Yahoo Tuning Groups Ultimate Backup tuning The Harmonic Divergence Hypothesis of Minorness and Musical "Feeling": No theory, just a listening example

Hi all,

Somewhere in the last interaction with Carl, I elucidated a theory of
what causes "minorness" that got lost completely. It's really just a
bit too complicated to fully explain in a way that sounds convincing
in few words, and endless paragraphs put people to sleep. so I figured
I'd start off with a listening example of some xenharmonic minor
chords instead.

Carl outlined what I saw as a brilliant theory of what causes
minorness that I had never considered before. He noted that it broke
down in some cases but held in a significant amount of others,
although with no psychoacoustic explanation. We also noted that
Partch's "utonal = minor" hypothesis held in a few cases, but broke
down eventually. I sought to come up with a hypothesis to explain the
successes of these when they work, the failings of when they don't,
and that has something of a psychoacoustic basis.

OK, now, the listening example. Very quickly, you should know that I'm
going to reference something called "pain." Pain is exactly what it
sounds like: pain, as in fingernails on a chalkboard type pain. I'm
also going to reference the concept of "negativity," which refers to
exactly what it sounds like. I will also reference things like
"sadness" and "minorness," which are both subjective, so you may feel
different labels are appropriate in cases when I use those terms.
Don't get too hung up on the specific choice of words I use for cases
like that. Suffice to say if your perceptual experience draws even an
abstract parallel in terms of pain, negativity, or even sadness, with
what the theory predicts, I will consider this somewhat of a success.

One final caveat: if you find that there's slightly more beating than
you'd like, note this: the theory predicts that beating doesn't matter
or contribute to the "minor" sound. If you choose to use ratios in the
"minor" vicinity that have less beating, you are free to do so; as
it's entirely a matter of personal taste. Just imagine I had picked
ratios with less beating.

http://www.mikebattagliamusic.com/music/minortest.mp3

The first two chords you hear will be 4:7:8:10:11, then its "minor" version.

After a few seconds of silence, you'll hear four more chords. First is
the otonal 7:9:11. Then you'll hear a minor version, which also
happens to be the utonal equivalent of 7:9:11. You'll then hear
"another" minor variant of it, and then you'll hear an alternate
version of the first minor chord, which I think didn't work as well as
the first.

Next we're going to sort of go up the "series of negativity." You'll
hear 8 chords. The first five chords start with minor and become
progressively more "negative" sounding, going far beyond minor into a
very painful sonority, then resolve back to major.

It starts with the familiar "sad" feeling of minor, then escalates
further into something even sadder, perhaps slightly like despair. It
then proceeds from there into further xenharmonic ultra-minor chords
that to me sound almost terrifying and painful. Stick with it through
all five chords.

After those five, you will hear a slow transition back into happiness.
I made this transition happen more rapidly than the first. The first
chord should be sort of a slight relaxing or "chilling out"; perhaps
it sounds like the pain is over, and you're resolving it. The next
chord should be an even more chilling out, still sounding "minor," but
becoming even less tense. The last chord, which is completely otonal
and it might perhaps sound something like "happiness." You may hear
that in some sense some of these chords "resolve" to the next one.

If you had an experience even remotely resembling this one, then
you've just experienced something sort of like xenharmonic modality,
and if you experienced the semi-resolution effect, something like
xenharmonic tonality.

Next we're going to play three chords that should sound -ROUGHLY- on
the same order of something like "sadness." Assign your own emotional
labels as you see fit. I personally heard the last one as roughly less
sad, perhaps, your mileage may vary.

Did anyone else have a perception that resembled this? If so, I'd be
happy to tell you what I'm doing, as it has a rough psychoacoustic
basis that I'm still fleshing out. If not, then I'm simply going to
assume that my brain is completely different than everyone else's. It
also draws a few parallel's from George Secor's ideas in his 17-tet
paper that the goals of harmony and melody are not in conflict, and
offers an explanation for why Partch's utonal hypothesis works at
first, and then gets progressively less useful.

