In a previous post I observed that this progression is not in fact "syntactically correct" because it shows the compromises necessary when melodic parsimony conflicts with a strictly maintained four-voice texture.
There is another problem with the progression, specifically with its status as a departure-return model. The first chord conflates a stable harmonic beginning point (I) with an already-departed melody (^3, not ^1). Now, this is presumably out of deference to Schenker, but that deference also means too much is taken for granted here. As Westergaard puts it near the beginning of Introduction to Tonal Theory, "pitch and time relationships are the primary stuff of the structure of any piece of tonal music" (11). If stable beginning is also to be point of departure, then only ^1 over I can be adequate to the task.
The properties are separable: in the examples below, all constitute points of departure (bar 1 of a waltz in D365) but only one of them constitutes a stable beginning (and even there, the tonally stable beginning is pre-empted metrically by the pickup note E5, which gains in salience from its mordent).
Although Westergaard, too, sets his rules out of deference to Schenker ("The first note of the basic step motion must be a tonic-triad member a third, fifth, or octave above the final note" (as cited in Peles 78)), he seems to recognize the problem at some level. Looking back at Introduction to Tonal Theory more than thirty years later, he says (59) that he would replace arpeggiation with borrowing as a basic operation (arpeggiation is melodic movement between triad members; borrowing accounts for these movements as drawing notes from other voices in the voice-leading web -- the same as Rothstein's imaginary continuo). Westergaard's example (60) begins (this is the crucial bit) with a closed position tonic triad, whose label reads "Four voices define a span by stating a C-major triad in closed position with the C doubled in the outer voices." Only after this, rearticulation confirms duration and borrowing lets the soprano claim ^3 from the tenor. Thus, the closed position triad, with ^1 at the top, is conceptually prior to the first chord of Agawu's "norm."
References:
Agawu, Kofi. Music as Discourse: Semiotic Adventures in Romantic Music. New York: Oxford University Press, 2009.
Westergaard, Peter. Introduction to Tonal Theory. New York: Norton, 1975.
Peles, Stephen. "An Introduction to Westergaard's Tonal Theory." In Theory Only 13/1-4 (1997): 73-94.
