Toward a first-year project.

[eirias]

Haven't updated this in ages; stopped doing much reading for a couple months and now that I've started again I really ought to try updating. It's a useful thing.

For now, though, I have a question. If you were going to invent a new system of music different from Western music, how would you do it? What kinds of regularities would you include?

Comments

[ukelele.livejournal.com]

Woot, hi!

Um...now I should say something useful, but I don't actually know very much about music. So instead I will babble about meter, which is vaguely related.

English uses a stress-based system system of meter ("accentual-syllabic," I think the jargon is, but don't quote me on that). The point is, poetic meter is based on repeating patterns of stressed and unstressed syllables. The base unit tends to be quite small and varies little or not at all when it is repeated.

Greek meter was based on the quantity of syllables -- a naturally long vowel, or a syllable ending in a consonant (usually), is long, whereas a syllable with neither of these properties is short. Meter is based on repeating patterns of longs and shorts. However, the base units can be extremely long and complicated (like, more than a line), and many of them permit a great deal of variation. (For example, the English iamb is dah-DAH; the Greek iamb is foo-long-short-long, where foo can be long or short, though it's more likely to be long in some places than in others in the line, not that I can remember which those are.) Latin meter was stolen from Greek meter, but with a lot of the complexity stripped out, and a lot of fudging because the language does not inherently take to this sort of meter.

Welsh meter, iirc, is based upon pitch. Not that I know how this works, but I can certainly see that, if your language involves rising and falling and whatever-else pitches, you could base a metrical scheme on that.

Japanese meter is based on syllable counts, but I don't know how. French meter is also based on syllable counts, what with French not having a lot of natural stress variation among syllables (or, really, as far as I can tell, much variation of any kind, including, say, of the basic sounds involved).

Alright, poetry is kinda like music, so you can take it over from here ;).

[eirias.livejournal.com]

*nods* Yeah, a lot of that stuff has come up in talks I've had with Jen; psycholinguists refer to it as "stress-timed" vs. "syllable-timed." Japanese is syllable-timed, hence the superiority of haiku in its native tongue :). A fellow grad student in my lab does stuff with prosody and word segmentation. Apparently trochaic words are much easier for infants to learn than iambic words, at least in Dutch & English. Unfortunately, there are too few iambic languages for us to really be sure what that means.

The things I'm most concerned about replicating, though, are below-surface things like grammar. I'm interested in the logical relationships between musical "chunks" - can we create something that's equivalent to a false grammar using musical stimuli?

I'm going to talk to Orkid about this later and see what she has to say. Seeing how another musical system (Indian classical music) differs from my own may be a good starting point

[ukelele.livejournal.com]

Uh...that's weird. I tend to think of there being not really any difference between iambs and trochees (and man, I'm gonna have to hurt myself for that grammar). I mean...in extended blocks of speech, there aren't. Though I guess single words you could tell the difference...summer versus ...uh, OK, I'm having real problems thinking of an iambic word here...maybe that would be why...okay I'm seeing your point...select. Hum. Now I'm very curious about the etymologies here -- I wonder if non-trochaic words disproportionately from far-off bits of the big ol' linguistic tree? I mean, "select" has to be Latin, but I don't know where "summer" is from.

*hrm* I noticed the other day, watching soccer on Univision ;), that Spanish tends very strongly to accent every other syllable...I mean, way more than English does...I wonder if that's part of why it sounds so fast to a lot of English speakers, the unbroken pattern of quick meter. (That and the extremely pure consonants, which I imagine are rather faster to say than all those wacky diphthongs other languages have.)

*hrm* I wonder how people's reactions to, say, (a) languages they don't know (b) nonlanguages compare to their reactions to (a) musical systems they don't know (b) there is no b. I mean, when I am in a Chinese restaurant and hear some wacky pentatonic thing going on, my innate reaction is "that is not music." My brain has to say, no, no, it is, but my emotions are really unconvinced, because the rules are so foreign.

*hrm* Wasn't a whole lot of twentieth-century music all about breaking the rules of counterpoint? Yeah, twelve-tone, baby. Not that I want to condemn you to a life of Schoenberg or anything.

Uh, yeah. Back to what you said. Exploring another musical system. Very clever. Your research. Bitchin' cool. Me. Envious.

[dolohov.livejournal.com]

Japanese is entirely syllabic, and so counting syllables is trivial to a native speaker. (Non-native speakers have a little trouble with this, since the syllables, "nu", "nyu", "u", and "n" all have the same length, which gets tricky)

[ukelele.livejournal.com]

Oh, it wasn't counting syllables that was the bit that left me totally confused. It was the thing where I asked the native Japanese speaker with a lot of formal poetry knowledge "so, I hear Japanese is syllabic meter, what's up with that?" and he said something like "uh...it's not what people think it is with 5-7-5 or whatever...uh...it's complicated."

[mokatz.livejournal.com]

There's always the Eastern musics, who, of course, DON'T base their music on any sort of Doric, mixolydian, or otherwise scales. Then there are even the quarter tones, where the octave is broken up into even MORE than 12 evenly spaced steps.

Something that might be interesting, but probably horrible sounding, would be to break up the octave into 16 step, thus being a nice power of 2, sort of mirroring the whole, half, quarter, eigth, etc. of meter. Each step would be (Frequency of previous note) * 2^(1/16) ~= F*1.04423, as opposed to F*(2^(1/12)), for the western scale.

Thusly for an A440, one would have frequencies of 440, 459.48, 479.82, 501.07, 523.25, 546.42, 570.61, 595.87, 622.25, 649.8, 678.57, 708.62, 739.99, 772.75, 806.96, 842.69, 880.

It might also be interesting making some type of frequency ratios relating to pi (http://mathworld.wolfram.com/Pi.html), e (http://mathworld.wolfram.com/e.html) or the Golden Ratio (http://mathworld.wolfram.com/GoldenRatio.html) (this one ought to be interesting), or maybe even the Fine Structure Constant (http://scienceworld.wolfram.com/physics/FineStructureConstant.html). Either as the basis of the "octave", or as the ratio between notes.

my $0.02

[rshruti.livejournal.com]

So I clearly have a lot of knowledge on a non-Western system of music, and one that's pretty comprehensive on the melodic end of things, too. I could probably give you more specific suggestions about how to structure a false grammar if I knew why you wanted to do it. What things about it are you trying to test? Because I can guarantee that if you take, for instance, a group of different scales, and made up little melodies in each one, children would be able to tell which scale is which (if by no other way than by saying that one melody reminds them of a different melody in the same scale). If you play a melody that starts in one scale and switches to a different one halfway through, a child familiar with the scales will know when you switched. In my experience, kids are pretty sensitive to that sort of thing. I don't know how sensitive they are to meter; if you switched from 5 to 7 I don't know that they'd have any idea. But if you switched from 3 to 5 that's a bit more obvious, so maybe they'd pick it up. Children have also been known to make up their own melodies... start them off in a scale and tell them to finish it and they will stay within the scale.

I wonder if any studies have been done on this sort of thing. I'm speaking purely from observation of kids who have parents interested in music...