The Clarinet BBoard

Re: vibrato of clarinet

Author: Jack Kissinger  Date:   2011-06-29 06:17 Buster, In fairness to Mr. Fujita, I don't think that he is saying that he agrees with Jonathan Cohler or FTM any of the other posters he quotes. In fact, he doesn't even quote Cohler. He quotes Chris Ondaatje's evaluation of Cohler's article. My interpretation is that he is not citing these comments to support a position for or against vibrato. (In fact, I don't think he has taken a position one way or the other on its artistic merit.) As I interpret him, he is simply citing these various messages as evidence that the use of vibrato in clarinet playing is controversial. I, for one, and I suspect most of the rest of us would be willing to concede that point as common knowledge but he is trying to be thorough in providing evidence that the controversy exists or perhaps he has misunderstood earlier criticisms of his comments that, in turn, misinterpreted what he was trying to say. It's a vicious cycle not helped by the language difference. You criticize his math continuously but I think your criticism misses the mark. His hypothesis is that there are certain characteristics inherent in the design of a clarinet (closed cylinder) that make vibrato more difficult to produce on the clarinet than other woodwind instruments. This means players must work harder to produce (pleasing) vibrato on the clarinet than, e.g., on the flute, oboe or saxophone. To develop theory to support his hypothesis, he defines a mathematical model of the clarinet then manipulates this model logically (mathematically). Note that there are two issues here: (1) is his derivation correct within the parameters of his model, i.e., has he made any logical errors in the derivation [internal validity], and (2) is his model a sufficiently accurate representation of real-world clarinets to warrant claiming that his analysis is applicable to real-world clarinets (external validity). The first pass in evaluating his theory should be, IMO, to verify the internal validity of his derivations without regard to whether his model represents a real-world clarinet. The math doesn't look outrageously difficult but I can't evaluate it because Mr. Fujita has been a bit lazy in that he hasn't defined his variables. He may be using standard notation from the fields he incorporates but, since I have never paid much attention to those fields, I don't know what the variables all represent. If he wants to present his work to people who don't have background in his field, IMO, he has an obligation to define the variables. You have made the point that there are errors in his math, that certain operations he has performed are incorrect but, it seems to me, that you are saying they don't apply because his model is not an accurate representation of the clarinet. If that's the case, I would submit that you are really saying that his operations are incorrect because they don't apply to your ideal model of the clarinet. That criticism confuses internal and external validity. On the other hand, if there are errors in his derivation, those need to be pointed out. So far, though, I haven't recognized any of his critics' addressing that issue. I will say that I find his attempt to apply wave particle duality to sound waves novel and I have to wonder if it's valid considering the magnitudes (relative to light and atoms) involved. As for Heisenberg's uncertainty principle, I think it provides an explanation for the difference in ensemble between professional and community orchestras. We can never know precisely both where the conductor's baton is in a beat and the baton's velocity (tempo). Professionals are simply better at measurement. Internal validity problems would allow us to reject Mr. Fujita's findings out of hand but, as I've already said, I don't think anyone has found any of those yet. (But keep looking.) That brings us to the second question and the one that I think is really bothering most of the respondents to this thread: Is Mr. Fujita's model a sufficiently accurate representation of a real-world clarinet to allow generalizations from his model to a real-world clarinet. I think it safe to conclude from your comments that you do not think it is and your primary evidence is the difference between real clarinet wave forms and the wave form of Mr. Fujita's "ideal" clarinet. You may be right. Mr. Fujita has largely based his model on what I might call first order conditions (I'm probably not using the term entirely correctly but I hope you get my point). To obtain a more accurate representation, he needs to incorporate higher order conditions. Those might change the outcome. On the other hand, maybe you are using the wrong criterion. All models are abstractions from reality. The model builder selects those parameters that s/he thinks are important but, in the interest of tractability (and sometimes simple ignorance), excludes other potentially important parameters -- sometimes many more than are included, sometimes critical ones. But even a simple model can sometimes give us insight into the real world. For example, the model of an "ideal" clarinet gives useful information on which harmonics should be present and the fact that opening the register key will produce a 12th instead of an octave. Newtonian physics is right 99.9% of the time (OK, I made that last statistic up). So I can't rule out the possibility that Mr. Fujita's finding is also valid and, potentially, useful. In any case, I think most of the people who have responded rather harshly to his comments have been unfair to him -- though the fact that English is not his first language hinders his ability to express himself clearly makes it easy to misinterpret what he is saying. As I read his comments, his main points are: 1. Whether or not vibrato is appropriate on clarinet is a controversial issue. The controversy hinges on two points: (1) whether "good" vibrato has any artistic merit, and (2) whether it's possible/easy to achieve "good" vibrato on the clarinet. 2. Mr. Fujita has no opinion on the first point. None. Zip. Nada. On the second point, he concedes that "good" vibrato is possible (based on reports/observations of some clarinetists who have been able to achieve it) but concludes, based on the relative (in)frequency of such clarinetists, that "good" vibrato may be more difficult to achieve on the clarinet than other woodwind instruments. 3. Observing that "good vibrato" seems to be more difficult to achieve on the clarinet, he develops a perhaps novel theory to explain part (but not necessarily all) of the difficulty based on the inherent characteristics of the clarinet. If anyone is still with me (wait a minute while I don my flame-retardant underwear), I respectfully submit that those of you who have attacked Mr. Fujita on the grounds that vibrato does or does not (take your pick) have artistic merit have shifted the discourse from why he thinks vibrato is more difficult to produce on the clarinet to your own views of the value of vibrato. This is probably not entirely your fault. Mr. Fujita's efforts to explain himself in English are sometimes hard to interpret. And we all have our own maps. Those who have argued that "good" vibrato is possible and cited examples of players who have achieved it aren't actually in disagreement with him (though, in fairness, I think that some of you have brought him around). Mr. Fujita is not (or at least is no longer) saying that "good" vibrato is impossible --just difficult so we should recognize and celebrate the abilities of clarinetists who have been able to achieve it. The comment that clarinetists can't achieve "good" vibrato because they don't practice it enough is totally consistent with his theory because, according to that theory, achieving "good" vibrato on the clarinet should require even greater effort (practice) than on other instruments. As for the physics, I would love to see someone verify or refute his derivations but without the confusion between internal and external validity. Of course, assuming the derivations are correct, there is still a significant piece of the puzzle missing -- verification of the theory through application in experiments involving real clarinets. Without those tests the theory remains in the theory plane with no verified real-world application. One final point. I have checked and double-checked the Bulletin Board rules and NOWHERE do I find any requirement that threads begin with a question. Best regards, jnk