Where Theory Falls Short

First, a question: how many of you have played or otherwise demonstrated alternative intonational systems to non-microtonalists and/or non-musicians? Those of you that have, have you noticed any trends in the responses?

What's been nagging me for a while is that for all the theories that get discussed here, there seems to be very little consensus both on list and off over "what sounds good"--even when "stylistic" considerations are accounted for. On this list alone, there is a huge diversity of tuning preferences and one person's "out of tune" is another person's "perfect harmony".

In the "real world", one thing I notice with overwhelming frequency: if I sit someone down (usually a non-musician) and play them some chord that differs blatantly from a familiar 12-TET chord--even if it's a 5-limit JI major triad--my listener thinks it sounds "out of tune". Yet if I play them a complete piece of microtonal music, they *don't* think it sounds out-of-tune, even if it really is (i.e. something in Mavila). They think it sounds "weird" or maybe "ethnic", but not "out-of-tune".

Another thing I notice, specifically with guitarists, is that they don't give a damn how allegedly "in-tune" a system is, they really only care how many frets there are, because to them anything non-12 is all lumped together and all they care about is how scary the guitar looks. Of the five or six guitarists to pick up and play my microtonal guitars, even after I've demonstrated the properties of each, not a single person has preferred 19 to 16. I have good reason to suspect that 15 would go over even better, and 13 or 14 maybe better still.

I know Mike has mentioned that his music friends usually seem less than turned on by microtonal music, even really in-tune stuff like 22-EDO. This is the real crux for me: if psychoacoustics is really important, shouldn't there be almost universal preference for more in-tune systems and almost universal aversion to not-so-in-tune systems? In reality it seems like psychoacoustic considerations rarely do better than random chance at predicting alternate tuning system preferences, though they do corroborate the overwhelming preference for 12-TET and/or meantone.

Over the years, theorists have claimed varieties of different tuning systems as the "natural" evolution from 12-TET; we've got the Inteas/Pentadecaphonic people claiming it's 15, Armodue claiming it's 16, the whole 19-TET cult of Yasser et al., Huygens-Fokker et al. with 31-TET, and maybe we can even count Paul Erlich with 22-EDO (there may be others I'm missing). Oh, let's not forget the quarter-tone people (Carrillo, Haba, etc.), or BP, or John O'Sullivan. Looked at together, these tunings all have very little in common--so why the diversity? 12-TET took the world by storm with very little resistance, suggesting it dovetailed snugly with the preferences of a vast majority, so why is consensus on non-12-TET tunings so rare? Has anyone ever wondered about this?

-Igs

On Thu, May 12, 2011 at 6:03 PM, cityoftheasleep
<igliashon@...> wrote:
>
> First, a question: how many of you have played or otherwise demonstrated alternative intonational systems to non-microtonalists and/or non-musicians? Those of you that have, have you noticed any trends in the responses?

Musicians hate it more often than non-musicians. People with perfect
pitch hate it more than people without perfect pitch. People like
blackwood more than 14-equal. People like familiar things that fit
together differently rather than unfamiliar things that fit together
awkwardly.

> What's been nagging me for a while is that for all the theories that get discussed here, there seems to be very little consensus both on list and off over "what sounds good"--even when "stylistic" considerations are accounted for. On this list alone, there is a huge diversity of tuning preferences and one person's "out of tune" is another person's "perfect harmony".

I don't ever think music theory will tell us with 100% certainty what
will sound good for an arbitrary listener. In fact, I think it will
predict its own limitations in this regard. I do wish that we spent
more time trying to understand the specific ways in which we're
accultured so we can work on extending those ways without blasting the
listener with too much novelty. I think we're just about getting to
that point now.

> In the "real world", one thing I notice with overwhelming frequency: if I sit someone down (usually a non-musician) and play them some chord that differs blatantly from a familiar 12-TET chord--even if it's a 5-limit JI major triad--my listener thinks it sounds "out of tune". Yet if I play them a complete piece of microtonal music, they *don't* think it sounds out-of-tune, even if it really is (i.e. something in Mavila). They think it sounds "weird" or maybe "ethnic", but not "out-of-tune".

When my friends say that, they generally are saying politely that the
whole thing sounds out of tune. Maybe your friends are more
open-minded.

> I know Mike has mentioned that his music friends usually seem less than turned on by microtonal music, even really in-tune stuff like 22-EDO. This is the real crux for me: if psychoacoustics is really important, shouldn't there be almost universal preference for more in-tune systems and almost universal aversion to not-so-in-tune systems? In reality it seems like psychoacoustic considerations rarely do better than random chance at predicting alternate tuning system preferences, though they do corroborate the overwhelming preference for 12-TET and/or meantone.

You know I'm going to respond with the Frere Jacques cite, so I'll
leave this for everyone else to discuss.

> Over the years, theorists have claimed varieties of different tuning systems as the "natural" evolution from 12-TET; we've got the Inteas/Pentadecaphonic people claiming it's 15, Armodue claiming it's 16, the whole 19-TET cult of Yasser et al., Huygens-Fokker et al. with 31-TET, and maybe we can even count Paul Erlich with 22-EDO (there may be others I'm missing). Oh, let's not forget the quarter-tone people (Carrillo, Haba, etc.), or BP, or John O'Sullivan. Looked at together, these tunings all have very little in common--so why the diversity? 12-TET took the world by storm with very little resistance, suggesting it dovetailed snugly with the preferences of a vast majority, so why is consensus on non-12-TET tunings so rare? Has anyone ever wondered about this?

At the risk of bombarding you on two separate forums with lengthy
explanations simultaneously, I'll give it a go and try to be brief: I
don't think there is a "natural" evolution from 12-TET anymore. We've
been utilizing both 128/125, 648/625, 50/49, and 81/80 puns for over a
century now. I forget what the diaschisma is right now, but we've
probably been using that too. The only tuning that supports all of
those vectors is 12-equal, so for us to naturally evolve past 12-equal
would basically require giving one of those up. Since 81/80 is by far
the most important to western tonal structure, the rest can go, so I
guess if you're willing to compromise then 19-TET or maybe 31-TET are
the evolutions of 12-equal. But once you're willing to compromise and
accept a system that means you CAN'T do as much as you can already do,
for the sake of new things - hell, at that point, why not go all the
way? At that point you might as well consider things like 15 and 16
that are completely novel. It depends on your tolerance.

Actually, I take it back - there is one more ET that might do the
trick, and it really is the "natural" evolution of 12-EDO: 24-EDO.
24-EDO renders all existing music playable, adds new harmonic
resources, and doesn't have too many notes. 24-EDO is, however, of
course, unusable, because the 7/4 is 18 cents flat, and it subdivides
12 into 2, which is bad. Which is a shame, because it's a killer
2.3.5.11.13.17.19 tuning, supports island/semaphore, mohajira,
demolished, sensi, has 8-equal in it, etc.

-Mike

--- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
> I don't ever think music theory will tell us with 100% certainty what
> will sound good for an arbitrary listener.

No, but right now we don't even have 50% certainty, despite having a multiplicity of sophisticated and experimentally-verified theories of concordance.

> When my friends say that, they generally are saying politely that the
> whole thing sounds out of tune. Maybe your friends are more
> open-minded.

One of my friends--the only one who's spent more than a few minutes playing one of my micro guitars--has traveled extensively in Madagascar, and has recorded lots of music of the native folks there. They use 12-TET instruments often, especially guitars, but don't usually bother to tune them.

> You know I'm going to respond with the Frere Jacques cite, so I'll
> leave this for everyone else to discuss.

What does that cite have to do with tuning preferences?

> At the risk of bombarding you on two separate forums with lengthy
> explanations simultaneously, I'll give it a go and try to be brief: I
> don't think there is a "natural" evolution from 12-TET anymore.

But other people do, or at least did, and they all had various "good reasons". I think it's a valid and interesting question as to why none of them caught on, while 12-TET easily took the world by storm. Why does the psychoacoustic superiority of an alternate tuning fail to guarantee it popularity?

> We've
> been utilizing both 128/125, 648/625, 50/49, and 81/80 puns for over a
> century now. I forget what the diaschisma is right now, but we've
> probably been using that too. The only tuning that supports all of
> those vectors is 12-equal, so for us to naturally evolve past 12-equal
> would basically require giving one of those up. Since 81/80 is by far
> the most important to western tonal structure, the rest can go, so I
> guess if you're willing to compromise then 19-TET or maybe 31-TET are
> the evolutions of 12-equal. But once you're willing to compromise and
> accept a system that means you CAN'T do as much as you can already do,
> for the sake of new things - hell, at that point, why not go all the
> way? At that point you might as well consider things like 15 and 16
> that are completely novel. It depends on your tolerance.

But why? There are lots of "new" things in other tuning systems that are, by all psychoacoustic and group-theoretic considerations, better than what can be done in 12-TET. The fact that they fail to appeal the masses suggests something kind of terrible: that familiarity has more appeal than "superior quality".

> Actually, I take it back - there is one more ET that might do the
> trick, and it really is the "natural" evolution of 12-EDO: 24-EDO.
> 24-EDO renders all existing music playable, adds new harmonic
> resources, and doesn't have too many notes. 24-EDO is, however, of
> course, unusable, because the 7/4 is 18 cents flat, and it subdivides
> 12 into 2, which is bad. Which is a shame, because it's a killer
> 2.3.5.11.13.17.19 tuning, supports island/semaphore, mohajira,
> demolished, sensi, has 8-equal in it, etc.

LOL, Ron Sword sells more 24-EDO guitars than any other tuning, apparently. The only major label band to embrace a microtonal tuning is that horrible Nu Metal band "M.A.N.", and they used 24-EDO. Lots of people seeking to approximate middle eastern music are getting quarter-tone frets added to their guitars. Maybe 24-EDO will be the new 12, for all the reasons you say. Psychoacoustics be damned. More evidence that familiarity is attractive, since most of the new intervals in 24-EDO are at or near maxima of HE.

-Igs

On Thu, May 12, 2011 at 6:54 PM, cityoftheasleep
<igliashon@...> wrote:
>
> No, but right now we don't even have 50% certainty, despite having a multiplicity of sophisticated and experimentally-verified theories of concordance.

We don't have a way to model how learned factors influence things. If
we did, then we'd be able to. Here's an approach that'll probably
cover most people - take the HE curve, split it into 12-equal parts,
and smear everything together within each part, so you end up with an
upside down cityscape of 12 skyscrapers. That should handle most of my
friends.

> > You know I'm going to respond with the Frere Jacques cite, so I'll
> > leave this for everyone else to discuss.
>
> What does that cite have to do with tuning preferences?

It has to do with musical preferences because it suggests that people
don't even get access to these more complex harmonic sounds, because
they just end up hearing out of tune diatonic sounds, which are the
VFs that they're conditioned to look for a la Frere Jacques. Once
again, Igs, I'm not sure why you continue to put out there that
learning dominates over psychoacoustics, and then when I cite studies
that agree with you, you disagree. It is profoundly unpleasant to
converse with someone who seems to just like to play devil's advocate
with whatever it is that you say.

> > At the risk of bombarding you on two separate forums with lengthy
> > explanations simultaneously, I'll give it a go and try to be brief: I
> > don't think there is a "natural" evolution from 12-TET anymore.
>
> But other people do, or at least did, and they all had various "good reasons". I think it's a valid and interesting question as to why none of them caught on, while 12-TET easily took the world by storm. Why does the psychoacoustic superiority of an alternate tuning fail to guarantee it popularity?

Because 12-TET is good enough, easy to play, added new musical
possibilities while being compatible with 100% of existing music, etc.

> But why? There are lots of "new" things in other tuning systems that are, by all psychoacoustic and group-theoretic considerations, better than what can be done in 12-TET. The fact that they fail to appeal the masses suggests something kind of terrible: that familiarity has more appeal than "superior quality".

You know, I was reading some studies that show how familiarity can
actually completely change the way harmonic series detection in the
brain works. Perhaps they're relevant.

> > Which is a shame, because it's a killer
> > 2.3.5.11.13.17.19 tuning, supports island/semaphore, mohajira,
> > demolished, sensi, has 8-equal in it, etc.
>
> LOL, Ron Sword sells more 24-EDO guitars than any other tuning, apparently. The only major label band to embrace a microtonal tuning is that horrible Nu Metal band "M.A.N.", and they used 24-EDO. Lots of people seeking to approximate middle eastern music are getting quarter-tone frets added to their guitars. Maybe 24-EDO will be the new 12, for all the reasons you say. Psychoacoustics be damned. More evidence that familiarity is attractive, since most of the new intervals in 24-EDO are at or near maxima of HE.

24-EDO is actually great psychoacoustically; like I said, it hits all
primes within about 10-15 cents in the 2.3.5.11.13.17.19 subgroup. It
just has a bad rep in the microtonal community because it's become a
poster child for half-assed microtonal exploration.

-Mike

--- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:

> It has to do with musical preferences because it suggests that people
> don't even get access to these more complex harmonic sounds, because
> they just end up hearing out of tune diatonic sounds, which are the
> VFs that they're conditioned to look for a la Frere Jacques. Once
> again, Igs, I'm not sure why you continue to put out there that
> learning dominates over psychoacoustics, and then when I cite studies
> that agree with you, you disagree. It is profoundly unpleasant to
> converse with someone who seems to just like to play devil's advocate
> with whatever it is that you say.

I'm not sure I see where I'm disagreeing. I haven't changed my stated opinion at all. It seems to me like you are posting things that corroborate what I'm saying and then suggesting they refute what I'm saying. But if we're both actually in agreement then we should probably just shut up on the matter.

> Because 12-TET is good enough, easy to play, added new musical
> possibilities while being compatible with 100% of existing music, etc.

I understand why 12-TET won out over meantone and various well-temperaments (one word: halberstadt), since both evolved out of the basic pythagorean/diatonic framework. But I still wonder why pythagorean/diatonic won out so easily over the panoply of alternative melodic frameworks that coexisted with it? The greeks had so many tetrachords, and any of them could have led to different temperaments and harmonic structures.

> You know, I was reading some studies that show how familiarity can
> actually completely change the way harmonic series detection in the
> brain works. Perhaps they're relevant.

To the extent that we misperceive a pure harmonic series when it's served up on a silver platter, because we're used to a detuned one?

> 24-EDO is actually great psychoacoustically; like I said, it hits all
> primes within about 10-15 cents in the 2.3.5.11.13.17.19 subgroup. It
> just has a bad rep in the microtonal community because it's become a
> poster child for half-assed microtonal exploration.

Primes 11, 13, 17, and 19 are all basically maxima of 2HE if they're in the 1st octave, and I don't think 3HE makes them look very good in triads, either.

I should try writing some 24-EDO stuff...so far I've only written one piece in it, using the "demolished" scale (it's the first track on my forthcoming "Transcendissonance" album, which Tony Dubshot is mastering right now), and I actually liked it--it "failed to suck", despite my belief that it would fight my playing tooth-and-nail. I wonder how your friends would like 24?

-Igs

On Thu, May 12, 2011 at 7:47 PM, cityoftheasleep
<igliashon@...> wrote:
>
> > It has to do with musical preferences because it suggests that people
> > don't even get access to these more complex harmonic sounds, because
> > they just end up hearing out of tune diatonic sounds, which are the
> > VFs that they're conditioned to look for a la Frere Jacques. Once
> > again, Igs, I'm not sure why you continue to put out there that
> > learning dominates over psychoacoustics, and then when I cite studies
> > that agree with you, you disagree. It is profoundly unpleasant to
> > converse with someone who seems to just like to play devil's advocate
> > with whatever it is that you say.
>
> I'm not sure I see where I'm disagreeing. I haven't changed my stated opinion at all. It seems to me like you are posting things that corroborate what I'm saying and then suggesting they refute what I'm saying. But if we're both actually in agreement then we should probably just shut up on the matter.

That sure beats

Igs: I think learning trumps psychoacoustics
Mike: Here's a study that agrees and gives insight into how they interact
Igs: I disagree. I think learning trumps psychoacoustics

repeat forever

> > Because 12-TET is good enough, easy to play, added new musical
> > possibilities while being compatible with 100% of existing music, etc.
>
> I understand why 12-TET won out over meantone and various well-temperaments (one word: halberstadt), since both evolved out of the basic pythagorean/diatonic framework. But I still wonder why pythagorean/diatonic won out so easily over the panoply of alternative melodic frameworks that coexisted with it? The greeks had so many tetrachords, and any of them could have led to different temperaments and harmonic structures.

Because Pythagoras was the first person to figure out what a perfect
fifth was. You see, they didn't have air conditioners blasting out 7/4
and refrigerators blasting out 13/4 back in the day. They didn't have
helicopters flying overhead diffracting all of the harmonics as they
passed by. I'm pretty sure they didn't have Scala. They could maybe
hear glimmers of 3 and 5 in ordinary sounds if they concentrated, but
they barely had exposure to 11 and 13 at all. They had to go through
years of intense meditation to hear crap like that and give it a name
like "ohmmm."

I guess when they started playing with strings and monochords they got
access to some higher harmonics, but as Paul said until the
utonal/otonal paradigm shift I doubt they even knew those sounds were
"in" the string to begin with.

> > You know, I was reading some studies that show how familiarity can
> > actually completely change the way harmonic series detection in the
> > brain works. Perhaps they're relevant.
>
> To the extent that we misperceive a pure harmonic series when it's served up on a silver platter, because we're used to a detuned one?

That wasn't what I was getting at. I was actually facetiously getting
at Frere Jacques again. But while you mention it, I did note that lots
of my musician friends actually prefered 12-tet dominant 9 chords to
4:5:6:7:9, because they said they thought the 7 sounded flat.

> > 24-EDO is actually great psychoacoustically; like I said, it hits all
> > primes within about 10-15 cents in the 2.3.5.11.13.17.19 subgroup. It
> > just has a bad rep in the microtonal community because it's become a
> > poster child for half-assed microtonal exploration.
>
> Primes 11, 13, 17, and 19 are all basically maxima of 2HE if they're in the 1st octave, and I don't think 3HE makes them look very good in triads, either.

Whatever. They work well enough in 4:5:6:9:11:13:17:19. For someone
who claims to not like HE all that much, you sure seem to put more
stock in it than I do.

> I should try writing some 24-EDO stuff...so far I've only written one piece in it, using the "demolished" scale (it's the first track on my forthcoming "Transcendissonance" album, which Tony Dubshot is mastering right now), and I actually liked it--it "failed to suck", despite my belief that it would fight my playing tooth-and-nail. I wonder how your friends would like 24?

I think they'd dig it, as much of it would be immediately
comprehensible to them, but with new stuff added as well.

-Mike

--- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
> Igs: I think learning trumps psychoacoustics
> Mike: Here's a study that agrees and gives insight into how they interact
> Igs: I disagree. I think learning trumps psychoacoustics

We're really only disagreeing about the relative importance of psychoacoustics in explaining anything. I keep asking you questions but I think you keep thinking I'm being rhetorical.

> I guess when they started playing with strings and monochords they got
> access to some higher harmonics, but as Paul said until the
> utonal/otonal paradigm shift I doubt they even knew those sounds were
> "in" the string to begin with.

But you don't need to know what harmonics are to know that two strings tuned in simple ratios sound good. Heck, they had some far-out scales, like all the stuff Archytas came up with...what about the Enharmonic genus? As Petr has pointed out, you can take a chord of simple ratios, subdivide them in different ways, and end up with different small intervals that could lead to different temperaments. Okay, I know this question is unanswerable, but why didn't anyone discover any temperaments besides meantone? And why did people go from hearing Pythagorean 3rds as dissonances to hearing them as consonances--I understand meantone got its popularity from the 5-limit 3rds, but 12-TET basically undid that and took us back to Pythagorean, except people thought it sounded just fine all of a sudden. And then we have:

> I did note that lots of my musician friends actually prefered 12-tet dominant 9 chords to
> 4:5:6:7:9, because they said they thought the 7 sounded flat.

