> We don’t start our investigations at some arbitrarily chosen point in the past; we begin where we are, from our current burning passion. That reveals a probl

> We don’t start our investigations at some arbitrarily chosen point in the past; we begin where we are, from our current burning passion. That reveals a problem with learning backwards-- rampant epicism in which students would rarely choose to learn about, say, dead ideas like sonata form. Rather than requiring the student to learn linearly through time, the professor ought to select a handful of musical periods and/or case studies and then randomize the order in which they are taught. Make a big spectacle of picking numbered ping pong balls from a bingo machine in the lecture hall. In the U.S. these are usually elective classes so they are essentially an institutional form of entertainment, anyway. In other words-- it shouldn't matter how you pick a period/style/composer. What should matter is that students improve their ability to use their ears and their growing knowledge to investigate any music, no matter its origin or time period. Publicly using randomness to select from possible periods/cases would show good faith that the professor is skilled enough to leave the chosen topic up to fate. That would make it harder for profs who aren't fully comfortable with all the material, which is probably why nobody teaches this way. :) Edit: clarification