Re: [tuning] Digest Number 1969

On 3/15/02 5:32 AM, "tuning@yahoogroups.com" <tuning@yahoogroups.com> wrote:

> Message: 8
> Date: Fri, 15 Mar 2002 05:38:36 -0000
> From: "paulerlich" <paul@stretch-music.com>
> Subject: Re: Jerries: a conclusion or two
>
> --- In tuning@y..., Gerald Eskelin <stg3music@e...> wrote:
>> On 3/14/02 3:58 PM, "tuning@y..." <tuning@y...> wrote:
>>
>>> Message: 4
>>> Date: Thu, 14 Mar 2002 04:28:31 -0000
>>> From: "paulerlich" <paul@s...>
>>> Subject: Re: Jerries: a conclusion or two
>>>
>>> --- In tuning@y..., Gerald Eskelin <stg3music@e...> wrote:
>>>
>>>> Yes, it does. But singing a pitch that "agrees" with the apparent high
>>>>third *also* sounds good.
>>>
>>> what does that mean? aren't you *hearing* your voice, just as you're
>>> *hearing* the ji third in the recording?
>>
>> Yes. Of course it does. That's the point. My voice seems to be "in tune"
>> with the JI illusionary third.
>>>>
>>> wouldn't singing an
>>> actual high third result in a psychoacoustic "even higher" third, and
>>> therefore sound bad?
>>>>
>> No. That doesn't happen. That's also what makes it so interesting.
>> Apparently, the illusionary high third only appears when the three elements
>> (root, fifth and third) are in perfect JI tuning (2:3:5, 4:5:6 and possibly
>> other arrangements). So, when an actual high third is sung into a perfectly
>> tuned fifth (2:3) it doesn't produce a "higher" third. It simply "sounds
>> good."
>>>
>>> i can't think of any known or postulated psychoacoustic effect that
>>> would explain an observation like this.

Did you remember what I said earlier? It doesn't seem to happen when the
elements are not well-tuned, as when played on a 12t-ET keyboard. Don't you
think that's significant?
>>
>> Okay. So let's postulate one. I'm just telling you what I have experienced.
>
> not exactly -- you're talking about singing 'into' a heard pitch as
> if it takes place without hearing, via the same hearing mechanism,
> the sung pitch. this sounds like a logical fallacy. and it's in your
> *interpretation* of what you experienced -- the experience itself had
> to involve a single, coherent hearing mechanism.

Are you saying that one can't hear two things at the same time? The phrase
"singing in tune" implies adjusting one's voice to "fit" with another sound
source. That's what I did in this case. I repeat: when I sang "my version"
of the JI third with jerry00 there was no pitch difference in our thirds
when the root and fifth dropped out; but when I sang the high third (which
*also* tuned with the apparent high third being produced by the JI triad)
there was a strong dissonance when the root and third dropped out. Clearly,
I was singing a *different* frequency than the one sounding in jerry00.

Does it help to mention that the high third was "heard" in both cases? The
difference in the third I chose to sing was only evident by *intent* before
the jerry00 third entered and by dissonance and actual pitch difference
after the root and fifth dropped out.

What are we missing here? I did two different things which produced two
different results. Are you suggesting that I didn't? If so, what created the
dissonance in the second case? Interpretation? I don't think so.

>> Are Joe and I (and likely thousands of other "believers") simply mistaken?
>
> joe didn't offer this interpretation at all. he didn't even try
> singing into a heard pitch, or anything like that. thousands?

Okay. I give. That wasn't a rational statement. But it might prove
interesting if Joe (and others) tried it.

'Thousands' is obviously conservative. I probably went that way because I'm
surprised that you and others don't hear it. I've heard it all my life.
Perhaps *millions* of singers and string players "prefer" the high third
(which has been widely noted). I'm just curious about *why* they do.
Apparently, you are, too.
>
>> Interesting to me that you "remember" hearing the high third
>
> i didn't say that. i meant i remember hearing illusory pitch-shift
> effects along the lines of what you heard when the root and fifth
> dropped out.

