Mujitsu and Tairaku's Shakuhachi BBQ

Chris Moran wrote:

edosan wrote:

NOT ANY REPORTED DIFFERENT SEE EVEN THOUGH PLAYED HEAR.

Acrostically speaking, is this what you mean?

Yes. And furthermore: HEAR, HEAR!!

In the '70's when the famous "material doesn't matter" experiment was done, it was common for people be amused by how thier minds could be tricked to perceive things that aren't really there. It was pretty much common knowledge that psychosomatic illnesses existed due to a focus on the subject by popular TV shows. I'm not sure what the doctors thought back then, but these days I know for sure that doctors do not chalk up much as being psychological. If all the tests for conditions that would result in the symptoms turn out negative, they typically look for some other explanation, and wouldn't even think of suggesting it's all in the patients mind. So why, and in this day and age, would anybody doubt when almost every very highly trained musician says there's a difference in the sound when different materials are used.

When someone get's a double-blind set up using robotics or something and every highly trained musician tested can't tell the difference between music played on similar instruments with identical dimensions, that's when I'll start doubting the perceptions of the master players.

The fly in the ointment here is the seemingly innocuous phrase "identical dimensions". Let me tell you a little bedtime story about another famous materials experiment. Once upon a time, a famous player and professor of flute named Joan Lynn White conducted an experiment with the goal of answering the question of whether identical Boehm flutes in different materials sound and respond differently, and how much. For this she and the experimenters obtained five handmade Prima Sankyo flutes. Sankyos are extremely high-end flutes, and being handmade meant the highest possible dimensional consistency, as they are not simply stamped out of metal--very tight tolerances. They had one in platinum, one in 9K gold, one in 14K gold and two silver flutes.

The reason for having two silver flutes was as a control, one of those very commendable things about the scientific method. The idea was to establish a baseline for response differences based on small dimensional variations. Once the researchers established this, they would know just how much variation they might expect from the inevitable small differences in the bores of these supposedly-identical flutes, and take this into account when comparing differential responses.

Surprise, surprise! They found that the spectral differences between those two "identical" flutes was so great, as to completely obviate the possibility of measuring any difference due to the material. After very precise tests, they could only say that perhaps there was a small difference in the seventh partial in the 14K gold flute. However this partial is weak even in the lower octave, and completely disappears in the second octave, so they concluded that there is no functional difference in flutes due to the metal.

Now, the moral of this tale is as follows: if two metal flutes with completely cylindrical bores that do not vary more than +- .02 mm at most show such obvious differences, how can you possibly pretend that two shakuhachi with hand formed bores would not show a much more pronounced variation in response? How, given this fact, can you possibly be sure that the differences that you experience and hear are due to different materials and not to subtle differences in the bore?

Another thing: you claim that "almost every highly trained musician" says that there is a difference based on materials. How do you know, have you polled them? I know plenty who do not say this.

As far as highly trained musicians not being able to tell the difference--this is exactly what both Smith's experiments with trombones and the Linortner experiment with flutes demonstrated. The Coltman experiment which you so heavily criticize was especially clever. In case you have not yet found the time to read the paper: Coltman made three identical short heads in delrin and attached them to three different tubes--wood, copper and silver. He had a number of good musicians play these tubes. Most all the players had definite preferences for one or the other, and could describe the differences they felt. Then Coltman mounted all three flutes on a spindle that rotated, with the tubes enclosed so that the different textures couldn't be felt. Each player was asked to play one of the tubes, then the spindle was rotated and the player asked to find the same flute out of the three again. None could do so. If that isn't double-blind, nothing is.

I find it hard to believe that anyone would dimiss the observations of so many credible and brilliant musicians based on old, obviously flawed (not double-blind) experiments where the experimenter explained away anyone saying that they can hear that his theory is wrong by saying they are being misled by pre-concieved notions and that they aren't in actuality hearing any difference.
Excuse me, but you are setting up straw men here. First of all, if anything (and as I have pointed out in the fMRI experiment with wines), the understanding of the effects of the psychological on perception, experience and even physiology has broadened and deepened. Coltman's experiment has never been disproved, and remains a classic in the field after some 35 years. You say "obviously flawed" but this is disingenuous. Just how and which of the many peer-reviewed studies on this subject are "flawed"? I'm quite happy to discuss this, but we must try to avoid demagoguery.
I'm only referring to the often referenced and "famous" '70's experiment. Has this one been duplicated double-blind? The only way I can see doing that is with robotics that exist now. I'd also think it would be important to see it done with some really top musicians in the test, I mean best of the best, like Galway.

