Music/ Language Relation, Frequency of Pitches, Intervals, Composition Flow Chart

DOFTS Books on creativity say to use an abstract metaphor. That music is like a language was taken. Also, it was easy to find a list of letters ranked by percent of use. It's rather difficult to find Phonemes ranked by frequency and probably even more difficult to use.

What groups do you imagine were monotonic? Were in fact any groups discussed except a group of letters and notes. Then you imagine that it was "mistook" and not intentional. Or just said that so as have something to post in order to contradict what was posted. And yes, that's what the scientists were referring to. Duh!

I put up photos in my Music Composition folder under pics- the voice-leading probabilites of every diatonic note. Taken from 1,000s of melodies. They graphically represent the likelihood that C, for example, will go to D or that G will go to A etc... There's also a tentative rhythmic probability chart. MySpace.com - Greg - Las Vegas /Bali, Indonesia - Ambient / Electronica / Experimental - www.myspace.com/gongchime

Originally Posted by Gongchime

DOFTS Books on creativity say to use an abstract metaphor. That music is like a language was taken. Also, it was easy to find a list of letters ranked by percent of use. It's rather difficult to find Phonemes ranked by frequency and probably even more difficult to use.

What groups do you imagine were monotonic? Were in fact any groups discussed except a group of letters and notes. Then you imagine that it was "mistook" and not intentional. Or just said that so as have something to post in order to contradict what was posted. And yes, that's what the scientists were referring to. Duh!

I probably spoke over your head, Oh well. When you compare sets you form groups, group theory 101. When you see relationship between group 1(music) and group2 (language,) you are doing a study of group theory. You then broke each group into subgroups. The subgroups you choose are defined to be monotonic because their exist an upper and lower bond within the set. More so it is increasing or non-decreasing.

You shouldn't even be operating on list of what appears more often. If you want to express a true relationship, it's often better to do an analyst of the development of language and what music was doing during that time.

For example, when people spoke old english, how did music sound? What relationship occurred? Yes, I know it's tough work to answer those questions and it actually takes time and effort. But that's research.

Every result you posted earlier is trivial. It's like saying, pi is irrational because it isn't rational. Well yes...

Ah trivial in the context of logic but not trivial in the context of Creativity in which case any abstract metaphor whether of early English or late, early music or late would take us somewhere interesting. Requiring that there be a correspondence is overly rigid and would not create a sufficiently abstract metaphor which is the point. You are trying to impose something you know about onto something that you don't in an attempt to appear to know what you're talking about. In fact, for creativity, the more abstract the metaphor the better. Don't ask people trained in car design to come up with a more novel type of car. Ask someone completely on the fringe of car design such as someone engaged in hydraulics to come up with a new design and you will get a car the accelerates from zero to sixty without turning on the engine because all of the energy normally wasted in deceleration was stored as pressure in hydraulic fluid.

Originally Posted by Gongchime

I put up photos in my Music Composition folder under pics- the voice-leading probabilites of every diatonic note. Taken from 1,000s of melodies. They graphically represent the likelihood that C, for example, will go to D or that G will go to A etc... There's also a tentative rhythmic probability chart. MySpace.com - Greg - Las Vegas /Bali, Indonesia - Ambient / Electronica / Experimental - www.myspace.com/gongchime

You might want to dig a bit about the use of transition tables in algorithmic composition. What you described there is a first order transition table: A table that shows the probabilty of a specific note with knowledge of exactly one note before it. A table that shows the probabilities of notes regardless of the note before it would be a zeroth order transition table; a table that takes into account -two- notes before the current one would be a second order transition table, and so on.

Using such methods in music was first done in the 1950s by Harry F. Olson, who analysed songs by the composer Stephen Foster and created probability tables for the pitches. He also did the same with rhythms. Then he generated melodies with these (zeroth order) transition tables, but soon realized that without taking into account previous notes those algorithmic compositions didn't resemble Foster's songs very much. Similar experiments with higher order transition tables generate results that sound less and less random, however the amount of data rises extremly fast with higher orders, so transition tables with very high orders are quite impractical. And while it is indeed possible to create quite "human sounding" compositions with transition tables of high orders, this generally only works for very simple music, such as folk songs. For more complex music other types of "artificial intelligence" provide better results.

Interesting stuff, in any case!

Right Gardener, that is the next step. Like second order Markov chains.

I'm still waiting for an abstract.