-MIke

🔗genewardsmith <genewardsmith@...>

🔗Mike Battaglia <battaglia01@...>

On Sat, Sep 18, 2010 at 1:16 PM, genewardsmith
<genewardsmith@...> wrote:
>
> --- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
>
> > Did anyone else have a perception that resembled this? If so, I'd be
> > happy to tell you what I'm doing, as it has a rough psychoacoustic
> > basis that I'm still fleshing out. If not, then I'm simply going to
> > assume that my brain is completely different than everyone else's.
>
> Sorry. all I heard were some xenharmonic chords. The difference between otonal and utonal was apparent, but I wonder if the emotional effect is as powerful as it is with 5-limit triads.

Hi Gene,

Thanks for your input. You should know that not all of the chords are
really xenharmonic. The first two sets of chords are xenharmonic and
shouldn't really be as intense as major/minor. It isn't otonal/utonal
stuff that's going on though. The next set of chords should start with
two non-xenharmonic chords that are just intoned nicely in JI - minor
and minor maj7. Most people hear these chords as "sad" and "more sad"
(or perhaps "despair") respectively. It then adds a few more
xenharmonic extensions that I hear as continuing the pattern of
negativity and heighten the feeling - you don't hear these as more
@#$*$# up than the last ones? And then I just try to slowly resolve it
back into fully otonal to explore the range of moods in between.

I'm really surprised about the last 3 - these aren't really too
xenharmonic, just 12-tet chords that I'm intoning a little bit
differently. Most people I know hear major9#11 chords as sounding
somewhat "sad," or lydian mode in general having a bit of "sadness" to
it...

-Mike

🔗cityoftheasleep <igliashon@...>

Yeah, I had an experience like you suggested, but I also listened to it while reading your description. It's hard to say what I would have heard if I was "listening blind". Though FWIW, the 7:9:11 major/minor examples definitely worked. To bad there aren't any low EDO's that give both a good 7:9 and 7:11...maybe there's a good ED3 tuning for those chords? It's interesting that the 7:9:11 chords are more conspicuously major and minor to my ears than the 3:5:7 or 5:7:9 chords of BP, as well as 4:5:7 chords.

Also, what is the purpose in these listening examples of using such extended chords? Wouldn't it be more sensible to start by comparing the emotional qualities of triads first?