Quote from Galway: "As you probably have noticed I have never really recommended one metal over another when discussing flutes. The fact is that I cannot tell the difference when I am listening to someone play."

As mentioned, the Coltman experiment was certified double-blind. It is quite easy to use an artificial embouchure to eliminate unconscious player adjustment, and it has been done. Linortner's expeiment relied on a statistical technique, of using many players. As far as using top players--just how high do we have to go? Do we pick the musician with the best tone, or the best technique, or the most popular or the richest? Does that mean that merely professional musicians would not do?

The Smith experiment was quite interesting for one reason. First of all, he used really top professional trombonists, just to satisfy that requirement of yours. At first in the dark, the musicians were able to identify the different bells. Then Smith realized that they had different weights. People are extemely sensitive to small variations in weight, so he counterweighted the bells so that they all had the same heft. After that, all the musicians suddenly became unable to tell them apart! And this even though there were clearly measurable sonic differences between them that should have been perceptible.


As far as "credible" and "brilliant" musicians goes: There are plenty of credible and brilliant people in all fields who get it totally wrong. And again, no one is saying that these musicians are not sensitive to differences between instruments, only that--given the complexities--they do not have the means of analyzing and isolating the causes for the effects that they observe.

Research is continuing on the question of materials, and every new experiment seems to support the conclusion that it doesn't make a perceptible difference, to the point where this is the universally-accepted scientific consensus. The French experiment I described earlier is the latest in a continuing series of these. Again, I have an abundance of material at hand.
Well, my only real problem is with the famous '70's experiment you mentioned. I'm not familiar with the others. And I'll concede that the '70's experiment could be used as evidence that supports the argument that material doesn't matter. My problem is with his conclusion because the experiment is still inconclusive until you get rid of the problem that the player of the instrument knows what material it is made of, and so it can not be a true double-blind.

I think you really need to read the paper, because I think you have it backwards.

And please, don't give me the tired argument that musicians are more "sensitive" than scientists, or that human perception beats any machine yet devised. As I mentioned, in a study of the spectral differences between trombone bells, not one of ten top professional players could detect any difference between several bells in different thicknesses, even though there was an easily-measured 2 dB difference between their response at the position of the players' ears.

Those players, BTW, could easily tell the difference when the test wasn't double-blind ;^}
OK, those musicians couldn't tell the difference, even with the advantage that it wasn't a true double blind because the player the subjects listened to wouldn't necessarily play each instrument the same way. How "top" were these 10 "top professional" players? Were they equivalent to Galway in the trombone world (tippy top), or first trombonists in major orchestras (still pretty darn good subjects), or something less? It's a tired argument, as you say, but a good one. I pretty much know for a fact that I don't hear the things the same way as more skilled musicians.

Don't make no nevermind how "tippy-top" the musicians were. This is the exact same situation with our shakuhachi players. How "tippy-top" do our shakuhachi players have to be before we believe them when they say that they can feel a difference between materials?

It kind of sounds like my favorite definition of "scientific method": collect as many data points as you can, toss out the ones that do not fit your theory as "anomalous outliers" and forget about them.
I must say that I would not have expected such a luddite sentiment here. It shows how little you apparently know about how science is conducted. Every time you fly in an airplane (or ride in a car for that matter), among numberless other things,  you can thank the scientific method.
I know enough about science to know that it's less numbers crunching than common-sense experimentation (think myth-busters), and I know that at least the famous '70's experiment wasn't double-blind (understandable, the experimenter didn't have the technology because it didn't exist) and double-blind is crucial to weighing psychological influences.

If you think that myth busters is modern science then there is not much more for us to discuss.  My brother has a PhD in high-speed particle physics from Cal Tech and Stanford. He probably had no more than a couple of hours to run his experiments on the Stanford Linear Accelerator, and the rest of the 9 years was crunching numbers.

Take a closer look at your excerpt:
"...the musician cannot, under normal playing circumstances, dissociate his personal preferences and prejudices from the question at hand...
Even if the experiment was reconducted with the flute-playing robot (double-blind) and the results were the same, how did the experimenter determine that the psychological problem was an inabiltiy to dissociate personal preferences and prejuduces from the question?

