Origins of music: new theory

Subject: Origins of music: new theory
From: Steve Parkes
Date: 07 Aug 03 - 11:41 AM

A report in New Scientist suggests that "Key universal features in world music may be based on acoustic
quirks of the human vocal tract, suggests a new study".

Interesting stuff for you musical fundamentalists!

Steve

Subject: RE: Origins of music: new theory
From:
Date: 07 Aug 03 - 11:46 AM

Wow! That is extremely interesting.

Subject: RE: Origins of music: new theory
From: Steve Parkes
Date: 07 Aug 03 - 11:47 AM

This is the piece in full (in case it disappears in a few days):

Musical roots may lie in human voice
13:28 06 August 03 NewScientist.com news service

Key universal features in world music may have their roots in the ever-present sound of the human voice during the course of evolution, suggests a new study.
The analysis of thousands of recorded speech samples found peaks in acoustic energy that precisely mirror the distances between important notes in the twelve-tone scale, the system that forms the foundation of almost all music.
"The mysteries of music have a biologically principled explanation," says Dale Purves, at Duke University, North Carolina, and lead author of the study. "A reasonable speculation is that we hear these tonal relationships because they are involved in our interpretation of each other's speech."
As a slide whistle shows, it is possible to change seamlessly the pitch of a sound from low to high and back again. But for making music, human cultures have sliced the pitch dimension into twelve distinct tones.
This twelve-tone "chromatic scale" can be heard by starting at any piano key and then playing the next dozen white and black keys in succession. On the thirteenth note, the scale begins again, one octave higher.

Pythagoras's theorem
Different musical traditions have characteristic sound because many cultures have devised scales from a subset of the full chromatic scale, with different distances, or "intervals," between the tones. Chinese music is based on five-tone scales, while scales common in Western music have seven tones.
But all cultures favour certain intervals from the chromatic scale, and listeners judge these same intervals to create the most harmonious combinations of two tones. Pythagoras proposed that such preferences could be predicted from mathematical relationships between tones, but these approaches have yet to provide a complete explanation.
The Duke researchers randomly extracted over 100,000 speech samples, each 0.1 second long, from recordings of thousands of English sentences. Acoustic analysis of the combined samples revealed 10 frequency peaks that match the most significant intervals used in musical scales worldwide.

Mandarin and Farsi
Moreover, the relative heights of the peaks backed numerous studies in which listeners ranked the harmoniousness of intervals. Speech in other languages - Mandarin, Farsi, and Tamil - also displayed the same pattern.
The frequency peaks are caused when a sound wave from the vocal cords is shaped by resonances of the throat and oral cavity. The researchers say that, aside from animal calls, speech emanating from oscillations of the human vocal cords is virtually the only natural sound that we hear as tones.
This fact, combined with the new finding that preferred musical intervals are better predicted by the acoustic quirks of the human vocal tract than by mathematics, leads the scientists to argue that the structure of music is rooted in our long exposure to the human voice over evolutionary time.

Journal reference: Journal of Neuroscience (vol 23, p 7160)
Peter Farley


Heres a link to some cool stuff at Duke Uni

Steve

Subject: RE: Origins of music: new theory
From: JohnInKansas
Date: 07 Aug 03 - 12:24 PM

The most important thing in the notice is that they actually did and experiment:

The Duke researchers randomly extracted over 100,000 speech samples, each 0.1 second long, from recordings of thousands of English sentences. Acoustic analysis of the combined samples revealed 10 frequency peaks that match the most significant intervals used in musical scales worldwide.

The result confirms (if test conditions and data reduction methods were appropriate) what has been generally assumed, and has been found in other tests.

It is generally also held (with some experimental confirmation) that the same result would be obtained from a similar test of bird calls, animal grunts, and virtually any other "natural" complex source.

The "interval matching" observed by the Duke researcers has, in fact, been applied to studies of "communication" between animals. In very simplistic terms, if two animals "harmonize" it is (by some researchers) assumed that they are communicating with each other, if there is no "common interval" relationship, then it's more likely just that they felt like making a noise.

