Re: Re: Re: Re: cryogenic tuba freezing


[ Follow Ups ] [ Post Followup ] [ TubeNet BBS ] [ FAQ ]

Posted by Rick Denney on February 02, 2001 at 11:41:29:

In Reply to: Re: Re: Re: cryogenic tuba freezing posted by Cryotuba on February 02, 2001 at 01:23:48:

With all due respect, I am an engineer and scientist and therefore a great believer in cause and effect. There are two ways to approach science. One is to formulate a model based on a clear understanding of cause and effect, and design according to that model with careful testing along the way to make sure the model considers all factors. The other way is to just try things and measure the results. You are advocating the second approach. For it to work, you have to rigorously control for other effects, especially if your subjective response to the horn is the effect being measured. A scientific approach that could lead to general advice (i.e. "All tuba freezing is good") would require a study where a number of identical instruments would be play tested, and found, through statistical analysis, to be identical. Then, some of the instruments would be frozen, and others not. Then, the testers would play all the horns and rate them on criteria hypothetically affected and not affected by freezing, without any knowledge of which horn was frozen and which not or which test is relevant and which not. In fact, the persons administering the test should also not know which is which, but should identify the horns by an unrelated code that is resolved into the frozen and unfrozen samples after the testing by different people. This eliminates potential bias. Then, you run statistical analysis on the ratings and show that the sample of frozen horns has a significantly different rating than the sample of unfrozen horns. By "significantly different" I mean different enough not be attributable to chance, for which there are a number of tests that are the stuff of life to statisticians.

Here's the important point: Even if you showed a correlation in playing characteristics between the frozen and unfrozen samples, you'd have to explain why. You cannot show correlation without explaining the cause-and-effect mechanism. This is carved on the forehead (or should be) of freshman statistics students.

That's what nobody has done in terms understandable to anyone with knowledge of materials. Instead, the descriptions of the process are couched in technical-sounding terms that are meaningless to those who understand such terms but impressive to people who don't. There is a scientific term for this: It is called a "snow job."

Now, if I've been snowed into believing that it works, then it is entirely likely that it will work for me, at least in the short term. Why? Because so much of playing an instrument is perceptual, and is filtered through our expectations. I can't play that high G in tune, because I've never been able to before. But I add a gadget that I've been persuaded will make that high G sing, and suddenly discover that I can play it. What has changed? Maybe the gadget works, maybe not. After a month or two, I might miss the note enough times to allow those old doubts to creep back in, and I settle back to my former expectations. This process seems to be a more likely explanation of the results of freezing than any "technical" explanation I've heard.

This process happens all the time in all aspects of life, which is why those of us with extensive technical training and experience tend to be the most skeptical of processes that are poorly explained. For example, I've written articles on bicycle technology for Triathlete Magazine debunking all sorts of myths, such as the myth that aluminum is stiff and unyielding (when it is less stiff than any other metal used in bicycles). But tell that to cyclists who have been persuaded that aluminum bikes are "harsh!"

But you cannot tell those of us who are skeptical that we cannot comment on the process if we have never tried it ourselves. I could turn that around and say that those who have spent hundreds of dollars in freezing their tubas have been persuaded of its effects thoroughly enough to make an investment in that belief, and therefore are the least objective to make an evaluation.

Rick "not planning to freeze his tuba" Denney


Follow Ups: