<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=windows-1252"
http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
<font size="-1"><font face="Georgia">The relation between function
theory and Roman numerals is a complex matter. It depends, in my
opinion, on one's conception of a harmonic function.<br>
The problem not only is that the same function may be projected by
harmonies built on different scale degrees, but also that the same
scale degree may project different functions. In this respect, it is
striking that Goetschius, whose rules quoted by Richard otherwise
faithfully echo Simon Sechter's description (<i>Die richtige Folge der
Grundharmonien</i>, 1853, p. 12-13), nevertheless include the IV-V
progression which Sechter (and Bruckner or Schoenberg after him)
wouldn't consider a progression properly speaking because of the
absence of common tones. [R. Wason, <i>Viennese Harmonic Theory</i>,
p. 154 n. 10, notes that Sechter's <i>Grundsätze</i> were known in
American translation as early as 1871; 12th edition in 1912.]<br>
<br>
That the function of IV in IV-V may be considered different from that
in IV-I has been a major concern of harmonic theory since Rameau's <i>double
emploi</i>. <br>
In order to make IV a subdominant (S) in both cases, Riemann is
compelled to view IV-V as conceiling a IV-(I)-V progression; this
remains a major weakness of his theory: Riemann cannot imagine a simple
path from one of the opposites of his dualist construction to the
other, from <i>Unterdominante </i>to <i>Oberdominante.<br>
</i>Sechter, on the contrary, considers that IV in IV-V stands for
II. If a function were to be assigned to IV-V = II-V, however, it would
be a dominant function (in a sense inherited from Rameau), not a
subdominant as in IV-I. </font></font><font size="-1"><font
face="Georgia">As Wason states (p. 35), Sechter's "notion of the
subdominant is completely unrelated to its use as <i>dominant
preparation</i>".</font></font><br>
<font size="-1"><font face="Georgia"><br>
The question then, is whether a harmonic function is a "chord quality",
independent from context (as in Riemann), or arises from the context
(as in Sechter, or more generally in theories based on cadential
formulas). There is a tendency, in Germany, to consider Roman numerals
as expressing a theory based on six (or seven) different harmonic
functions. This may be true in some cases, but Roman numerals more
often convey a conception of contextual functions. The distinction
between the two is not always clear, however.<br>
<br>
Cordialement,<br>
<br>
Nicolas Meeùs<br>
<a class="moz-txt-link-abbreviated" href="mailto:nicolas.meeus@paris-sorbonne.fr">nicolas.meeus@paris-sorbonne.fr</a><br>
<a class="moz-txt-link-freetext" href="http://www.plm.paris-sorbonne.fr">http://www.plm.paris-sorbonne.fr</a><br>
<br>
<br>
</font></font><br>
Richard Porterfield a écrit :
<blockquote cite="mid:BLU113-W29DC6BCB81934D6868EC42D5EB0@phx.gbl"
type="cite">
<style>
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
</style>[...] <br>
Although some theorists such as Riemann track functions (T, D, S) which
might be projected by a harmony built upon one or another scale degree,
while others track scale degrees without regard to function,
the historical association I speak of is longstanding and not limited
to Schenker. <br>
[...] Regarding the North American theoretical tradition specifically,
Bernstein refers to "a roman numeral style of harmonic analysis in
America during the nineteenth century" of which Percy Goetschius is a
prime mover (787-88). Consider the following passage from Goetschius's <em>The
Theory and Practice of Tone-Relations </em>first published in Boston
in 1892 (I quote from page 25 of the 1917 edition [New York: G.
Schirmer], available on Google Books): <br>
<p align="center"><strong>Rule I:</strong> The tonic triad can
progress, under all harmonic circumstances, into any other chord of its
own, or of any other, key. This is its prerogative as chief of the
harmonic system. Therefore I-V and I-IV are good.</p>
<p align="center"><strong>Rule II:</strong> The subdominant triad
(IV) may progress either into the I or the V. Thus: IV-I or IV-V. </p>
<p align="center"><strong>Rule III:</strong> The dominant triad (V)
may progress, legitimately, <strong>only into the tonic chord.</strong>
Therefore V-I is good; but V-IV must be avoided [emphasis original]. </p>
There may well be a paper or even a Ph.D. in the subject of
Goetschius's influence on American theory having made conditions
favorable for the reception of Schenker's theories and methods later in
the twentieth century. <br>
<br>
Regards, <br>
Richard Porterfield<br>
Instructor, Mannes College of Music<br>
Ph.D. Candidate, <br>
<hr id="stopSpelling">
From: <a class="moz-txt-link-abbreviated" href="mailto:Paul.Sheehan@ncc.edu">Paul.Sheehan@ncc.edu</a><br>
To: <a class="moz-txt-link-abbreviated" href="mailto:porterfr@hotmail.com">porterfr@hotmail.com</a><br>
CC: <a class="moz-txt-link-abbreviated" href="mailto:jcovach@mail.rochester.edu">jcovach@mail.rochester.edu</a>; <a class="moz-txt-link-abbreviated" href="mailto:smt-talk@societymusictheory.org">smt-talk@societymusictheory.org</a><br>
Date: Fri, 4 Sep 2009 01:03:06 -0400<br>
Subject: Re: [Smt-talk] I - II- IV as a progression (counterpoint)<br>
<br>
Dear Readers,<br>
<br>
Re. Richard Porterfield's statement:
<p class="EC_MsoNormal"><span
style="font-size: 10pt; color: rgb(68, 68, 68); line-height: 115%; font-family: 'Verdana','sans-serif';">"That’s
what the Roman numerals have been for, historically, not only
identifying scale-steps but also their function in a tonal context." </span></p>
<span
style="font-size: 10pt; color: rgb(68, 68, 68); line-height: 115%; font-family: 'Verdana','sans-serif';">
<p class="EC_MsoNormal"><span
style="font-size: 10pt; color: rgb(68, 68, 68); font-family: Verdana;">I
don't mean to be overly fussy, but is it the case that Roman numerals
have _not_ historically entangled with function? I am under the
impression that, until Schenker, scale step theory (Roman numerals) was
used as an analytical tool independently of function
theory. Furthermore, I am under the impression that Schenker in
particular combined scale degree theory with function theory in a way
that now seems almost automatic to many theorists and other interested
parties (at least in North American academic culture). Witness, e.g.,
many textbook treatments of such matters since Aldwell and Schachter,
inclusive. Do any historians of theory care to comment?</span></p>
</span>
<p class="EC_MsoNormal"><span
style="font-size: 10pt; color: rgb(68, 68, 68); line-height: 115%; font-family: 'Verdana','sans-serif';">Paul
Sheehan</span></p>
<p class="EC_MsoNormal"><span
style="font-size: 10pt; color: rgb(68, 68, 68); line-height: 115%; font-family: 'Verdana','sans-serif';"><<a
moz-do-not-send="true" href="mailto:Paul.Sheehan@NCC.edu">Paul.Sheehan@NCC.edu</a>></span><br>
<br>
</p>
</blockquote>
</body>
</html>