On 15 Feb 2006 at 11:57, Dennis Bathory-Kitsz wrote:
> At 11:18 AM 2/15/06 -0500, Andrew Stiller wrote:
> >I basically agree with David Fenton that
> >more than 8 or 9 different levels are impossible for the ear to
> >distinguish.
>
> That can't be true. The very notion of shaping a line has to do with
> changing its dynamics slightly.
I would actually have to agree with you that Andrew's statement is
incorrect, and not a valid rewording of my point.
My point is not about what gradations instruments are capable of
producing -- those are infinite, within the dynamic range that a
particular instrument is capable of producing.
My point is about how the notation indicates instructions to the
performers. In that regard, I don't think that the dynamic markings
beyond ppp and fff have any reliable interpretation that makes any
sense.
If, for instance, a passage is at ppp and you want to get softer, I
think I'd just write "softer" and if even more is needed, "softer
still." These are clearly relative dynamic indications, without the
stepped gradations implied by adding p's to the dynamic indication.
I think that's my problem with pppp and ppppp. It's just not clear to
me that there can be any difference between the two except in a
context, and there is no rationally definable difference (in the
mathematical sense of the word "rational," not the logical sense)
between the two. To me, the notation with specific definable
gradations implied in it is at odds with what is surely intended,
because the variation between performers (and between different
performances by the same performer(s)) is going to be greater than
the difference between the two distinct dynamic marks.
The problem is not that we can't perceive minute differences nor that
we can't perform minute dynamic gradations, but that the system we
use for notating dynamics breaks down at its extremes.
There's also the very real issue of how one interprets the range of
dynamic marks, which is raised by Andrew's point about the dynamic
limitations of certain instruments. What's to be done in a piece with
a range of ffff to pppp as compared to a piece with a range of ff to
pp? Should the dynamic markings used in each piece by mapped onto the
dynamic ranges of the instruments for which the music is written,
with the ffff in the one piece being the loudest the instrument can
play, and the ff in the other piece being the same loudness?
Of should the ff in the second piece be the same loudness as the ff
in the first piece?
To me, this is a very hard question, and to me, neither answer is
satisfactory.
The first one is unsatisfactory first of all because it will depend
on the piece (I'm assuming two pieces from the same style period, so
there's no issue of historical performance practice; and I'm not even
sure that I agree that Mozart's ff is different from Brahms's or
anyone else's -- it's highly dependent on the instruments, since ff
on Mozart's piano is not nearly as loud as ff on Brahms's; but if
you're not playing on Mozart's piano, then I don't think that fact is
relevant at all; but I digress). But assuming for a moment that it
would be reasonable in the two pieces to interpret the dynamic
markings as applying relative to each other to the full dynamic range
of the instrument, then it would mean that the central region of the
dynamic range of the piece with the broader palette of dynamic
markings will, of necessity, be narrower, since the extreme dynamic
markings have to be accomodated. That means that ff in the piece with
the larger number of markings is not as loud as ff in the other
piece, which is problematic.
On the other hand, if you attempt map the dynamic markings to some
absolute dynamic level, you get other problems that can lead to
relative dynamic imbalances within either of the two pieces.
Secondly, in order to accommodate the wider dynamic range of the one
piece, you're limiting the range of the other piece.
One solution to this would be to use a "dynamic response" curve to
adjust the degree of gradations, analogous to a frequency response
curve one might use in a mixer. If you interpret the gradations
between the main dynamic markings as being larger than the gradations
at the extreme, you'd basically be able to keep the common dynamic
markings basically the same with only a small contraction of the
louder and softer markings, while making room to accommodate the
outer extremes. That is, you could keep f to p the same, and make ff
and pp slightly less extreme in order to accommodate fff, ffff, ppp
and pppp.
But the downside of that is that it reduces the absolute range of
gradation for those extreme markings to such an extreme that there's
not going to be much real difference between fff and ffff.
(I actually think that this "dynamic response" adjustment is what we
actually do with these kinds of dynamic marks, so that instead of:
| | | | | | | | | | | |
ppppp pppp ppp pp p mp mf f ff fff ffff fffff
we would have something more like:
| | | | | | | | | | | |
ppppp pppp ppp pp p mp mf f ff fff ffff fffff
I'm limited here by spacing limitations of trying to graph this with
a fixed-pitch font, but I think you get the idea.
My guess is that this is really closer to what we do when
interpreting dynamic markings. And of course, there's the issue of
the logarithmic nature of the production of sound that I haven't even
dealt with here)
So, I see no good solution here, other than to use the limited set of
dynamic marks from fff to ppp and then to use markings to shade them
in order to indicate subtleties.
The real problem with the extreme dynamic markings is, as someone
said, an apparent overspecificity, and the problem with that is that
as a practical matter, that specificity can never be reached in
performance.
I learned the pitfalls of overspecificity while observing a
dissertation defense in my department. The subject was the pianos of
the Viennese piano builder, Graf. The dissertation included lots of
measurements of the parts of the instruments down to hundredths of a
millimeter, all taken with a micrometer. In the defense, the writer
of the dissertation was asked two telling questions:
1. did the workmen in Graf's shop have tools that could reliably
produce parts accurate to a hundredth of a millimeter? AND
2. have the wooden parts shrunk with age, so that the current
measurements are unlikely to represent the original size?
This was a case where a granularity of .5mm probably would have been
sufficient to capture all the useful information. This is not to say
that the measurements were inaccurate -- they surely were, given the
modern tools used to make them -- only that the accuracy was of no
real utility in determining how the instruments were actually
constructed.
Likewise, it seems to me that in the case of dynamics, while we as
listeners can perhaps perceive gradations to a "hundredth of a pp",
the notation can't really indicate that fine a gradation using the
traditional dynamic markings.
Now, Dennis makes another argument about setting down as specific a
score as possible and not worrying too much about the imperfectness
of each individual performance in capturing the exact intentions
indicated in the score. I am fine with that idea as a principle, but
I don't actually think that traditional dynamic markings *can*
clearly indicate fine gradations (I keep wanting to type
"degradations") sufficiently clearly to capture those kinds of
"ideal" intentions. Perhaps a score supplemented by a recording of a
synthesized performance illustrating exactly how the composer wants
things interpreted could bridge the gap between the specificity of
conception and the messiness of the notation we have available to us.
Now, philosophically, I put more faith in performers to understand
the musical content and the intent of the composer than to ever be
that specific myself, and also doubt my own ability to fully
understand everything about my own compositions, but many people are
much better composers than I am and are, perhaps, much better
qualified to be directing performers to do such specific things.
Me, well, I like it when performers have different ideas, even if I
may reject some specific ideas about interpretation.
But that returns us to the discussion that Dennis and I have been
having on this list for many years.
--
David W. Fenton http://dfenton.com
David Fenton Associates http://dfenton.com/DFA/
_______________________________________________
Finale mailing list
[email protected]
http://lists.shsu.edu/mailman/listinfo/finale
