Ever since I learned to play an instrument, I was fascinated with music theory. I found it addictive because so many of my favorite songs could have their structure described by it. It was like seeing the code behind the matrix; the deeper pattern that held everything together.
After I started working with computers a few years later, the two interests combined in experiments with music composed by the computer. It came from a beautiful idea: if music theory is the DNA of what’s happening in all of your favorite songs, maybe you could write a program to arrange that DNA in new ways, into great things the world had never heard.
It turns out that while you can compose music with computers, the results are never as good as you expect. It took about another decade to realize why this was, and what it all meant. In the process I learned a lot about theory and creativity, and I want to try to capture that to clarify my own thinking, and share in case it’s useful for others.
Does a theory tell us what is right, or just what is?
One lesson that I learned was the difference between descriptive claims, and normative claims. A descriptive claim is one that describes something that’s happening. Normative theories tell you what ought to be, what is correct.
Let’s look at your language’s dictionary as an example. Some people will tell you that dictionaries are normative, that is, they say what’s correct. But it isn’t so! Language develops over time and dictionaries are always running to catch up. So we would say that dictionaries are descriptive of the English language.
If you’re writing a computer program to generate new sentences, and you don’t understand the spirit of how people create new words like “Brexit”, “mansplain”, or “snitty” then you’ll miss a lot of what’s going on. Similarly, music theory helps explain what’s happening in a song, but it doesn’t help you see how artists color outside of the lines, or even play with your perception of where those lines are. And the more you look into music, the more you find that’s where the most sublime moments are found.
This is the first important idea:
Most theories outside of the hard sciences are descriptive, not normative, and descriptive theories about the world have limits. Not all true things are described by them no matter how good the theory.
Does theory help creativity and innovation, or hinder it?
Suppose you could peer into the code of what is “good writing” or “good music”. You’ve got a book by your favorite author that identifies all of the practices to help do it well. But you know that when it comes to music or writing, you have to do it in your own way, not simply copy someone else’s style. Does all of that theory knowledge help you or hurt you?
I have found that there is no consensus here at all. Some people feel that theory helps get them out of a rut, and identify new possibilities of where a song could go next. Others feel like knowing too much about theory really hurts you, because you’re too stuck in the rules, it hurts your ability to break them. This goes back to the normative versus descriptive view; if you look at theories as normative, they will constrain you.
Once you start looking, you can see this confusion in many places. In music, there are theoretically excellent musicians who turn out something that sounds impressive, but somehow soulless; missing something you can’t quite put your finger on. Then there are raw talents who have zero training who turn out beautiful music. But just as easily you’ll find great musicians who will tell you that understanding what’s really happening underneath of music deepened their love for it and expanded their possibilities. This is the second idea:
Descriptive theories are not absolute truth, they are just words on a piece of paper; they can be used as a lens to see some things, but what’s most important is what you see with them, and what you make of it.
Every Person’s Mind is a New Universe
When we use a technical word like “theory”, it makes many people think in terms of the scientific. That the stuff we’re dealing with here is like a lab; a neat row of beakers full of exact amounts of pure chemicals. But the moment we get into something with an aesthetic angle, like music, writing, cooking, or anything related to art, we aren’t in the lab at all.
In short, theory outside of the hard science world is almost always descriptive, not normative.
When thinking about music, I wanted to make “good music”. But it took years to understand what that meant. In the end, it’s about creating an emotional impact in the listener. It’s not about impressing them with how great of a player you are, and it isn’t even about communicating a grand idea. But how do you create that emotional impact?It turns out that you don’t! The listener does it for you.
Inside of every head is a complex mix of different things, even that person cannot identify or measure for you. And in terms of theory’s ability to create “good music”, this is pretty much a death blow: theory doesn’t have a way of talking about people’s internal state, or even acknowledging that they exist.
In my 20s, my all-time favorite band released a new album. I already had 4 of them, and I was really well primed to love this new one. After an extremely stressful week at work, I put it on for a listen and was shocked at how mediocre it was. A total disappointment. This was an album with rave reviews, in my favorite style, everything about it on paper ought to have been perfect but it flopped for me, and it puzzled me. That same weekend, a family member told a story about how he had a great time at a new restaurant, catching up with an old friend. He raved about how great the food was, but then said he should have known it would be great, because “what you put on the seats is more important than what you put on the plates”.
Considering the mediocre album, or the excellent restaurant, does it matter whether my favorite band knows anything about music theory, or whether the chef at that restaurant studied well? The more I thought about this, the more I questioned the whole idea of reviews of things. This is the third idea:
we’re all at the center of our own universe, people project on to anything with aesthetics, and the creator can’t control most of how the work arrives. There is no goodness measure where some people score high and others low. There is only how the work arrives.
Forthright 