Academic music theory is descriptive. But it's a bit like Latin. You don't learn Latin to speak it, you learn it to understand the foundations of other related languages. It makes it easier to learn them because you have a ready made meta-template for grammar and vocabulary.
Informal music theory is absolutely prescriptive. You cannot write music in a recognisable genre without following the rules of that genre for instrumentation, production, use of rhythm, harmonic colour, melodic form (and sometimes specific melodic cliches), vocal/instrumental stylings, arrangement choices, and decorations.
All of those are invariants for genre. Some of the options cover a wide space of potential choices, but anything outside a genre boundary is very obvious and most people can hear it instantly.
Of course these rules are rarely written down, and most musicians pick them up by ear.
The only difference with Bach etc is that attempts were made to write down the rules. These became Music Theory™.
But in fact the rules don't come close to describing what Bach etc were doing, so they're mostly a poor and misleading attempt. Recently people like Gjerdingen have been expanding on traditional academic theory by going back to the original historical sources - not just the music itself - and examining what and how composers of that period were taught. And sometimes why.
Meanwhile naive statistical analysis is pointless and even stupid. Baroque music is tightly structured and all the elements interlock. So saying "The bass moves by step 50% of the time" is a non-fact.
So - yes it does. But the problem is knowing why at that particular point in that particular piece. And statistical analysis won't tell you that.
Informal music theory is absolutely prescriptive. You cannot write music in a recognisable genre without following the rules of that genre for instrumentation, production, use of rhythm, harmonic colour, melodic form (and sometimes specific melodic cliches), vocal/instrumental stylings, arrangement choices, and decorations.
All of those are invariants for genre. Some of the options cover a wide space of potential choices, but anything outside a genre boundary is very obvious and most people can hear it instantly.
Of course these rules are rarely written down, and most musicians pick them up by ear.
The only difference with Bach etc is that attempts were made to write down the rules. These became Music Theory™.
But in fact the rules don't come close to describing what Bach etc were doing, so they're mostly a poor and misleading attempt. Recently people like Gjerdingen have been expanding on traditional academic theory by going back to the original historical sources - not just the music itself - and examining what and how composers of that period were taught. And sometimes why.
Meanwhile naive statistical analysis is pointless and even stupid. Baroque music is tightly structured and all the elements interlock. So saying "The bass moves by step 50% of the time" is a non-fact.
So - yes it does. But the problem is knowing why at that particular point in that particular piece. And statistical analysis won't tell you that.