, the serious classical artist doesn't generally record or perform in recital just a single movement of a Beethoven sonata or Mozart symphony. I mentioned a few exceptions to that rule, but for better or for worse, the bias against excerpting movements is pretty strong. (It's always one of the chief complaints about the less high-minded classical radio stations that they dare to play unattached movements - I should know, having made this complaint many times!)
So, if pulling movements out on their own is frowned upon, I'd guess it would be even more controversial to re-order movements within a tidily assembled work or - horror! - to patch various movements into a new whole. But, quietly geeky radical that I am, I do find myself having such thoughts, three of which I'll share here - all of them, curiously, having to do with violin repertoire. (I can't really explain why that is, other than that I love the violin repertoire.) The first two "thought experiments" are pretty straightforward, and mainly noteworthy because I think they'd 1) work really well and 2) would probably never be seriously considered by anyone. The third one has spawned both the title of this post and something altogether more interesting.
Switcheroo #1: Back in September, I twittered about my idea that Brahms' amazingly wonderful third violin sonata might be just a tad more amazingly wonderful if the 2rd and 3rd of its four movements were swapped. Honestly, my reasoning is more intuitive than anything else (and, I know, who am I to put my intuition up against Brahms'?), but the song-like 2nd movement has such a radiantly tender, calm-before-storm quality that it feels right to me for it to precede the violently stormy finale. Meanwhile, the nervous hesitations of the scherzo seem to fit nicely after the weary ending of the passionate first movement. I'm perfectly content for others to disagree with me about this, and it's always hard to argue against the rightness of familiarity, but I think it would be a lovely idea to try this 1324 order out in performance - and yet, innocent and simple though the change would be, it feels like it would be some sort of violation. Here's a helpful playlist in which you can try out my version...if you dare!
Switcheroo #2: This is maybe even more radical, though no less innocently simple in concept, but the truth is I've never been completely in love with any of the Mozart violin concerti. Not that they need my love or anything, and sure, he was a young composer when he wrote them, etc., etc. I do think the 5th and final concerto is mostly ideal except it just doesn't have a great second movement. Or, let me put it more bluntly: it's second movement is not as perfect as the slow movement of the 3rd concerto. That slow movement is as divinely inspired as anything Mozart ever wrote. For whatever reason, the slow movement of Concerto #5 always sounds to me like a melody in search of...well, in search of that tune from the Concerto #3.
So, if I were somehow transformed into an internationally-acclaimed violin virtuoso asked to play a Mozart concerto for an upcoming gig, I would seriously want to propose my own little hybrid concerto. The key relationships work out just fine: A Major, D Major, A Major. (The "original" 2nd mvt of Concerto #5 is in E Major; up a 5th, down a 5th. Same difference.) You can try it out with this little playlist. It's not like I'm proposing first movement from Sibelius, second from Bruch, and third from Brahms. Just three Mozart movements which would complement each other beautifully. One of his piano sonatas (K. 533/494) is basically a hybrid as well, its third movement having been composed first and then added to two other movements at the request of a publisher. Maybe I'm not supposed to say this, but a lot of Mozart and Haydn movements are somewhat interchangeable - but we never get to play around with them. Pianists get to make their own multi-movement composite Scarlatti sets all the time (since his so-called sonatas are all single-movement works), but convention denies us that creative option with most other composers.
My last example is a bit different, as here it's more a problem of a work that has proportional issues. (By the way, if you think it's impudent of me to be challenging the final shape given to works by Brahms and Mozart, then imagine my trepidation at finding fault with Bach - and not just any Bach, but one of his revered works for unaccompanied violin.) Well, a few months ago, my violinist daughter was assigned her first-ever solo Bach - a deeply meaningful milestone to me. Curiously, her marvelously old-school Russian teacher decided to start with what to me is the most musically severe and uncompromising of these works, the Partita No. 1 in B Minor. It consists of eight binary-form dance movements, but it's really four dances, each of which is each then followed by a variation called a Double. Curious things, these Doubles. Each has the exact same formal and harmonic shape as the dance that precedes it, but the Doubles generally feature faster and more evenly flowing note values, meaning the dance character is sublimated a bit. All of the dances are in B minor (an austere key for Bach), and all are rather severe in shape and gesture - dances, yes, but not quite like listening to the Nutcracker Suite.
