Universals and birdsongs...

[Musicologo]

http://www.pnas.org/content/early/2014/10/29/1406023111

So, It seems some birds also like overtone-series based scales, not only some human cultures... humm...

Anyone is up for more debate on the "universal" constraints versus the "cultural" constraints?

Will we reach the point where we can entangle a series of parameters and realize that after all "oh, we use these kind of scales and intervals and ranges because of biological constraint X..." like we already know that "oh we use tempi around 40-200 and we feel moderate is around 60-100 because we have a heart that beats around those values as well"...

[jancivil]

We've been here before and duked it out, albeit not in relation to birdsong.

Many human musical scales, including the diatonic major scale prevalent in Western music, are built partially or entirely from intervals (ratios between adjacent frequencies) corresponding to small-integer proportions drawn from the harmonic series.


This is not an auspicious beginning in my opinion.
‧ What scales are not built from intervals?

‧ Prevalent in western music is that 12 equal divisions of the octave. some of these do correspond more-or-less with the harmonic series, but some do not so much *correspond*. Let's say you're a jazzer and you go for a b13th 'dominant 7th' color. What is there in the harmonic series to 'correspond' with it? 13:8 is ~840.53 cents.
I'm pretty confident that that choice has to do with other things, music historically, harmonic context within tempered practice.
Although I would note that '#11' is closer, and as such available through P5th x P5th at a point, and if you drive a fundamental hard enough you'll hear it clearly. So it doesn't have to be too simple.

So the tones 'western musicians' (and for that matter Indian classical, or Chinese traditional musicians) prefer don't so much take 'harmonic series' as a direct basis for 'scales'. {People temper things according to their instrument! But in this thesis, birds don't, they're thinking and obeying maths but let's lose how their instrument is made.}
The scales in fact belong to convention and to some extent culture; one could state their agreement with 'harmonic series' basically (cf., the first few 'partials') but actual music exceeds 'harmonic series' almost always. Now it is true that instruments are built observing acoustical principles; eg. an open horn IS giving you harmonic series.
But what does the singer prefer to produce? I'm in jeopardy of an argument from teleology a bit, but my POV is that creatures that make music have physicality suited for that, and the planet earth is suited to sound, and hearing music. So, does the voice, or larynx of a bird, not agree with acoustical principles? I could be jumping the gun but the thesis seems confused now.* Hopefully this means they aren't after a dichotomy, 'it's both biological and cultural', because the germ of both arguments is present.

So.

Scientists have long debated the extent to which principles of scale generation in human music are biologically or culturally determined.

Scientists debate cultural determinants. I don't think the people that wrote the abstract fully know what they're talking about.

But I would like to see them "show that this frequency selection results not from physical constraints governing peripheral production mechanisms but from active selection at a central level." I'm very skeptical of this (here *is* a dichotomy), but I'm going to keep an open mind.

<These data provide the most rigorous empirical evidence to date of a bird song that makes use of the same mathematical principles that underlie Western and many non-Western musical scales, demonstrating surprising convergence between human and animal “song cultures.”>
I'm not confident of their premise, really, as I indicate. They are at the same time trying to assert 'harmonic series' maths, and appear to have a contradictory premise, having decided to vacate how that works in the mechanism of producing sound. I have kind of an allergic reaction to what I'm reading so far.

*: Is the question being begged, the premise of bird choices resembling human choices is desirable and a conclusion is about to be manufactured?

[jancivil]

Will we reach the point [] because of constraint X..."

I would prefer to avoid unnecessary dichotomies. Yes, we aren't going to perceive things past a certain rate of frequency. But the birds perception is probably at a different rate. So, I don't know, I don't think it's anything to act like I know. I think that birds are making music and I think some things are the same but I don't need to shoehorn things into a preconception of laws. It looks like the central concern is this dichotomy of 'the instrument determines what notes' vs 'their intention determines what notes', and I think that is not very compelling as a way to investigate music and what I read seems to show that as problematic. It's because maths, but it's because culture. It's because both, for humans. If the idea is to show, through this investigation, that's how music works I tend to agree.

[rockstar_not]

On the tempo side of things, there is a sort of range of beats that is associated with syllable-rate in human speech. Some reading to get you started: http://content.time.com/time/health/art ... 77,00.html

I am a mid-western American English native speaker, and speak German with fluency, and some Swedish with a little less fluency.

I have noticed that those that speak English as a second language with more apparent speed than native English speakers are those hailing from Netherlands. I have a hard time understanding them sometimes because it seems that they speak English so fast - at least each word seems fast.