-Igs

--- In tuning@yahoogroups.com, Mike Battaglia <battaglia01@...> wrote:
>
> Hi all,
>
> Somewhere in the last interaction with Carl, I elucidated a theory of
> what causes "minorness" that got lost completely. It's really just a
> bit too complicated to fully explain in a way that sounds convincing
> in few words, and endless paragraphs put people to sleep. so I figured
> I'd start off with a listening example of some xenharmonic minor
> chords instead.
>
> Carl outlined what I saw as a brilliant theory of what causes
> minorness that I had never considered before. He noted that it broke
> down in some cases but held in a significant amount of others,
> although with no psychoacoustic explanation. We also noted that
> Partch's "utonal = minor" hypothesis held in a few cases, but broke
> down eventually. I sought to come up with a hypothesis to explain the
> successes of these when they work, the failings of when they don't,
> and that has something of a psychoacoustic basis.
>
> OK, now, the listening example. Very quickly, you should know that I'm
> going to reference something called "pain." Pain is exactly what it
> sounds like: pain, as in fingernails on a chalkboard type pain. I'm
> also going to reference the concept of "negativity," which refers to
> exactly what it sounds like. I will also reference things like
> "sadness" and "minorness," which are both subjective, so you may feel
> different labels are appropriate in cases when I use those terms.
> Don't get too hung up on the specific choice of words I use for cases
> like that. Suffice to say if your perceptual experience draws even an
> abstract parallel in terms of pain, negativity, or even sadness, with
> what the theory predicts, I will consider this somewhat of a success.
>
> One final caveat: if you find that there's slightly more beating than
> you'd like, note this: the theory predicts that beating doesn't matter
> or contribute to the "minor" sound. If you choose to use ratios in the
> "minor" vicinity that have less beating, you are free to do so; as
> it's entirely a matter of personal taste. Just imagine I had picked
> ratios with less beating.
>
> http://www.mikebattagliamusic.com/music/minortest.mp3
>
> The first two chords you hear will be 4:7:8:10:11, then its "minor" version.
>
> After a few seconds of silence, you'll hear four more chords. First is
> the otonal 7:9:11. Then you'll hear a minor version, which also
> happens to be the utonal equivalent of 7:9:11. You'll then hear
> "another" minor variant of it, and then you'll hear an alternate
> version of the first minor chord, which I think didn't work as well as
> the first.
>
> Next we're going to sort of go up the "series of negativity." You'll
> hear 8 chords. The first five chords start with minor and become
> progressively more "negative" sounding, going far beyond minor into a
> very painful sonority, then resolve back to major.
>
> It starts with the familiar "sad" feeling of minor, then escalates
> further into something even sadder, perhaps slightly like despair. It
> then proceeds from there into further xenharmonic ultra-minor chords
> that to me sound almost terrifying and painful. Stick with it through
> all five chords.
>
> After those five, you will hear a slow transition back into happiness.
> I made this transition happen more rapidly than the first. The first
> chord should be sort of a slight relaxing or "chilling out"; perhaps
> it sounds like the pain is over, and you're resolving it. The next
> chord should be an even more chilling out, still sounding "minor," but
> becoming even less tense. The last chord, which is completely otonal
> and it might perhaps sound something like "happiness." You may hear
> that in some sense some of these chords "resolve" to the next one.
>
> If you had an experience even remotely resembling this one, then
> you've just experienced something sort of like xenharmonic modality,
> and if you experienced the semi-resolution effect, something like
> xenharmonic tonality.
>
> Next we're going to play three chords that should sound -ROUGHLY- on
> the same order of something like "sadness." Assign your own emotional
> labels as you see fit. I personally heard the last one as roughly less
> sad, perhaps, your mileage may vary.
>
>
>
> Did anyone else have a perception that resembled this? If so, I'd be
> happy to tell you what I'm doing, as it has a rough psychoacoustic
> basis that I'm still fleshing out. If not, then I'm simply going to
> assume that my brain is completely different than everyone else's. It
> also draws a few parallel's from George Secor's ideas in his 17-tet
> paper that the goals of harmony and melody are not in conflict, and
> offers an explanation for why Partch's utonal hypothesis works at
> first, and then gets progressively less useful.
>
> -MIke
>

🔗Mike Battaglia <battaglia01@...>

On Sat, Sep 18, 2010 at 5:27 PM, cityoftheasleep
<igliashon@...> wrote:
>
> Yeah, I had an experience like you suggested, but I also listened to it while reading your description. It's hard to say what I would have heard if I was "listening blind".

Ha! Well, I'm glad there's such healthy skepticism around here. The
truth is that the examples worked for me, and I simply found a pattern
that predicted when I'd hear a chord as happy, sad, sinister,
alarming, etc. I just posted it here to see if other people heard it
the same way. To be honest, if everyone else heard it completely
differently from me, then that would be real reason for me to assume
that musical feeling emerges from cognitive aspects of the music,
rather than psychoacoustics. There's plenty of room for cultural stuff
even assuming the hypothesis I'm making, though.

But I think you'd have heard it somewhat similar if you were listening
blind, especially the "series of negativity part." I mean, just remove
it from a musical context. If I hadn't told you to keep listening as
it gets worse and worse, you might have just been like "Damn it this
is an irritating noise, Mike's examples suck, I need to turn this
sound the hell off, beep beep boop organism must take action to remove
painfully discordant intrusive stimulus."

Another Zen koan!

> Though FWIW, the 7:9:11 major/minor examples definitely worked. To bad there aren't any low EDO's that give both a good 7:9 and 7:11...maybe there's a good ED3 tuning for those chords? It's interesting that the 7:9:11 chords are more conspicuously major and minor to my ears than the 3:5:7 or 5:7:9 chords of BP, as well as 4:5:7 chords.

