What’s the difference between good & bad music?

How to predict the next note?

When I asked her if we could predict the next note in the music based on previous notes, my piano teacher said a hard “No!”. Although I agree (because she’s trustworthy), my mind can’t make sense out of it. So here I am, making you think about it (it doesn’t matter if you are interested in music; if you are reading this sentence, you’ve got to read the whole thing now).

A thought experiment

Consider yourself Einstein, and let’s do a thought experiment. You are sitting in a room, empty, with just a piano. There are 88 keys on it. You start with the note E (if you want to imagine, it’s the first repeating note in the Interstellar theme music). You have nine of your fingers empty and 87 keys to try out. You can either press with all your ten fingers or just choose one for the next note/chord.

Now the probability is simple to calculate, so it’s not worth wasting our time there. Let’s just say it’s in the trillions (and if we keep “time signatures” into account, it’s pretty much quadrillion+). So, if there’s no structure to it, it’s practically impossible to predict the next note in the sequence in probabilistic terms. Keep this in mind.

Effect of Evolution on Music

The next point: we, as humans, learn music based on evolution. We found “beat” because we started walking. If we were limbless animals, we would have never found beat. We found melody, obviously because of the birds and the sound of the wind. This is all intuitive. I will take that for granted. So, when we listen to Interstellar, the next note in that piece is A. Hans Zimmer is a genius, but in order to make it sound beautiful, he has to follow human intuition, which is again based on evolution. So, as much as he wants to experiment with music, good music is not that random; it has to be ‘good’ sounding for the brain, a product of evolution. But the probability of the next note is still not very low, still in the trillions. Keep this in mind.

Physics of Music

Sound waves are the result of back-and-forth motion of air molecules (it’s false but enough for our discussion). Now these waves create harmonics/superposition/overtones/undertones. Basically, when the two waves combine in a certain way, they create a new wave altogether! A single note on the piano, let’s say a wave, and if we play a bunch of notes together (or one after another while the other note still can be heard), we create these overtones/undertones/superpositions. This makes a song, well, music. Keep this in mind.

Neuroscience of Music

Our brain is a neural net. It is just a calculating math machine. To decide something is good (pleasant to the senses) or not. When we hear the song, our brain does some fancy calculations (although it’s never prevalent in our ‘thinking’ style). As we hear more and more songs, better and better songs, combined with our evolutionary knowledge, we tend to understand music better, we tend to hate garbage music. And it doesn’t require education just to have a good taste of it (though it helps).

How do we ‘understand’ music?

What I want to convey is that there are trillions of possibilities for writing/playing the next note, yet (some) humans do it better than others. This ‘better’ or not depends on evolution and an individual’s experience with good taste. Put trillions of possibilities next to a dinky few-kilogram human brain. How the heck does it claim to ‘understand’ music? That’s my first question.

There are many AI models that can produce songs now. A dumb idiot like me could build a model. But all of these models are trained on human music, with a trillion probability, not quadrillion. Again, these are classical computers; a Quantum computer will be able to handle quadrillions very effectively. So, we wouldn’t have to worry about data in that sense. If that’s true, any art form, be it music, painting, poetry, or writing, is it a waste of time? When elitists boast about having good taste, they think they know better, but in reality, there’s no good or bad probability (?).

In search of good Music

This debate extends to anything related to truth—what we care about and what we don’t. As breathing, living humans, we resist confinement to boxes that define us. If someone finds ‘tiktok‘ music appealing, they have every right to enjoy it. Likewise, those with musically superior minds expressing themselves through a classical symphony deserve recognition. My question isn’t directed at those who appreciate this art form; rather, it’s aimed at those who comprehend its intricacies.

People frequently discuss the depth in music, and I’m curious about what they mean. They often attribute it to those who think deeply and listen to mature music. Can we reverse this process? Can listening to mature music lead to a mature mind? If so, I am genuinely in search of this mature music.

---

One of the moments of happiness for me is listening to good music. Honestly speaking, I never had a good taste in music (and I still don’t think I do). However, due to the influence of musically gifted friends and Spotify, I have recently started widening my musical horizon. It’s a new alley for me, as I was never exposed to this kind of learning method. This is the first time I am considering it as a hobby. My aim is to someday find joy in it, not to make money or have any practical returns. The following thought experiment comes from my studies in neuroscience, physics and probability. This is not intended to disrespect anyone. I am an ultra-beginner student of music, with the IQ of a monkey 🙂

Book recommendations:

1. Musical Human by Michael Spitzer