<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Le 15/07/2011 19:26, Dmitri Tymoczko a écrit :
<blockquote
cite="mid:FD86182F-F1D8-41AB-B87B-73DDC1AECEBF@princeton.edu"
type="cite">I've always thought that the notion of "prolongation"
is freighted with really serious philosophical complications.
Anyone who could remove these complications, and come up with a
non-circular definition of "prolongation," would be doing the
field a great service ...<br>
</blockquote>
I think that Schenker's own notion of "prolongation" is without much
problem in Schenker's own terms – which must be read in the original
German.<br>
<blockquote
cite="mid:FD86182F-F1D8-41AB-B87B-73DDC1AECEBF@princeton.edu"
type="cite">
<pre wrap="">I've actually been thinking about this issue off and on for years. My current best guess is that "prolongation" is an intentional, rather than a grammatical concept: to say "X prolongs Y" is (roughly speaking) to make an intentional claim about how the composer used X as a means of getting to (or doing) Y. This contrasts with the formal or grammatical or descriptive concepts (e.g. "V7 chord") familiar from other regions of musical discourse.
</pre>
</blockquote>
To say that "X prolongs Y" fully contradicts Schenker's own
description, in that is presupposes, as you say, that X preexists as
"a means of getting to (or doing) Y". What is given is Y, at first a
mere abstraction of a chord, a mere image of the series of harmonic
partials. In order to become a musical chord, Y must be inscribed it
time, "prolonged", which may be done by a mere arpeggiation –
producing no subordinate chord. The prolongation, however, may
involve filling in the "tonal space" (the empty intervals) of the
initial chord and this, is special cases, may happen to produce a
subordinate chord X. This X by no means is "doing Y"; it merely is a
by-product of Y itself.<br>
I think that Roland Barthes' description of the construction of
discourses proposes something similar, which I didn't reread
recently enough to remind his own terms. A discourse is a succession
of nodes, each of which may receive an ornamentation (I think his
term is <i>catalyse</i>)<i> </i>which certainly smooths the
succession and which possibly gives rise to secundary nodes. This
also has to do with how a discourse can develop from a deep
structure, in Shomsky's terms.<br>
Schenker's conception of the "tonal space" is developed in a
text titled <i>Erläuterungen</i>, that he published twice in the
last volumes of <i>Der Tonwille</i> and twice in the two first ones
<i>Das Meisterwerk in der Musik –</i> this fourfold publication
might suffice to stress how important this short text was for him.<br>
<br>
Prolongation indeed is intentional – not in the sense that the
composer used X as a means to doing Y, but that Y served as a means
to produce X. This is the intentionality of the work itself, not
necessarily of the composer. And, obviously, Schenkerian analysis
has nothing in common with a mere descriptive labeling of chords
with roman numerals. I am not sure that this can be viewed in terms
of intentionality <i>vs </i>grammaticality, as you suggest. Roman
numeral analysis is lexical at best, and has little to say about
grammar; I do not see how a grammar could be conceived without a
level of intentionality.<br>
<blockquote
cite="mid:FD86182F-F1D8-41AB-B87B-73DDC1AECEBF@princeton.edu"
type="cite">
<pre wrap="">Interestingly, my digression couldn't be included in a Wikipedia article, since it hasn't been published anywhere. (Wikipedia has strict standards about avoiding "original" or "unpublished" research.) But nor could Wikipedia include a simple, uncontroversial summary of what "prolongation" means, because we in the field don't really agree about the issue at all. So about the best you could hope for is a survey of the various proposals that have been made, one that gets updated as the discussion proceeds.
</pre>
</blockquote>
Well, I would be tempted to consider that proposals based on
Schenker's own writings should gain preeminence, at least in
articles devoted to Schenkerian concepts (prolongation, as a
concept, may have an existence as a non-Schenkerian, or
post-Schenkerian, or neo-Schenkerian concept, but that is not my
concern here).<br>
<br>
Yours,<br>
Nicolas Meeùs<br>
Université Paris-Sorbonne<br>
</body>
</html>