Oh yes it is! :) Try 4:5:7 chords where the 5 is flattened by about 70
cents. Should sound sadder. Now either flatten or sharpen the 7 by
another 70 cents. Should sound even sadder. Add the 6 back in,
shouldn't make much of a difference. Flatten the 6 by 70 cents, gets
worse, etc. Note how consonance morphs into sadness morphs into fear
and despair and eventually becomes
fingernails-on-a-chalkboard-shut-that-damn-alarm-off bad.

Note 70 cents isn't a hard and fast rule; play by ear and find where
the sound is most discordant. Introduce discordance in small,
controlled, measured, predictable quantities, and you have
"minorness." I just detailed the space between "minor" and "utter
painful discordance." The space between "minor" and "concordance,
which sounds pretty cool" I think is far more interesting to explore,
however.

Also, once you detune one of the notes by this maximally discordant
melodically optimal, you might choose to retune things into some JI
near-equivalent to remove beating for cultural reasons; if you do
this, you have effectively created a form of regular mapping in which
this irrational, maximally discordant interval, is a "prime" that
becomes "tempered" to align with the JI ratio of your choosing.

> Also, what is the purpose in these listening examples of using such extended chords? Wouldn't it be more sensible to start by comparing the emotional qualities of triads first?

I don't think so. I simply tried to create an amount of discordance
with each sonority, regardless of the human-imposed limit of how many
notes are in a chord. I think there's a deeper principle at work here,
which I have partially revealed above.

I'm going to come up with a piece of counterpoint next that uses this
as a compositional technique; I think it's instrumental in setting up
something like "tonality."

One last hint: 12-tet's 100 cent semitone functions pretty well as
this "maximally discordant interval." 13-tet from this perspective, is
a bit better. Look at 13-tet from this perspective and note that every
detuned JI ratio you see is a "minor" note, in a very, very abstract
sense. 70 cents detuned from JI seems to be the peak of the "pain"
that is introduced. You can add concordant intervals in order to undo
some of this "pain," if you create too much.

Have fun with it and see what you can come up with. Here's an
interesting sonority that lies somewhere between "classic minor" and
"major" on the "pain" spectrum, to my ears: 0 5 7 10 13 17 in 13-tet.
There's a little bit of stimulative intrusiveness, I hear it as
slightly "unresolved" (another zen koan) yet exciting, although
slightly unsettled. My personal, subjective interpretation. From the
perspective I outlined above, it's a detuned 4:5:6:7:8:10.

-Mike

🔗Carl Lumma <carl@...>

🔗Mike Battaglia <battaglia01@...>

🔗Carl Lumma <carl@...>

🔗Mike Battaglia <battaglia01@...>

On Mon, Sep 20, 2010 at 5:33 AM, Carl Lumma <carl@...> wrote:
>
> I wrote:
> > It's seriously hard to listen into one long file like this.
>
> Also it isn't presented as a test. You ask for comments
> after telling us how you hear it. You should be asking us
> binary questions such as "Which of these sounds more minor?"

It wasn't meant to be a rigorous test, just an introductory listening
example to stimulate discussion or feedback. I'm not trying to prove
anything rigorously, yet, because I only have a very basic concept
that I'm still developing. I'll do things properly, write it up in a
paper, do blind listening tests, and all of that later. I'm trying to
throw this together as I prepare to volunteer for 3 weeks in Haiti
starting the day after tomorrow. so I'm doing the best I can here.

That being said, a while ago I pointed out that 7:9:11 and 1/(7:9:11)
sounded "happy/sad," and you agreed. This wasn't really double blind
either, but served as a useful informal data point for further
thought. In that light, did you hear the examples anything like as I
described?

-Mike

🔗Chris Vaisvil <chrisvaisvil@...>

Let me throw in here:

Not Only Music forum supports making polls, mp3 uploads, and the streaming
of the uploads.

Perhaps the best way to do an experiment like this is to put up a poll and
examples at NOM and ask people to rate "minorness" or whatever property you
are after.