The research performed, and the result obtained, appear to be a worthwhile bit of work. The anthropomorphization evidenced in the "announcement," by trying to "prove" that humans are "the source of everything," went out of style several hundred years ago. What is really "proved," perhaps, is that the human vocal tract has evolved to produce, with some facility, the intervals that are "easiest" to produce and that are easiest to distinguish (and hence best for carrying information content?).

Of course, the tone of their announcement is well suited to attract attention, perhaps, from a broader range of persons. It's a forgivable bit of "adspeak," if their full publication shows accurately what they did and what result they obtained.

John

Subject: RE: Origins of music: new theory
From: Bert
Date: 07 Aug 03 - 12:28 PM

But it could have been the other way around, that the patterns in speech are the result of generations of exposure to music.

Subject: RE: Origins of music: new theory
From: JohnInKansas
Date: 07 Aug 03 - 12:55 PM

More likely the middle way forth:

Both music and speech (and grunts, whistles, and screetches) likely evolved from the same physical "fact" that sounds with certain fairly specific intervals are much easier to produce, at the same time or in close sequence, than "unrelated" intervals.

Or, if you're of other persuasion, "she made it that way to be nice to us."

John

Subject: RE: Origins of music: new theory
From: GUEST,Guest
Date: 07 Aug 03 - 03:44 PM

They say a lot, but give no results that can be compared with other studies, and there is a lot of real data to be compared.

Experimental results on harmonizing, as recognized by the human ear, are given in a graph on p. 167 of the 3rd edition of Juan G Roederer's 'The Physics and Psychophysiics of Music', 1995. [See also his bibliography, where there's a lot of real data noted.]

What do any number of normal speach patern's of any miodern language have to do with speech capacities of the originators of music? They may have discovered harmonization by beating sticks or femurs on hollow logs of various diameters and lengths and material constants
before they had any language.

Subject: RE: Origins of music: new theory
From: Malcolm Douglas
Date: 07 Aug 03 - 08:53 PM

I've heard it suggested that language developed from music, rather than the other way around, but in the end I don't see that it's of any importance. Language and music are essentially the same thing. What does surprise me is that something so self-evident might seem surprising or worthy of remark. Presumably there was a worthwhile research grant in it, though, so someone at least has had their rent paid for a year or two on the strength of it.

Subject: RE: Origins of music: new theory
From: GUEST,Q
Date: 07 Aug 03 - 09:24 PM

Guest (above) with his remarks about beating femurs and sticks on objects is closer to what is in my racial memory, except that other objects besides logs were sounded. The sound of the thunk on the skulls of the enemy my ancestors ate compared with the thunk on the head of a recalcitrant female may have contributed to the development of low notes....

Subject: RE: Origins of music: new theory
From: Teadoir
Date: 08 Aug 03 - 02:42 AM

The above sugestion that harmonization by beating what ever with what ever may have preceded langauge seem a bit odd to me. Possible, but one doesn't find many examples in the animal kingdom does one? Especialy amoungst the primates. Show me primate that does such, with out vocalisations, and I'll be flabergasted. but maybe you were joking?

Subject: RE: Origins of music: new theory
From: JohnInKansas
Date: 08 Aug 03 - 02:59 AM

I hate to agree with you, Malcolm, but the "language and music are essentially the same thing" seems fairly accurate and appropriate here.

One of the problems of what passes for "science" is that what seems obvious often doesn't get tested. The measurements they purport to have made are the boring sort of thing that sometimes does need to be done. (And solid "science" depends on many people being able to do the same thing and get the same result.) If their data, and documentation of the method, support the result that they found, then others may be able to do some other experiment instead of that one.

My main objection is that they attempted an unfounded (from their results) extrapolation to a cause/effect relationship that is apparently more a result of their "wish to show" something than it is a result of their experimental result. I'd write that off as the attempt for a "snappy headline," ... perhaps.

As to the preceding annnonnymingous comment - "They say a lot, but give no results..." you have to consider the source. New Scientist is a place where people report that they've done something. While it does, occasionally, give some useful information, it's not the place where you expect to see the full report of what and how (far too boring for the subscription base). Hopefully, the real report will be in some obscure journal to be seen only by those who can actually use it. Certainly the authors can (and maybe even will) be called for discussion with those who are professionally interested, now that they've announced the existence of their result.

