I often tell people that the only things I truly know are secondary dominants and how to reverse a Python list.
It is a joke, but not entirely. There is a specific comfort in those two domains. A secondary dominant operates within a defined harmonic structure: given a key and a target chord, its function is precise and verifiable. Reversing a list in Python offers the same sort of clarity. reverse() rearranges the elements in place and returns nothing. [::-1] leaves the original untouched and produces a new list. The difference is small, and it is entirely knowable.
During my time at Howon University, a professor asked a student if he knew Al Jarreau’s Breakin’ Away. The student replied that he did. The professor then asked, “Have you memorized the chords?” When the student admitted he hadn’t, the professor said, “Then you don’t know it.”
At the time, the remark felt somewhat strict, but it pointed toward a specific definition of knowledge: partial familiarity is not knowledge. If you cannot reconstruct the structure precisely, you remain outside it. The demand was also proportional. She was speaking to students who had chosen the field and were capable of reconstructing it. The standard was strict because we were in a position to meet it.
In another jazz harmony lecture with the same professor, we debated whether a four-bar passage in a jazz standard constituted a genuine modulation or merely a temporary tonicization. I raised counterexamples, trying to find a boundary that would classify it. The professor listened and eventually said, “If you try to find deterministic rules in popular music, you will fail.”
For a long time I did not know what to do with these two remarks. One told me that anything short of exact reconstruction was not knowledge. The other told me that the exactness I was reaching for did not exist. Between them, I could not place my own uncertainty. When a passage refused to resolve, I had no way of telling whether I had failed to study it closely enough, or whether there was nothing there to find. The difference mattered, because one reading required more work and the other required stopping. I settled on neither, which in practice meant continuing without an answer.
I did look for a deeper rule before giving that up. The overtone series seemed like the place to find one, since it is a physical fact and not a convention: it explains why a dominant pulls where it does. Later I spent some time with George Russell’s Lydian Chromatic Concept of Tonal Organization, an attempt to deduce an entire tonal order from that same acoustic ground. Both taught me a great deal. Neither told me whether those four bars had modulated. They described the gravity in the passage without deciding where it had landed.
What I eventually noticed is that the two remarks do not answer the same question. The chords of Breakin’ Away are a fixed object. They were decided before I arrived, and I either hold them or I do not. The four-bar passage is not an object of that kind. What was in dispute there was not whether I had the material, but whether a rule existed that could decide the case at all. One remark is about what is given. The other is about what can be derived from it. They appear to conflict only because the same word, precision, is doing different work in each.
They are not two points on a spectrum, then. They are two layers, and the lower one does not become the upper one by being extended.
To know the chords of a song, or to know exactly which of the two reversals you have written, is to comprehend the mechanics. It is the groundwork that allows one to enter the space in the first place. Without that rigour, any claim to understanding remains lazy: a mere illusion of familiarity.
However, comprehension of mechanics does not guarantee a grasp of the whole. A song is more than its chord progression, just as an application is more than the logic inside it. The logic can be correct while the feature is wrong for the person using it, and no amount of inspection will reveal this, because the error lies outside the code. When these components meet the unpredictable nature of human experience, deterministic rules begin to bend or fray. If you insist on treating the entire world as a rigid set of instructions, you miss the music.
This is why I keep coming back to secondary dominants and Python lists. They are the small, deterministic grounds I can stand upon with certainty. But they are useful as something to push off from, not something to hold on to. Knowing where they end is what allows me to look beyond them.
Strict mechanics and irreducible ambiguity are not mutually exclusive. The first gives me a place to start, and the second gives me a world to inhabit. The harder part is what happens where the first runs out. The four bars still have to be played one way and not another, and the rule will not decide which. I no longer wait for it to. For now, I play the passage as though the interpretation were settled, and stay ready to hear it otherwise.
Related: Uncertainty and Participation, Control and the Uncontrollable, Rigour as Paralysis, Firmness Without Finality, Weight and Assent, What I Hope to Be Able to Say