This is probably not what Bach intended.

I'm going to be uploading the next movement of the third cello suite next Friday, and it's going to be . . . a little different. It'll still be very recognizably Bach, but I'm taking liberties with it, many more than I have in the other movements, and many more also than I'd take if I were playing the piece in an audition for judges who would doubtless know the score. Normally I'd just put it out there as-is with minimal comment, but this touches on a larger issue that I have a lot of thoughts about, so I'm going to take this post to justify what iI've done with the Sarabande.

Those of you familiar with how performance practices have changed over time may suspect that I'm going to appeal to the much greater interpretive freedom that performers had in the past. It's certainly true that performers used to have much broader latitude in making interpretive choices than we commonly afford them today [The Conversation blog], up to and including flat-out improvisation. With the rise of the Historically Informed Performance movement, there's been a ton of research into how performers interpreted scores in the days before recordings, and much of that work has revolutionized the way we approach works from the time Bach was writing.

I think that work is super important, and I think every musician should have exposure not just to their results but also their methodologies. (I also think their work is applicable to more than just Baroque music, but that's a post for another day.) That said, I don't actually think historical performance practice is a good justification for what I'm doing. Ultimately, the specific choices I'm making have very little to do with how someone would have played this movement at the time it was written. Instead, I'm dealing with something deeper.

One of the stranger pages on TV Tropes is the Fiction Identity Postulate. It's not itself a trope, but rather a principle: All fiction is equally real. That is to say: The version of Romeo and Juliet where Juliet wakes up in time to stop Romeo from offing himself really happened to exactly the same extent as the one where she doesn't and they both die, i.e. not at all. Now, only one of these is the version that Shakespeare actually wrote, and that's the one that pretty much every production, adaptation, and re-telling is going to follow, but still. They're both made up. Neither of them happened. The version that's officially canon isn't any less fictitious than the version that's not. It's just more canon.

What does any of this have to do with music? Well. On the page in front of me, I have the Sarabande as Bach wrote it. (For now we're just going to pretend that there is an error-free transmission process from his pen to the printer that spat out the copy of the score that I own. In truth, there are numerous questions about what exactly should be on the page, but for our purposes we can ignore the critical apparatus entirely. Sorry, editors . . . ) That's canon. That's the official version. 

But as with Shakespeare, that's not the only way it could've gone. There is no law of the universe that says this version is the only version of the Sarabande that could possibly exist. The score is a set of instructions for producing a sequence of sonic events, but the specific sequence it denotes isn't any more "real" than any number of variations on it; versions with re-arranged articulations, distorted melodies, and alternate harmonizations are all just as real once they've been played. Bach didn't write any of them — they're not canon — but they exist as potential pieces just as much as the version he did write does.

Now, this is not to say that all of these possible variants sound equally good. Most of them sound appalling. There's no law of the universe forcing Bach's hand, but there are rules of harmony and taste. And when you're working up a piece, it can often be very instructive to ask why the composer made the choices they did. Very few of them are truly arbitrary, and while many will be obvious, exploring the more confusing ones can lead the way to profound discoveries about how the piece in question works. This note has an accent because it's fulfilling this function in the harmonic scheme. The composer is telling you to get softer here because they sneakily hit a structural closure several bars before you would have expected them to. Even when it's "merely" a matter of phrasing, composers are often deeply attuned to the optimal way of delivering a phrase due to the sheer amount of time we spend turning the things over and mining them for musical material.

Still, taste is a tricky thing, and the optimal way of delivering a phrase can be as subjective and arbitrary as anything else. Something I find elegant and finely balanced someone else might think stilted and unnatural, while what they think graceful and lithe I might find bloated and galumphy. This is a good thing! If we all had identical taste, music would be dreadfully boring indeed. Difference of opinion, even radical difference of opinion, is an absolute requirement for a vibrant, thriving art world. 

In the western concert music world, we like to pretend that there's a nice bright line separating interpretative choices from compositional ones, but there really isn't. We can shade smoothly from deciding just how loud "forte" really is to supplying articulation markings to sprinkling in ornaments to playing a variation of a phrase on the repeat . . . and suddenly we're dealing with matters that are indisputably in the realm of composition. There is no such thing as a literal reading of a piece of sheet music. All performance requires interpretation, and interpretation can be intensified directly into composition*. 

We don't really have a way of dealing with this in the western concert music world. When someone says they're playing a piece by Bach, we expect them to hew closely, even exactingly to the printed score; we don't have an easy vocabulary for "I am playing a bunch of music that is very closely tied to what Bach wrote, but with interpretive deviations that may at times rise to the level of actual composition.". We don't have a term for it, but that doesn't mean the results can't be great art**.

This is not to say that all inquiry into compositional intent should be forbidden, that composers should abandon all expressive markings and throw our works to the wind to be re-written willy-nilly. Far from it. I think it's entirely legitimate for a performer to take the stance that the version of a work that's closest to the composer's intentions is the most compelling, and I think it behooves composers to be as clear as possible about our intentions since a) as said above we do often have pretty deep insights into what makes our own music tick and b) performers of new works seldom have the luxury of time to delve deeply into an unfamiliar score, so clear guidance is helpful.

Still, I also think it's legitimate for a performer to disregard compositional intent if they find a version of the piece they think makes more compelling art. (In general, I think performers have an artistic duty to create the most compelling performance that they can.) I think this is especially true when the piece is something famous and well established in the repertoire. No matter how strange my interpretation of the Sarabande, I'm not going to ruin Bach for anyone. You will hear Bach played again by people very devoted to hewing to his intentions. My one recording of this piece is not going to be the make-or-break moment for it, the one and only exposure that some people will have to this work.

For more obscure pieces, especially ones by living composers, I think there is at least some obligation to give the composer a fair hearing. I don't think it's an overriding obligation — feel free to put Spohr thru a blender — but I do think it's there. Also if you're working with a living composer, it's probably not the best idea to bluntly tell them that you're changing a bunch of stuff in their piece because you don't like the way they wrote it. An artistic obligation to make good art doesn't instantly and automatically override every other concern that might possibly arise***. Art is a balancing act, and different considerations are going to take priority at different times.

But for Bach, this isn't a concern. I'm not going to hurt Bach's feelings by getting a little experimental, nor am I going to damage his reputation with audience members if my version of this movement is a flop. So when you hear me going off script next week, know that it's not because I think this best captures what the composer intended. It's because I like it. And that's reason enough for me.

*I am not particularly interested in parsing out where, precisely, the line between interpretation and composition is, where performing slips over into creating a derivative work. The legal aspects of this get very messy very quickly, and I'm not going to spend time trying to hash them out in this post.

**The popular music world, of course, has the notion of covering, which is exactly equivalent to what I'm talking about here. Some covers of classic songs are abysmal, but some of them are incredible, better even than the originals. 

***This is also leaving aside cases where representation or deep personal or political significance get involved. There are artistic choices that are not only tasteless, but also immoral — just because you can do something doesn't mean you should