Well, I tried to put words into your mouth. Damn. Didn't work.

I'm very much aware of the need to move pitches "on the fly" as harmonic
context changes. For example, the Bb in a well-tuned C7 chord wants to move
upward when the E is flatted ‹ apparently to accommodate the sounding of two
perfect fifths. I don't think that's what we are experiencing here.

In this case, a well-tuned combination of specific pitches corresponding to
the partial series seems to produce the illusion of a different pitch. Did
your early experience with illusory pitch shift require acoustically
well-tuned combinations? I think that's an important factor important here.
>
>> (Paul, why do my nice neat paragraphs turn into chopped liver when you
>> repost them on the list? What browser are you using?
>
> netscape or internet explorer -- it doesn't seem to matter. sorry
> about the mess.
>
All right. Since I'm retired now I guess I can afford the extra time to sort
them out. :-)

Jerry

--- In tuning@y..., Gerald Eskelin <stg3music@e...> wrote:

> >>> i can't think of any known or postulated psychoacoustic effect
that
> >>> would explain an observation like this.
>
> Did you remember what I said earlier? It doesn't seem to happen
when the
> elements are not well-tuned, as when played on a 12t-ET keyboard.
Don't you
> think that's significant?

it may be significant, but i still can't see how to explain it, aside
from categorical perception at work. more tests will help.

> >>
> >> Okay. So let's postulate one. I'm just telling you what I have
experienced.
> >
> > not exactly -- you're talking about singing 'into' a heard pitch
as
> > if it takes place without hearing, via the same hearing mechanism,
> > the sung pitch. this sounds like a logical fallacy. and it's in
your
> > *interpretation* of what you experienced -- the experience itself
had
> > to involve a single, coherent hearing mechanism.
>
> Are you saying that one can't hear two things at the same time?

sure you can -- but you use the same auditory mechanism to hear both
things. so any psychoacoustical effects which affect one will have to
affect the other as well.

> The phrase
> "singing in tune" implies adjusting one's voice to "fit" with
another sound
> source. That's what I did in this case. I repeat: when I sang "my
version"
> of the JI third with jerry00 there was no pitch difference in our
thirds
> when the root and fifth dropped out; but when I sang the high third
(which
> *also* tuned with the apparent high third being produced by the JI
triad)

this is what is strange. are you saying that if you sang anything
*between* these two values, it *wouldn't* fit with the chord, but
these two values (about 18 cents apart) both *do* fit with the chord?
if true, we're both very far from postulating any kind of reasonable
explanation for this. we should at least conduct more tests to verify
this (if you believe it's true). right now, i'm hesitant to believe
that the heard interval is a non-monotonic function of the actual
interval, or anything as 'twisted' as that.

> What are we missing here? I did two different things which produced
two
> different results. Are you suggesting that I didn't? If so, what
created the
> dissonance in the second case? Interpretation? I don't think so.

i think we're talking past each other. i'm not doubting that you sang
a lower and then a higher third. obviously, the higher third would
have produced beating against the .wav files' lower third. before we
get mad, let's remember that communication is not always that easy
here.

> 'Thousands' is obviously conservative. I probably went that way
because I'm
> surprised that you and others don't hear it. I've heard it all my
life.
> Perhaps *millions* of singers and string players "prefer" the high
third
> (which has been widely noted). I'm just curious about *why* they do.
> Apparently, you are, too.

i'm sticking with my categorical perception hypothesis for now,
essentially the same hypothesis i offered two years ago.

