Jump to content
Maestronet Forums

Double blind test of Strads and Guarneris


pahdah_hound

Recommended Posts

Research from the National Academy of Sciences finds it tough for expert violinists to tell fine old Italians from modern instruments based on sound.

http://www.npr.org/blogs/deceptivecadence/2012/01/02/144482863/double-blind-violin-test-can-you-pick-the-strad

For those who don't like links, here is the text of the NPR article. On the site, there are also sound clips.

"In the world of violins, the names Stradivari and Guarneri are sacred. For three centuries, violin-makers and scientists have studied the instruments made by these Italian craftsmen. So far no one has figured out what makes their sound different. But a new study now suggests maybe they aren't so different after all.

OK, here's a test. Clip one is a musical phrase from Tchaikovsky's Violin Concerto in D Major. Clip two is the same phrase. The same musician plays both. But one is on a Stradivarius violin, the other on a violin made in 1980. See if you can tell the difference.

It's a tough choice. But a professional violinist could tell the difference, right?

Well, a research team recently tried to find out. They gathered professional violinists in a hotel room in Indianapolis. They had six violins — two Strads, a Guarneri and three modern instruments. Everybody wore dark goggles so they couldn't see which violin was which.

Then the researchers told the musicians: These are all fine violins and at least one is a Stradivarius. Play, then judge the instruments.

Joseph Curtin, a violin-maker from Michigan, was one of the researchers. "There was no evidence that people had any idea what they were playing," he says. "That really surprised me."

Curtin says of the 17 players who were asked to choose which were old Italians, "Seven said they couldn't, seven got it wrong, and only three got it right."

Claudia Fritz designed the experiment. She's an acoustics physicist from France's National Center for Scientific Research — and a flute player, by the way. She says this test was more rigorous than previous ones because it was "double-blind" — no one knew which instrument was which until after the test. That rules out the kind of bias that might creep in when a musician judges an instrument he or she knows is 300 years old and maybe played by someone like Fritz Kreisler or Henryk Szeryng.

And this experiment asked seasoned violin players, not listeners, to choose.

Fritz says some of the players told her they were certain which were the new violins and which were the old Italians.

"'Ah, it's just a bit too new for me'," she recalled one musician saying. "And it was a Strad. Another one said, 'Ah, I love the sound of this one, it really has the sound of an old Italian, ah, just so warm.' And it was a brand new violin. "

When Fritz asked the players which violins they'd like to take home, almost two-thirds chose a violin that turned out to be new. She's found the same in tests with other musical instruments. "I haven't found any consistency whatsoever," she says. "Never. People don't agree. They just like different things."

In fact, the only statistically obvious trend in the choices was that one of the Stradivarius violins was the least favorite, and one of the modern instruments was slightly favored.

Now, what does that mean for all the years of research studying the old violins — the design, the wood, the varnish, even the glue? If no one can tell the difference, what's the point?

Well, Fritz says maybe researchers should focus more on people than old wood.

"People looked at the violin, tried to understand how it vibrates, what are the mechanics behind it," she says of past research. "But nobody has really looked at the human side." She says her research is aimed at determining how people choose what they like, and what criteria they use.

Curtin has spent years trying to capture the quality of old-world instruments. But he's not discouraged by the results. "If new violins get better, it doesn't mean old ones get worse," he says. "The question is, Can the sound be gotten from a new instrument, as well as an old one?"

The old Italians certainly sound great, but not necessarily better or even that different from the best new ones, he says. It's more in the mind, or ear, of the listener.

Dale Purves, a professor of neuroscience at Duke University, says the research "makes the point that things that people think are 'special' are not so special after all when knowledge of the origin is taken away."

The research appears in the Proceedings of the National Academy of Sciences.

Related NPR Stories

CT Scans Re-Create 307-Year-Old Violin Nov. 30, 2011

How To CAT-Scan (And Hot-Rod) A Stradivarius Nov. 9, 2008"

Link to comment
Share on other sites

++++++++++++++++

Quoted " Dale Purves, a professor of neuroscience at Duke University, says the research "makes the point that things that people think are 'special' are not so special after all when knowledge of the origin is taken away." "

It is a good point to make.