John

Subject: RE: Origins of music: new theory
From: Steve Parkes
Date: 08 Aug 03 - 03:15 AM

The thing about "real" science (as distinct from, say, Social Science) is the concept of disprovability: it's not possible to prove a theory is correct, but a theory should be capable of disproof if it is incorrect. What this means in practice is the theory is published along with supporting evidence, experimemntal data, references, etc. so that anyone else can recreate exactly the same experiments and check the results, and devise new experiments to test the theory further; if anyone can disprove the theory in this way, then we need a better theory.

I'm not sure if what I called a "new theory" is simpy a new hypothesis -- sorry if I'm misleading eveyone! But it's provoked discussion, which is what I intended. Can anyone think of ways to test the hypothesis/theory?

Steve

Subject: RE: Origins of music: new theory
From: GUEST,Q
Date: 08 Aug 03 - 04:42 PM

Ah, the scientific method. It always comes back to multiple working prejudices.

Subject: RE: Origins of music: new theory
From: Frankham
Date: 09 Aug 03 - 12:07 PM

The idea of language being the basis of music in not new. An analysis of speech reveals that (I read it somewhere) the minor third is the most comfortable interval that is uttered, originally as a young baby calls "mama". Kids use it in "nyah nyah nyah nyah nyah nyah". The minor third is the second and third string of the guitar in combination. It's at the root of harmonization in the Western Musical tradition. The start of music in the Western Civilization was basically vocal. When instrument technology increased, it becomes more involved harmonically and contrapuntally.

Inflections of speech often mirror musical phrases. It can be said that music is a language and that the use of good language is music.

The voice is the most personal instrument and a musician should be able to sing the notes that he/she plays although maybe not in the same octave as the instrument he/she plays.

The pyramiding of thirds constitutes what we have grown to know as consonance in music. Inverted, they are sixths, also considered consonant.

Frank Hamilton

Subject: RE: Origins of music: new theory
From: GUEST,Guest
Date: 09 Aug 03 - 08:41 PM

The minor third is a factor of 1.2 in a just intonation scale, and a major third is a factor of 1.25. From the graph in Roderer's book noted above, the most consonant interval is about 1.27 for which there isn't a close scale note, but 1.25 is pretty close. That minor third is about the minimum consonant interval. 4 of them successively is the diminished 7th.

Subject: RE: Origins of music: new theory
From: McGrath of Harlow
Date: 10 Aug 03 - 07:29 AM

...human cultures have sliced the pitch dimension into twelve distinct tones.

The researchers don't appear to have taken any note of the existance of those musical cultures that go in for half-tones and quarter tones. But then they also seem to have ignored the fact that most people don't speak English. Any research which exclusively concentrates on those would need to be extremely cautious indrawing any general conclusions.

I tend to assume that singing of one sort or another came before speech anyway. The same way it does with children.

Subject: RE: Origins of music: new theory
From: Steve Parkes
Date: 11 Aug 03 - 03:18 AM

Kevin: Speech in other languages - Mandarin, Farsi, and Tamil - also displayed the same pattern.

Steve

Subject: RE: Origins of music: new theory
From: Pied Piper
Date: 11 Aug 03 - 07:23 AM

Spot on McGrath.
The assumption that there are twelve ET semitones, that are sort of out there and human cultures are limited to these for musical purposes is nonsense.
As for consonance, the most consonant intervals are the ones with the lowest numbers in their frequency ratios, the most consonant being 2/1 the octave.
I can certainly tune a 9/8 Major second interval, and these are used as harmonies in Eastern European vocal music.
Also the 7/4 Harmonic seventh interval is used in Barber Shop harmony.
8/7(7/4 up), 15/8 and 16/15, are used against tonic drones in many cultures.
It's hardly surprising that the same physics applies to musical instruments and the human sound making apparatus.
I think hearing at a very primitive level (in the sea) probably evolved first but sound generating must have come along almost immediately, and then the two systems co-evolved.
The ear assumes sounds are generated by oscillating systems that produce integer multiple frequency harmonic series, and is evolved to detect these.
The more interesting question is how these intervals are used to create a psychological dynamic and convey emotional meaning.

All the best PP