> >
> >> Interesting to me that you "remember" hearing the high third
> >
> > i didn't say that. i meant i remember hearing illusory pitch-shift
> > effects along the lines of what you heard when the root and fifth
> > dropped out.
>
> Well, I tried to put words into your mouth. Damn. Didn't work.
>
> I'm very much aware of the need to move pitches "on the fly" as
harmonic
> context changes. For example, the Bb in a well-tuned C7 chord wants
to move
> upward when the E is flatted ‹ apparently to accommodate the
sounding of two
> perfect fifths. I don't think that's what we are experiencing here.

no, it's sort of the opposite. perceiving a change when there really
isn't one, as opposed to necessitating a real change even when one
isn't perceived.

On 3/15/02 5:32 AM, "tuning@yahoogroups.com" <tuning@yahoogroups.com> wrote:

> Message: 9
> Date: Fri, 15 Mar 2002 07:22:30 -0000
> From: "paulerlich" <paul@stretch-music.com>
> Subject: Re: Jerries: a conclusion or two
>
> --- In tuning@y..., "paulerlich" <paul@s...> wrote:
>> --- In tuning@y..., Gerald Eskelin <stg3music@e...> wrote:
>>
>>> BTW, I have the time now (since I recently retired from my college gig) to
>>> do some reading. Do you remember the sources of your info about "categorical
>>> perception"? Who knows? I might even understand some of it.
>>
>> back at college, i used to sit in the library all day, every day,
>> reading this stuff -- psychology journals, acoustics journals,
>> as well as everything in the music library -- unfortunately, this
>> particular area was not one in which i retained many names and dates.
>
> let me give you an example of 'categorical perception' at work in a
> different context.
>
> the japanese language has no 'l' sound or 'r' sound, but rather a
> single sound in-between the two.
>
> people who grow up speaking japanese can't hear the difference
> between the english 'l' and 'r' -- they simply hear both as the
> single sound they're familiar with.
>
> it's extremely difficult to train native japanese speakers to hear
> the difference between what, to us, are completely different sounds.

Thanks, Paul. This helps me to know what you mean by the phrase. I wonder if
a speaker of Japanese would hear a difference (contrast, dissonance,
whatever) if he/she sustained the sound of the Japanese vowel while hearing
an American R or L. In our experiment, the focus is largely on simultaneous
sounds rather then simply recognizing different ones in isolation.

It's fairly obvious to me that we tend to "force" perceptions to fit into
whatever "concepts" we have collected over a lifetime. You know...the old
"all (pick an ethnic group) look alike" nonsense. That's likely what makes
it possible for many "musicians" to reduce everything they hear to 12t-ET.
That's also why many readers of my books have no frigging idea what I'm
talking about.

The "cure" then would be a matter of assimilating more and better "concepts"
with which to perceive the differences we're talking about. I think there
are a great many pitch-sensitive people out there who know they are hearing
something other than 12t-ET. Many of these send me emails telling me how
welcome my writing is to them. Very gratifying. What's more, I haven't found
the student yet who can't learn to hear good JI tuning. It's the
stuck-in-the-mud teachers who are frightened to death that they will have to
abandon their precious and well-worn course outlines and favorite textbooks
if they allow themselves to take me seriously.

But I digress. Are you suggesting that I (and evidently scores of others)
may have developed a categorical perception when it comes to hearing the
high third? If so, the "concept" by which I perceive it is a rather specific
one since the jerries vary in their closeness to it. I'll be interesting to
see if one of your next series of jerries "nails" it.
>
> here's a paper that mentions categorical perception and a bit about
> its musical implications for musicians vs. nonmusicians:
>
> http://cognitrn.psych.indiana.edu/rgoldsto/pdfs/perlearn.pdf
>
> here's an interesting paper that relates the possible effect of
> categorical perception in a different culture, one where musicians
> recognize five, rather than twelve, distinct categories within an
> octave:
>
> http://www.inform.umd.edu/EdRes/ReadingRoom/Newsletters/EthnoMusicolo
> gy/Archive/cooke-paper

I have downloaded these, Paul, and will get to them right away. Thanks.

Jerry