Acculturation to 12-TET does seem to make a lot of people think JI sounds out-of-tune. And yet here we weirdos are, judging tunings according to their proximity to JI. Could it be that we have it ass-backwards?

> Whatever. They work well enough in 4:5:6:9:11:13:17:19. For someone
> who claims to not like HE all that much, you sure seem to put more
> stock in it than I do.

I'm not saying I put stock in it. I'm just reiterating what HE says about those harmonics. Frankly I think chords in 24-EDO sound peachy, especially in Barbados temperament.

> I think they'd dig it, as much of it would be immediately
> comprehensible to them, but with new stuff added as well.

Maybe we should work out ways to use 24 as a "gateway" tuning to ease people into the more esoteric stuff. Once they can get a handle on the new scales and subdivisions in 24, 19 might not seem so far-out.

-Igs

On Thu, May 12, 2011 at 10:30 PM, cityoftheasleep
<igliashon@...> wrote:
>
> --- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
> > Igs: I think learning trumps psychoacoustics
> > Mike: Here's a study that agrees and gives insight into how they interact
> > Igs: I disagree. I think learning trumps psychoacoustics
>
> We're really only disagreeing about the relative importance of psychoacoustics in explaining anything. I keep asking you questions but I think you keep thinking I'm being rhetorical.

OK, then for the record, I think

1) Let's say you're in a forest, and you have a bow and arrow.
Whenever you shoot the bow and arrow, if it hits a tree, a certain
feeling is produced.
2) You, unfortunately, can only aim the arrow in one of 12 possible
fixed directions.

That's how I think most of the Western world hears music. Ways to
solve this problem

3) Train yourself to aim it in new directions (Porcupine, let's say)
4) Rotate yourself slightly so that the normal directions end up being
slightly warped (Father[8], Machine[6])
5) Warp the fabric of space and time so that if you shoot an arrow up
and to the left, it somehow ends up coming back and shooting you from
behind (comma pumps are more pleasant than this analogy entails)
6) Get clever enough to write a piece of music that activates an
entirely different set of pitch cues than the usual Western ones
(imagine an orchestra sounding like an air conditioner or something
that isn't music)

Or probably the best of all

7) Find some kind of opening that allows you to expand your current
way of doing things outward into new musical territory (we've been
doing this since music invented and it's the holy grail of microtonal
music)

We'll get to #7 once we learn more about how all of this internalized
pitch cue stuff works, which is to say we'll learn the "rules."

> But you don't need to know what harmonics are to know that two strings tuned in simple ratios sound good. Heck, they had some far-out scales, like all the stuff Archytas came up with...what about the Enharmonic genus? As Petr has pointed out, you can take a chord of simple ratios, subdivide them in different ways, and end up with different small intervals that could lead to different temperaments. Okay, I know this question is unanswerable, but why didn't anyone discover any temperaments besides meantone? And why did people go from hearing Pythagorean 3rds as dissonances to hearing them as consonances--I understand meantone got its popularity from the 5-limit 3rds, but 12-TET basically undid that and took us back to Pythagorean, except people thought it sounded just fine all of a sudden. And then we have:

You asked about why Pythagorean tuning dominated. The reason is that
nobody knew that two strings tuned in simple ratios sounded good,
because Pythagoras was the first guy to figure out that a perfect
fifth was a 3/2 ratio, that a fourth was 4/3, and that the difference
between them was the vile and dissonant 9/8. Then a bunch of other
stuff happened, and now you're fortunate to be a part of the emerging
Great Renaissance of Music Theory, or at least I'd like to be able to
say that once this last bit of the puzzle is figured out.

Alright, I'll give a clue as to how I think it works:

http://www.youtube.com/watch?v=XSfnyr1MhXE

Ignore my mumbly ass screwing up the description of the comma pump.
But the idea is that every individual chord movement is something
really familiar that we're used to - it's Cm7 -> Ebmaj7 -> Ebm7 ->
Dbmaj/Eb -> Cbm7 -> Bbmaj/Db -> Cbm7, except since you're in porcupine
the Cb at the end is a C again. So every individual aspect of that
movement is something that's already intelligible to us - i.e. I went
with chord movements that everyone is familiar with, like Cm7 ->
Ebmaj7 -> Ebm7 - instead of for some reason limiting myself to
porcupine[8]. The thing is, althouhg each individual step is
intelligible, the whole thing returns you to the tonic in an
unexpected way because of the comma pump. Most of my friends described
that they could hear it as being microtonal, but they could hear "some
sort of logic in it," and that it was "pleasant." Before this, they
had said my stuff was like "well it's cool man, I guess I just need to
spend more time with it, you know I'm new to this stuff," aka they
heard it as being out of tune.

Non-musicians often didn't realize it was microtonal, and super
advanced musicians couldn't handle it because they were able to follow
the progression mentally all the way through and it screwed their map
up so strongly that it actually became an unpleasant experience for
them. But even the super advanced musicians seemed to get that there
was some kind of order that was going on.

> > I did note that lots of my musician friends actually prefered 12-tet dominant 9 chords to
> > 4:5:6:7:9, because they said they thought the 7 sounded flat.
>
> Acculturation to 12-TET does seem to make a lot of people think JI sounds out-of-tune. And yet here we weirdos are, judging tunings according to their proximity to JI. Could it be that we have it ass-backwards?

I think it means that we held up JI as this ideal, and as a result
many of us have developed an internal map to now fit JI just like we
have one that fits 12-equal. Gene has to be on top of this by far;
he's writing 11-limit comma pumps and seems to get them as
comprehensible, whereas I only recently felt like I could grasp the
logic of normal 5-limit JI. But I really like Gene's stuff, actually,
and I enjoy listening to it and trying to wrap my head around it.
However, there's still some kind of learning phase that needed to take
place before I could get away from 12-equal (and into JI).

There are at least three meanings of the word "out of tune" that I can think of

1) Roughness/beating - mavila fifths played with brass instruments
will sound "out of tune that way"
2) Violating the internal structure that we have of how music lays out
and how things connect to one another (READ: COMMA PUMPS) - a chord
progression in 5-limit JI will sound "out of tune" for someone
expecting a I-iii-vi-ii-V-I
3) Entropy being too high - if a fifth gets sharp enough, it'll start
sounding like 8/5 and not like 3/2 anymore. This is what we're saying
that learning influences.

> > Whatever. They work well enough in 4:5:6:9:11:13:17:19. For someone
> > who claims to not like HE all that much, you sure seem to put more
> > stock in it than I do.
>
> I'm not saying I put stock in it. I'm just reiterating what HE says about those harmonics. Frankly I think chords in 24-EDO sound peachy, especially in Barbados temperament.

8HE would say the chord I laid out above is really consonant. But who
cares about HE anyway?

> > I think they'd dig it, as much of it would be immediately
> > comprehensible to them, but with new stuff added as well.
>
> Maybe we should work out ways to use 24 as a "gateway" tuning to ease people into the more esoteric stuff. Once they can get a handle on the new scales and subdivisions in 24, 19 might not seem so far-out.

Two of my sax player friends are trying to learn it now. Apparently 23
of the 24 notes are playable on sax, but the 11/8 above C isn't (god
damn). Semaphore is definitely a good start, and so is 8-equal.
10:12:15 can be turned into 10:11:12:13:14:15 without much thought. I
think 24-tet needs some kind of do-over in terms of how the community
sees it.

-Mike

--- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
>
> 1) Let's say you're in a forest, and you have a bow and arrow.
> Whenever you shoot the bow and arrow, if it hits a tree, a certain
> feeling is produced.
> 2) You, unfortunately, can only aim the arrow in one of 12 possible
> fixed directions.

Man, you come up with some funky metaphors some times. This is up there with Graham's "Alternate Vegetables List".

> That's how I think most of the Western world hears music. Ways to
> solve this problem

I'm not trying to speculate on how to solve this "problem". I'm trying to speculate on why it is a problem and why it has so far failed to be solved.

> You asked about why Pythagorean tuning dominated. The reason is that
> nobody knew that two strings tuned in simple ratios sounded good,
> because Pythagoras was the first guy to figure out that a perfect
> fifth was a 3/2 ratio, that a fourth was 4/3, and that the difference
> between them was the vile and dissonant 9/8. Then a bunch of other
> stuff happened, and now you're fortunate to be a part of the emerging
> Great Renaissance of Music Theory, or at least I'd like to be able to
> say that once this last bit of the puzzle is figured out.

http://en.wikipedia.org/wiki/Archytas
http://www.ex-tempore.org/ARCHYTAS/ARCHYTAS.html
http://en.wikipedia.org/wiki/Ptolemy#Music
http://www.nonoctave.com/tuning/scales/Ptolemy's_Scale_Catalog.html

These guys knew JI as well as we do, and we're basically hashing out the same shit they did, except now we have fancy-pants scientific explanations involving psychoacoustics. But it amounts to the same basic stuff. Strict Pythagorean tuning certainly did not dominate right from the get-go. So my question stands: why did it beat out all of these other approaches to tuning? Why did nothing "competitive" come out of these tunings until a few millennia later?

> But the idea is that every individual chord movement is something
> really familiar that we're used to - it's Cm7 -> Ebmaj7 -> Ebm7 ->
> Dbmaj/Eb -> Cbm7 -> Bbmaj/Db -> Cbm7, except since you're in porcupine
> the Cb at the end is a C again. So every individual aspect of that
> movement is something that's already intelligible to us - i.e. I went
> with chord movements that everyone is familiar with, like Cm7 ->
> Ebmaj7 -> Ebm7 - instead of for some reason limiting myself to
> porcupine[8]. The thing is, althouhg each individual step is
> intelligible, the whole thing returns you to the tonic in an
> unexpected way because of the comma pump. Most of my friends described
> that they could hear it as being microtonal, but they could hear "some
> sort of logic in it," and that it was "pleasant." Before this, they
> had said my stuff was like "well it's cool man, I guess I just need to
> spend more time with it, you know I'm new to this stuff," aka they
> heard it as being out of tune.

That's why so much of what I write is based on 5-limit common-tone progressions. If you always use common-tone progressions, you're guaranteed to always use familiar movements, more or less.

> I think it means that we held up JI as this ideal, and as a result
> many of us have developed an internal map to now fit JI just like we
> have one that fits 12-equal. Gene has to be on top of this by far;
> he's writing 11-limit comma pumps and seems to get them as
> comprehensible, whereas I only recently felt like I could grasp the
> logic of normal 5-limit JI. But I really like Gene's stuff, actually,
> and I enjoy listening to it and trying to wrap my head around it.
> However, there's still some kind of learning phase that needed to take
> place before I could get away from 12-equal (and into JI).

Yeah, I've internalized a LOT over the years. Even 13-EDO sounds "normal" to me now. So it can be done, at least be some members of the population.

> There are at least three meanings of the word "out of tune" that I can think of

And I've gotten really good at ignoring all of them.

> 8HE would say the chord I laid out above is really consonant. But who
> cares about HE anyway?

Mmmm...I dunno, people who put stock in psychoacoustics in general?

> Two of my sax player friends are trying to learn it now. Apparently 23
> of the 24 notes are playable on sax, but the 11/8 above C isn't (god
> damn). Semaphore is definitely a good start, and so is 8-equal.
> 10:12:15 can be turned into 10:11:12:13:14:15 without much thought. I
> think 24-tet needs some kind of do-over in terms of how the community
> sees it.

I agree. I'll get right on it.

-Igs

On Thu, May 12, 2011 at 11:55 PM, cityoftheasleep
<igliashon@...> wrote:
>
> > That's how I think most of the Western world hears music. Ways to
> > solve this problem
>
> I'm not trying to speculate on how to solve this "problem". I'm trying to speculate on why it is a problem and why it has so far failed to be solved.

Oh. Well, the answer to why it's a problem, is that it's just a
general tendency of the brain to extract order from the information
stream that it's being fed. As for why it's failed to be solved thus
far, I'd say because it's 2011 and not 2012. My question is how to
solve it, which I think I gave some good insight into in the parts of
my reply that you cut out.

> http://en.wikipedia.org/wiki/Archytas
> http://www.ex-tempore.org/ARCHYTAS/ARCHYTAS.html
> http://en.wikipedia.org/wiki/Ptolemy#Music
> http://www.nonoctave.com/tuning/scales/Ptolemy's_Scale_Catalog.html
>
> These guys knew JI as well as we do, and we're basically hashing out the same shit they did, except now we have fancy-pants scientific explanations involving psychoacoustics. But it amounts to the same basic stuff. Strict Pythagorean tuning certainly did not dominate right from the get-go. So my question stands: why did it beat out all of these other approaches to tuning? Why did nothing "competitive" come out of these tunings until a few millennia later?

Because they couldn't figure out how to use the 5th harmonic in music?
I dunno, what are you getting at? That what we call 5/4 is actually
81/64 or something?

> > 8HE would say the chord I laid out above is really consonant. But who
> > cares about HE anyway?
>
> Mmmm...I dunno, people who put stock in psychoacoustics in general?

I put plenty of stock into psychoacoustics, but I am well aware of
HE's flaws and wouldn't use it as a measure to actually predict how
good or bad things will sound. It's an ultra-simple proof of concept
model that aims to test the hypothesis that 5/4 colors the sound of
81/64 and similar stuff like that. That's about it. It doesn't take
into account learning, and Paul has never claimed that it did.

Things I took from the HE model
- Intervals have fields of attraction
- There's no point trying to learn to hear the schismatic major third
as different from 5/4
- Attempting to do so because of a contrived ideal that the key to new
musical territory lies in discriminating finer and finer JI ratios
will yield continued diminishing returns
- That's about it

Things I don't think the HE model really predicts
- The exact spot when a mistuned interval will start sounding "bad"
for all human beings on planet earth
- The synergy that happens with 6:7:9, e.g. where it could sound like
3:6:7:9 or 1:6:7:9
- The exact predictions the curve makes over my own subjective
experience of hearing
- How learned factors will tie into it

The subject of learning is right now dealt with, in my opinion, a
handful of wildly inaccurate ideas that I am unsatisfied with but
don't know how to improve on. I have a love/hate relationship with
Rothenberg, and I have a hate/hate relationship with this "Tritone
hypothesis," and I have a hate/hate/hate relationship with the
"tonality coming from scales" concept in general. Until we can figure
out how to establish a "key" in novel tuning systems, and even better,
until we can figure out how to establish a concrete sense of "mode,"
we'll be forever running up against a wall in our exploration of novel
temperaments. The modes we establish will be unlike anything anyone
has ever heard, and once we learn how to cue a key, we'll be able to
really go crazy with it.

It will probably require zeroing in on what these learned factors of
tonality are, in regular ol 12-TET, and then applying them to novel
temperaments. V7-I is a strong cue, so that's one simple way to start.
Of course, people think that's cheating. Either way, I think Petr has
stumbled on some kind of epic and mindblowing discovery with what he's
doing with comma pumps. He's all about establishing a sense of "mode,"
whereas I was all about establishing a sense of "key." If we could
harmonize the two, I think we'd hit the sweet spot. But I'm just being
silly, really, because I already have the concept in my head, I just
need to get my computer set up and keep cranking examples out and stop
arguing about psychoacoustics.

-Mike

--- In tuning@yahoogroups.com, "cityoftheasleep" <igliashon@...> wrote:
>
> First, a question: how many of you have played or otherwise demonstrated alternative intonational systems to non-microtonalists and/or non-musicians? Those of you that have, have you noticed any trends in the responses?

I find definite "trends", presenting alternative tuning systems regularly.

>
> What's been nagging me for a while is that for all the theories that get discussed here, there seems to be very little consensus both on list and off over "what sounds good"--even when "stylistic" considerations are accounted for. On this list alone, there is a huge diversity of tuning preferences and one person's "out of tune" is another person's "perfect harmony".

Perception of any art is heavily influenced, sometimes almost dictated, by context. (Finding or creating the right contexts is most of the battle.) An individual's perception of "out of tune" is contextual, too. The ambient soundscape I've got running in a gallery at the moment is going over great- it's tuned to successive golden sections (phi, phi of phi, phi of phi of phi, etc., in the frequency realm, ie. it's a differentially coherent and wind-chimey tuning).

>
> Another thing I notice, specifically with guitarists, is that they don't give a damn how allegedly "in-tune" a system is, they really only care how many frets there are, because to them anything non-12 is all lumped together and all they care about is how scary the guitar looks.

:-) Death metallers should be competing in this arena then.

>Of the five or six guitarists to pick up and play my microtonal >guitars, even after I've demonstrated the properties of each, not a >single person has preferred 19 to 16. I have good reason to suspect >that 15 would go over even better, and 13 or 14 maybe better still.

Almost every guitarist I've ever come across is pattern-oriented. I would guess that 12-tET muscle-memory doodling makes interesting sounds in 13, 14, 15, but sounds wonky in 19-edo. Something of an educated guess, from playing fretless and saz, and hearing shear gak and complete lack of comprehension when highly skilled guitarists try to play my saz (fretted close to 17-edo). In contrast a friend of mine who is only a moderately skilled guitarist but has an excellent ear takes only seconds to make decent sounds in all kinds of frettings (we've been cruising for axes together in Turkey and China).

Notice that this does not mean that once the mind were to let the fingers go to better places in 19-edo than they tend to go by habit, that 19-edo would then sound better than the more exotic tunings. And of course you did demonstrate aspects of each tuning, so it couldn't be simply a matter of the other persons playing habits.

My seven-year-old gripes about flat fifths and, regardless of whether this is because of cultural or physical reasons, or some combination of the two, I bet his perception is common.

If you're looking for common reactions, I'd say that high fifths vs. low fifths making a very big difference in feel is definitely a consistent reaction I've found. But there's context again- the 26-edo music I did in a "water...voyage..." kind of context went over very well, though the 3*(8/7) "fifths" are concretely flat of 3:2.

>
> I know Mike has mentioned that his music friends usually seem less than turned on by microtonal music, even really in-tune stuff like 22-EDO. This is the real crux for me: if psychoacoustics is really important, shouldn't there be almost universal preference for more in-tune systems and almost universal aversion to not-so-in-tune systems? In reality it seems like psychoacoustic considerations rarely do better than random chance at predicting alternate tuning system preferences, though they do corroborate the overwhelming preference for 12-TET and/or meantone.

I find that people are turned on microtonal music and consistently immediately recognize the mellow/trippy/soft/organic/ancient etc quality of rational tuning, up to ratios of the 13th partial. I include very accurate and micro- temperaments under "rational", and differentially-coherent tunings a la Dudon also get similar reactions. Some people dig the very inharmonic tunings like 11 and 13. The only tunings that I find to consistently inflict cringing and laughs are the "good" tunings like 22, but I'll have to more systematically verify this, because these are only inflicted upon folks in demonstrations, not in my actual music.

>
> Over the years, theorists have claimed (snip)

hm yes. I try to give more credence to theory that is directly tied to my actual experiences personal and social, but as you know I'm a clueless idiot. :-)

Up to and including ratios of the 13th partial I should have said.