Chris

On Mon, Sep 20, 2010 at 5:51 AM, Mike Battaglia <battaglia01@...>wrote:

>
>
> On Mon, Sep 20, 2010 at 5:33 AM, Carl Lumma <carl@...<carl%40lumma.org>>
> wrote:
> >
> > I wrote:
> > > It's seriously hard to listen into one long file like this.
> >
> > Also it isn't presented as a test. You ask for comments
> > after telling us how you hear it. You should be asking us
> > binary questions such as "Which of these sounds more minor?"
>
> It wasn't meant to be a rigorous test, just an introductory listening
> example to stimulate discussion or feedback. I'm not trying to prove
> anything rigorously, yet, because I only have a very basic concept
> that I'm still developing. I'll do things properly, write it up in a
> paper, do blind listening tests, and all of that later. I'm trying to
> throw this together as I prepare to volunteer for 3 weeks in Haiti
> starting the day after tomorrow. so I'm doing the best I can here.
>
> That being said, a while ago I pointed out that 7:9:11 and 1/(7:9:11)
> sounded "happy/sad," and you agreed. This wasn't really double blind
> either, but served as a useful informal data point for further
> thought. In that light, did you hear the examples anything like as I
> described?
>
> -Mike
>
>

🔗Carl Lumma <carl@...>

Chris wrote:

> Perhaps the best way to do an experiment like this is to put
> up a poll and examples at NOM and ask people to rate "minorness"
> or whatever property you are after.

There are polls here too, but probably NOM's are better.
Some kind of polling would definitely be an improvement.

I was gonna suggest to Mike & Igs, that if we can agree on
a set of chords to test, and Mike or Igs were willing to
synthesize the examples, I am willing to buy Mechanical Turk
time to get results and then maybe the three of us could
put a paper together.

I agree with Igs we should restrict ourselves to triads for
now. The form I'm thinking of is trials each containing 3
triads -- an initial triad and 2 menu triads. Subjects are
asked to choose which of the menu triads sounds most like
"a sad version of" the initial triad. We test all
3-combinations of the set of entire set triads, and for
each combination, perhaps even three versions with each
triad in the initial position. Mike and I would predict
the outcome of each trial in advance...

-C.

🔗Chris Vaisvil <chrisvaisvil@...>

🔗Carl Lumma <carl@...>

🔗Mike Battaglia <battaglia01@...>

On Mon, Sep 20, 2010 at 5:07 PM, Carl Lumma <carl@...> wrote:
>
> > Perhaps the best way to do an experiment like this is to put
> > up a poll and examples at NOM and ask people to rate "minorness"
> > or whatever property you are after.
>
> There are polls here too, but probably NOM's are better.
> Some kind of polling would definitely be an improvement.

I want to organize my thoughts first and nail down the pattern as much
as possible. Then I'll be able to make concrete predictions. I am
absolutely 100% confident that I will be able to do so after some more
time organizing my thoughts, and that they will predict musical
feeling significantly more than average.

The problem I'm having now is coming up with a way to actually ask the
question, since I have seen people put different labels onto different
things that are probably perceptually equivalent from person to
person. That is, one person might say that a chord is "sad," whereas
one might be more descriptive and say "this chord doesn't really sound
'sad' per se, perhaps more like 'longing.'" And another person might
imagine the same chord in a musical context and say "this actually
sounds brighter, because I can imagine it resolving this way." So the
test should incorporate questions about ones musical training and such
too, I think.

> I was gonna suggest to Mike & Igs, that if we can agree on
> a set of chords to test, and Mike or Igs were willing to
> synthesize the examples, I am willing to buy Mechanical Turk
> time to get results and then maybe the three of us could
> put a paper together.

I'm going to write some of my own papers on related ideas as well, but
if you want to kick it off with a joint paper about minorness I would
be down with that.

> I agree with Igs we should restrict ourselves to triads for
> now. The form I'm thinking of is trials each containing 3
> triads -- an initial triad and 2 menu triads. Subjects are
> asked to choose which of the menu triads sounds most like
> "a sad version of" the initial triad. We test all
> 3-combinations of the set of entire set triads, and for
> each combination, perhaps even three versions with each
> triad in the initial position. Mike and I would predict
> the outcome of each trial in advance...

