In a wonderful interview between Andre Previn and Oscar Peterson (you can watch it here),  the discussion — as it inevitably does between two jazz pianists — turns to Art Tatum. Peterson brings up a pianist he met once that played by taking everything from Tatum recordings. An astounding feat of musicianship, to be sure, but in the jazz world, doing something just like someone else isn’t something that is valued (except perhaps as a curiosity.) In the classical world, however, doing it “like they did it back then” is something that is extolled. Curiously, there are a dearth of recordings from before the 1920s, so the folk in the HIP [Historically Informed Performance] world may be speculating a little on exactly how it was done.

To clarify the butter here, think of the following question: how did Oscar Peterson play “Stomping’ at the Savoy?” It’s a hard question to answer. I mean, we can talk about some stylistic things that he tends to do. We could even listen to a couple of recordings. But, it’s an impossible question to answer. For every recording we have, he probably played it hundreds of times that weren’t recorded — and every one was different. He didn’t play it a certain way because jazz is improvisatory in nature. And, so the story goes in conducting class, classical music is not. Conductors are not trained to include improvisation because it’s not part of the style.

Here’s where it gets confusing. When we are talking about classical music not being an improvisatory art form, what do we mean?

  1. Well, we don’t really mean Medieval music. At least if the Musica enchiriadis is to be trusted. It gives rules for improvising organum.
  2. We don’t really mean early Renaissance music because Zarlino talks about improvising a third vocal part over a two-voice piece and provides an example of how he would do it for a Josquin piece.
  3. We aren’t talking about Palestrina, because we know that the singers – at least when they sang one on a part, were expected to add flourishes and “diminutions” to the writing.
  4. We aren’t talking about Baroque music. Those guys were doing something much more akin to writing jazz charts. Here is one of my favorite examples from Telemann. The top line is the line that Telemann wrote in the score. The second line is one example of how he might play the top line in real life.

We aren’t, of course, talking about the Classical Era. Mozart and Beethoven were famed improvisers. They were famous for improvising not only their cadenzas for concerti, but also embellishing the written score on the spot. Did you ever wonder about that curious practice they had of writing a grace note 16th, an 8th, and two 16ths.

My teachers always told me, divide the 8th note in half, and play it as four 16ths. “Well, why didn’t they write it as four 16ths?” I would ask. None of my teachers knew. And then, I read through Leopold Mozart’s treatise. People were improvising so much — even in the orchestral violin sections(!) — that if they wrote four 16ths, it might get improvised into all sorts of nonsense. They wrote the grace note-8th note pattern to say, “really just play the four 16ths here and don’t improvise!”

6. We aren’t talking about the Romantics. Paganini, Liszt, and Chopin were all famed improvisors. One of the big problems in Chopin scholarship is that there are different editions with different notes. At least one prominent theory suggests that the reason for so many variants is that he didn’t always play it the same way. There wasn’t a set way to play a piece with them any more than there was for Oscar Peterson.

So when we are talking about Western “Classical” music not being an improvisatory style, we aren’t really talking about music history as it was practiced. We are talking about something that was largely an aesthetic of the 20th century that is projected backward onto the actual history of music as it was performed. However, conductors and performers aren’t trained as if that were the case. They are trained to respect the score even though in actual practice the score was not performed as it was written. It was often just a framework for improvisation. I’m the only pianist that I know that sometimes improvises on Chopin when I play his music. I do it because he did it like that, but now being faithful to Chopin means playing it — not like he did, but by meticulously following the written page in a literal sense like a religious fundamentalist.

When did the big change occur? When did the musical culture shift from the idea that conductors and performers were actually supposed to interpret and improvise? When did we become fundamentalists that slavishly stick to a literal reading of the text in a positivist sense as if a score were a scientific document? Again, I think I will side with Taruskin and lay most of the burden at the feet of Stravinsky. He was the one that wanted people to just follow the score and not “interpret” it. True to his beliefs, he practiced what he preached. He was deeply committed to the most boring and uninspired playing possible. If you don’t believe me, listen to this recording of Mozart with his son. It’s hard to imagine a more unimaginative and slavish commitment to dullness.

Naturally, he didn’t always do this. He insisted on others following his metronome markings exactly, but he made four different recordings of the Rite all at different tempi. So, strict literal readings for the hoi polloi, but a little more flexibility for Igor.

The musical fundamentalism that’s taught in our institutions hasn’t limited itself to a literal reading of the text (something that would have clearly offended Telemann.) It has flattened out interpretation across the board. If I played a recording of a Mozart’s Eine Kleine Nachtmusik by Bruno Walter and another by Wilhelm Furtwängler, you would quickly hear the differences and learn to recognize each conductor’s idiosyncrasies and thoughts. You would learn where each one chose to speed up and slow down in places were there aren’t accel’s or rit’s marked in the score. They assumed that the so called, “flexible tempo” was one of their main jobs as an interpreter. If I took ten recordings from the past ten years from ten different conductors, you would have a much harder time telling them apart. Part of that is from the economics and rehearsal schedule of the modern orchestra (gone are the days of Carlos Kleiber asking for and getting 17 rehearsals or an Audi), but part of it is because of the way conductors are being prepared as musical fundamentalists. If a conductor in a modern conducting class started an accelerando where one wasn’t written in a Mozart score, they’d likely be accused of “over-Romanticizing” the text.

The result is vast battalions of conductors emerging from our academic institutions that all “interpret” music in the same way. Then they teach their students to do the same thing. Ever the empiricist, I took my theory for a test drive a few years ago in a class being taught by someone I love and trust. A controversy about how to conduct a bit of early music came up. The grad student asked me a question. I asked back, “Why are you doing it like that?” The student said, “Because one of my conducting teachers told me to do it like that.” I said, “Your teacher told you to you to do it like that because his teacher told him to do it like that. He hasn’t actually read Baroque treatises or thought about the problems of interpretation in a philosophical sense. He’s just repeating what he was told in his conducting class.” The entire class looked to my friend that was teaching. The response: “Yeah. That’s pretty much correct. That’s how we’re taught to conduct.”