So my problem has always been that, as a listening experience, the eight movements back-to-back-to-back-etc. can be a bit too much. [...dodging lightning bolt...] What vexes me particularly is hearing each dance played with repeats and then hearing the Doubles traverse the same territory, with the same repeats. In fact, the first time I saw the score, I assumed the Doubles were intended as alternate ways to play the repeats, and I still think this could be an interesting approach. But, it seems the most common approach is to be a good soldier and play everything as written. If the violinist is good enough, the experience can be richly rewarding, as this is some of the most profound and centering music ever written...but still. I can't help but help wonder if Bach would ever have imagined hearing this music presented in recital in such a way. Again, if I could suddenly be a great violinist, I'd be tempted to perform the work either with no repeats or, better yet, with the Doubles acting as the repeats. I have heard of pianists interspersing works by Schoenberg amongst the movements of Bach sets as a way of opening ears to the cross-century connection*; I wish artists would be as open to re-imagining ways of presenting a single composer's works.
But my favorite discovery out all of this came from thinking of another way of dealing with those Doubles. It occurred to me that they could perhaps be played simultaneously with their dances as duets. A quick experiment with the opening Allemanda was somewhat disappointing as there was too much literal doubling between the two parts. However, the Courante and its Double proved to be a revelation. I imagine violinists must have tried this out on occasion - maybe with teacher and student in lessons - but my not-extensive Googling hasn't turned anything up in the way of recordings. So, I went to work with a couple of the virtual violinists residing in my computer and, voilà, a strikingly successful duet.** It's almost as if Bach intended these movements to be "mashed up," as they mostly move in contrary motion to each other, the slashing sixteenth notes of the Double parrying the more angular eighth-notes of the original Courante. My favorite piece of music in the whole world is probably Bach's Double Violin Concerto, known far and wide as "The Bach Double," but it's rather satisfying to have stumbled on this new "Bach Double" (a much bigger success than this bizarre "Bach Double" I created a while back.)
A couple of final points: I'm rather proud of the way the video above traces the two separate scores. In an ideal world, I'd have taken the time to format both movements nicely into duet form, but it's kind of fun trying to follow both scores at once. [Here's another Bach animation of mine that lets you do something similar.] It's a reminder that counterpoint is always a kind of "mash-up." One of the joys of counterpoint is the experience of experiencing multiple distinct strands simultaneously; mash-ups can often be engaging for the same reason. In these multi-tasking times, perhaps Bach's music is more timely than ever.
* Note that Andrew Rangell, the pianist who interspersed Bach with Schoenberg for a conference recital, is quoted as saying he'd never do that in a "real" concert. What a crazy world we live in!
** As it happens, another work my daughter is studying now is a Wieniawski Etude-Caprice, which comes with its own accompanying part for a second violinist. So, I guess I have had violin duets on the brain...
P.S. I'll be posting duet versions of the Sarabande and Bourree in the near future; they don't work as well as the Courantes, but they are worth hearing.
UPDATE: A Twitter colleague, the very knowledgeable JoseSPiano, mentioned hearing violinist Daniel Heifetz perform the Partita No. 2 in D Minor WITHOUT the concluding Chaconne. The 15-minute Chaconne is perhaps Bach's most monumental achievement, but it dwarfs the four much briefer dance movements that precede it. I've no doubt that the Allemande, Courante, Sarabande, and Gigue make a wonderful set on their own, and this practice would allow them to be heard differently, much like what I said about movements by Schubert and Ives in my last post.