The random page (191) describes how a black man found the word “niggardly” offensive despite the word meaning “miserly”. The offended guy thought the word was related to “nigger”, which is understandable, since “niggardly” is such an obscure word. Also, another black dude was offended when being addressed by his first name only, which was the normal way the offender addressed people. The offended guy thought it was a reference to times of slavery when black people were known only by their first name.
Links To Music Composition
The offended people mentioned here are particularly sensitive to racial slurs. American society seems a little over-sensitive to this kind of thing. Knowing this, it could well be that the “niggardly” example was a deliberate stealth insult, knowing the likely misinterpretation. A deliberate manipulation of emotion knowing the biases and likely ignorance of the target (and most people, in this case).
Perhaps it is possible to catalogue a bunch of musical “recipes” that induce specific emotional responses from audiences. This would be an interesting list to try to make. It would be facinating to read, and its existence would offend a lot of people that think there is something magical and intangible about music and it’s effects on us. I’ve struck similar reactions from people who react very strongly to the idea that a computer program could ever create music.
How could such a catalogue be assembled? Existing music could be categorised by the emotional response of audiences, although this may be different for different people. Perhaps there is a natural experiment that could provide some kind of input to this process. In popular music, there are often lists of top 40, hottest 100 and similar rankings. Comparing popular songs to unpopular ones might give a way of identifying some characteristics that make a song rank higher on these lists. This would not be an emotional response measurement, but might possibly be something that can be measured. I’m not sure how to go about this kind of analysis in a way that would yield something that was actually useful to anyone wanting to make music.
Music theory should be like scientific theory, you should be able to use it predict things and do experiments that have an outcome that is predicted by the theory. The experiments should be able to be repeated. Most music theory, in my limited experience, consists of finding patterns that exist in music that people already like. These patterns certainly exist, but it is usually pretty easy to come up with music that contains this pattern, but sounds terrible.
Paul Hindemith wrote an interesting book on composition that I read a while ago, The Craft of Musical Composition. It outlines his system of composition via a large set of rules. I even went to the extreme of writing a program that generated random note sequences that followed all of his rules. There were more than 50 rules. This is far too many for mere mortals to use. The program I wrote generated things that I wasn’t that fond of, so I listened to some of his music, and found that I didn’t really like his music either. So I figured that the best I would get out of these rules was music that sounded like Hindemith’s, so I stopped developing the program. Another problem that cropped up, was that sometimes the random algorithm I was using to generate the notes got into a dead end. It found situations where there was no allowable next note, according to the rules. It was probably a bug that I could have fixed, but it is possible that the rules themselves have a contradiction in them somewhere.
Hindemith’s book did have a great section on his classification of chords, ranking them into dissonance classes. He also had a method of working out the root of any combination of notes. His chord classification system included all combinations of notes. I even implemented this in a web page (with a dodgy interface, it works on Chrome).