Music creates context like language

[gnu23]

For the technically minded, an article in Nature:

http://www.nature.com/nsu/040614/040614-11.html

and D.H. Zanette's complete paper:

http://xxx.arxiv.org/abs/cs.CL/0406015

Enjoy!

[aMUSEd]

Very interesting. I wonder though if one went on to analyse later atonal works would you find the same applies? Zanette only analysed one, very early Schoenberg piece - too small a sample to extrapolate from I think - especially when compared to such a representative selection of tonal works. You could argue that the differences he found were attributable not to the atonal nature of the music per se but to the fact that Schoenberg was essentially creating a new musical vocabulary and therefore that it needed (and probably still does) time to fully develop before it could be compared meaningfully to the way more mature musical forms that have had generations and generations to evolve create context. Also meaning - whether musical or linguistic, evolves within a cultural context. I wonder if Zipf's "law" is a absolute as is being presented here once you go beyond western cultures?

[Sepheritoh]

In my opinion, this guy is f#$%ck stupid to waste his time prooving the obvious. Tonal music in a certain key would obviously have more frequencies in common than atonal music. It is only his personnal opinion if those pieces he annalysed are better or more soothing than other pieces. Any music with a higher veriety of tones will give a "flatter" graph as he refers to it. I don't need maths to proove that.

[peejunk]

Also meaning - whether musical or linguistic, evolves within a cultural context. I wonder if Zipf's "law" is a absolute as is being presented here once you go beyond western cultures?

Actually, in non-western musical traditions, I beleive this is even more so. Songs usually stay strongly on one scale, even if the scales aren't exactly the same thing like western tonalities. Also there is a lot of repetition and the music is much more organic and natural.

That's why I'm still strong on my prejudice that, in some aspects not so clear even to me, atonal and simmiliar avantgarde are somewhat of an unnatural intelectual excursion in an interesting but essentially (carefull with that axe Eugene) unmusical direction (hope I will survive to see the errors of my ways after saying that) if used per se.

Tho I have recently been shown some atonal works that are indeed an enjoyable listening expirience. I still think that rules of dodecaphonics, for example, are even more unnaturally restrictive than those of, say, functional harmony. I think that one should avoid all rules, and especially the rules invented to avoid other rules (a lot of what's harsh in atonal music is a consequence of trying to avoid any trace of tonality, as I've been told), and be natural in seeking whatever it is you wish to achieve with music.

And I also strongly beleive that music should carry a message, like they quote Schoenberg -- to be comprehersible (I'd just add accessible aswell).

I think that music without repetition and some set reference would be a musical equivalent of talking gibberish. This is just what I think, and this article somewhat supports my opinion.

[aMUSEd]

Maybe - that was just speculation - although you are probably right that many cultures stick to one scale some are more "tonal" than others to my ears.

But my first point - that perhaps what has been discovered is merely an artifact of the relative immaturity of a new musical language is may have more truth.

[peejunk]

But my first point - that perhaps what has been discovered is merely an artifact of the relative immaturity of a new musical language is may have more truth.

I think the problem with atonal (tho I honestly don't know much about it outside basic rules of dodecaphonics and few works I've heard, and sonic/rhythmic experimentations of the likes of Stockhauzen and Cage) is that it strongly tries to be different from it's antipod. Which is a silly way to go in such an unfair relation since traditional western music is strongly based on emotional, whereas atonal is much more based on the intelectual.

So this maturation will probably be in the direction of a holistic, omnious musical übersystem that would incorporate even harmony and melodic traditions of the near and far east etc., from which we would extract miniature systems for use on one or more works. I allready know few (formally educated, non-pop/dance/jazz) composers that work somewhat in this way.

[gnu23]

In my opinion, this guy is f#$%ck stupid to waste his time prooving the obvious. Tonal music in a certain key would obviously have more frequencies in common than atonal music. It is only his personnal opinion if those pieces he annalysed are better or more soothing than other pieces. Any music with a higher veriety of tones will give a "flatter" graph as he refers to it. I don't need maths to proove that.