If you take a look at that article above, you will find that the syllable rate in syllables per second, corresponds well with 1/4 notes at roughly 75-120 bpm.

This also corresponds somewhat well with the range of sensitivity for the psychoacoustic metric of Fluctuation Strength.

I think I actually saw some kind of analysis like this in Levitin's "This is Your Brain on Music". My copy is boxed up somewhere.

[MadBrain]

We've been here before and duked it out, albeit not in relation to birdsong.

Many human musical scales, including the diatonic major scale prevalent in Western music, are built partially or entirely from intervals (ratios between adjacent frequencies) corresponding to small-integer proportions drawn from the harmonic series.


This is not an auspicious beginning in my opinion.
‧ What scales are not built from intervals?

‧ Prevalent in western music is that 12 equal divisions of the octave. some of these do correspond more-or-less with the harmonic series, but some do not so much *correspond*. Let's say you're a jazzer and you go for a b13th 'dominant 7th' color. What is there in the harmonic series to 'correspond' with it? 13:8 is ~840.53 cents.
I'm pretty confident that that choice has to do with other things, music historically, harmonic context within tempered practice.
Although I would note that '#11' is closer, and as such available through P5th x P5th at a point, and if you drive a fundamental hard enough you'll hear it clearly. So it doesn't have to be too simple.

So the tones 'western musicians' (and for that matter Indian classical, or Chinese traditional musicians) prefer don't so much take 'harmonic series' as a direct basis for 'scales'. {People temper things according to their instrument! But in this thesis, birds don't, they're thinking and obeying maths but let's lose how their instrument is made.}
The scales in fact belong to convention and to some extent culture; one could state their agreement with 'harmonic series' basically (cf., the first few 'partials') but actual music exceeds 'harmonic series' almost always. Now it is true that instruments are built observing acoustical principles; eg. an open horn IS giving you harmonic series.
But what does the singer prefer to produce? I'm in jeopardy of an argument from teleology a bit, but my POV is that creatures that make music have physicality suited for that, and the planet earth is suited to sound, and hearing music. So, does the voice, or larynx of a bird, not agree with acoustical principles? I could be jumping the gun but the thesis seems confused now.* Hopefully this means they aren't after a dichotomy, 'it's both biological and cultural', because the germ of both arguments is present.

So.

Scientists have long debated the extent to which principles of scale generation in human music are biologically or culturally determined.

Scientists debate cultural determinants. I don't think the people that wrote the abstract fully know what they're talking about.

But I would like to see them "show that this frequency selection results not from physical constraints governing peripheral production mechanisms but from active selection at a central level." I'm very skeptical of this (here *is* a dichotomy), but I'm going to keep an open mind.

<These data provide the most rigorous empirical evidence to date of a bird song that makes use of the same mathematical principles that underlie Western and many non-Western musical scales, demonstrating surprising convergence between human and animal “song cultures.”>
I'm not confident of their premise, really, as I indicate. They are at the same time trying to assert 'harmonic series' maths, and appear to have a contradictory premise, having decided to vacate how that works in the mechanism of producing sound. I have kind of an allergic reaction to what I'm reading so far.

*: Is the question being begged, the premise of bird choices resembling human choices is desirable and a conclusion is about to be manufactured?

I'd say it's the harmonic series is a great start. But for western music, the Pythagorean scale built from the 3/2 ratio is even more relevant:
- Our tuning system is essentially a slightly modified Pythagorean scale
- Except for triads, a lot of our core musical concepts revolve around stacked fifths (diatonic and pentatonic scales in particular)
- Our major 3rd behaves like a hybrid of a 81/64 third and a 5/4 third
- You can say that our octave is divided in 12 parts... but IRL it's really more like 12 consecutive fifths, slightly adjusted to loop endlessly.
- The jazz "b13" you mention is is presumably part of an "alt7" chord. It's often referred to as a "#5", and is part of a series of fifths in the fully extended "alt7" chord: #11 - b9 - #5 - #9 - 7 (they're not written as 5ths but the interval between each one is 7 semitones).

The "cultural" part is that not every musical tradition builds so heavily on stacked 5ths: Indian music is really more about the simple integer ratios, so its major 3rd is really more 5/4 than 81/64, plus the scales often have "gaps" which means that not all scale notes can be reached by 5ths (ex: 1 b2 3 4 5 b6 7, the 5ths are 4-1-5, 3-7 and b2-b6, so to go from 1 to 3, you have to make a non-5th jump). Note that it's still fairly close to the overtone series, in a different way than western music.

