I don’'t know how feasible this project is, but I think it’ll be interesting to try, and hopefully in the process educate myself on the history of music. Let’s try to make a taxonomic tree of music from the western tradition. The top level domain I want to consider in this thread is western music, because trying to include all music from every culture out there seems like too big a task, and lumping in traditional music from various unrelated societies into one big category would also be incorrect.
Let’s start at the kingdom level, and maybe get down to the phylum level as well. Should this project turn out to be feasibly, I think other threads for lower levels, especially once we get past the phylum level, might make sense. The two big ones are of course small c classical music and small p pop music. What other styles of music are different enough that they should have their own kingdom rather than being grouped under classical or pop? The first style that comes to mind is jazz. What other genres are different enough from classical and pop music that they should be at this level, meaning that they don’t fit into either the classical or pop category?
I have seen a couple infographic books with a detailed tree-like map of all music genres as well. It would be easy to add titles like kingdom, etc. but I don’t think they did that. Once you get to modern techno variations, for example, the difference between subtypes is very modest.
Exactly! I’m going to need a bigger monitor to really get into all the details .
ETA: Or new glasses. At first I thought the breakbeat category said breakfast. How can there have been a genre of breakfast music all these years that I’ve never heard of? That was my first thought .
I’ve shown my cartography students this riff (so to speak) on the famous London tube (subway) map, where each tube line is a genre and each station a group or artist:
Just rock, pop, and related genres (and almost all English-speaking) since about 1955, though.
Very much not my field: but isn’t “western music” the result of having been influenced by just about every culture out there? How would ignoring that produce an accurate tree?
That’s a great site, thanks! I have a t-shirt with a similar, but rudimentary taxonomic chart of the history of rock. To match the map on the site I’d have to wear a circus tent.
Being influenced by something does not make the music genetically non-Western. I can vaguely imagine a vastly extended tree where all the influences are taken into account; e.g. Romantic + Hungarian folk + Carnatic → ??
Even better, we should have the computer construct it for us. Maybe someone has already done it, though? Getting the data science/music informatics right is not trivial, though, especially if we want to avoid guiding the computer as to musical features or genres and have it figure that all out for itself.
I suppose you could generate a directed graph where every genre, or every song even, is connected to its influences and what it influenced. But there’s no reason to expect that that graph would be planar, which would mean that it’d be a pain to put it on a piece of paper, or a static document on a screen.
That is kind of what I had in mind, except I wonder if the “genes”
can also be determined from the data. E.g., I do not recall the context but in one classification problem it turned out there was a difference between music in major keys and music in minor keys— that transpired naturally and was not fed in as a parameter in advance.
If the number of edges is several times the number of vertices then it starts to get messy. However, if we limit the number of immediate influences then the crossing number may not be too bad.
.
.