I think the purpose of the paper was to provide a mathematical, or perhaps scientific underpinning for at least some of the "how" in "how do humans interpret music". There's a lot more to the subject, of course, such as "why do humans like music", "why do humans create music" and so on.

It would be interesting (for me anyway) to see how other works/genres of music come out of such analysis.

At any rate, I didn't post up the links to offend - just thought it was an interesting view on the subject we all come here to talk about.

[VicDiesel]

In my opinion, this guy is f#$%ck stupid to waste his time prooving the obvious. Tonal music in a certain key would obviously have more frequencies in common than atonal music. It is only his personnal opinion if those pieces he annalysed are better or more soothing than other pieces. Any music with a higher veriety of tones will give a "flatter" graph as he refers to it. I don't need maths to proove that.

I think you've rather misunderstood the article. The guy considers two Cs as two different notes. For a flat graph you don't just have to write atonally, you'd have to play as much in the top octave as in the middle octave as in the bottom octave. Which is not going to happen.

That said, the article strikes me as rather trivial. The curve for the Schoenberg piece has a very strange "plateau" look. It's a missed opportunity that he didn't explain that. Even stranger, the piece that looks the most like it, is the Bach piece, not the Debussy, as you might have guessed.

V.

[nuffink]

Stop press.

Scientist finds statistical correllation between 12tet and western language patterns. Extrapolates to the whole of humankind. Asian musicians unimpressed.

[Sepheritoh]

Sorry if I offended It is just my opinion that music is too subjective in these kind of studies. The person who do the test must start off by making a personnal judgment on the "goodness" of music. This type of thing was very popular in the 70's when everybody wanted to proove scientifically that rock music was crap.

In this case it seems a bit obvious to me that tonal music within a certain key will have on average more of certain frequencies. That does not make it better music. After all, a constant sine wave for 10 minutes would produce a very nice looking graph, but I would not find it very soothing

[Alive In Chernobyl]

It is very obvious what this article says.

I know many here have not been educated in music schools, but they teach you what is the same as musical grammar. Every phrase has puncuation in some way. Phrases must be complete ideas. The classical music is taught to be a language. Very true the language of classical music is taught to be german too. You can hear the national difference in composer in the ways the language they speak is given to the music they compose.

If you were to teach a person these rules with the movement of the body this can be analyzed to be a language too. This is a language we have formed

This is stating very obvious things Can I be a doctor too? I know many obvious things, and I am very skilled with a dictionary to give me big words! I can make graphs, and talk with a foreign accent too! How much more can you respect a doctor than a crazy russian?

[aMUSEd]

Stop press.

Scientist finds statistical correllation between 12tet and western language patterns. Extrapolates to the whole of humankind. Asian musicians unimpressed.

Exactly - does seem such a small sample to base such a "law" on.

[peejunk]

This is stating very obvious things Can I be a doctor too? I know many obvious things, and I am very skilled with a dictionary to give me big words! I can make graphs, and talk with a foreign accent too! How much more can you respect a doctor than a crazy russian?

Dunno really. Maybe in America it would work, here where I live everyone has a Slavic accent when speaking English, and anything that can remotely be connected with communism, isn't as popular as it used to be

But big words work everywhere, and if you can say big words in Russian (and throw in a latin or a german one here and there, like omnius and eigensystem) the audience is yours.

[nuffink]

I'm not knocking it. It seems self evident that langauge influences music influences language ad infinitum.*
Since serialism was the result of an intellectual experiment rather than the product of a cultures need to communicate. I would expect it to mathematically deviate from that cultures language.
Or you could argue that, since mathematics is the lingua franca of the modern world, serialism was also a cultural necessity.


*How else do you explain the fact that white men can't rap? {Ducks}

[peejunk]

*How else do you explain the fact that white men can't rap? {Ducks}

Because they use too many latin words?

Yes, about the rest, you've pretty much said what I've wanted to say without boring everyone to death and in a precise way.