Similarly, Arabic music is not really all about stacked 5ths either (except for the more diatonic Maqams which are basically the same as in western music). I'm not an expert but I think it's really more about the small intervals and tetrachords, so the relationship to the overtone series is definitely more complex, especially when quarter tones come into the picture. The neutral 3rd is often considered as if you started with a 81/64 major third, which is then lowered to get you the 5/4 major third, except it's more lowered than it should be (afaik, that's actually how it's explained in Turkish music). Other explanations do use simple integer ratios (such as 11/9).

Thai xylophone music kindof goes the other way - it's very much based around the pentatonic scale so it's all based around the 5th, but the intonation is way different due to splitting the octave into 7 equal tones. This difference seems to be related to the difference in tone caused by using xylophones and bells - other traditions that heavily use chromatic percussion often have strange intonation as well (some African traditions, and especially Gamelan). I admit that this is quite harder to explain in terms of whole number ratios... my theory is that it's still the same base ratios (which is why it uses the pentatonic scale), but they are stretched to fit with the inharmonic characteristics of the instrumental tones.

The western intervals are not universal. BUT the underlying principles that have led to them are very much universal:

- The human propensity to quantize sounds into categories. For instance, there's an infinity of vowels between "EE" (/i/) and "A" (/a/), but every language splits this into a few separate, well defined vowels (for instance, English splits this range into 5, giving "reed", "rid", "rate", "red", "rad"). This explains why we have scales, instead of free pitch variation.
- The human ear's bias towards harmonic sounds. This might have evolved to help speech recognition (speech is harmonic except for voiceless consonants), or it might be a common trait in a lot of animals (this seems likely to me but I have no idea really). This explains why the scales almost always clusters around harmonic ratios.

The fact that we've found birdsongs that cluster around overtones probably means that we've found another species where both these principles are at play.

[jancivil]

The "cultural" part is that not every musical tradition builds so heavily on stacked 5ths: Indian music is really more about the simple integer ratios

WHAT? ICM, ragas are totally based in the perfect fifth. It is a structural feature in determining the shape of lines, in that your plateaus, 'vadi and samvadi', dominant and subdominant (not so much V and IV, it depends on the raga) definitely have a relationship with 'Sa', or 'tonic' at the Perfect Fifth.

You're actually going to indicate 'integer ratios' while saying 'not stacked fifths'? That's nonsensical.

We've been here before, more than once. I should reiterate my received knowledge about this particular, I reckon.

Confer Danielou 22-tone gamut. 12, JI five-limit; with an additional 10 at a syntonic comma [80:81] distance from their neighbor.

The reasoning is to obtain 3:2 where it is compromised in that JI 12:
{16:15; 9:8; 6:5; 5:4; 4:3; 45:32; 3:2; 8:5; 5:3; 9:5; 15:8; 2:1}.
16:15 x 3:2 = 8:5 ('Db -> Ab')
8:5 x 3:2 = 6:5 (back within 2:1) ('Eb')
6:5 x 3:2 = 9:5 " " ('Bb')
9:5 x 3:2 = 27:20 " " ('F')
IE: NOT 4:3. So 80:81 x 27:20 = 4:3.
4:3 x 3:2 = 2:1 ('C').
1 x 3:2 ('G')
3:2 x 3:2 = 9:8 ('D')
9:8 x 3:2 = 27:16 ('A')
NOT 5:3.
27:16 x 3:2 = 81:64 ('E')
NOT 5:4. Use the syntonic comma again for the other Ga, 5:4.
(IE: 5:3 x 3:2 = 5:4)

So we have imperfect fifths to correct.

5:4 x 3:2 = 15:8 ('E -> B')
15:8 x 3:2 = 45:32 ('F#')
45:32 x 3:2 = 135:128 ('C#')
{Compare 64:45}
64:45 x 3:2 = 16:15

This is just one way to look at the thing and is really 'theory' but the thinking is solidly within traditional thinking.

[jancivil]

As per '"Indian Music" does 5:4, not 81:64': Where there is a desire to have a best concordance with eg., the 27:16 'Dha' (A vis a vis C), it does prefer it; there are different expressions of tones a western musician may like to call microtonal. ICM has musicians that are seriously IN TUNE and they care about this kind of thing, whatever school or way of describing it they might stick with.