--- In tuning@yahoogroups.com, "lobawad" <lobawad@...> wrote:
>
>
>
> --- In tuning@yahoogroups.com, "cityoftheasleep" <igliashon@> wrote:
> >
> > First, a question: how many of you have played or otherwise demonstrated alternative intonational systems to non-microtonalists and/or non-musicians? Those of you that have, have you noticed any trends in the responses?
>
> I find definite "trends", presenting alternative tuning systems regularly.
>
> >
> > What's been nagging me for a while is that for all the theories that get discussed here, there seems to be very little consensus both on list and off over "what sounds good"--even when "stylistic" considerations are accounted for. On this list alone, there is a huge diversity of tuning preferences and one person's "out of tune" is another person's "perfect harmony".
>
> Perception of any art is heavily influenced, sometimes almost dictated, by context. (Finding or creating the right contexts is most of the battle.) An individual's perception of "out of tune" is contextual, too. The ambient soundscape I've got running in a gallery at the moment is going over great- it's tuned to successive golden sections (phi, phi of phi, phi of phi of phi, etc., in the frequency realm, ie. it's a differentially coherent and wind-chimey tuning).
>
>
> >
> > Another thing I notice, specifically with guitarists, is that they don't give a damn how allegedly "in-tune" a system is, they really only care how many frets there are, because to them anything non-12 is all lumped together and all they care about is how scary the guitar looks.
>
> :-) Death metallers should be competing in this arena then.
>
> >Of the five or six guitarists to pick up and play my microtonal >guitars, even after I've demonstrated the properties of each, not a >single person has preferred 19 to 16. I have good reason to suspect >that 15 would go over even better, and 13 or 14 maybe better still.
>
> Almost every guitarist I've ever come across is pattern-oriented. I would guess that 12-tET muscle-memory doodling makes interesting sounds in 13, 14, 15, but sounds wonky in 19-edo. Something of an educated guess, from playing fretless and saz, and hearing shear gak and complete lack of comprehension when highly skilled guitarists try to play my saz (fretted close to 17-edo). In contrast a friend of mine who is only a moderately skilled guitarist but has an excellent ear takes only seconds to make decent sounds in all kinds of frettings (we've been cruising for axes together in Turkey and China).
>
> Notice that this does not mean that once the mind were to let the fingers go to better places in 19-edo than they tend to go by habit, that 19-edo would then sound better than the more exotic tunings. And of course you did demonstrate aspects of each tuning, so it couldn't be simply a matter of the other persons playing habits.
>
> My seven-year-old gripes about flat fifths and, regardless of whether this is because of cultural or physical reasons, or some combination of the two, I bet his perception is common.
>
> If you're looking for common reactions, I'd say that high fifths vs. low fifths making a very big difference in feel is definitely a consistent reaction I've found. But there's context again- the 26-edo music I did in a "water...voyage..." kind of context went over very well, though the 3*(8/7) "fifths" are concretely flat of 3:2.
>
> >
> > I know Mike has mentioned that his music friends usually seem less than turned on by microtonal music, even really in-tune stuff like 22-EDO. This is the real crux for me: if psychoacoustics is really important, shouldn't there be almost universal preference for more in-tune systems and almost universal aversion to not-so-in-tune systems? In reality it seems like psychoacoustic considerations rarely do better than random chance at predicting alternate tuning system preferences, though they do corroborate the overwhelming preference for 12-TET and/or meantone.
>
> I find that people are turned on microtonal music and consistently immediately recognize the mellow/trippy/soft/organic/ancient etc quality of rational tuning, up to ratios of the 13th partial. I include very accurate and micro- temperaments under "rational", and differentially-coherent tunings a la Dudon also get similar reactions. Some people dig the very inharmonic tunings like 11 and 13. The only tunings that I find to consistently inflict cringing and laughs are the "good" tunings like 22, but I'll have to more systematically verify this, because these are only inflicted upon folks in demonstrations, not in my actual music.
>
> >
> > Over the years, theorists have claimed (snip)
>
> hm yes. I try to give more credence to theory that is directly tied to my actual experiences personal and social, but as you know I'm a clueless idiot. :-)
>

--- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
> > I'm not trying to speculate on how to solve this "problem". I'm trying to speculate on
> >why it is a problem and why it has so far failed to be solved.
>
> Oh. Well, the answer to why it's a problem, is that it's just a
> general tendency of the brain to extract order from the information
> stream that it's being fed.

Man, you're not picking up what I'm putting down at all.

The "problem" is that about the only tuning system everyone on this list (and in the general public) can agree "sounds good" is Meantone (and to a very slightly lesser extent, 12-TET). Every single other tuning system that gets discussed has at most a handful of devotees and maybe a modicum of casual practitioners--this despite whatever definitive theoretical "advantages" any of them have over all other tuning systems. The "problem" (or "the big masturbatory theoretical question" as I have taken to thinking of it) is why was it possible to reach a massive consensus on one tuning system, but not any more than that?

> Because they couldn't figure out how to use the 5th harmonic in music?
> I dunno, what are you getting at? That what we call 5/4 is actually
> 81/64 or something?

What I'm getting at is Ancient Greece was sort of the "primordial soup" of tuning, out of which came the precursors to meantone and 12-TET in the form of Pythagorean tuning. There were many possible roads that the multitudes of scales (such as those of Ptolemy and Archytas) Western civilization could have gone down. Hell, several could have been pursued at once...but they weren't. Despite the fact that there was ample knowledge of Just intonation, all tunings eventually fell before the juggernaut of the Pythagorean diatonic scale.

I guess what I'm really getting at is, if psychoacoustics (as we understand it) is really so fundamental, what kept people from basing scales on other simple ratios--or rather, what kept those scales from catching on? Nowadays we can point to the various 12-TET-based infrastructures and bemoan the cultural inertia that makes them so stuck, but back in the ancient times, there were no infrastructures. There were just peoples' preferences.

If you look around the world at all the various indigenous musical tunings, pretty much the only tunings that are supported by psychoacoustic principles are 12-TET/meantone/Pythagorean. Maybe some middle-eastern scales too, but not so much the scales of the South-East (Thailand especially), or South America, or Africa (the Chopi come to mind), or the South Pacific (the Sundanese/Balinese/Javanese gamelan traditions). Doesn't it seem odd, if you believe that the "naturalness" of simple ratios explains the evolution of meantone, that no one ever discovered Porcupine or Magic or Srutal (*especially* Srutal)? Sure, meantone is psychoacoustically "better" than all of those, but all of them seem more psychoacoustically "valid" than the various scales of the non-Western world. I mean, if the things we claim to understand about the psychoacoustic basis of Western scales really hold true across humanity, why don't any of the traditional non-Pythagorean systems fit similar patterns? Why should an exo-temperament like Mavila be more-or-less "found in the wild" while Porcupine isn't? Why is Thai traditional music in 7-EDO? Why are there neutral intervals in middle-eastern music?

If psychoacoustics cannot help us answer these questions, I don't think it will be very useful as a guide to finding new tunings that people will like.

> I put plenty of stock into psychoacoustics, but I am well aware of
> HE's flaws and wouldn't use it as a measure to actually predict how
> good or bad things will sound. It's an ultra-simple proof of concept
> model that aims to test the hypothesis that 5/4 colors the sound of
> 81/64 and similar stuff like that. That's about it. It doesn't take
> into account learning, and Paul has never claimed that it did.

So...what pyschoacoustic theories do you put more stock in, as far as explaining peoples' musical preferences?

> The subject of learning is right now dealt with, in my opinion, a
> handful of wildly inaccurate ideas that I am unsatisfied with but
> don't know how to improve on. I have a love/hate relationship with
> Rothenberg, and I have a hate/hate relationship with this "Tritone
> hypothesis," and I have a hate/hate/hate relationship with the
> "tonality coming from scales" concept in general.

Ditto...except that I pretty much have a hate/hate relationship with Rothenberg, too.

> But I'm just being
> silly, really, because I already have the concept in my head, I just
> need to get my computer set up and keep cranking examples out and stop
> arguing about psychoacoustics.

Yeah, I agree. It's probably detrimental to both of our actual contributions to the field for us to keep having discussions like this. I think the take-away point of my above ramble is that input from non-microtonalists and non-musicians is probably a better guide than input from other microtonalists, even though I think a lot of us tend to ignore or discount the opinions of our friends and family because they don't "get it". In reality, their opinions are probably more illuminating that all the psychoacoustic theories you can shake a stick at.

-Igs

On Fri, May 13, 2011 at 1:15 PM, cityoftheasleep
<igliashon@...> wrote:
>
> Man, you're not picking up what I'm putting down at all.
>
> The "problem" is that about the only tuning system everyone on this list (and in the general public) can agree "sounds good" is Meantone (and to a very slightly lesser extent, 12-TET). Every single other tuning system that gets discussed has at most a handful of devotees and maybe a modicum of casual practitioners--this despite whatever definitive theoretical "advantages" any of them have over all other tuning systems. The "problem" (or "the big masturbatory theoretical question" as I have taken to thinking of it) is why was it possible to reach a massive consensus on one tuning system, but not any more than that?

I like it. The BMTQ, I christen it. I think everyone on this list has
come up to the same question, and that's why in addition to all of the
stuff about psychoacoustics, Rothenberg keeps getting thrown around.
Rothenberg, the tritone hypothesis, etc are all basically initial,
preliminary attempts to answer that question. They do not address
psychoacoustics at all, but attempt to address the complex underlying
system of pattern recognition that belies the very cognitive structure
of music that you keep referencing.

I tend to treat Rothenberg as a fascinating misfire. I can never quite
put my finger on what it is that I don't like about it. The tritone
hypothesis, I think is far too simplistic. But these are all an
attempt to address exactly what you're talking about. I think that
where they all fail is that they overemphasize "scale structure,"
while not providing enough insight into the underlying mechanisms of
pattern recognition that scale structures can encapsulate.

The even bigger masturbatory theoretical question that I have is
whether or not this itself is built on something more fundamental, or
if it's all arbitrary. Since we seem to keep "discovering new order"
in music (like the blues), and certain things really do just "work," I
think there is something more fundamental at work, and it may not be
psychoacoustics. I note that lots of my friends jumped onto Blackwood
really easily (assuming a suitable timbre), for instance, but they
hate porcupine.

> I guess what I'm really getting at is, if psychoacoustics (as we understand it) is really so fundamental, what kept people from basing scales on other simple ratios--or rather, what kept those scales from catching on? Nowadays we can point to the various 12-TET-based infrastructures and bemoan the cultural inertia that makes them so stuck, but back in the ancient times, there were no infrastructures. There were just peoples' preferences.

Pythagoras based everything around the 3/2, and then Archytas added
5-limit stuff to it, and then it was Ptolemy centuries later that
decided music was more about tetrachordal structure and less about JI,
if I read everything correctly. So I guess the tuning list is in the
Pythagoras stage now, and we need to get into the Ptolemy stage. I
don't know if tetrachordal structure is really what it's all about,
however.

> Doesn't it seem odd, if you believe that the "naturalness" of simple ratios explains the evolution of meantone, that no one ever discovered Porcupine or Magic or Srutal (*especially* Srutal)?

When you say "Srutal," are you talking about the 10-note MOS of
Srutal, or just the diaschismatic tempering in general? Because if
it's the latter, then they definitely did discover it, that's what
tritone substitutions are all about.

> Sure, meantone is psychoacoustically "better" than all of those, but all of them seem more psychoacoustically "valid" than the various scales of the non-Western world. I mean, if the things we claim to understand about the psychoacoustic basis of Western scales really hold true across humanity, why don't any of the traditional non-Pythagorean systems fit similar patterns? Why should an exo-temperament like Mavila be more-or-less "found in the wild" while Porcupine isn't? Why is Thai traditional music in 7-EDO? Why are there neutral intervals in middle-eastern music?

I think someone's tolerance to mistuning can vary, and hence so can
that of an entire culture. They liked Mavila, and didn't mind the
blown fifths, so they worked around them. I think that JI is just one
sound that people can learn to like, as can they learn to like Mavila.
So I think when people claim that Mavila is a "bad tuning" because of
the flat fifths, I think they're basically just saying they don't want
to learn to get used to it. But this doesn't mean that the concept of
harmonic series detection doesn't apply at all to the Mavila folks.

> So...what pyschoacoustic theories do you put more stock in, as far as explaining peoples' musical preferences?

I put stock in basic psychoacoustic ideas and concepts in general,
which means I prefer to critically analyze HE for its strengths and
weaknesses, use it for its strengths, and not think too much about its
weaknesses. I think you tend to apply a more rigid interpretation of
the HE data than many on the list. But then, after applying this more
rigid interpretation, you criticize it. I treat HE as an interesting
model that was a step in the right direction, not the end-all be-all
of music cognition.

> > But I'm just being
> > silly, really, because I already have the concept in my head, I just
> > need to get my computer set up and keep cranking examples out and stop
> > arguing about psychoacoustics.
>
> Yeah, I agree. It's probably detrimental to both of our actual contributions to the field for us to keep having discussions like this.

I do think it's good to discuss theory, but I want to make sure it's
balanced with actual music at this point, or at least musical
examples. It would be really great if, in arguments about music
theory, people tried to win by constantly outdoing one another with
better musical examples to demonstrate their points. Then everyone
wins. I bet discussions here would get less heated, too.

> I think the take-away point of my above ramble is that input from non-microtonalists and non-musicians is probably a better guide than input from other microtonalists, even though I think a lot of us tend to ignore or discount the opinions of our friends and family because they don't "get it". In reality, their opinions are probably more illuminating that all the psychoacoustic theories you can shake a stick at.

Sure, although I think what you're really reacting against is
extremism, not psychoacoustics, and I think this attitude will also
lead to some extremist outcomes. But I do note that people seemed to
like my porcupine comma pump, and to not like my 17-tet etude.

-Mike

Hi Igs,

--- In tuning@yahoogroups.com, "cityoftheasleep" <igliashon@...> wrote:
>
> First, a question: how many of you have played or otherwise demonstrated alternative intonational systems to non-microtonalists and/or non-musicians? Those of you that have, have you noticed any trends in the responses?

I've played a lot of stuff to a non-micro musician friend. The response is almost always reasonably positve and: "if you hadn't told me it was microtonal I wouldn't have guessed". The only exception was "Stick Men" by Elaine Walker, in Bohlen Pierce 13ED3: "sounds out of tune".

Sometimes I wonder if in some circumstances we just don't tell people it's microtonal, so as to not give them a reason to dislike it.

MatC

Igs> "The "problem" is that about the only tuning system everyone on this
list (and in the general public) can agree "sounds good" is Meantone
(and to a very slightly lesser extent, 12-TET). "

  What about Mohajira and some of 12-tone systems Gene has posted?  At least I recall a handful of them sounding good where others noted very useful properties for those scales (if not outright praise for them).  Same goes with Wilson's Hexanies scale...I have shown it to a bunch of people and rarely heard any complaints.  Same goes with Blackwood.  Igs (if you could please post a link to your last guitar album...I lost it when my last hard disk died) I think your last album is a fleeting example of just how well at many times "non-Pythagorean" scales can work...I showed a handful of people that and even if the aren't "followers of those tunings"...all of them liked it.

>"So I guess the tuning list is in the Pythagoras stage now, and we need to get into the Ptolemy stage. I don't know if tetrachordal structure is really what it's all about,

however."
  Speaking of Ptolemy...from what I've shown people I've composed in his scales...people don't seem to find it "out of tune" at all (unlike, say, Mohajira)...I've heard descriptions like "the same painting with different tint and contrast".  And (as I've said before)...a lot of this seems to go back to using 12/11-like neutral semitones often adding up to more familiar sounding 3rds...even though the 5ths are often off (think 16/11)...the way the thirds add up often give people enough "priming" so the emotional "gist" of the chords hold up as if there were a fifth.

   To me that's the obvious challenge we should be dealing with...how to make "tolerable" chords that can live without perfect fifths...and how to make scales that contain such chords.  As long as we're stuck with optimizing (near)perfect fifths, to a large extent, we're only making "beautified Pythagorean scales", even if they aren't technically Pythagorean.  In fact, coming down to it...it seems virtually the whole of Western music is "beautified Pythagorean"...the only difference is things like the ability to modulate between scales in a tuning has come a long way.  We really do seem stuck with v"ariations on Pythagorus"...why not try something else?

--- On Fri, 5/13/11, Mike Battaglia <battaglia01@...> wrote:

From: Mike Battaglia <battaglia01@...>
Subject: Re: [tuning] Re: Where Theory Falls Short
To: tuning@yahoogroups.com
Date: Friday, May 13, 2011, 11:12 AM

On Fri, May 13, 2011 at 1:15 PM, cityoftheasleep

<igliashon@sbcglobal.net> wrote:

>

> Man, you're not picking up what I'm putting down at all.

>

> The "problem" is that about the only tuning system everyone on this list (and in the general public) can agree "sounds good" is Meantone (and to a very slightly lesser extent, 12-TET). Every single other tuning system that gets discussed has at most a handful of devotees and maybe a modicum of casual practitioners--this despite whatever definitive theoretical "advantages" any of them have over all other tuning systems. The "problem" (or "the big masturbatory theoretical question" as I have taken to thinking of it) is why was it possible to reach a massive consensus on one tuning system, but not any more than that?

I like it. The BMTQ, I christen it. I think everyone on this list has

come up to the same question, and that's why in addition to all of the

stuff about psychoacoustics, Rothenberg keeps getting thrown around.

Rothenberg, the tritone hypothesis, etc are all basically initial,

preliminary attempts to answer that question. They do not address

psychoacoustics at all, but attempt to address the complex underlying

system of pattern recognition that belies the very cognitive structure

of music that you keep referencing.

I tend to treat Rothenberg as a fascinating misfire. I can never quite

put my finger on what it is that I don't like about it. The tritone

hypothesis, I think is far too simplistic. But these are all an

attempt to address exactly what you're talking about. I think that

where they all fail is that they overemphasize "scale structure,"

while not providing enough insight into the underlying mechanisms of

pattern recognition that scale structures can encapsulate.

The even bigger masturbatory theoretical question that I have is

whether or not this itself is built on something more fundamental, or

if it's all arbitrary. Since we seem to keep "discovering new order"

in music (like the blues), and certain things really do just "work," I

think there is something more fundamental at work, and it may not be

psychoacoustics. I note that lots of my friends jumped onto Blackwood

really easily (assuming a suitable timbre), for instance, but they

hate porcupine.

> I guess what I'm really getting at is, if psychoacoustics (as we understand it) is really so fundamental, what kept people from basing scales on other simple ratios--or rather, what kept those scales from catching on? Nowadays we can point to the various 12-TET-based infrastructures and bemoan the cultural inertia that makes them so stuck, but back in the ancient times, there were no infrastructures. There were just peoples' preferences.

Pythagoras based everything around the 3/2, and then Archytas added

5-limit stuff to it, and then it was Ptolemy centuries later that

decided music was more about tetrachordal structure and less about JI,

if I read everything correctly. So I guess the tuning list is in the

Pythagoras stage now, and we need to get into the Ptolemy stage. I

don't know if tetrachordal structure is really what it's all about,

however.

> Doesn't it seem odd, if you believe that the "naturalness" of simple ratios explains the evolution of meantone, that no one ever discovered Porcupine or Magic or Srutal (*especially* Srutal)?

When you say "Srutal," are you talking about the 10-note MOS of

Srutal, or just the diaschismatic tempering in general? Because if

it's the latter, then they definitely did discover it, that's what

tritone substitutions are all about.

> Sure, meantone is psychoacoustically "better" than all of those, but all of them seem more psychoacoustically "valid" than the various scales of the non-Western world. I mean, if the things we claim to understand about the psychoacoustic basis of Western scales really hold true across humanity, why don't any of the traditional non-Pythagorean systems fit similar patterns? Why should an exo-temperament like Mavila be more-or-less "found in the wild" while Porcupine isn't? Why is Thai traditional music in 7-EDO? Why are there neutral intervals in middle-eastern music?

I think someone's tolerance to mistuning can vary, and hence so can

that of an entire culture. They liked Mavila, and didn't mind the

blown fifths, so they worked around them. I think that JI is just one

sound that people can learn to like, as can they learn to like Mavila.