Link to comment
Share on other sites

Research from the National Academy of Sciences finds it tough for expert violinists to tell fine old Italians from modern instruments based on sound.

Now, what does that mean for all the years of research studying the old violins — the design, the wood, the varnish, even the glue? If no one can tell the difference, what's the point?

It means that all those years of research, etc., etc., have been worthwhile & have produced great results. The point is: That no one can tell the difference.

Link to comment
Share on other sites

It means that all those years of research, etc., etc., have been worthwhile & have produced great results. The point is: That no one can tell the difference.

Excellent point. Please understand that I have quoted the article in its entirety and neither the comments nor opinions are mine. But how many world class professional violinists choose to play a newer violin-especially in concert? It is as much about the status of owning and using something old, rare and fine as it is the tone produced.

Jesse

Link to comment
Share on other sites

But how many world class professional violinists choose to play a newer violin-especially in concert? Jesse

The Emerson String Quartet has been performing and recording using modern instruments by Sam Zygmuntowicz. Christian Tetzlaff performs on a modern violin by Peter Greiner. I'm sure there are many professional orchestra players who use modern instruments. Wasn't that one of the reasons for antiquing? So that a player could use a modern instrument without prejudice? Another reason why many professionals use modern instruments is cost, of course.

Link to comment
Share on other sites

She's found the same in tests with other musical instruments. "I haven't found any consistency whatsoever," she says. "Never. People don't agree. They just like different things."

...

"People looked at the violin, tried to understand how it vibrates, what are the mechanics behind it," she says of past research. "But nobody has really looked at the human side." She says her research is aimed at determining how people choose what they like, and what criteria they use.

The part I've quoted is the one that says everything. I wonder whether she's aware that what she's researching is Love and its drivers, or that there's been quite a lot of pretty solid research done in that area already.

Love -pretty much regardless of its object- expresses a concatenation of needs, desires, and perceptions, and a lot of post-hoc rationalisation goes on in which we "explain" why we love this person/fiddle/house rather than that other one. When sources of information are blocked off, so that we have to do in fact what we claim we always do (e.g., judge a fiddle by its sound, playability, etc), what we reveal about ourselves can be quite embarrassing.

(edit: corrected a typo)

Link to comment
Share on other sites

The part I've quoted is the one that says everything. I wonder whether she's aware that what she's researching is Love and its drivers, or that there's been quite a lot of pretty solid research done in that area already.

Love -pretty much regardless of its object- expresses a concatentation of needs, desires, and perceptions, and a lot of post-hoc rationalisation goes on in which we "explain" why we love this person/fiddle/house rather than that other one. When sources of information are blocked off, so that we have to do in fact what we claim we always do (e.g., judge a fiddle by its sound, playability, etc), what we reveal about ourselves can be quite embarrassing.

Very profound indeed. I have found a great deal of variation in taste in fiddles among even top professional performers. In fact, it is common among performers to hear such things as "John's Storioni sounds thin and muddy" or Michael's Vuillaume is so strident and shrill". The choices those players make in their instruments are very different and there is little consensus among the most highly trained ears as to what constitutes "good".

However, taste changes over time, develops and sometimes improves. I wonder whether a top performer who has played the same Strad for 25 years could pick out another Strad(of the same period and model, with identical set ups)by sound alone, from a group of similar quality violins. I would speculate that the expert's average will be no better than anyone else.

Jesse

Link to comment
Share on other sites

Doesn't NPR's report of the research results contradict the conclusion that differences are not discernible?

If three instruments were selected entirely at random from the six provided, without respect to order, there are twenty possible combinations, only one of which includes all three Italian masters. The probability of one person correctly selecting all three is therefore only .05, but 3 of the 10 participants who attempted to identify all three italians did so correctly (7 of the 17 opted out).

Am I missing something or is NPR? (I may be!)