But for western music, the Pythagorean scale built from the 3/2 ratio is even more relevant:
- Our tuning system is essentially a slightly modified Pythagorean scale
- Except for triads, a lot of our core musical concepts revolve around stacked fifths (diatonic and pentatonic scales in particular)
- Our major 3rd behaves like a hybrid of a 81/64 third and a 5/4 third
- You can say that our octave is divided in 12 parts... but IRL it's really more like 12 consecutive fifths, slightly adjusted to loop endlessly.
- The jazz "b13" you mention is is presumably part of an "alt7" chord. It's often referred to as a "#5", and is part of a series of fifths in the fully extended "alt7" chord: #11 - b9 - #5 - #9 - 7 (they're not written as 5ths but the interval between each one is 7 semitones).

Pythagorean tuning is based in study of a string vibrating and extrapolating 3:2 and discovering that it does not agree with 2:1.

I said 'b13th' in order to show the difference between a CONVENTION and the PHYSICS which this paper calls mathematics. There are other things out of the Harmonic Series which are not 'diatonic', nor do they really help you make functional harmony 'V7 b13' kind of things. You will arrive at something close enough for jazz by multiplying 3:2 8 steps, but I haven't noticed that one 'resounding' like say #11.

But, you're right that Arabic music became interested in ratios per se, even as it totally is out of all of that 'Pythagorean' bidness. It IS different than certain ICM in this particular interest. Cf., al Farabi. I think they are after expressing 'Arabic character' or like that with the 'neutral' thirds for instance. BUT, there are small 'semitones' in Hindustani music; heard more on Sarod (fretless) than say Sitar music. There seems to be a lot more 'theory' in Arabic music than Indian. Indian culture prefers to hand things down directly through a person and an oral tradition than to notate everything.

I very much agree with 'the human ear's preference for harmonicity' and I think that the production mechanisms for music agrees with that.

[MadBrain]

The "cultural" part is that not every musical tradition builds so heavily on stacked 5ths: Indian music is really more about the simple integer ratios

WHAT? ICM, ragas are totally based in the perfect fifth. It is a structural feature in determining the shape of lines, in that your plateaus, 'vadi and samvadi', dominant and subdominant (not so much V and IV, it depends on the raga) definitely have a relationship with 'Sa', or 'tonic' at the Perfect Fifth.

You're actually going to indicate 'integer ratios' while saying 'not stacked fifths'? That's nonsensical.

We've been here before, more than once. I should reiterate my received knowledge about this particular, I reckon.

Confer Danielou 22-tone gamut. 12, JI five-limit; with an additional 10 at a syntonic comma [80:81] distance from their neighbor.

The reasoning is to obtain 3:2 where it is compromised in that JI 12:
{16:15; 9:8; 6:5; 5:4; 4:3; 45:32; 3:2; 8:5; 5:3; 9:5; 15:8; 2:1}.
16:15 x 3:2 = 8:5 ('Db -> Ab')
8:5 x 3:2 = 6:5 (back within 2:1) ('Eb')
6:5 x 3:2 = 9:5 " " ('Bb')
9:5 x 3:2 = 27:20 " " ('F')
IE: NOT 4:3. So 80:81 x 27:20 = 4:3.
4:3 x 3:2 = 2:1 ('C').
1 x 3:2 ('G')
3:2 x 3:2 = 9:8 ('D')
9:8 x 3:2 = 27:16 ('A')
NOT 5:3.
27:16 x 3:2 = 81:64 ('E')
NOT 5:4. Use the syntonic comma again for the other Ga, 5:4.
(IE: 5:3 x 3:2 = 5:4)

So we have imperfect fifths to correct.

5:4 x 3:2 = 15:8 ('E -> B')
15:8 x 3:2 = 45:32 ('F#')
45:32 x 3:2 = 135:128 ('C#')
{Compare 64:45}
64:45 x 3:2 = 16:15

This is just one way to look at the thing and is really 'theory' but the thinking is solidly within traditional thinking.

Ah, sorry then. I have agree here. I guess I should have said that it was less close to stacked 5ths than western music (essentially because it uses less harmony), not that it was based strictly on simple ratios.

[jancivil]

That's an easy misconception to arrive at, say through the word 'stacked'; but it isn't less so just because their music isn't concered with simultaneous vertical manifestation of that fifth; the melodic conception is after the very same concordance, 'perfect fifth'. As though the mind recalls the 'harmony' linearly, not requiring the vertical statement of it.