So I think when people claim that Mavila is a "bad tuning" because of

the flat fifths, I think they're basically just saying they don't want

to learn to get used to it. But this doesn't mean that the concept of

harmonic series detection doesn't apply at all to the Mavila folks.

> So...what pyschoacoustic theories do you put more stock in, as far as explaining peoples' musical preferences?

I put stock in basic psychoacoustic ideas and concepts in general,

which means I prefer to critically analyze HE for its strengths and

weaknesses, use it for its strengths, and not think too much about its

weaknesses. I think you tend to apply a more rigid interpretation of

the HE data than many on the list. But then, after applying this more

rigid interpretation, you criticize it. I treat HE as an interesting

model that was a step in the right direction, not the end-all be-all

of music cognition.

> > But I'm just being

> > silly, really, because I already have the concept in my head, I just

> > need to get my computer set up and keep cranking examples out and stop

> > arguing about psychoacoustics.

>

> Yeah, I agree. It's probably detrimental to both of our actual contributions to the field for us to keep having discussions like this.

I do think it's good to discuss theory, but I want to make sure it's

balanced with actual music at this point, or at least musical

examples. It would be really great if, in arguments about music

theory, people tried to win by constantly outdoing one another with

better musical examples to demonstrate their points. Then everyone

wins. I bet discussions here would get less heated, too.

> I think the take-away point of my above ramble is that input from non-microtonalists and non-musicians is probably a better guide than input from other microtonalists, even though I think a lot of us tend to ignore or discount the opinions of our friends and family because they don't "get it". In reality, their opinions are probably more illuminating that all the psychoacoustic theories you can shake a stick at.

Sure, although I think what you're really reacting against is

extremism, not psychoacoustics, and I think this attitude will also

lead to some extremist outcomes. But I do note that people seemed to

like my porcupine comma pump, and to not like my 17-tet etude.

-Mike

"cityoftheasleep" <igliashon@...> wrote:

> The "problem" is that about the only tuning system everyone on
> this list (and in the general public) can agree "sounds good"
> is Meantone (and to a very slightly lesser extent, 12-TET).

They do? This list used to be a list of confirmed 12-ET
haters. Not sure if/when that changed.

> Every single other tuning system that gets discussed has at
> most a handful of devotees

I'm not a practitioner, but I like 'em all.

> Doesn't it seem odd, if you believe that the "naturalness"
> of simple ratios explains the evolution of meantone, that no
> one ever discovered Porcupine or Magic or Srutal

Howabout harmony period? Only one culture seems to have
discovered it. And it took hundreds of years and a major
cross-cultural event to get them to accept 7th chords.

Why didn't medieval people use internal combustion engines?
All the parts were available.

> If psychoacoustics cannot help us answer these questions,
> I don't think it will be very useful as a guide to finding
> new tunings that people will like.

Tunings we see around the world are consistent with a
badness-weighted random sampling of the regular temperament
universe. That's a bold claim I have no interest in
rigorously backing up. Partly it's true because we fitted
the theory to the data, and partly because there's so
little data it's consistent with a lot of theories.

> I think the take-away point of my above ramble is that input
> from non-microtonalists and non-musicians is probably a better
> guide than input from other microtonalists, even though I think
> a lot of us tend to ignore or discount the opinions of our
> friends and family because they don't "get it". In reality,
> their opinions are probably more illuminating that all the
> psychoacoustic theories you can shake a stick at.

You must be joking! Just look at how vehemently people's
tastes differ on what music sounds good (country, rap,
classical...) even when it's in the SAME tuning. I mean,
are you living on the same planet as me?

We could go back to 1300 and shop around a desktop Stirling
engine. What reactions do you think we'd get?

http://amzn.to/m6RamV

No doubt some clever Islamic polymath did just that around
1300, and got the gee-whiz reaction you'd expect. Because
the adoption of the engine required changes in society, and
society had to be ready, and it had to go all-in to realize
those changes to realize any use for the engine.
And I hate to say it but the prospects for this sort of
thing in music are poor and trending worse, given the state
of music literacy in the population. The 7th chords didn't
replace triads and even triads are becoming an endangered
species. In the '90s I was very hopeful, but trends since
then have not been good and I also seem to have been raised
in a place with a strong local culture of music literacy that
was not representative of the general population. Trends
can still reverse of course. . .

Since you asked (yeah right) I'll tell you bluntly: listen
to Mike's porcupine w/higher-limit extensions demo and you'll
hear awesome sounds that are considerably more awesome than
found anywhere in your music. The difference is, your music
is music, and Mike's demo is just a demo. Music is
compelling, and demos are gee-whiz.

Now please, get your mind right and stop having a failure
to communicate about masturbation and theory and glaven.
I like masturbating and I like theory, so it's a natural
fit for me. Please don't spoil it for me.

-Carl

Carl>"They do? This list used to be a list of confirmed 12-ET haters. Not sure if/when that changed."

  The leading psychoacoustic theories tossed around, to say the least, tend to focus on preserving the fifth foremost.  Hence the indirect focus on Pythagorean. 

>"How about harmony period? Only one culture seems to have

discovered it. And it took hundreds of years and a major

cross-cultural event to get them to accept 7th chords."

I think it's fair to say the Middle Eastern culture discovered their own type of harmony...which is, in many ways, more along the lines of what Ptolemy was doing.  I certainly don't believe Western culture is the only one that "discovered harmony".

Igs>non-musicians is probably a better guide than input from other microtonalists
Agreed...as music is ultimately a form of emotional communication...not an art with a goal of conforming to formulas to the highest degree possible.  And don't get me wrong...extracting formulas and patterns is a sweet form of mental engineering...but it's also one that, so far as music, often poses more useless limitations than emotionally useful possibilities.

To finish this off: Why didn't you use porcupine with higher-
limit extensions for a piece? Because you didn't. Can you go
back and retune one of your pieces to it? Maybe, but the
result probably wouldn't be good. The tuning you used was part
of your process and it made a piece I like, therefore it was
a good tuning. Likewise, if porcupine with higher-limit
extensions is somehow preventing Mike from finishing a piece
of music, it's a bad tuning for him.

I'm going to get a bit crazy and claim that I've exhibited
more microtonal music, of more different kinds, to more people
from more walks of life than anyone reading this. The
reactions I've gotten are as varied and contradictory as
you'd expect. Here are three memorable ones:

* When I first heard Partch I thought it was just noise.
Denny told me a lot of people have that reaction. Now I
think it's pure greatness.

* When I first heard Junta, Relayer, and several other of
the best albums ever recorded, I thought they were crap.
Wait- that's not microtonal.

* In 2002 I played David Doty's Paradigms Lost for a
friend - she's a music lover and sometime singer in an
early music octet.
http://lumma.org/tuning/xenmusic/ParadigmsLost.mp3

In 1998 I thought this piece was sent by God to open my
mind, and I still think it sounds great. She made me turn
it off after about a minute, claiming it sounded horribly
out of tune.

-Carl

I wrote:
> Since you asked (yeah right) I'll tell you bluntly: listen
> to Mike's porcupine w/higher-limit extensions demo and you'll
> hear awesome sounds that are considerably more awesome than
> found anywhere in your music. The difference is, your music
> is music, and Mike's demo is just a demo. Music is
> compelling, and demos are gee-whiz.

Michael <djtrancendance@...> wrote:

> I think it's fair to say the Middle Eastern culture discovered
> their own type of harmony...which is, in many ways, more along
> the lines of what Ptolemy was doing.  I certainly don't believe
> Western culture is the only one that "discovered harmony".

When "harmony" is suitably defined, they were. :)
There was a HUGE discussion on this on MMM recently, so I'll
refer to that:

/makemicromusic/topicId_25238.html#25357
/makemicromusic/topicId_25132.html#25187
etc.

-Carl

--- In tuning@yahoogroups.com, "cityoftheasleep" <igliashon@...> wrote:

> What I'm getting at is Ancient Greece was sort of the "primordial soup" of tuning, out of which came the precursors to meantone and 12-TET in the form of Pythagorean tuning. There were many possible roads that the multitudes of scales (such as those of Ptolemy and Archytas) Western civilization could have gone down. Hell, several could have been pursued at once...but they weren't. Despite the fact that there was ample knowledge of Just intonation, all tunings eventually fell before the juggernaut of the Pythagorean diatonic scale.

I think the fact that the Greek theory in the West was filtered through Boethius, whereas in the Islamic world and India it wasn't, may explain a lot of the difference.

I'd be interested in what people here have to say about this with respect
the BMTQ

http://www.thehistoryblog.com/archives/2326

35,000 yr old flute. It doesn't sound like it is far from... 12.

I mined the direct link to the mp3

http://www.thehistoryblog.com/wp-content/uploads/2009/07/flute_replica.mp3
<view-source:http://www.thehistoryblog.com/wp-content/uploads/2009/07/flute_replica.mp3>

I'll be putting it into v-vocal shortly to see what I can learn about the
tuning.

Chris

On Fri, May 13, 2011 at 4:56 PM, genewardsmith
<genewardsmith@...>wrote:

>
>
>
> --- In tuning@yahoogroups.com, "cityoftheasleep" <igliashon@...> wrote:
>
> > What I'm getting at is Ancient Greece was sort of the "primordial soup"
> of tuning, out of which came the precursors to meantone and 12-TET in the
> form of Pythagorean tuning. There were many possible roads that the
> multitudes of scales (such as those of Ptolemy and Archytas) Western
> civilization could have gone down. Hell, several could have been pursued at
> once...but they weren't. Despite the fact that there was ample knowledge of
> Just intonation, all tunings eventually fell before the juggernaut of the
> Pythagorean diatonic scale.
>
> I think the fact that the Greek theory in the West was filtered through
> Boethius, whereas in the Islamic world and India it wasn't, may explain a
> lot of the difference.
>
>
>

sounds different here

http://www.youtube.com/watch?v=xn5ZygWxKmw

On Fri, May 13, 2011 at 5:51 PM, Chris Vaisvil <chrisvaisvil@...>wrote:

> I'd be interested in what people here have to say about this with respect
> the BMTQ
>
> http://www.thehistoryblog.com/archives/2326
>
> 35,000 yr old flute. It doesn't sound like it is far from... 12.
>
> I mined the direct link to the mp3
>
> http://www.thehistoryblog.com/wp-content/uploads/2009/07/flute_replica.mp3
>
>
> I'll be putting it into v-vocal shortly to see what I can learn about the
> tuning.
>
> Chris
>
>
> On Fri, May 13, 2011 at 4:56 PM, genewardsmith <
> genewardsmith@...> wrote:
>
>>
>>
>>
>> --- In tuning@yahoogroups.com, "cityoftheasleep" <igliashon@...> wrote:
>>
>> > What I'm getting at is Ancient Greece was sort of the "primordial soup"
>> of tuning, out of which came the precursors to meantone and 12-TET in the
>> form of Pythagorean tuning. There were many possible roads that the
>> multitudes of scales (such as those of Ptolemy and Archytas) Western
>> civilization could have gone down. Hell, several could have been pursued at
>> once...but they weren't. Despite the fact that there was ample knowledge of
>> Just intonation, all tunings eventually fell before the juggernaut of the
>> Pythagorean diatonic scale.
>>
>> I think the fact that the Greek theory in the West was filtered through
>> Boethius, whereas in the Islamic world and India it wasn't, may explain a
>> lot of the difference.
>>
>>
>>
>
>

Wow, they are definitely different in timbre and slightly different in
register, it seems. We don't know which fingerings were chosen for the 5
holes, I guess, but they sure follow different tunings. We might guess
something like a tin-whistle type fingering.

Mainly, though, we don't even know if these are the same reproductions, do
we? or what exactly they reproduce about the original and to what
specification, or are these just similar-looking vulture bones with holes in
the same places? I haven't yet found any good info on the Internetz about
the pitches of the original. :/

In the folder are 3 screen shots that gives you the pitch vs 12 equal -
the first 2 from the mp3 and the last from the video - I deleted the part
where they talk over the sound of the flute.

http://micro.soonlabel.com/various/35000flute/

On Fri, May 13, 2011 at 7:25 PM, Daniel Nielsen <nielsed@uah.edu> wrote:

>
>
> Wow, they are definitely different in timbre and slightly different in
> register, it seems. We don't know which fingerings were chosen for the 5
> holes, I guess, but they sure follow different tunings. We might guess
> something like a tin-whistle type fingering.
>
> Mainly, though, we don't even know if these are the same reproductions, do
> we? or what exactly they reproduce about the original and to what
> specification, or are these just similar-looking vulture bones with holes in
> the same places? I haven't yet found any good info on the Internetz about
> the pitches of the original. :/
>
>

Very nice! gotta check these out later, Chris. (Just btw, you probably know
this, but these sorts of flutes normally have an "octave" that generally
slightly exceeds 2/1)

On Thu, May 12, 2011 at 11:55 PM, cityoftheasleep
<igliashon@...> wrote:
>
> > Two of my sax player friends are trying to learn it now. Apparently 23
> > of the 24 notes are playable on sax, but the 11/8 above C isn't (god
> > damn). Semaphore is definitely a good start, and so is 8-equal.
> > 10:12:15 can be turned into 10:11:12:13:14:15 without much thought. I
> > think 24-tet needs some kind of do-over in terms of how the community
> > sees it.
>
> I agree. I'll get right on it.

As a second note, 48-equal might be even sicker than 24. Once people
get used to 24, it's only a matter of time before they start thinking
about 48-equal. Not only does 48 contain 12 and 24, but it also
contains 16, but in a way that relates to 12. This assumes we find
ways to actually play instruments like that.

So you can merge the two systems - there are just two major thirds,
and two fifths (actually three, one of them is a good father fifth),
just one of them fits into the mavila/magic structure, whereas the
other fits into the meantone/augmented structure. You also get negri
and tetracot, which use the 16-equal major third and the 12-equal
fifth. There's some other stuff too.

I can easily see everyone going the 12->24->48 route, and then the
48->16->everything route. But let's hope we can get a more direct
route then that out there.

-Mike

Re Chris:
Since several flutes were apparently found at the site, I wonder if their
tunings correlate much. This must have been studied or understood, but I
have no idea where to find the answers.

On Fri, May 13, 2011 at 4:56 PM, genewardsmith
<genewardsmith@...> wrote:
>
> --- In tuning@yahoogroups.com, "cityoftheasleep" <igliashon@...> wrote:
>
> > What I'm getting at is Ancient Greece was sort of the "primordial soup" of tuning, out of which came the precursors to meantone and 12-TET in the form of Pythagorean tuning. There were many possible roads that the multitudes of scales (such as those of Ptolemy and Archytas) Western civilization could have gone down. Hell, several could have been pursued at once...but they weren't. Despite the fact that there was ample knowledge of Just intonation, all tunings eventually fell before the juggernaut of the Pythagorean diatonic scale.
>
> I think the fact that the Greek theory in the West was filtered through Boethius, whereas in the Islamic world and India it wasn't, may explain a lot of the difference.

So do we have any idea what ancient Greek music actually sounded like?
Is it basically lost to us now?

-Mike

On Fri, May 13, 2011 at 7:25 PM, Daniel Nielsen <nielsed@...> wrote:
>
> Wow, they are definitely different in timbre and slightly different in register, it seems. We don't know which fingerings were chosen for the 5 holes, I guess, but they sure follow different tunings. We might guess something like a tin-whistle type fingering.
> Mainly, though, we don't even know if these are the same reproductions, do we? or what exactly they reproduce about the original and to what specification, or are these just similar-looking vulture bones with holes in the same places? I haven't yet found any good info on the Internetz about the pitches of the original. :/

That is hip. Sounds pretty 12-equalish to me, but then again you have
to keep in mind that even if this is a faithful replication, they had
to hand it off to a modern, western, 12-equal flautist to make
anything happen from it. Thus it makes sense that they'd attempt to
play something that sounds intelligible to western ears on it.

-Mike

--- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:

> So do we have any idea what ancient Greek music actually sounded like?
> Is it basically lost to us now?

We've got some idea. We have fragments of music, recovery and reconstructions of instruments, and quite a lot of theoretical writing. The enharmonic, chromatic and diatonic genera in use during the Classical Period (Plato, et al) is a place to start.

--- In tuning@yahoogroups.com, "genewardsmith" <genewardsmith@...> wrote:

> The enharmonic, chromatic and diatonic genera in use during the Classical Period (Plato, et al) is a place to start.
>

We could use these as an excuse to drag temperament into it, particularly since it can be argued that meantone grew out of the diatonic genus. The diatonic genus divides up a fourth into two tones and a semitone, and if the tones are the same size, we can get Pythagorean, but also of course meantone. The chromatic has a fourth divided as a minor third and two semitones; making the semitones the same size and approximately 21/20, we could end up with octagar temperament, tempering out 4000/3969. The enharmonic genus divides the fourth into a major third and two small intervals; if we conflate 28/27 with 36/35, we temper out 245/243, and we can take the enharmonic genus to be in sensamagic (245/243) temperament.

Taking all three commas at once leads to the 14c val, which of course has nothing much to do with Greek anything. 81/80 and 245/243 leads to godzilla, and the attempt, I suppose, to play Greek music in 19edo. 81/80 and 4000/3969 together leads to injera temperament, another fun idea. Even more interesting, 245/243 with 4000/3969 leads to octacot temperament, which is pretty accurate, and related to the 88 cent business some people like. It would be interesting to see what could be done with what remains of Greek music in the enharmonic and chromatic generi, harmonized in octacot.

Just following the thread now...

Re: why 12 equal has dominated, the reasons in the west are well known:
expediency; not too many notes to tune or deal with on keyboards or other
instrument's physical interfaces; the prevalence of harmonic and triadic
thinking, usually in the five limit, bringing about meantime and then well
temperaments and then 12 equal when remote modulation, especially in the
18th century.

The question about non Western music like Indonesia and Thai and African is
interesting--- here I think it's a matter of cultural preference for beating
and a certain enjoyment of spicy sourness or complexity in the intervals: in
Gamelan music for instance, the beating, even of unisons, was a way of
producing divine altered states of awareness.

AKJ
On May 14, 2011 10:34 AM, "genewardsmith" <genewardsmith@...>
wrote:
>
> --- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
>
>> So do we have any idea what ancient Greek music actually sounded like?
>> Is it basically lost to us now?
>
> We've got some idea. We have fragments of music, recovery and
reconstructions of instruments, and quite a lot of theoretical writing. The
enharmonic, chromatic and diatonic genera in use during the Classical Period
(Plato, et al) is a place to start.
>
>
>
> ------------------------------------
>
> You can configure your subscription by sending an empty email to one
> of these addresses (from the address at which you receive the list):
> tuning-subscribe@yahoogroups.com - join the tuning group.
> tuning-unsubscribe@yahoogroups.com - leave the group.
> tuning-nomail@yahoogroups.com - turn off mail from the group.
> tuning-digest@yahoogroups.com - set group to send daily digests.
> tuning-normal@yahoogroups.com - set group to send individual emails.
> tuning-help@yahoogroups.com - receive general help information.
> Yahoo! Groups Links
>
>
>

this is an important part of history - I wonder why this isn't being talked
about - or even studied.

On Sat, May 14, 2011 at 10:52 AM, Daniel Nielsen <nielsed@...> wrote:

>
>
> Re Chris:
> Since several flutes were apparently found at the site, I wonder if their
> tunings correlate much. This must have been studied or understood, but I
> have no idea where to find the answers.
>
>
>
>

Haven't been following this whole thread, but...

> The question about non Western music like Indonesia and Thai and African is
> interesting--- here I think it's a matter of cultural preference for beating
> and a certain enjoyment of spicy sourness or complexity in the intervals: in
> Gamelan music for instance, the beating, even of unisons, was a way of
> producing divine altered states of awareness.

