Klarinet Archive - Posting 000214.txt from 2005/04

From: "Lelia Loban" <lelialoban@-----.net>
Subj: [kl] Test results
Date: Tue, 12 Apr 2005 11:12:37 -0400


I wrote,
>However, after seeing what was in the test, I don't
>think it could accomplish the stated purpose, because
>there are way too many variables *besides* the
>players' nationalities. The musicians play composers
>who write in vastly different styles, who come from
>all over the globe, and who live in different historical
>periods, too. Some of the players share their composer's
>nationality; others don't. The size and type of the ensemble
>varies. Moreover, as some people have mentioned on the
>list, volunteers downloaded these files on miscellaneous
>computers with miscellaneous software and then listened
>on miscellaneous good, bad and indifferent sound systems.
>
>Dan, I hope you try this test again under more controlled
>conditions, at a ClarinetFest, for instance, where you can
>sit volunteers down all together in a room and play them
>a tape of samples. If what you want to know is simply
>whether or not trained listeners can distinguish the nationality
>of a clarinet player, then I suggest that you eliminate a lot
> more variables by choosing samples of different clarinet
>players all playing the same passage. One brief passage
>from the Mozart concerto would be good, because
>that's the piece that just about everyone has recorded.

Dan Leeson wrote,
>>Lelia, I see your comment as 20-20 hindsight. The
>>argument that was under examination was the a national
>>sound existed. German players were capable of being
>>identified by the German national sound character, and
>>it did not matter what they played. da capo ad nauseam
>>for American, French, and English players.
>>
>>Now you change the rules that you had no hand in setting
>> up and that were unambiguously specified. Of course the
>>players play the music of composers who wrote in vastly
>>different styles.

As you say, I had no hand in making the rules. At the time you organized
the test, you didn't seem very open to suggestions, although back then, if
you recall, several of us did raise some points that you chose to ignore.
In any case, I can hardly ask to change the rules now that the test is
finished. No, I'm suggesting you do a different test with different rules,
because the test you did can't yield results that either support or refute
the thesis that there are national styles of playing. My complaint is
simply that your test doesn't follow scientific method.

The problem with hearing all these musicians play all sorts of different
compositions is that one can expect *any* musicians to play Mozart
differently from the Blank Concerto. It tells me a lot about two players
if I can compare how they both play the same passage of Mozart or Blank.
It tells me much less if I hear them interpret such radically different
compositions, because I have no way of knowing what stylistic differences
in the performance result from the players' perceptions that these
composers *require* different styles of playing versus what differences
only from the muscians' own personal, ethnic or national styles and thus
might carry over from composer to composer. Hearing David Niethamer play
Blank can tell me very little about how he would play Mozart, let alone how
his Mozart would compare with a German clarinet player's Mozart. For all I
know, David might use an American style for playing Blank and a German
style, with a different instrument, for playing Mozart. In fact, I'm
fairly sure he wouldn't record the Mozart concerto on his bass clarinet!
That's what I mean by "too many variables." I would need to hear whether
or not nationality seems to create a *consistent* approach to music in
general: Do most or all of the German players do something similarly, that
players from other countries don't do? In order to know that, I'd have to
hear everybody trying to play the same thing.

Would watching me eat an apple tell you anything about how I eat spaghetti?
Not really, because spaghetti and apples require different eating
techniques. At most, you might be able to determine that I'm either a slob
or a neatnik.

If you watched me eat spaghetti and watched Tony Pay eat an apple, would
you expect to be able to determine whether there are English and American
styles of apple-eating and spaghetti-eating? From watching me eat
spaghetti, would you be able to predict accurately whether I might eat the
apple whole, including the stem and the core? Would you be able to
predict, on the basis of how Tony eats an apple, whether he would be more
likely to use a spoon to roll spaghetti around a fork, or would he be more
likely to take the cut-and-shovel approach? If you can't even predict how
these two individuals would behave with different food, then you certainly
couldn't extrapolate how other people of their nationalities would behave.
In order to succeed with this type of reasoning, you would need, for
instance, to watch a number of different English people eat apples, to
determine whether or not English people generally eat the apple core. You
would need to watch a number of Americans eat apples to determine whether
or not we generally eat the core. Then you would need to watch the same
people eat spaghetti. Then and only then you might be able to determine
that core-eaters tend to be spaghetti-rollers (or not) and that English
people are more or less likely than Americans to eat cores and/or roll
spaghetti.

>>Comments were made that the more astute listener
>>could detect the nationality of the player independent
>>of what was being played. It sounds to me that you
>>suggest identification of a German player should be
>>done by having him play "Ach du lieber Augustine."

Obviously I suggested no such thing.

>>No madame, it was suggested that a German player
>>could be identified by the character of his/her playing
>>no matter who the composer was.

I'll ignore the gratuitously insulting way you phrased that comment, but
let's clarify your use of passive voice there: I wasn't among those who
suggested that this identification is possible. As I wrote in an earlier
message, I think that national characteristics *may* have existed in the
past but if so, they've probably blurred together in the modern era.

>>Now you are welcome to believe whatever you wish,
>>but the argument was about sound character independent
>>of composition style or type.

Yes, that's right, but I stand by my argument that you can only prove this
point by testing different musicians playing the work of *one composer at a
time*. You would need several series of tests to make that point. First
musicians 1, 2, 3, 4, 5, etc. play some Mozart. *Then* the same musicians
play some Scriabin, and so forth. You have to hear all of the musicians
play all of the samples in order to make a telling comparison. And that's
leaving aside the whole question of making sure your listeners hear the
same samples, unaffected by differences in our computer systems.

>>What I hear from you is the same thing I heard after the
>>Shroud of Turin was tested for age. When it failed the
>>test, those who were not in agreement with the answer
>>suggested that the test was biased because it did not test
>>the right thing.

Whoa, wait a minute--I didn't say your test was biased. "Biased" is a
loaded word, because, although it has a technical meaning in experimental
science, people also use it in a social context to refer to unfairness.
Combining that word with your use of religious imagery loads your argument
in a way that *is* biased, in the social sense: it misrepresents what I
said. I didn't accuse you of bias. I said your test was unscientific.
There's a difference.

Lelia Loban
Who watches the watchers?

-------------------------------------------------------------------
Klarinet is a service of Woodwind.Org, Inc. http://www.woodwind.org