...In the case of the three 'flutes' I constructed, nearly every player who picked them up and tried them had a preference for one or the other. Often he would describe his impressions - the wooden flute has a 'fuller' tone, the silver one 'projects' much better, etc. He was then usually baffled to find that he could not identify any of the instruments under the 'blindfold' conditions I described.
OK, I see, the musicians played the flute and heard a fuller tone, or it projected better. Then, when someone played it back for them without them they couldn't hear the difference. Seems to me that it could be that there was some vibration in the material that caused differences in bone conduction that resulted in the perception of a fuller tone, or maybe some other reason I'm missing right now. Oh, but wait... that's just my luddite perspective, the scientific perspective is that the musician has an inability to drop his preconcieved notions and listen objectively. yea, right...

The plain facts are that his judgment is influenced by preconceived notions and mental associations of tone quality with other properties of the material.
Yup, that was just a plain fact. How foolish of me.

This is a normal human reaction, intensified in the case of those trained to incorporate feeling into their art, and to whom the instrument becomes, in effect, an extension of their own body and personality. It is just not suited for answering narrow, objective questions like the one I posed - namely: can the material of which a flute is made directly influence the tone quality produced? "
Why the heck is it this guy is so set against believing these musicians actually heard a difference when they tried the instruments? They said they heard the difference, it's just that they didn't hear it when it was played back to them by another player. Did this other player also believe that there was a difference and tried to play fuller? Or maybe the other player tried to play it more blandly. Or maybe the experimenter played the test subject's trial back to him on a tape recorder and he didn't hear the difference anymore, then the difference could've been lost due to the recording technology. It's kind of obvious the double-blind is neccessary to give this experiment some real credibility.


You've really got to wonder about the "famous" experiments of that era. Another one is the Milgram authority figure experiment where an actor pretended he was in pain as an authority figure instructed subjects to shock the actor with a fake device they were told would inflict pain. The wonderous thing about this experiment is that the subjects almost without fail would follow the authority figures orders to inflict pain. The experimenter goes on to use the results to suggest how easily a nazi-germany like situation could happen again when people will hurt other people so easily just because an authority figure tells them to. He may have a point, by why the heck wasn't everyone in the scientific community jumping up and down crying foul because the subjects might have at some level (maybe even unconsious one) perceived that those were the screams of an actor, and probably a student actor at that, and not of a person actually getting hurt. Not that I think they could have ethically done much better with that experiment, as it is the ethics of deluding innocent subjects into thinking they were hurting someone has been heavily debated, but in my opinion the results have that much less credibility the way the experiment was conducted.


So, Toby, show me the link where this experiment was done properly, with a flute playing robot for a true double-blind, normal materials like nickel-silver, sterling silver, and gold, with top musicians like Galway or Paula Robeson as subjects, and I'll readily believe that there is absolutely no difference in the sound the an observer hears in those materials. Wood and bamboo would be more difficult due to varying dimensions (hard for the robot to do the exact same thing), but the results of the metals that are pretty much commonly believed to produce different sound in the flute world would go a long way to help your argument.

You really need to read the Coltman and Linortner papers so that we can get back on track here, because what you ask for is exactly what has been done. You have misunderstood both the premise and the methodology of the experiments.

Toby

Last edited by Toby (2009-12-22 10:33:58)

fouw wrote:

Scientific method says: don't take seriously a person who quotes Madam Blavatsky

Any better now?

Toby wrote:

You really need to read the Coltman and Linortner papers so that we can get back on track here, because what you ask for is exactly what has been done. You have misunderstood both the premise and the methodology of the experiments.

Toby

I just read the Coltman papers. I see what's going on. In his scientific paper part of his conclusion is "Of course, it is possible that individuals exist whose discriminatory senses are keen enough to find a distinction, but if so, they are certainly not common." All well and good in my opinion, and a good scientific conclusion. It turns out that his comment about it just being fact that musicians can't dissociate from their preconceived notion was from a letter written to Woodwind World magazine in response to some article where a very unscientific writer went about answering the material/tone question by polling musicians to see what they said. The context isn't scientific, Coltman expressed his non-scientific opinion there and it was possibly exagerrated some by anger since the method of polling musicians probably was appalling to Coltman who must have been proud of the excellent work he did.

I don't think it's a stretch at all from Coltman's conclusion to entertain the notion that there are some rare individuals out there who truly can perceive a difference. He only tested 27 subjects, I think you could test that many musicians for perfect pitch and find that not one has it. Perfect pitch is rare, many who look like they have it really just honed their relative pitch and remember how a certain note sounds on their instrument and compare it. But it's pretty well accepted that some people have perfect pitch. I don't see why it is so hard to believe that some people may also be able to discern the very small tonal changes associated with material just because I can't nor have met anyone who can. Same thing with perfect pitch, I've never met anyone who has it, when I thought I did once he said he didn't have it and told me about the relative pitch trick. There's no tricks to hearing small tonal differences due to material. All Coltman would need to do is keep testing with his device until he finds someone who can score consistently better than random or gives up at some high number where you can say that if this special person is so rare he or she is not worth looking for. He's probably given up already because he's happy enough having shown that if there is any difference most people can't hear it.