If I remember correctly, Sethares's analysis of the gamelan spectra crossed with harmonic spectra (which do exist in gamelan ensembles in the voice and sometimes string instruments) led to sensory dissonance minima (i.e. minimal beating) in both the 5-tone (approximately) equal and pelog scales used in gamelan ensembles.

The 7-tone (approximately) equal balafon (wooden xylophones) tunings of Africa also line up with sensory dissonance minima using their spectra, this time without having to be crossed with harmonic spectra.

The preference for nonwestern intervals still seems to be motivated by a lack of beating, the lack of beating just ending up at different intervals given different spectra.

John M

--- Mike Battaglia <battaglia01@...> wrote:

> So do we have any idea what ancient Greek music actually
> sounded like? Is it basically lost to us now?

We don't really know what Beethoven sounded like. We can
reconstruct the instruments of the day with the materials
of the day and read the notes and try to interpret them.
It's the same with ancient Greek music, except it's further
in the past and they didn't write down the notes. The
closest thing we have today is maqam music.

A long time ago, John Chalmers provided this link:
http://www.oeaw.ac.at/kal/agm/

Probably there is youtube gold to be had also. -Carl

--- In tuning@yahoogroups.com, "jlmoriart" <JlMoriart@...> wrote:

> If I remember correctly, Sethares's analysis of the gamelan
> spectra crossed with harmonic spectra (which do exist in
> gamelan ensembles in the voice and sometimes string
> instruments) led to sensory dissonance minima (i.e. minimal
> beating) in both the 5-tone (approximately) equal and pelog
> scales used in gamelan ensembles.

Also, I have a beautiful oceanfront timeshare in Arizona
you might like!

> The 7-tone (approximately) equal balafon (wooden xylophones)
> tunings of Africa also line up with sensory dissonance minima
> using their spectra, this time without having to be crossed
> with harmonic spectra.

Timeshares are a cost-effective way to vacation!

-Carl

I found this via wikipedia

http://www.myspace.com/mvsicaromana

--- In tuning@yahoogroups.com, "Carl Lumma" <carl@...> wrote:

> Probably there is youtube gold to be had also. -Carl

On Sat, May 14, 2011 at 12:22 PM, Chris Vaisvil <chrisvaisvil@...>wrote:

>
>
> this is an important part of history - I wonder why this isn't being talked
> about - or even studied.
>

Well, I sent a brief inquiry to the maker of the popular online Flutomat
calculator (http://www.cwo.com/~ph_kosel/designs.html). Re:

"I have no idea why you sent this to me. It has nothing to do with flutomat
and I am not an antiquarian. Why not ask the people at some second-hand
shop where you live instead? I'm an old man and don't have time for this
nonsense."

Well, at least he replied. The response was a bit confusing considering I
explained my motivation, explicitly asked if he had any interest in this
sort of thing at all, and noted I was just an inquiring stranger. Assuming
his advice was sincere, I can't think anyone at any of the junk shops around
here would be of any help in answering the question.

I want to apologize for being dismissive. Needless to say I
wasn't convinced by Sethares' gamelan analysis. I don't know
about the balafon research but I very much doubt they naturally
have 7-ET partials. There are more compelling reasons 7-ET
might be chosen, and it's possible balafon optimize their
instruments for it, though I am skeptical even of this version.
-Carl

--- In tuning@yahoogroups.com, "Carl Lumma" <carl@...> wrote:
>
> --- In tuning@yahoogroups.com, "jlmoriart" <JlMoriart@> wrote:
>
> > If I remember correctly, Sethares's analysis of the gamelan
> > spectra crossed with harmonic spectra (which do exist in
> > gamelan ensembles in the voice and sometimes string
> > instruments) led to sensory dissonance minima (i.e. minimal
> > beating) in both the 5-tone (approximately) equal and pelog
> > scales used in gamelan ensembles.
>
> Also, I have a beautiful oceanfront timeshare in Arizona
> you might like!
>
> > The 7-tone (approximately) equal balafon (wooden xylophones)
> > tunings of Africa also line up with sensory dissonance minima
> > using their spectra, this time without having to be crossed
> > with harmonic spectra.
>
> Timeshares are a cost-effective way to vacation!
>
> -Carl

Typo: It's possible balafon __makers__ optimize their
instruments (timbre for 7-ET to reduce beating) but I'm
skeptical even of this. -Carl

I wrote:

> I want to apologize for being dismissive. Needless to say I
> wasn't convinced by Sethares' gamelan analysis. I don't know
> about the balafon research but I very much doubt they naturally
> have 7-ET partials. There are more compelling reasons 7-ET
> might be chosen, and it's possible balafon optimize their
> instruments for it, though I am skeptical even of this version.
> -Carl

"Carl Lumma" <carl@...> wrote:
> I want to apologize for being dismissive. Needless to say
> I wasn't convinced by Sethares' gamelan analysis. I don't
> know about the balafon research but I very much doubt
> they naturally have 7-ET partials. There are more
> compelling reasons 7-ET might be chosen, and it's
> possible balafon optimize their instruments for it,
> though I am skeptical even of this version. -Carl

Right, the gamelan analysis (in Tuning, Timbre, Spectrum,
Scale) isn't entirely convincing because he got to choose
which instruments he measured the spectra of -- as well as
it only being a small sampling of gamelan orchestras. But
it's over 10 years now and I haven't seen an opposing
analysis. The working hypothesis has to be that tuning and
timbre are correlated in gamelans.

I don't know about balafons. I didn't know there was an
analysis of balafons. Does anybody have a citation?

The Thai analysis is in the second edition of Tuning,
Timbre, Spectrum, Scale. I happen to have this because one
of the colleges I worked at had a subscription to Springer
online.

Figure 15.1 in page 305 (page 3 of the PDF) shows a
spectrum. It's for a pong lang, "a wooden xylophone-like
instrument from Northeast Thailand". There are two
unmistakable peaks at 436 and 2393 Hz. The interval between
them is around 5.48:1 or 548:100.

Here are some temperaments you get by adding 548/100 to the
5-limit spectrum:

http://x31eq.com/cgi-bin/pregular.cgi?limit=2.3.5.548%2F100&error=5

The first equal temperament listed -- meaning the one with
lowest badness -- has 7 steps to the octave. The first
four rank 2 temperaments also involve 7, so it's something
of a standout. The weightings are arbitrary but that's
good -- it shows the result is stable over different
weightings. Sethares used a different method to find
7-EDO, using his spectral dissonance function.

The choice of the 5-limit as the harmonic basis was
arbitrary. Maybe the harmonic timbres are only important
up to the 3-limit. No problem! 7-equal is still optimal:

http://x31eq.com/cgi-bin/pregular.cgi?limit=2.3.548%2F100&error=5

The target of 5 cents error is an underestimate. 7-equal
comes out with an adjusted error of over 15 cents in the
augmented 5-limit. Choose a badness that favors it, and
7-equal completely dominates:

http://x31eq.com/cgi-bin/pregular.cgi?limit=2.3.5.548%2F100&error=15

But, hey, we don't have to do that. This mixed timbre
really does point strongly towards 7-equal.

(You could choose the target error in a more scientific
way, by measuring how much energy is outside the peaks of
the spectral plot. I didn't.)

In an ideal bar, the added partial comes out as 5.40/1
instead of 5.48/1.

http://x31eq.com/cgi-bin/pregular.cgi?limit=2.3.5.54%2F10&error=5

Oh dear, now 7 doesn't dominate so much, and Porcupine is
no longer the optimal linear temperament for Thai music.
Maybe the timbre of this pong lang was optimized for
7-equal. The important thing is that it didn't have to
be. Music involving near-ideal bars will naturally
gravitate towards near-equal 7 note scales.

There are serpents in this Eden. Firstly, if an instrument
like a Javanese gambang strongly favors 7-equal, why are
Javanese scales so unlike 7-equal? At 10 cents adjusted
error, at least 5 and 9 make the shortlist:

http://x31eq.com/cgi-bin/pregular.cgi?limit=2.3.5.54%2F10&error=5

The next strongest partial of the pong lang is at 1246 Hz.
It makes a ratio of 2.85:1 with the fundamental. Add that
to the mixed timbre and you get:

http://x31eq.com/cgi-bin/pregular.cgi?limit=2.3.5.548%2F100.285%2F100&error=5

Oh dear! 7-equal is no longer listed. This partial breaks
the pattern. Now we do have to choose the weights more
carefully. The peak at 1246 Hz is much weaker than that
at 2393 Hz, but 2.85:1 gets undue weight as a small
interval. Let's push it up a couple of octaves to get
285/25:

http://x31eq.com/cgi-bin/pregular.cgi?limit=2.3.5.548%2F100.285%2F100&error=5

7-equal is now second, and Porcupine (for what difference
that makes) is top of the rank 2 list again. Because 7 is
simpler than the other equal temperaments in the list, we
can push it back to first place:

http://x31eq.com/cgi-bin/pregular.cgi?limit=2.3.5.548%2F100.285%2F100&error=10

Including this partial, 7-equal dominates more for the
ideal bar than the measured pong lang timbre:

http://x31eq.com/cgi-bin/pregular.cgi?limit=2.3.5.54%2F10.276%2F25&error=5

The analysis in the book uses the ideal bar. So, there's a
certain amount of cherry-picking going on. Despite that,
a range of different choices will give qualitatively
similar results. And, yes, there are other reasons for
using 7-EDO, like there are reasons besides 5-limit harmony
for using a fifth-generated diatonic. We still have
another example of a reasonably good correlation between
tuning and timbre.

Graham

Graham wrote:

> Right, the gamelan analysis (in Tuning, Timbre, Spectrum,
> Scale) isn't entirely convincing because he got to choose
> which instruments he measured the spectra of -- as well as
> it only being a small sampling of gamelan orchestras. But
> it's over 10 years now and I haven't seen an opposing
> analysis. The working hypothesis has to be that tuning and
> timbre are correlated in gamelans.

I'll have to re-read it. But as I recall, the analysis
was based only on intervals measured up from the tonic
(like most of the scales in the book). In other words,
most of the intervals in the scale aren't included in the
analysis.

Also, according to Paul, the algorithm gives wildly
different minima when the overall amplitude of the inputs
is changed (even if their relative amplitudes are
the same!). If so, that casts doubt on the specificity
of any analysis based on the algorithm.

Finally, as you mention, these cultures all have ensembles
of different instruments with wildly different spectra.
Not that all the instruments in these ensembles are tuned
identically, but as far as things like the number of
notes/octave which they typicall share, I think we need to
be looking for other explanations.

> Figure 15.1 in page 305 (page 3 of the PDF) shows a
> spectrum. It's for a pong lang, "a wooden xylophone-like
> instrument from Northeast Thailand". There are two
> unmistakable peaks at 436 and 2393 Hz. The interval between
> them is around 5.48:1 or 548:100.
> Here are some temperaments you get by adding 548/100 to the
> 5-limit spectrum:
> http://x31eq.com/cgi-bin/pregular.cgi?limit=2.3.5.548%2F100&error=5
> The first equal temperament listed -- meaning the one with
> lowest badness -- has 7 steps to the octave. The first
> four rank 2 temperaments also involve 7, so it's something
> of a standout. The weightings are arbitrary but that's
> good -- it shows the result is stable over different
> weightings. Sethares used a different method to find
> 7-EDO, using his spectral dissonance function.

It's much more likely the bars were shaved to meet the
tuning, rather than the tuning being chosen on account of
the bars.

> Maybe the timbre of this pong lang was optimized for
> 7-equal.
> The important thing is that it didn't have to
> be. Music involving near-ideal bars will naturally
> gravitate towards near-equal 7 note scales.

Or maybe the fact that Thai music modulates a lot is
important.

> The analysis in the book uses the ideal bar. So, there's a
> certain amount of cherry-picking going on. Despite that,
> a range of different choices will give qualitatively
> similar results. And, yes, there are other reasons for
> using 7-EDO, like there are reasons besides 5-limit harmony
> for using a fifth-generated diatonic. We still have
> another example of a reasonably good correlation between
> tuning and timbre.

I'm late for church but after working with this stuff for
many years the conclusion that spectral dissonance has
essentially no explanatory power in music theory is almost
inescapable. The best way of arguing this point at the
moment is unfortunately escaping much better.

-Carl

On Sun, May 15, 2011 at 1:52 PM, Carl Lumma <carl@...> wrote:
>
> Also, according to Paul, the algorithm gives wildly
> different minima when the overall amplitude of the inputs
> is changed (even if their relative amplitudes are
> the same!). If so, that casts doubt on the specificity
> of any analysis based on the algorithm.

Are you talking about his dissonance algorithm in general? Do you know
where I could find an implementation of such an algorithm? I've had my
eye on revisiting Sethares for quite a while now.

> I'm late for church but after working with this stuff for
> many years the conclusion that spectral dissonance has
> essentially no explanatory power in music theory is almost
> inescapable.

I think that is far too strong a claim. Timbre has a very profound
effect on everything you do. Picking a suitable timbre can make
something like father[8] sound amazing, whereas the GM reed organ
patch will make it sound terrible. If I play stuff with sine waves,
almost anything sounds good - local maxima of harmonic entropy in fact
sound really good, as they sound like they're blending the intervals
in the surrounding field of attraction together.

Igs has described the 522 cent interval as sounding like a wobbly and
drunken fourth, which has a unique character all its own, and I'd
agree. That is my qualitative experience of an interval being in the
field of attraction of 4/3 (a "fourth") and being of higher entropy
("wobbly and drunken"). You can use it for great artistic effect, and
I'd like to explore more of it. But you'd better be careful what
timbre you use with an interval like that, because if you pick one
that's too harsh it'll make you hate everything. I have always thought
Igs has applied this concept masterfully, and Knowsur's 14-tet album I
think also did a good job with it.

-Mike

We have to be careful to not assume that the Western ideal of minimizing
beating isn't assumed to be universal for all cultures. In the gamelan
tradition, beating is actually a desirable trait, as are complex, impure
intervals. Although this may have its origins in inharmonic spectra, we
shouldn't necessarily assume that there should be a tight match between
scales and gamelan instrument's spectra.

For gamelan musicians, beating is associated with liveliness and awareness,
as well as conveying a rhythmic pulse within the texture.

AKJ
On May 14, 2011 12:34 PM, "jlmoriart" <JlMoriart@...> wrote:
> Haven't been following this whole thread, but...
>
>> The question about non Western music like Indonesia and Thai and African
is
>> interesting--- here I think it's a matter of cultural preference for
beating
>> and a certain enjoyment of spicy sourness or complexity in the intervals:
in
>> Gamelan music for instance, the beating, even of unisons, was a way of
>> producing divine altered states of awareness.
>
> If I remember correctly, Sethares's analysis of the gamelan spectra
crossed with harmonic spectra (which do exist in gamelan ensembles in the
voice and sometimes string instruments) led to sensory dissonance minima
(i.e. minimal beating) in both the 5-tone (approximately) equal and pelog
scales used in gamelan ensembles.
>
> The 7-tone (approximately) equal balafon (wooden xylophones) tunings of
Africa also line up with sensory dissonance minima using their spectra, this
time without having to be crossed with harmonic spectra.
>
> The preference for nonwestern intervals still seems to be motivated by a
lack of beating, the lack of beating just ending up at different intervals
given different spectra.
>
> John M
>
>
>
> ------------------------------------
>
> You can configure your subscription by sending an empty email to one
> of these addresses (from the address at which you receive the list):
> tuning-subscribe@yahoogroups.com - join the tuning group.
> tuning-unsubscribe@yahoogroups.com - leave the group.
> tuning-nomail@yahoogroups.com - turn off mail from the group.
> tuning-digest@yahoogroups.com - set group to send daily digests.
> tuning-normal@yahoogroups.com - set group to send individual emails.
> tuning-help@yahoogroups.com - receive general help information.
> Yahoo! Groups Links
>
>
>

"Carl Lumma" <carl@...> wrote:
> Graham wrote:
>
> > Right, the gamelan analysis (in Tuning, Timbre,
> > Spectrum, Scale) isn't entirely convincing because he
> > got to choose which instruments he measured the spectra
> > of -- as well as it only being a small sampling of
> > gamelan orchestras. But it's over 10 years now and I
> > haven't seen an opposing analysis. The working
> > hypothesis has to be that tuning and timbre are
> > correlated in gamelans.
>
> I'll have to re-read it. But as I recall, the analysis
> was based only on intervals measured up from the tonic
> (like most of the scales in the book). In other words,
> most of the intervals in the scale aren't included in the
> analysis.

Do an analysis that includes all the intervals, then. I
don't think it makes a difference. Good matches to equal
temperaments (9-equal in this case) don't depend on the
choice of tonic.

> Also, according to Paul, the algorithm gives wildly
> different minima when the overall amplitude of the inputs
> is changed (even if their relative amplitudes are
> the same!). If so, that casts doubt on the specificity
> of any analysis based on the algorithm.

Why does it cast doubt? That's entirely consistent with
human hearing. And if you don't like it you can do your
own analysis. It shouldn't depend on artifacts of the
dissonance curves. From what I remember, it doesn't.

> Finally, as you mention, these cultures all have ensembles
> of different instruments with wildly different spectra.
> Not that all the instruments in these ensembles are tuned
> identically, but as far as things like the number of
> notes/octave which they typicall share, I think we need to
> be looking for other explanations.

Yes, they have different spectra. When somebody measures
them, we can talk about them. The evidence we have now
points to a tuning/timbre connection.

> It's much more likely the bars were shaved to meet the
> tuning, rather than the tuning being chosen on account of
> the bars.

Probably it is. Why does it matter?

> > Maybe the timbre of this pong lang was optimized for
> > 7-equal.
> > The important thing is that it didn't have to
> > be. Music involving near-ideal bars will naturally
> > gravitate towards near-equal 7 note scales.
>
> Or maybe the fact that Thai music modulates a lot is
> important.

Of course it's important. Why do you have to state the
obvious and pretend it's a contradiction? Maybe for your
next trick you can find some evidence for modulation in
European music and ponder whether there's any connection
between 12-equal and harmonic timbres.

> > The analysis in the book uses the ideal bar. So,
> > there's a certain amount of cherry-picking going on.
> > Despite that, a range of different choices will give
> > qualitatively similar results. And, yes, there are
> > other reasons for using 7-EDO, like there are reasons
> > besides 5-limit harmony for using a fifth-generated
> > diatonic. We still have another example of a
> > reasonably good correlation between tuning and timbre.
>
> I'm late for church but after working with this stuff for
> many years the conclusion that spectral dissonance has
> essentially no explanatory power in music theory is almost
> inescapable. The best way of arguing this point at the
> moment is unfortunately escaping much better.

We're waiting for your evidence. It's been over 24 hours
now.

Graham

Aaron Krister Johnson <aaron@...> wrote:
> We have to be careful to not assume that the Western
> ideal of minimizing beating isn't assumed to be universal
> for all cultures. In the gamelan tradition, beating is
> actually a desirable trait, as are complex, impure
> intervals. Although this may have its origins in
> inharmonic spectra, we shouldn't necessarily assume that
> there should be a tight match between scales and gamelan
> instrument's spectra.

We don't have to assume anything. We have evidence.

Temperament at the level we're talking about isn't about
minimizing beating, is it? You need the consonances to be
almost pure to get audible beats, and here we're talking
about 7- or 9-equal being used with ensembles involving
harmonic timbres. The beats come from mistuned unisons
between pairs of instruments.

Graham

Mike wrote:

> Are you talking about his dissonance algorithm in general?
> Do you know where I could find an implementation of such an
> algorithm? I've had my eye on revisiting Sethares for quite
> a while now.

