The argument rages (or simmers, depending) in academic circles whether the music theory canon is or isn't a good idea. I am definitely on the side that says it is not.
I would replace it with musical logic, a function of listening rather than of identifying elements and functions. Listening is a complex activity which, at its best, eschews repetition and predictability. That musical logic is innate to the musical mind was amply demonstrated back in the 50's by Viktor Zuckerkandl. His work made it evident to me that I could teach music as a way of deepening my own experience while imparting the basics to very young as well as to older learners.
His point is that mistakes reflect sensible alternatives to the composer's designated choice. That has meant to me that attending to mistakes reveals the essence of musical composition. Even better, attending to mistakes builds confidence on the part of the student, who thus learns to recognize and trust her own musical intuition at work.
I recall being praised for my playing as a teenager, though I had no idea what the praise was based on. My intuitions were probably interesting; they were certainly strong. But to what did they correspond? I had no clue.
Until Brahms. I was given a volume of Brahms pieces, Peters Edition, at the time considered "the best" edition. Dutifully observing the phrasings and articulations, I found Brahms to be a most unmusical composer. Imagine my double shock when, in college, I stumbled on the Kalmus edition, at the time considered "the worst," of the same pieces. The phrasings and articulations, all radically different from the Peters, turned out to be from the Authentic Edition, prepared by Brahms himself.
It was pure musical logic at work. I sometimes wonder what might have happened to my development had someone taken the trouble to analyze with me the differences between the two editions as they spoke to my musical intuition in such contradictory ways.