The reason that I don't want to stick to triads is that in my
perspective, detuning higher harmonics leads to less of an effect.
This is why, for the last example in my listening test, I took a triad
with one note detuned (minor) and pitted it against two pentads with
three notes detuned. I don't think we'll ever hear a triad as sounding
"equivalent" to C-Eb-G in "pain," and when you get to 7:9:11 and its
detuned variant the difference is subtle. I also think that the
distinction between triads and pentads breaks down on the
psychoacoustic level that I'm talking about.

If you want to stick to triads, I might be able to figure something
out, but it'll have to wait until I get my head on straight with
exactly how to best make predictions. It might have to wait until I
calculate tetradic entropy, which I want to try and accomplish in my
free time in the next few weeks. (You might actually find my approach
somewhat interesting, as I think I can get it down to nlogn time for
any amount of notes.).

-Mike

🔗cityoftheasleep <igliashon@...>

Hi Carl,

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

That's not a bad idea. I had no idea such a thing as Mechanical Turk existed (and where the heck did they come up with that name?), but that seems like a great idea. We might want to see if we can find someone to consult on our experimental design...and of course, we need to really nail down a single hypothesis to test, first. But I'd be happy to synthesize the examples, provided I'm given instruction for an experimentally-ideal timbre. Or should we include timbre as a variable as well? I'll admit I'm kind of curious if sawtooth waves are actually capable of obscuring the "emotional" content of a chord.

-Igs

🔗Chris Vaisvil <chrisvaisvil@...>

here

http://en.wikipedia.org/wiki/The_Turk

it was a fake robot in 1770

On Mon, Sep 20, 2010 at 7:59 PM, cityoftheasleep <igliashon@...>wrote:

>
>
> Hi Carl,
>
>
> --- In tuning@yahoogroups.com <tuning%40yahoogroups.com>, "Carl Lumma"
> <carl@...> wrote:
>
> > I was gonna suggest to Mike & Igs, that if we can agree on
> > a set of chords to test, and Mike or Igs were willing to
> > synthesize the examples, I am willing to buy Mechanical Turk
> > time to get results and then maybe the three of us could
> > put a paper together.
>
> That's not a bad idea. I had no idea such a thing as Mechanical Turk
> existed (and where the heck did they come up with that name?), but that
> seems like a great idea. We might want to see if we can find someone to
> consult on our experimental design...and of course, we need to really nail
> down a single hypothesis to test, first. But I'd be happy to synthesize the
> examples, provided I'm given instruction for an experimentally-ideal timbre.
> Or should we include timbre as a variable as well? I'll admit I'm kind of
> curious if sawtooth waves are actually capable of obscuring the "emotional"
> content of a chord.
>
> -Igs
>
>
>

The Harmonic Divergence Hypothesis of Minorness and Musical "Feeling": No theory, just a listening example

🔗Mike Battaglia <battaglia01@...>

🔗genewardsmith <genewardsmith@...>

🔗Mike Battaglia <battaglia01@...>

🔗cityoftheasleep <igliashon@...>

🔗Mike Battaglia <battaglia01@...>

🔗Carl Lumma <carl@...>

🔗Mike Battaglia <battaglia01@...>

🔗Carl Lumma <carl@...>

🔗Mike Battaglia <battaglia01@...>

🔗Chris Vaisvil <chrisvaisvil@...>

🔗Carl Lumma <carl@...>

🔗Chris Vaisvil <chrisvaisvil@...>

🔗Carl Lumma <carl@...>

🔗Mike Battaglia <battaglia01@...>

🔗cityoftheasleep <igliashon@...>

🔗Chris Vaisvil <chrisvaisvil@...>

🔗Michael <djtrancendance@...>

🔗Mike Battaglia <battaglia01@...>

🔗Michael <djtrancendance@...>

🔗Mike Battaglia <battaglia01@...>