Yes. He has the Matlab on his website.

> > I'm late for church but after working with this stuff for
> > many years the conclusion that spectral dissonance has
> > essentially no explanatory power in music theory is almost
> > inescapable.
>
> I think that is far too strong a claim. Picking a suitable
> timbre can make something like father[8] sound amazing,
> whereas the GM reed organ patch will make it sound terrible.

Timbre choice isn't something music theory usually deals
with. It's the art of arrangers, conductors, producers,
and sound engineers. There's no hint that Sethares' model
can help with it either. Instead this model has been put
forward as a way to explain scale choice given the timbre.

This is a non-starter if you think about it. The cultures
with the inharmonic timbres and scales that might be based
on them don't use harmony as we know it. They have rapidly
moving lines too fast for beating to be noticeable.
Harmony, where it occurs at all, typically occurs between
instruments in an ensemble, often having very different
spectra. Indeed the metallophones play right along with
wind instruments and human voices that have perfectly
harmonic spectra, with no problems.

In Indonesia, there's said to be considerable variation in
tuning between ensembles or villages. Right off the bat
this suggests the music isn't terribly intonation-sensitive,
or if it is, that it's a matter of local style with about
as much deep significance as the features on this map:
http://bit.ly/iRVZyD

Guitarists will go on and on about whether mahogany or
rosewood sounds better in a solid-body electric guitar.
Violin buffs go on and on about the magic lacquers of
Stradivarius. Wine critics, audiophiles, and so on. It's
a general phenomenon so when somebody tells you the master
gamelan makers are minimizing Plomp-Levelt roughness in a
way that Western piano builders don't even bother with,
you should hold on to your wallet.

-Carl

--- Graham Breed <gbreed@...> wrote:

> Do an analysis that includes all the intervals, then. I
> don't think it makes a difference.

If you constrain yourself to ETs, it obviously doesn't.
Otherwise- of course it does!!

> > Also, according to Paul, the algorithm gives wildly
> > different minima when the overall amplitude of the inputs
> > is changed (even if their relative amplitudes are
> > the same!). If so, that casts doubt on the specificity
> > of any analysis based on the algorithm.
>
> Why does it cast doubt? That's entirely consistent with
> human hearing. And if you don't like it you can do your
> own analysis. It shouldn't depend on artifacts of the
> dissonance curves. From what I remember, it doesn't.

So they use different scales when they play louder?

> > I'm late for church but after working with this stuff for
> > many years the conclusion that spectral dissonance has
> > essentially no explanatory power in music theory is almost
> > inescapable. The best way of arguing this point at the
> > moment is unfortunately escaping much better.
>
> We're waiting for your evidence. It's been over 24 hours
> now.

It's been less than 5 hours since I wrote what you're
replying to here, in case that's important.

-Carl

I got a little more feedback on the "first flutes found" issue:

Old men get grumpy, and think about dyin a lot. Besides, my memory's shot
and I don't remember nuthin about how my brother fingered that bone flute
back when we were kids, before he lost it in that stupid cave.

...
On Sun, May 15, 2011 at 9:14 PM, Peter H. Kosel <ph_kosel@...> wrote:

> I believe it's "hohle fels", not "hohlefels":
> http://www.google.com/search?hl=en&q=hohle+fels+flute
>

I probably should have written Hohle Fels, since that's more common than
Hohlefels. Articles claim these to be the first man-made instruments found (
http://en.wikipedia.org/wiki/Hohle_Fels).

I think this one is older:
> Neanderthal flute
> http://www.greenwych.ca/fl-compl.htm
>
> Now you know as much as I do.
>

Oh, I didn't know of this! That Fink article is interesting. I wouldn't have
thought much of a musical analysis could be made from one piece of one
flute, but he does a good job. Apparently there is controversy over whether
the holes are man-made and whether the piece is intended to be played as a
flute (http://en.wikipedia.org/wiki/Divje_Babe_flute#Neanderthal_flute).
Thank you for the info and for the information on your site.

"Carl Lumma" <carl@...> wrote:
> --- Graham Breed <gbreed@...> wrote:
>
> > Do an analysis that includes all the intervals, then. I
> > don't think it makes a difference.
>
> If you constrain yourself to ETs, it obviously doesn't.
> Otherwise- of course it does!!

Of the three scales we're talking about, two are
constrained to ETs: Slendro is 5-equal and the Thai scale
is 7-equal. Pelog is different, but because my software's
set up for regular temperaments, I'm analyzing it in terms
of regular temperaments, and the best correlation is with
9-equal. Sethares does talk about the fine structure, so
if you have a better analysis, let's see it.

> > > Also, according to Paul, the algorithm gives wildly
> > > different minima when the overall amplitude of the
> > > inputs is changed (even if their relative amplitudes
> > > are the same!). If so, that casts doubt on the
> > > specificity of any analysis based on the algorithm.
> >
> > Why does it cast doubt? That's entirely consistent with
> > human hearing. And if you don't like it you can do your
> > own analysis. It shouldn't depend on artifacts of the
> > dissonance curves. From what I remember, it doesn't.
>
> So they use different scales when they play louder?

I don't know. Should they?

> It's been less than 5 hours since I wrote what you're
> replying to here, in case that's important.

I haven't accused anybody of fraud, and within 9 hours of
your apology for doing so I provided a lot of analysis.

Graham

"Carl Lumma" <carl@...> wrote:

> > > I'm late for church but after working with this stuff
> > > for many years the conclusion that spectral
> > > dissonance has essentially no explanatory power in
> > > music theory is almost inescapable.
<snip>
> Timbre choice isn't something music theory usually deals
> with. It's the art of arrangers, conductors, producers,
> and sound engineers. There's no hint that Sethares' model
> can help with it either. Instead this model has been put
> forward as a way to explain scale choice given the timbre.

So timbre has no explanatory power in music theory because
music theory's defined such that timbre's abstracted out.
Suddenly that claim sounds a lot weaker.

> This is a non-starter if you think about it. The cultures
> with the inharmonic timbres and scales that might be based
> on them don't use harmony as we know it. They have
> rapidly moving lines too fast for beating to be
> noticeable. Harmony, where it occurs at all, typically
> occurs between instruments in an ensemble, often having
> very different spectra. Indeed the metallophones play
> right along with wind instruments and human voices that
> have perfectly harmonic spectra, with no problems.

Right, they don't use harmony as we know it. They use
harmony as they know it. I'm glad that one's out of the
way.

The evidence says that Thai xylophones work together with
harmonic timbres to give 7 roughly equal steps to the
octave. Bill Sethares specifically analyzed a pong lang,
apparently because it plays a solo on a record he has. I
think it's reasonable that an instrument that solos at the
start of a piece of music helps to determine the harmonic
structure. Does harmony, as Thais understand it,
contradict that?

The evidence is that the bonang timbre correlates with
5-equal (and therefore slendro) and the saron timbre
correlates with pelog. Is that consistent with harmony as
Javanese understand it? I don't know.

I plugged in a gender timbre, and get a reasonable
correlation with 9-equal. That suggests it'll work with
pelog. I haven't checked to see if it gets any closer
than that. 5-equal comes up in the shortlist, so maybe it
isn't horribly out of tune in slendro either. This is a
420:845:1079:1701:2016 timbre with a target error of 10
cents (adjusted).

> In Indonesia, there's said to be considerable variation in
> tuning between ensembles or villages. Right off the bat
> this suggests the music isn't terribly
> intonation-sensitive, or if it is, that it's a matter of
> local style with about as much deep significance as the
> features on this map: http://bit.ly/iRVZyD

The evidence we have suggests that the variation in tuning
tracks the variation in timbre. That only comes from
measuring two gamelans, so it doesn't prove a general
rule. If you have more evidence, let's see it.

There was a big variation in keyboard tunings in Europe in
the 18th Century. There wasn't a corresponding variation
in timbre that I'm aware of. Does that establish that
common practice music isn't terribly intonation-sensitive?

> Guitarists will go on and on about whether mahogany or
> rosewood sounds better in a solid-body electric guitar.
> Violin buffs go on and on about the magic lacquers of
> Stradivarius. Wine critics, audiophiles, and so on. It's
> a general phenomenon so when somebody tells you the master
> gamelan makers are minimizing Plomp-Levelt roughness in a
> way that Western piano builders don't even bother with,
> you should hold on to your wallet.

So what you do is look at the evidence.

Graham

On Sun, May 15, 2011 at 6:43 PM, Carl Lumma <carl@...> wrote:
>
> Mike wrote:
>
> > Are you talking about his dissonance algorithm in general?
> > Do you know where I could find an implementation of such an
> > algorithm? I've had my eye on revisiting Sethares for quite
> > a while now.
>
> Yes. He has the Matlab on his website.

Alright, thanks. And the criticism is that the amplitude input of the
notes leads to wildly different curves? Does he have it like that
intentionally?

> > I think that is far too strong a claim. Picking a suitable
> > timbre can make something like father[8] sound amazing,
> > whereas the GM reed organ patch will make it sound terrible.
>
> Timbre choice isn't something music theory usually deals
> with. It's the art of arrangers, conductors, producers,
> and sound engineers. There's no hint that Sethares' model
> can help with it either. Instead this model has been put
> forward as a way to explain scale choice given the timbre.

Timbre choice hasn't factored into harmonic music theory because we
haven't had to deal with it thus far. When you're playing in 12-equal,
the error of each successive prime is inversely correlated with that
prime's energy in the kind of timbres we play around with. So for a
dominant 7 chord, the prominent 3rd partial will experience almost no
beating, the 5th will beat but be much softer, the 7th will beat but
be even softer, etc. This pattern continues for the 11th and 13th
harmonics as well.

A decently similar result applies to meantone tunings, except for some
of those 5/1 is more accurate than 3/1. However, meantones still have
exceptionally decent error for 5/4 and 3/2 in general, so the general
trend is true enough anyway; we never had to think much about timbre,
because the most prominent partials (and hence the most likely to
beat) where spectral roughness might become really apparently were
also the most in tune.

The end result is that we have the notion ingrained that we can pretty
much play any chord on any instrument and not have to think about
timbre as an actual part of music theory. We also don't think much
about chord voicings, whereas for a tuning like 15-equal it's
noteworthy that 3/2 goes over way better than 3/1.

All of the above is a lucky exception that doesn't apply to many
tunings. It doesn't apply to tunings where there's serious spectral
dissonance in the third (or maybe fifth) harmonic(s), which is when
the consideration of timbre starts to play a larger role. What will
happen is that one of us on the list will screw around with 15-equal
long enough to figure out how to satisfactorily "hide" the error of
the 3/2, with vibrato or clever choice of timbre or perhaps voicing
the chords a certain way, or maybe by using chorusy gamelan-esque
timbres, and subsequent generations will figure out whatever we did
and just teach that. If we were all playing in 15-equal or 16-equal,
orchestrators might teach that you don't want to give fifths to brass
instruments, or to always play them with vibrato, etc. Piano-makers
might adjust the timbres of their pianos to hide the 3rd partial.

> This is a non-starter if you think about it. The cultures
> with the inharmonic timbres and scales that might be based
> on them don't use harmony as we know it. They have rapidly
> moving lines too fast for beating to be noticeable.
> Harmony, where it occurs at all, typically occurs between
> instruments in an ensemble, often having very different
> spectra. Indeed the metallophones play right along with
> wind instruments and human voices that have perfectly
> harmonic spectra, with no problems.

I wasn't really talking about inharmonic spectra. One idea I'd
suggested in the past is to design timbres in which the partials that
are most likely to beat have their amplitudes lowered to be just under
the masking curve of the nearest match in the tuning. So in 15-equal,
we'd use timbres where the 3/1 gets attenuated to fit under the
masking curve at 1920 cents. If you're willing to do this
electronically, you could make this attenuation only happen when the
relevant notes are played, and at full volume otherwise.

> In Indonesia, there's said to be considerable variation in
> tuning between ensembles or villages. Right off the bat
> this suggests the music isn't terribly intonation-sensitive,
> or if it is, that it's a matter of local style with about
> as much deep significance as the features on this map:
> http://bit.ly/iRVZyD

I am rather distraught that most of America refers to soda as "pop."

> Guitarists will go on and on about whether mahogany or
> rosewood sounds better in a solid-body electric guitar.
> Violin buffs go on and on about the magic lacquers of
> Stradivarius. Wine critics, audiophiles, and so on. It's
> a general phenomenon so when somebody tells you the master
> gamelan makers are minimizing Plomp-Levelt roughness in a
> way that Western piano builders don't even bother with,
> you should hold on to your wallet.

Gamelan tunings require more minimizing of roughness than Western
piano builders do, which is a point I hope I've made clear by my above
analysis. As for guitarists who claim that their 1950s vintage Martin
acoustic has an amazing tone just because it's vintage, I tend to
agree. Likewise with audiophiles who claim to be able to distinguish
320K MP3s from CD quality, or 48KHz sampled audio from analog.

I will note that there is a lot that goes into making a guitar that
does matter, and that the hardness of the wood can play a role in the
timbre by changing the compliance of the soundboard and resonant
cavity in general, although bracing has a lot to do with it as well.
I'm fortunate to have a good friend that's gone into luthiering where
he's quickly discovering for himself what aspects of guitar craft are
actually rooted in reality and which aspects are pseudoscientific
bullshit.

I'd also say that Stradavarius violins don't fit into the same
category either, as the timbre is different, and noticeably so. I got
to hear some sound samples and an acoustical analysis at the ASA
convention a few years ago - they really sounded amazing and quite
different. My initial reaction was that there was some boosting around
the 3-7th harmonics in such a way that some kind of "formant"-like
effect was produced, but I couldn't tell you what vowel it was. I
remember being frustrated that they didn't put that hypothesis forward
and decided instead to focus on the lacquers, as you said, and that
they did some bullshit analysis that had to do with the phase response
of the resonant cavity.

-Mike

--- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
>
> On Sun, May 15, 2011 at 6:43 PM, Carl Lumma <carl@...> wrote:
> > Timbre choice isn't something music theory usually deals
> > with. It's the art of arrangers, conductors, producers,
> > and sound engineers. There's no hint that Sethares' model
> > can help with it either. Instead this model has been put
> > forward as a way to explain scale choice given the timbre.
>
> Timbre choice hasn't factored into harmonic music theory because we
> haven't had to deal with it thus far. When you're playing in 12-equal,
> the error of each successive prime is inversely correlated with that
> prime's energy in the kind of timbres we play around with. So for a
> dominant 7 chord, the prominent 3rd partial will experience almost no
> beating, the 5th will beat but be much softer, the 7th will beat but
> be even softer, etc. This pattern continues for the 11th and 13th
> harmonics as well.

But we have dealt with timbre choice, compare how 12-equal sounds
with pianos and with harpsichords. The instruments and the tunings
have co-evolved. I also suspect lot of the conventions of synth
programming are there to hide the shortcomings of 12-equal.

Kalle

On Mon, May 16, 2011 at 6:36 AM, Graham Breed <gbreed@...> wrote:
>
> So timbre has no explanatory power in music theory because
> music theory's defined such that timbre's abstracted out.
> Suddenly that claim sounds a lot weaker.

I'd say the biggest resistance to the adoption of tunings like
15-equal and 16-equal, in fact, is probably that they're incompatible
with the abstracting out of timbre that you've listed above. Suddenly,
fifths sound way better if they're played with flutes than with
harpsichords, and much better if they're voiced as 3/2 than 3/1, which
is a challenge we've never had to deal with. I would imagine that the
solution for symphonic music will require something like

1) more finesse on the theory side of things, particularly with
timbral choice and chord voicings and the other stuff I laid out my
last response to Carl
2) alterations of the timbres that we use in orchestras; e.g. put more
vibrato on the violins, get hammers and plectra to attenuate the 3rd
partial for harpsichords and pianos, change the peak resonances of
brass instruments, maybe downplay certain instruments' roles entirely
3) probably other stuff we're not thinking about

Once people work that sort of thing out, there's not much to stop them
from playing harmonic music in 15- or 16-equal (or maybe 13 or
14-equal too) and having it sound great. The fact that music in
15-equal sounds great with sine waves, or with Rhodes or something,
but often sounds terrible with harsher timbres indicates just how
important spectral dissonance is in this regard.

> The evidence says that Thai xylophones work together with
> harmonic timbres to give 7 roughly equal steps to the
> octave. Bill Sethares specifically analyzed a pong lang,
> apparently because it plays a solo on a record he has. I
> think it's reasonable that an instrument that solos at the
> start of a piece of music helps to determine the harmonic
> structure. Does harmony, as Thais understand it,
> contradict that?

How exactly does he work that out? Does he take the Thai xylophone
timbre and a harmonic series, add the results together, run it through
his dissonance algorithm, and find out where the minima are? I don't
have TTSS, and can't drop $80 on it now.

-Mike

Mike Battaglia <battaglia01@...> wrote:

> How exactly does he work that out? Does he take the Thai
> xylophone timbre and a harmonic series, add the results
> together, run it through his dissonance algorithm, and
> find out where the minima are? I don't have TTSS, and
> can't drop $80 on it now.

I did my own calculation in this thread, that you can check.

What Bill did, here and in with the gamelan timbres, is to
take the inharmonic timbre and a representative harmonic
timbre, and calculate the dissonance between them. So you
choose the interval, add the two spectra together, and
calculate the dissonance. Then vary the interval, and
calculate the dissonance as a function of it.

I don't know which timbre was held constant, and which
varies. I think it matters, in that there's some symmetry,
but it doesn't go that far.

Graham

--- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...> wrote:
> I also suspect lot of the conventions of synth
> programming are there to hide the shortcomings of 12-equal.

LOL. I've heard this hypothesis before and I think it's hilarious. If you're talking about the sort of "motion" elements of a lot of synth patches (vibratos, LFOs, filter sweeps, etc.), they're there to make up for the fact that the basic synth timbres are static and people hate static sounds. Acoustic instruments produce sounds that morph and evolve in organic ways that synthesizers are only just now beginning to decently replicate. And in any case, if it's the beating that people hate, well, fast LFO pitch modulation basically IS beating, even more strongly than one gets when hearing 12-TET with static synth timbres. So you're saying that synth programming makes up for the "shortcomings" of 12-TET by making them worse. That's ridiculous. Most people I've talked to IRL seem to hate the sound of JI with a static synth timbre much more than they hate the same sound playing 12-TET; I get comments like "it sounds cold" or "grating" or "like an emergency broadcast system"--some couldn't even tell a chord was being played!--and that the 12-TET sounds "warmer" and "more organic". People like organic complex lively evolving sounds. That's all there is to it. Accurate temperaments like 19 and 22 might sound even better than 12 with these timbres, but mostly because they introduce slower beating patterns that actually add to the complexity of the sound, rather than eliminating it. JI definitely sounds the worst.

-Igs

On Mon, May 16, 2011 at 10:59 AM, cityoftheasleep
<igliashon@...>wrote:

> --- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...> wrote:
> > I also suspect lot of the conventions of synth
> > programming are there to hide the shortcomings of 12-equal.
> JI definitely sounds the worst.
>
>
I think this is too absolute a statement. I agree that it does in certain
situtations, but OTOH, I'm always amazed at how great it sounds on most
registers of the piano, for example. There's something definitely
"spiritual", for lack of a better word, about hearing pure harmonies in the
human voice, on a piano, in French horns......and again, it also depends on
what *kind* of JI we are talking about....5 limit almost never sounds bad,
but it often is rather limiting outside of really basic ethnic-folk like
situations.

Speaking of ethnic folk, rational Greek tetrachord scales on for example a
Kanun, they always sounds great to me, don't they to you?