Link to comment
Share on other sites

 

Doesn't NPR's report of the research results contradict the conclusion that differences are not discernible?

If three instruments were selected entirely at random from the six provided, without respect to order, there are twenty possible combinations, only one of which includes all three Italian masters. The probability of one person correctly selecting all three is therefore only .05, but 3 of the 10 participants who attempted to identify all three italians did so correctly (7 of the 17 opted out).

Am I missing something or is NPR? (I may be!) 

Ten is a very small sample size, and self-selected, so it's not impossible that 3 of the 10 would get the identifications right by chance or near-chance. And even if all three did actually recognise the fiddles, that doesn't really impeach the conclusion that most expert players cannot.

Link to comment
Share on other sites

Ten is a very small sample size, and self-selected, so it's not impossible that 3 of the 10 would get the identifications right by chance or near-chance. And even if all three did actually recognise the fiddles, that doesn't really impeach the conclusion that most expert players cannot.

The soundness of the methodology isn't really relevant to my point. The research results as reported do not support the conclusion in the NPR article.

As a matter of disclosure, I have no opinion in the underlying debate.

Link to comment
Share on other sites

i have noticed a lot of ignorance of statistics and probability on this forum, while it is not impossible it is extremely unlikely that three people out of ten would pick the old instruments entirely by chance, in fact the low probability of it happening by chance almost proves that it wasnt chance and that 3 out of ten could indeed tell the old from the new, this being most likely the case, it makes the authors conclusions both absurd and patently unscientific, as who should be fully aware of the laws of probability as an acoustics professor from the university of paris

Link to comment
Share on other sites

i have noticed a lot of ignorance of statistics and probability on this forum, while it is not impossible it is extremely unlikely that three people out of ten would pick the old instruments entirely by chance, in fact the low probability of it happening by chance almost proves that it wasnt chance and that 3 out of ten could indeed tell the old from the new, this being most likely the case, it makes the authors conclusions both absurd and patently unscientific, as who should be fully aware of the laws of probability as an acoustics professor from the university of paris

I'm sure I'd be shortlisted for "Least Likely Ever To Be A Useful Consultant On Numerical Questions", but consider what it means that the sample -the 17 professionals- was

  • tiny,
  • biased to begin with (they were invited) and
  • further biased to go on with (they accepted).

So right away there's this big presumption going in that if anyone can do it, they can.

The fact that only ten felt confident enough even to try, and only three succeeded suggests strongly that, in general, people can't.

No, it's not publishable, you're right about that. But it's good enough to win bar bets, which is a better measure of real-world validity.

Link to comment
Share on other sites

I'm sure I'd be shortlisted for "Least Likely Ever To Be A Useful Consultant On Numerical Questions", but consider what it means that the sample -the 17 professionals- was

  • tiny,
  • biased to begin with (they were invited) and
  • further biased to go on with (they accepted).

So right away there's this big presumption going in that if anyone can do it, they can.

The fact that only ten felt confident enough even to try, and only three succeeded suggests strongly that, in general, people can't.

No, it's not publishable, you're right about that. But it's good enough to win bar bets, which is a better measure of real-world validity.

Hello Everyone,Just an Observation; The one problem I have with this study is in the use of "professional players". I am a retired professional and even though having played on many great instruments I would never consider myself qualified for a test of this type. One may be a professional baseball player batting just over 200, or a really good player hitting a constant 300. So what happends if the professionl violinists used are all in the 200 range? What should be done in a test such as this is to find only the top ranked Concert Violinists who are familiar with concert violins as these violinists would be the only ones qualified to judge. I have heard 100s of professional violinists in many different settings and one might just be very surprised if you heard them in a solo setting trying to judge tone and the difference between old and new. The fact is most professional violinists do not know how to produce a really outstanding tone or even a really good tone. They may have the technique to play great works, but that is just about where it ends. I am not trying to offend anyone and am sure I will receive much criticism for my opinion, but this is what I have found to be true for over 50 years. OT

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.


×
×
  • Create New...