So, I'm happy. Are you?

Last edited by radi0gnome (2009-12-22 17:56:22)

Toby wrote:

As long as you are on a roll, you might consider perusing the Linortner flute study. Main points to observe are the extremely minor differences in the response of the flutes, the very large differences between the sound of the different players, and the fact that nobody listening could tell which flute was being played.

Toby

OK, I read it. You know, I'm really not surprised that there's a lot of evidence that flute material, particularly common metal flute materials, do not make a whole lot of difference. And apparently it's such a small amount of difference that a whole bunch of people, who often think they can hear a difference, can't perceive the difference in a blind test.

I can see where people can be easily deluded into thinking they can hear a difference. There are a lot of people in the flute world that say it does make a difference, and it can be tough to be the guy who says he can't hear it.

However, even if it's just because of what you mentioned about not being able to prove non-existence, there is still the possibility that the at least some of the highly respected teachers and individuals that claim they can hear a difference are not deluding themselves and can actually perceive a difference. Keep in mind too that bamboo is a lot softer than any of the metals used in the experiments and adds a bit more credibility to those in the shakuhachi world who maintain that there's a difference. I'd suggest to those that are 100% convinced by the science (which does leave open the possibility that some rare individuals can tell a difference) to keep your chuckling to yourself when someone with obvious skills say they can hear a difference. They just might really hear it.

edosan wrote:

radi0gnome wrote:

However, even if it's just because of what you mentioned about not being able to prove non-existence, there is still the possibility that the at least some of the highly respected teachers and individuals that claim they can hear a difference are not deluding themselves and can actually perceive a difference. Keep in mind too that bamboo is a lot softer than any of the metals used in the experiments and adds a bit more credibility to those in the shakuhachi world who maintain that there's a difference. I'd suggest to those that are 100% convinced by the science (which does leave open the possibility that some rare individuals can tell a difference) to keep your chuckling to yourself when someone with obvious skills say they can hear a difference. They just might really hear it.

[Shakes head, hangs head, places head in palm, shakes head again, slaps head sharply with palm to restore equanimity. Doesn't help....]

If that's your reaction to me, who never professed to be able to hear a difference, what's your reaction to those who have said they can hear it?

edosan wrote:

radi0gnome wrote:

edosan wrote:

[Shakes head, hangs head, places head in palm, shakes head again, slaps head sharply with palm to restore equanimity. Doesn't help....]

If that's your reaction to me, who never professed to be able to hear a difference, what's your reaction to those who have said they can hear it?

My reaction is simply to the fact that you somehow manage to miss the entire point of the discussion.

Then what was the point? I thought it was here in post #19 when Toby wrote on 11/17

Toby wrote:

I know that no one into the mysticism of it all wants to hear this, but it is completely, 100% about bore profile and smoothness. Sound is a physical quantity, and as such is determined by physical parameters. Like it or not, the bamboo itself doesn't vibrate enough to influence the sound in any perceptible manner.

I took exception to the premise when Toby pointed out how Dr. Coltman made a statement about musicians not being able to drop their preconceived notions. I felt that Toby was suggesting that anyone who still believed otherwise should just bow down and give into the science, it certainly doesn't sound like he was leaving room for error. The problem was definitively stating that material does not influence the sound in a perceptible manner and using science to back it up when the scientific paper admits that there may be individuals who can hear the difference. Toby conveniently didn't mention that aspect of the scientific paper's conclusion, although he did point out when the experimenter expressed his doubt's that anyone could hear a difference in a non-scientific letter to a woodwind magazine as an angry response to some very poor science that appeared in the publication.

And I'm curious, what is your reaction to those who still claim they can hear a difference despite all the evidence Toby has just presented them in this thread that they should not be able to? Or maybe that's a moot point because you two converted them all and now they see the error in their ways.

It's okay. Those of us who can hear a difference are clearly delusional.

Those who can hear need to fall in knees...

Those who can not, need to fall in knees and bit? No they hear along all body...

Anyway, in resume, lets ask everybody three features that can make the timbre better...



I will not say i will not change my opiniom... but i would say:


Utaguchi
Bore Shape
Holes shape


I think that there are others...

Just sitting back and observing the hawsers that bind coming from threads of interest.

thanks Toby. Lots of good info!