It looks like a blog, it reads like a blog, and yet somehow it isn't quite a blog. Odd, isn't it?
If you'd like to comment on anything here, feel free to contact me.
04/24/22
I was feeling a little down the other day, so I listened to “The Rite of Spring”, which I hadn’t done in a while. “Rite” is high on my personal list of all-time favorites. Every bar is staggeringly original. There’s nothing else like it, before or after. But as much as I love each of the episodes in the piece, I realized that there is one that really stands out. Somehow it just gives me a sense of joy every time I hear it.
It’s that one bar of 11/4, just before the “Glorification of the Chosen” scene. You know the one: “Wham! Wham! Wham! Wham! Wham! Wham! Wham! Wham! Wham! Wham! Wham!” Eleven times. OK, I know this is not exactly a joyful moment in the ballet, but that’s not the point. Taken on its own this bar is utterly inexplicable, but in context it’s brilliant. The first quarter-note is a definitive moment of arrival; the eleventh is an unstoppable moment of departure. In between, nine pounding heartbeats of anticipation.
This is also an amazingly liberating moment, compositionally. Four timpani and bass drum fortissimo, and all of the strings down-bowing but subordinate at mere forte. It’s an incredible gesture of defiance of tradition, the act of a headstrong young man. Stravinsky was in his late twenties, and his teacher Rimsky-Korsakoff had died a few years earlier. (I don’t imagine that NRK would have put up with this kind of nonsense had he still been alive.)
And look at what Stravinsky is telling the orchestra with that time signature: No, it’s not 4 plus 4 plus 3, it’s not 4 plus 3 plus 4, it’s not 3 plus 3 plus 3 plus 2 — it’s eleven. Just eleven. Who got taught at conservatory then (or even now?) to play eleven notes in a row in exactly the same way? Who knew how to conduct a bar of 11/4?
It’s as if all of your sophisticated musical education is held in abeyance for those eleven beats. It’s both prime and primal. Nothing matters, save for the ability to create a sound and beat in time with others. It’s a singularly primitive moment — and exactly what Stravinsky was trying to convey.
07/25/21
In the last 20 or so years, it's become possible for any composer with a computer to put together a computer-generated audio rendition of their music, regardless of whether they are competent performers on the instruments they've written for. If you're a composer who used to do it the traditional way but converted to using Finale, Sibelius, or more advanced tools, I hardly need to tell you what a monumental change this is. For the first time ever, it's possible to hear one's orchestration in advance, to try out different solutions, and to change one's mind — all before any other musician sees a single note. And it's possible, with work, to put together a computer-generated performance that's good enough for others to hear. Good enough to be an adequate demonstration of what the piece should sound like, or even, with more effort (and money), a legitimate performance of the piece.
Is that enough? I mean, now that it's possible to have orchestral music without an orchestra, is that a goal to strive for? After all, most film and video game composers do virtually all of their work in digital studios, and produce not just passable recordings, but the definitive recordings of their music, which become soundtracks. On this website I've put up two examples of what I would call adequate computer performances of my large-ensemble works. With a little luck, more people may hear my music through these recordings than ever heard a live performance; so I have to treat them seriously. I even have a recording of a live performance of one of those pieces that I won't put up on the site because of a single (but catastrophically ill-timed) wrong note. Computers don't ever play wrong notes, mis-count rhythms, or have poor intonation. With a few hundred more dollars and a few hundred more hours, I could indulge my perfectionist tendencies and create the ideal artificial performances.
And yet, that's not what I want to do.
An artificial performance is still artificial. A sampled note is a note devoid of context, devoid of shape, devoid of emotion; and a phrase assembled from sampled notes can't convey the same depth of expression that a human-performed phrase can. Moreover, even though I might be able to write an effective line for, say, violin, I'm not a violinist, so I will never understand exactly where a natural portamento will happen, where a natural hesitation will occur, or where the sound of the bow will come through. Not well enough to reproduce those effects on a computer, unless I learn to play the instrument at a high level myself. Sure, I can buy a keyboard with pressure-sensitive keys and practice until I'm fluent on the pitch wheel; and maybe soon we'll have AI-based "human playback" algorithms drawing from massive, multi-dimensional, contextual sample libraries. Or, I can find a real violinist.
Personally, I'd much rather interact with human musicians than instruct a robot, which is basically what one is doing with Finale — it's a machine with a limited instruction set which it interprets according to internal rules that are hidden from the user. Human musicians understand adverbs while robots don't. I can ask a human ensemble to play a phrase "gently" or "aggressively" and they'll immediately grasp what I'm asking for; the Finale robot doesn't have those words in its vocabulary, so to get the same effect I'd have to painstakingly micromanage every single note.
There's a universe of difference between an artificial performance on a computer and a real performance by human musicians. Humans are spontaneous, they listen, they interact and communicate in subtle ways that they themselves don't even fully understand. Human performances (good ones, anyway) are exciting and unpredictable. This is why people are are willing to buy tickets to live performances of music that they already know by heart from recordings. The thrill of eighty musicians coming together to create a unique realization of a musical work conceived by a single human mind is what makes large-ensemble music fantastic and amazing.
So I'll gladly take a human performance that is technically imperfect but exciting over a robotic one that is technically flawless but equally soulless. Computers are OK for demos, but real, living music comes from real, living people.
But please, people, if you don't mind... play the right notes, huh?