That said, I agree that for the most part, people prefer motion in their
tunings and timbres (if it's not TOO MUCH) to stasis.

AKJ

-Igs
>
>
>
> ------------------------------------
>
> You can configure your subscription by sending an empty email to one
> of these addresses (from the address at which you receive the list):
> tuning-subscribe@yahoogroups.com - join the tuning group.
> tuning-unsubscribe@yahoogroups.com - leave the group.
> tuning-nomail@yahoogroups.com - turn off mail from the group.
> tuning-digest@yahoogroups.com - set group to send daily digests.
> tuning-normal@yahoogroups.com - set group to send individual emails.
> tuning-help@yahoogroups.com - receive general help information.
> Yahoo! Groups Links
>
>
>
>

--
Aaron Krister Johnson
http://www.akjmusic.com
http://www.untwelve.org

--- In tuning@yahoogroups.com, Aaron Krister Johnson <aaron@...> wrote:

> I think this is too absolute a statement. I agree that it does in certain
> situtations, but OTOH, I'm always amazed at how great it sounds on most
> registers of the piano, for example.

I was talking *ONLY* about static synthetic timbres, like a pure square or saw wave. One of the reasons JI sounds good on acoustic instruments is because of the imperfections/uncertainties/"organic impurities" in the partials--or if you don't buy that, at least they have envelopes to them.

> Speaking of ethnic folk, rational Greek tetrachord scales on for example a
> Kanun, they always sounds great to me, don't they to you?

Sure. How do you like them with sustained sawtooth waves?

-Igs

--- Mike Battaglia <battaglia01@...> wrote:

> Alright, thanks. And the criticism is that the amplitude
> input of the notes leads to wildly different curves? Does he
> have it like that intentionally?

He has never responded to this criticism.

> Gamelan tunings require more minimizing of roughness than
> Western piano builders do, which is a point I hope I've made
> clear by my above analysis.

That's an absurd statement. There's no evidence anywhere
that they care about minimizing beating. The easiest way
to do so of course is to design a proper resonator that
turns your metallophone into a sine tone generator. The
American gamelan folks did this. I know because I helped
tune part of the Other Music gamelan.

> As for guitarists who claim that their 1950s vintage Martin
> acoustic has an amazing tone just because it's vintage, I tend
> to agree. Likewise with audiophiles who claim to be able to
> distinguish 320K MP3s from CD quality,

The best study I know of showed they can distinguish them
slightly better than by chance, but couldn't correctly
identify them better than by chance. That was almost ten
years ago and encoders have improved a lot. With certain
example inputs, artifacts can still be heard. But the
consensus on the forums is that nobody has ever demonstrated
the ability to ABX high-quality mp3s from CD for the vast
majority of inputs.

I've posted on this phenomenon before so there's no need
for me to rehash it. Every time wine critics are tested,
they fail miserably. Just last week, claims that expensive
HDMI cables have better picture quality were debunked.

> I will note that there is a lot that goes into making a guitar
> that does matter, and that the hardness of the wood can play a
> role in the timbre by changing the compliance of the soundboard
> and resonant cavity in general,

There's no soundboard or resonant cavity in a solidbody
electric guitar. Wood choice can still be significant of
course, but claims that woods of similar hardness and density
can be distinguished by sound alone are highly suspect.

> I'd also say that Stradavarius violins don't fit into the same
> category either, as the timbre is different, and noticeably so.
> I got to hear some sound samples and an acoustical analysis at
> the ASA convention a few years ago - they really sounded
> amazing and quite different.

You want me to believe that a bunch of violins made 300 years
ago and having survived multiple owners, restringings, new
tuning pegs and fingerboards and in many cases complete
restoration, share a typical sound quality found in no other
violin made since?

Let's have a look at wikipedia shall we?

"However, the many blind tests from 1817 to the present (as
of 2000) have never found any difference in sound between
Stradivari's violins and high-quality violins in comparable
style of other makers and periods, nor has acoustic analysis.[5]
In a particularly famous test on a BBC Radio 3 program in 1977,
the violinists Isaac Stern and Pinchas Zukerman and the violin
expert and dealer Charles Beare tried to distinguish among the
"Chaconne" Stradivarius, a 1739 Guarneri del Gesú, an
1846 Vuillaume, and a 1976 British violin played behind a
screen by a professional soloist. The two violinists were
allowed to play all the instruments first. None of the listeners
identified more than two of the four instruments. Two of the
listeners identified the 20th-century violin as the
Stradivarius.[6][7] Listening tests as recent as 2006 have also
failed to show differences; violinists and others have
criticized these tests on various grounds such as that they are
not double-blind (in most cases), the judges are often
not experts, and the sounds of violins are hard to evaluate
objectively and reproducibly.[7]"

-Carl

On Mon, May 16, 2011 at 2:36 PM, Carl Lumma <carl@...> wrote:
>
> > Gamelan tunings require more minimizing of roughness than
> > Western piano builders do, which is a point I hope I've made
> > clear by my above analysis.
>
> That's an absurd statement. There's no evidence anywhere
> that they care about minimizing beating. The easiest way
> to do so of course is to design a proper resonator that
> turns your metallophone into a sine tone generator. The
> American gamelan folks did this. I know because I helped
> tune part of the Other Music gamelan.

That was worded poorly on my part, I meant that a tuning like mavila
would require more minimizing of roughness than a tuning like
meantone. So the fact that the world's greatest piano makers don't
care about it is something I'd expect would change if we all adopted
16-equal.

> > As for guitarists who claim that their 1950s vintage Martin
> > acoustic has an amazing tone just because it's vintage, I tend
> > to agree. Likewise with audiophiles who claim to be able to
> > distinguish 320K MP3s from CD quality,
>
> The best study I know of showed they can distinguish them
> slightly better than by chance, but couldn't correctly
> identify them better than by chance. That was almost ten
> years ago and encoders have improved a lot. With certain
> example inputs, artifacts can still be heard. But the
> consensus on the forums is that nobody has ever demonstrated
> the ability to ABX high-quality mp3s from CD for the vast
> majority of inputs.
>
> I've posted on this phenomenon before so there's no need
> for me to rehash it. Every time wine critics are tested,
> they fail miserably. Just last week, claims that expensive
> HDMI cables have better picture quality were debunked.

This was also worded poorly, I meant to say that I agree with you
saying that all of the above claims are bullshit.

> > I will note that there is a lot that goes into making a guitar
> > that does matter, and that the hardness of the wood can play a
> > role in the timbre by changing the compliance of the soundboard
> > and resonant cavity in general,
>
> There's no soundboard or resonant cavity in a solidbody
> electric guitar. Wood choice can still be significant of
> course, but claims that woods of similar hardness and density
> can be distinguished by sound alone are highly suspect.

This was once again worded poorly, now I'm 0/3. Was talking about
acoustics, my friend is exclusively an acoustic luthier, I don't know
shit about electric luthiering, really need to wake up more before
posting on the tuning list, etc.

> > I'd also say that Stradavarius violins don't fit into the same
> > category either, as the timbre is different, and noticeably so.
> > I got to hear some sound samples and an acoustical analysis at
> > the ASA convention a few years ago - they really sounded
> > amazing and quite different.
>
> You want me to believe that a bunch of violins made 300 years
> ago and having survived multiple owners, restringings, new
> tuning pegs and fingerboards and in many cases complete
> restoration, share a typical sound quality found in no other
> violin made since?

They certainly had a sound quality that was found in no other violin
I've heard in the last 300 years.

> Let's have a look at wikipedia shall we?
>
> "However, the many blind tests from 1817 to the present (as
> of 2000) have never found any difference in sound between
> Stradivari's violins and high-quality violins in comparable
> style of other makers and periods, nor has acoustic analysis.[5]
> In a particularly famous test on a BBC Radio 3 program in 1977,
> the violinists Isaac Stern and Pinchas Zukerman and the violin
> expert and dealer Charles Beare tried to distinguish among the
> "Chaconne" Stradivarius, a 1739 Guarneri del Gesú, an
> 1846 Vuillaume, and a 1976 British violin played behind a
> screen by a professional soloist. The two violinists were
> allowed to play all the instruments first. None of the listeners
> identified more than two of the four instruments. Two of the
> listeners identified the 20th-century violin as the
> Stradivarius.[6][7] Listening tests as recent as 2006 have also
> failed to show differences; violinists and others have
> criticized these tests on various grounds such as that they are
> not double-blind (in most cases), the judges are often
> not experts, and the sounds of violins are hard to evaluate
> objectively and reproducibly.[7]"

Alrighty then.

-Mike

--- Mike Battaglia <battaglia01@...> wrote:

> I meant that a tuning like mavila
> would require more minimizing of roughness than a tuning like
> meantone.

It would admit to more roughness minimization, yes.

> This was also worded poorly, I meant to say that I agree
> with you saying that all of the above claims are bullshit.

Ah. :)

> > > I will note that there is a lot that goes into making a
> > > guitar that does matter, and that the hardness of the wood
> > > can play a role in the timbre by changing the compliance
> > > of the soundboard and resonant cavity in general,
> >
> > There's no soundboard or resonant cavity in a solidbody
> > electric guitar. Wood choice can still be significant of
> > course, but claims that woods of similar hardness and density
> > can be distinguished by sound alone are highly suspect.
>
> This was once again worded poorly, now I'm 0/3. Was talking
> about acoustics, my friend is exclusively an acoustic luthier,

I know. I thought my original said solidbody electric
so I was just clarifying. Certainly wood plays a greater
role in acoustic guitars or archtops.

Did I mention the one about hearing the difference between
Energizer and Duracell in active pickups? It's more
plausible than most of these but I still don't believe it. :)

-Carl

--- In tuning@yahoogroups.com, "Carl Lumma" <carl@...> wrote:
> Did I mention the one about hearing the difference between
> Energizer and Duracell in active pickups? It's more
> plausible than most of these but I still don't believe it. :)

I've heard that one, too. But I've also heard that brand and/or model of guitar/amp/effects doesn't make much of a difference, because a player's tone comes from his/her fingers. Funny creatures, guitarists.

-Igs

Graham wrote:

> The evidence is that the bonang timbre correlates with
> 5-equal (and therefore slendro) and the saron timbre
> correlates with pelog. Is that consistent with harmony as
> Javanese understand it? I don't know.

If Javanese music has harmony, the use of 5-ET may have a
more parsimonious explanation... pelog scales also appear
prominently in regular mapping without inharmonic spectra.

> There was a big variation in keyboard tunings in Europe in
> the 18th Century. There wasn't a corresponding variation
> in timbre that I'm aware of. Does that establish that
> common practice music isn't terribly intonation-sensitive?

It'd be more evidence that timbre doesn't matter, though I
don't know if I'd call that shift in keyboard tuning big.

> The evidence we have suggests that the variation in tuning
> tracks the variation in timbre. That only comes from
> measuring two gamelans,

I can see I'm going to have to make the 20 ft. walk - each
way I might add - to the bookshelf.

Figs. 10.2 & 10.3 show four spectra of saron keys (two each
from two gamelan). None of them are alike! What a joke.
Sethares claims certain partials are 'conserved' across
instruments in each gamelan, and lists their medians.
*cough*

It's not clear to me what Fig. 10.4 shows, but it it seems
to be the tremendous difference in spectrum for a single
gender key depending on how it is struck.

> The evidence says that Thai xylophones work together with
> harmonic timbres to give 7 roughly equal steps to the
> octave. Bill Sethares specifically analyzed a pong lang,
> apparently because it plays a solo on a record he has.
> Does harmony, as Thais understand it, contradict that?

I guess I'm waiting to hear some Thai music where xylophones
interact with harmonic timbres in the relevant way.
A demonstration of the sensitivity of the model to different
choices of "idealized renat" and "harmonic sound G" would
also be prudent.

Given a scale and timbre, some intervals are bound to be
rougher than others. Why not first test if the musicians
bother to play those less often? It's certainly the case for
Western music, with its harmony as Westerners understand it.
Sethares tries this (pg 309-310) and concludes that in fact,
the only way consonance and dissonance are created and varied
in Thai music is by changing the number of pitch classes
sounding (i.e. Fig. 15.5 simply shows the number of concurrent
pitch classes and is relatively insensitive to their timbre).

> Pelog is different, but because my software's
> set up for regular temperaments, I'm analyzing it in terms
> of regular temperaments, and the best correlation is with
> 9-equal. Sethares does talk about the fine structure, so
> if you have a better analysis, let's see it.

My gosh (pg 218)

* G now has five partials instead of six - why? And why
did it have six when it was an idealized renat? What are
the partials' amplitudes and how much does that matter?

* "only half of the curve contains scale steps of the
desired scale, so only that half is shown"

* Fig. 10.11 is an example of what I mentioned before --
considering intervals in only one mode of a non-ET. This
problem of TTSS was first noticed by Paul.

-Carl

--- In tuning@yahoogroups.com, "cityoftheasleep" <igliashon@...> wrote:
>
> --- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@> wrote:
> > I also suspect lot of the conventions of synth
> > programming are there to hide the shortcomings of 12-equal.
>
> LOL. I've heard this hypothesis before and I think it's
> hilarious. If you're talking about the sort of "motion" elements
> of a lot of synth patches (vibratos, LFOs, filter sweeps, etc.),
> they're there to make up for the fact that the basic synth timbres
> are static and people hate static sounds. Acoustic instruments
> produce sounds that morph and evolve in organic ways that
> synthesizers are only just now beginning to decently replicate.

Well, the harmonics in FM timbres can wiggle like crazy just by using
dynamic envelopes but the result is still preferably sent through
buttloads of detuning, chorusing, reverb and other kinds of smearing.
Why?

> And in any case, if it's the beating that people hate, well, fast
> LFO pitch modulation basically IS beating,

No, it's not.

> even more strongly than one gets when hearing 12-TET with static
> synth timbres. So you're saying that synth programming makes up for
> the "shortcomings" of 12-TET by making them worse. That's
> ridiculous.

And a straw man based on a false premise.

> Most people I've talked to IRL seem to hate the sound of JI with a
> static synth timbre much more than they hate the same sound playing
> 12-TET; I get comments like "it sounds cold" or "grating" or "like
> an emergency broadcast > system"--some couldn't even tell a chord
> was being played!--and that the 12-TET sounds "warmer" and "more
> organic".

But have you tried the same in a non-detuned but evolving sound?

P.S. Would you kindly stop calling everything I say or apparently say
hilarious and ridiculous? That doesn't add to your arguments.

Kalle

--- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...> wrote:
>
> Well, the harmonics in FM timbres can wiggle like crazy just by using
> dynamic envelopes but the result is still preferably sent through
> buttloads of detuning, chorusing, reverb and other kinds of smearing.
> Why?

Maybe because FM timbres are pretty harsh most of the time? In the days when warmer-sounding analog synths predominated, the "smearing" you speak of was a bit less common. Do you think that "smearing" would be *less* common if peoples' synths were playing in JI?

> > And in any case, if it's the beating that people hate, well, fast
> > LFO pitch modulation basically IS beating,
>
> No, it's not.

Compelling argument you made, there! What is it that's supposed to make beating unpleasant, if not the sensation of a rhythmic wobble in pitch?

> > even more strongly than one gets when hearing 12-TET with static
> > synth timbres. So you're saying that synth programming makes up for
> > the "shortcomings" of 12-TET by making them worse. That's
> > ridiculous.
>
> And a straw man based on a false premise.

Please, do explain.

> But have you tried the same in a non-detuned but evolving sound?

Why should that make a difference? If people don't prefer JI in the static sound, why would they be expected to prefer it in the evolving sound?

> P.S. Would you kindly stop calling everything I say or apparently say
> hilarious and ridiculous? That doesn't add to your arguments.

Nothing personal, Kalle. I've heard these arguments before from other people and I found them no less ridiculous then.

-Igs

--- In tuning@yahoogroups.com, "cityoftheasleep" <igliashon@...> wrote:
>
> --- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@> wrote:
> >
> > Well, the harmonics in FM timbres can wiggle like crazy just by
> > using dynamic envelopes but the result is still preferably sent
> > through buttloads of detuning, chorusing, reverb and other kinds
> > of smearing. Why?
>
> Maybe because FM timbres are pretty harsh most of the time? In the
> days when warmer-sounding analog synths predominated, the
> "smearing" you speak of was a bit less common.

What do you mean? Single oscillator synths were a rarity. Even the
single oscillator Juno-60 has a build-in chorus unit. Look, I'm not
saying all synth features are there to hide the roughness of
12-equal. I'm only saying that synth architectures and programming
conventions have been unconsciously gravitating towards particular
solutions partly due to the nature of 12-equal. Of course this is
unfalsifiable. But do you really think that 12-equal would have been
adopted anyway if harpsichord had remained the principal keyboard
instrument?

> Do you think that "smearing" would be *less* common if peoples'
> synths were playing in JI?

At least there would be less smearing, why tune to JI if you can't
hear it because of the smearing? But they would probably prefer some
pitch instability because phase-locked chords sound totally unnatural
and the common fate of partials is a strong cue in auditory
streaming.

> > > And in any case, if it's the beating that people hate, well,
> > > fast LFO pitch modulation basically IS beating,
> >
> > No, it's not.
>
> Compelling argument you made, there!

Well, they simply are different things.

> What is it that's supposed to make beating unpleasant, if not the
> sensation of a rhythmic wobble in pitch?
>
> > > even more strongly than one gets when hearing 12-TET with
> > > static synth timbres. So you're saying that synth programming
> > > makes up for the "shortcomings" of 12-TET by making them
> > > worse. That's ridiculous.
> >
> > And a straw man based on a false premise.
>
> Please, do explain.

It's not the beating that is unpleasant, but roughness!

http://www.mmk.ei.tum.de/persons/ter/top/roughness.html

> > But have you tried the same in a non-detuned but evolving sound?
>
> Why should that make a difference? If people don't prefer JI in
> the static sound, why would they be expected to prefer it in the
> evolving sound?

Well, didn't you say that your friends prefer 12-equal to JI with
static timbres because the result sounds more organic? If the sound
is already organic, then it's a completely different test, isn't it?

Kalle

Personally I feel far, far less need to detune unisons or use inharmonic timbres when using different tunings. Rational (or very near rational) tunings certainly don't need the same amount of effects to be mellow, inharmonic tunings don't need the effects to be bold and harsh.

I think the common tendency to make pitch more vague in 12-tET synthesis programming is probably as much about the over-familiarity of 12-tET as it is about the inharmonicity of the thirds and sixths in 12-tET, though.

--- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...> wrote:
>
>
>
>
>
>
> --- In tuning@yahoogroups.com, "cityoftheasleep" <igliashon@> wrote:
> >
> > --- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@> wrote:
> > >
> > > Well, the harmonics in FM timbres can wiggle like crazy just by
> > > using dynamic envelopes but the result is still preferably sent
> > > through buttloads of detuning, chorusing, reverb and other kinds
> > > of smearing. Why?
> >
> > Maybe because FM timbres are pretty harsh most of the time? In the
> > days when warmer-sounding analog synths predominated, the
> > "smearing" you speak of was a bit less common.
>
> What do you mean? Single oscillator synths were a rarity. Even the
> single oscillator Juno-60 has a build-in chorus unit. Look, I'm not
> saying all synth features are there to hide the roughness of
> 12-equal. I'm only saying that synth architectures and programming
> conventions have been unconsciously gravitating towards particular
> solutions partly due to the nature of 12-equal. Of course this is
> unfalsifiable. But do you really think that 12-equal would have been
> adopted anyway if harpsichord had remained the principal keyboard
> instrument?
>
> > Do you think that "smearing" would be *less* common if peoples'
> > synths were playing in JI?
>
> At least there would be less smearing, why tune to JI if you can't
> hear it because of the smearing? But they would probably prefer some
> pitch instability because phase-locked chords sound totally unnatural
> and the common fate of partials is a strong cue in auditory
> streaming.
>
> > > > And in any case, if it's the beating that people hate, well,
> > > > fast LFO pitch modulation basically IS beating,
> > >
> > > No, it's not.
> >
> > Compelling argument you made, there!
>
> Well, they simply are different things.
>
> > What is it that's supposed to make beating unpleasant, if not the
> > sensation of a rhythmic wobble in pitch?
> >
> > > > even more strongly than one gets when hearing 12-TET with
> > > > static synth timbres. So you're saying that synth programming
> > > > makes up for the "shortcomings" of 12-TET by making them
> > > > worse. That's ridiculous.
> > >
> > > And a straw man based on a false premise.
> >
> > Please, do explain.
>
> It's not the beating that is unpleasant, but roughness!
>
> http://www.mmk.ei.tum.de/persons/ter/top/roughness.html
>
> > > But have you tried the same in a non-detuned but evolving sound?
> >
> > Why should that make a difference? If people don't prefer JI in
> > the static sound, why would they be expected to prefer it in the
> > evolving sound?
>
> Well, didn't you say that your friends prefer 12-equal to JI with
> static timbres because the result sounds more organic? If the sound
> is already organic, then it's a completely different test, isn't it?
>
> Kalle
>

--- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...> wrote:
> I'm only saying that synth architectures and programming
> conventions have been unconsciously gravitating towards particular
> solutions partly due to the nature of 12-equal. Of course this is
> unfalsifiable.

Frankly I think people just like "special effects". People use phasers and flangers on all manner of non-harmonic sounds, drums even! Just listen to Led Zeppelin's "Kashmir". But this isn't really a matter of those kinds of effects, it's specifically pitch modulation that's the issue, and I do note that on the built-in GM synth on my MacBook, almost every single sustained sound devolves into vibrato if held for more than a second or two.

Clearly, this ubiquitous pitch modulation is making up for some short-coming, but where we disagree is whether it's a short-coming of tuning or of timbre.

> But do you really think that 12-equal would have been
> adopted anyway if harpsichord had remained the principal keyboard
> instrument?

Yes! As far as I've been educated, it was the proclivity to modulation in the compositional styles of composers that brought about the 12-TET revolution. Nowadays plenty of people tune harpsichords in 12-TET without thinking twice about it. A better question would be, would 12-TET have been adopted if keyboard instruments had never found favor?

> At least there would be less smearing, why tune to JI if you can't
> hear it because of the smearing?

I'm talking "hypothetical alternate reality" here, as if JI was the musical lingua franca instead of 12-TET. Suppose everyone was brought up playing in JI, and synthesizers evolved out of that tradition instead of 12-TET. Do you think reverb and chorus/vibrato effects would be horrendously unpopular?

> But they would probably prefer some
> pitch instability because phase-locked chords sound totally unnatural
> and the common fate of partials is a strong cue in auditory
> streaming.

So now it sounds like you're agreeing with me. Unless you think nominal JI played with unstable pitches produces a significantly different sonic gestalt than tempered pitches.

> It's not the beating that is unpleasant, but roughness!
>
> http://www.mmk.ei.tum.de/persons/ter/top/roughness.html

...which is also explained herein as a modulatory effect.

But in any case, no one I know outside this list seems at all aware of any roughness in 12-TET. Also I should note that in synth-based music, the ones that use modulatory effects the most tend to be the ones with the lowest level of harmonic complexity. Trance music, for instance, is known for its heavily-vibrato'd lead sounds.

Also, why do people like the sounds of choirs? You get quite a lot of roughness from all those detuned unisons, and there's no strict 12-TET going on there at all.

> Well, didn't you say that your friends prefer 12-equal to JI with
> static timbres because the result sounds more organic? If the sound
> is already organic, then it's a completely different test, isn't it?

"Organic" was just one adjective thrown out. But FWIW I've also done "tests" using my guitar instead of a synth, and the 5 naive listeners polled (friends/housemates and a few family members) preferred the 12-TET just as unanimously.

I'm not saying people always prefer the sound of 12-TET to JI, of course--in small ensembles of monophonic instruments or voices, JI (if it is in fact truly JI) often seems to occur naturally and meet with favorable response from listeners.

-Igs

On Mon, May 23, 2011 at 6:55 PM, cityoftheasleep
<igliashon@...> wrote:
>
> > But do you really think that 12-equal would have been
> > adopted anyway if harpsichord had remained the principal keyboard
> > instrument?
>
> Yes! As far as I've been educated, it was the proclivity to modulation in the compositional styles of composers that brought about the 12-TET revolution. Nowadays plenty of people tune harpsichords in 12-TET without thinking twice about it. A better question would be, would 12-TET have been adopted if keyboard instruments had never found favor?

I suggest yes. It doesn't take a genius to realize that three of these
"major third" things are very close to an octave. You'd generally have
two reactions to a concept like that

1) the reaction where people think that that's heretical because it's
a smearing out of the beautiful order they've learned to perceive in
music
2) the reaction where people think that such smearing out can be used
for novel and mindblowing artistic effect

Historically, #2 seems to have won out. I say this because #2 is
comprised of two historical stages - the development of 12-tet, and
the development of the regular mapping paradigm. So that's kind of
cheating.

> > It's not the beating that is unpleasant, but roughness!
> >
> > http://www.mmk.ei.tum.de/persons/ter/top/roughness.html
>
> ...which is also explained herein as a modulatory effect.

Which it is. You have absorbed much of psychoacoustics, young
grasshopper. Now if I could only get my head wrapped around
philosophy...

However, two things are worthy of note:

1) I do think that Kalle might be onto something when he says that the
use of synth effects might have been a lot different if we used JI.
For starters, all of the old analog synths that have really "fat" and
"warm" filters and all of that, and generally had some
nonlinearity/saturation (read: distortion) built in, would have
sounded epically insane with JI. And whenever I listen to deadmau5
play major chords with unfiltered sawtooths, I know, deep down, that
it would sound even more awesome if he'd just throw on Hermode tuning
in Logic. That is, the sterility of unfiltered timbres would be
attenuated somewhat because you'd also have the novel concept of being
able to explore trippy buzzing/fusing/timbre-meets-chord JI whenever
you want. That isn't to say that effects wouldn't still have found
use, but to suggest that a purer tuning might have made a lack of
effects a bit less annoying, because there'd be a new "effect" that JI
produces.

2) The difference between LFO and detuned notes is that LFO makes the
entire signal beat evenly, whereas plain old detuning causes all of
the partials to beat in a wildly irrational and chaotic pattern with
one another. A more adept analogy might be that a chorus pedal is like
timbral detuning, or some kind of mixture of AM and FM and/or chorus,
etc.

This doesn't matter too much for major thirds in practice, because
we're used to it, and because the major third just isn't that bright.
The difference is much more noticeable with detuned fifths.

All in all, I guess what I'm saying is that using JI would provide a
new "effect" for raw synth timbres with no effects on them, which is
that of perfect periodicity buzz and more apparent VF fusion and so
on. Throwing on some phatty filters would make said effect amazing.
Not that that wouldn't stop people from using phaser and such anyway,
but it's still something we can't really do now. It might take 7-limit
harmony to really make it work, fwiw.

> > Well, didn't you say that your friends prefer 12-equal to JI with
> > static timbres because the result sounds more organic? If the sound
> > is already organic, then it's a completely different test, isn't it?
>
> "Organic" was just one adjective thrown out. But FWIW I've also done "tests" using my guitar instead of a synth, and the 5 naive listeners polled (friends/housemates and a few family members) preferred the 12-TET just as unanimously.
>
> I'm not saying people always prefer the sound of 12-TET to JI, of course--in small ensembles of monophonic instruments or voices, JI (if it is in fact truly JI) often seems to occur naturally and meet with favorable response from listeners.

I remember the first time I heard 7/4, I thought it was flat as hell.
After some rumination, and being able to express what I disliked about
it, I figured out that it was that I couldn't split it into two
fourths. I disliked that strongly. In general, I couldn't run the same
mental circuit around it as the minor 7th in 12-equal - a circuit
which involved all kinds of similarly tempered relationships. My
awareness of how this interval could potentially fit with other notes
in such a way as to make music wasn't just off enough from what I was
expecting that it sounded "wrong" instead of "new." The uncanny
metaphor applies quite a bit here. Take that how you will.

Then I heard Ron Sword's 7-limit shrutar example on the old
xenharmonic alliance ning and was converted instantly. Funny how that
works sometimes.

-Mike

On Mon, May 23, 2011 at 5:55 PM, cityoftheasleep <igliashon@...>wrote:

>
> Also, why do people like the sounds of choirs? You get quite a lot of
> roughness from all those detuned unisons, and there's no strict 12-TET going
> on there at all.

If I'm like most people, the inherent instability of the human voice
ensembles is perceived as a "warm sound" provided it's limited in
effect--i.e. I can't stand a choir made of up fat vibrato "opera soloist"
types who don't really know what it means to blend---to limit the effects of
vibrato to nil, and allow the micro-instability of pitch because of basic
physical chaos to come to the fore.

> Well, didn't you say that your friends prefer 12-equal to JI with
> > static timbres because the result sounds more organic? If the sound
> > is already organic, then it's a completely different test, isn't it?
>
> "Organic" was just one adjective thrown out. But FWIW I've also done
> "tests" using my guitar instead of a synth, and the 5 naive listeners polled
> (friends/housemates and a few family members) preferred the 12-TET just as
> unanimously.
>

To really make this scientific, one should perform a really broad sample
with a bunch of different combinations that have been proposed: JI vs.
tempered; evolving vs. static; synthetic vs. acoustic, etc. etc.

In my own non-scientific studies, I've found people often prefer the
isolated sound of JI triads to tempered 12-tet triads, noting them to be
sweeter to the ear. This is true of both synth and acoustic sounds. I find
that usually, if the synth sound is crappy in JI, it's equally crappy in
12-tet! :)

>
> I'm not saying people always prefer the sound of 12-TET to JI, of
> course--in small ensembles of monophonic instruments or voices, JI (if it is
> in fact truly JI) often seems to occur naturally and meet with favorable
> response from listeners.
>
>
It's also worth noting that freely tunable and unfretted instruments are
easiest to perform with when the target tuning is JI--because the ear can
easily track those points of relative stability, both melodically and
particularly, harmonically, where sustained chords really make this true.

Best,
Aaron Krister Johnson
http://www.akjmusic.com
http://www.untwelve.org

"cityoftheasleep" <igliashon@...> wrote:
> --- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...>
> wrote:

> > But do you really think that 12-equal would have been
> > adopted anyway if harpsichord had remained the
> > principal keyboard instrument?
>
> Yes! As far as I've been educated, it was the proclivity
> to modulation in the compositional styles of composers
> that brought about the 12-TET revolution. Nowadays
> plenty of people tune harpsichords in 12-TET without
> thinking twice about it. A better question would be,
> would 12-TET have been adopted if keyboard instruments
> had never found favor?

I'd like to quote from one of the oldest recorded tuning
discussions. This is from a PDF of Bosanquet's
"Temperament or the Division of the Octave Part 2" dated
May 3, 1875. Nothing about harpsichords, unfortunately.

"""
Mr. Cummings though the harmonium (which Mr. Bosanquet had
referred to as the instrument by which he had made some of
his experiments) was not to be depended on for anything
whatever. It seemed never to give a chord fit to listen
to, and therefore he did not think any test of that kind
could be relied upon. It gave out so many harmonics that
you could not judge of any music fairly by its means.

Mr. Bosanquet said his harmonium had stood perfectly in
tune for two years. The reeds were at least twenty times
as sensitive as an organ-pipe, and had remained perfectly
in tune.

Mr. Cummings remarked that all the harmoniums he had ever
tried seemed to give out harmonics more freely than any
other instrument. If Mr. Bosanquet's was different in that
respect, he should be glad to her it.

The Chairman observed that the resultant sounds of an
harmonium were generally detestable.

Mr Bosanquet said the great defect in the harmonium was
that it was so sensitive to tuning, and that very fact made
it so valuable to him in his experiments, because
differences which the ear would not detect in another
instrument became evident at once on the harmonium. The
tuning of an harmonium was far more accurate than that of
organ-pipes; you could tune any quantity of the latter from
an harmonium, but you could not reverse the process. This
fact made it so valuable as a means of research.

The Chairman said he had had the advantage of hearing Mr.
Bosanquet's harmonium, and certainly some of the effects
were exceedingly beautiful: for instance, its powers were
singularly illustrated in a passage commencing the `Stabat
Mater' of Palestrina, which was frequently quoted as an
instance of the harshness of the old masters. As Mr.
Bosanquet played it, the effect was certainly altogether
different---much more delicate, aërial, /spirituelle/ than
he had ever heard it played. Of course four good singers
would unconsciously give the right intonation to it, as was
remarked by Burney in his account of the Pope's Chapel,
that the singers there unconsciously sing in what is no
doubt perfect tune. The point that remained, and the
question upon which musicians seemed never to have been
satisfied, was whether the mechanical difficulties of
obtaining these effects were not so great as to render them
useless for practical purposes. As he understood Mr.
Bosanquet, this organ required about six times the number
of keys and pipes of an ordinary instrument. Now, if the
organ of St. Paul's were multiplied by six of seven, it
would hardly go under the dome. Besides that, he thought
some of the combinations which had been put before them as
beautiful were hideous; some of the harmonic sevenths, for
instance, were extremely disagreeable.

Mr. Bosanquet said he had always found those notes
disagreeable to such persons as had a strong sense of
absolute pitch. The only way to like them was to listen to
the chords. He never knew anybody who had a good ear who
liked them at first.

The Chairman: Then comes the question, what is a good ear,
and who has it?

Mr. Bosanquet said he had frequent experience of all sorts
of people coming to hear his harmonium, and the result was,
that persons with acute ears, but not much musical
education, liked the chords, and always picked out the
effects which he liked best himself, as the result of long
custom; but persons who had the scale firmly in their
heads, as no doubt the Chairman had, did not like the
departure from the usual value of the notes. They did not
think of the consonance at all; the question with them
being, not whether it was smooth, but whether it was what
they were accustomed to. The question with him was simply
one of smoothness.

Mr. Cummings said that the chord of the sixth on Ab, at the
beginning of the second page of the example which Mr.
Bosanquet played, sounded to him very flat indeed.

Mr. Bosanquet said that was Ab /raised/. It was a curious
fact, that people with highly-educated ears almost always
singled out the true minor third of the chord as
disagreeable. The true minor third was always raised.
"""

Graham

--- In tuning@yahoogroups.com, Graham Breed <gbreed@...> wrote:

> I'd like to quote from one of the oldest recorded tuning
> discussions. This is from a PDF of Bosanquet's
> "Temperament or the Division of the Octave Part 2" dated
> May 3, 1875. Nothing about harpsichords, unfortunately.

[snip]

Sounds like in Bosanquet's day, the general attitude toward microtonality stood much as it does today. Which is to say it is the most highly trained musicians (esp. with absolute pitch) that tend to find JI disagreeable at first. It is interesting that the Chairman who so hated harmoniums did find some passages more agreeable on Bosanquet's, whereas others sounded "off" to him (like the minor 3rd and harmonic 7th, two intervals which I recall being somewhat contentious in a few discussions with Marcel "Mr. JI" DeVelde).

-Igs

What a perfectly clear record of this ancient discussion,
which has been so often repeated here. It could go directly
in the FAQ. -Carl

--- In tuning@yahoogroups.com, Graham Breed <gbreed@...> wrote:

> I'd like to quote from one of the oldest recorded tuning
> discussions. This is from a PDF of Bosanquet's
> "Temperament or the Division of the Octave Part 2" dated
> May 3, 1875. Nothing about harpsichords, unfortunately.
>
> """
> Mr. Cummings though the harmonium (which Mr. Bosanquet had
> referred to as the instrument by which he had made some of
> his experiments) was not to be depended on for anything
> whatever. It seemed never to give a chord fit to listen
> to, and therefore he did not think any test of that kind
> could be relied upon. It gave out so many harmonics that
> you could not judge of any music fairly by its means.
>
> Mr. Bosanquet said his harmonium had stood perfectly in
> tune for two years. The reeds were at least twenty times
> as sensitive as an organ-pipe, and had remained perfectly
> in tune.
>
> Mr. Cummings remarked that all the harmoniums he had ever
> tried seemed to give out harmonics more freely than any
> other instrument. If Mr. Bosanquet's was different in that
> respect, he should be glad to her it.
>
> The Chairman observed that the resultant sounds of an
> harmonium were generally detestable.
>
> Mr Bosanquet said the great defect in the harmonium was
> that it was so sensitive to tuning, and that very fact made
> it so valuable to him in his experiments, because
> differences which the ear would not detect in another
> instrument became evident at once on the harmonium. The
> tuning of an harmonium was far more accurate than that of
> organ-pipes; you could tune any quantity of the latter from
> an harmonium, but you could not reverse the process. This
> fact made it so valuable as a means of research.
>
> The Chairman said he had had the advantage of hearing Mr.
> Bosanquet's harmonium, and certainly some of the effects
> were exceedingly beautiful: for instance, its powers were
> singularly illustrated in a passage commencing the `Stabat
> Mater' of Palestrina, which was frequently quoted as an
> instance of the harshness of the old masters. As Mr.
> Bosanquet played it, the effect was certainly altogether
> different---much more delicate, aërial, /spirituelle/ than
> he had ever heard it played. Of course four good singers
> would unconsciously give the right intonation to it, as was
> remarked by Burney in his account of the Pope's Chapel,
> that the singers there unconsciously sing in what is no
> doubt perfect tune. The point that remained, and the
> question upon which musicians seemed never to have been
> satisfied, was whether the mechanical difficulties of
> obtaining these effects were not so great as to render them
> useless for practical purposes. As he understood Mr.
> Bosanquet, this organ required about six times the number
> of keys and pipes of an ordinary instrument. Now, if the
> organ of St. Paul's were multiplied by six of seven, it
> would hardly go under the dome. Besides that, he thought
> some of the combinations which had been put before them as
> beautiful were hideous; some of the harmonic sevenths, for
> instance, were extremely disagreeable.
>
> Mr. Bosanquet said he had always found those notes
> disagreeable to such persons as had a strong sense of
> absolute pitch. The only way to like them was to listen to
> the chords. He never knew anybody who had a good ear who
> liked them at first.
>
> The Chairman: Then comes the question, what is a good ear,
> and who has it?
>
> Mr. Bosanquet said he had frequent experience of all sorts
> of people coming to hear his harmonium, and the result was,
> that persons with acute ears, but not much musical
> education, liked the chords, and always picked out the
> effects which he liked best himself, as the result of long
> custom; but persons who had the scale firmly in their
> heads, as no doubt the Chairman had, did not like the
> departure from the usual value of the notes. They did not
> think of the consonance at all; the question with them
> being, not whether it was smooth, but whether it was what
> they were accustomed to. The question with him was simply
> one of smoothness.
>
> Mr. Cummings said that the chord of the sixth on Ab, at the
> beginning of the second page of the example which Mr.
> Bosanquet played, sounded to him very flat indeed.
>
> Mr. Bosanquet said that was Ab /raised/. It was a curious
> fact, that people with highly-educated ears almost always
> singled out the true minor third of the chord as
> disagreeable. The true minor third was always raised.
> """
>
> Graham

On Tue, May 24, 2011 at 4:31 PM, Carl Lumma <carl@...> wrote:
>
> What a perfectly clear record of this ancient discussion,
> which has been so often repeated here. It could go directly
> in the FAQ. -Carl

Upvote that. Wow.

-Mike