Volume 14, Number 2, June 2008
Copyright © 2008 Society for Music Theory

Olli Väisälä*

Rhetoric and Justification in Analysis: A Commentary on Eric Wen’s Commentary


REFERENCES: Samuel Ng, “Review of Peter H. Smith, Expressive Forms in Brahms’s Instrumental Music: Structure and Meaning in His Werther Quartet,” Music Theory Online 13.4; Eric Wen, “Commentary on Samuel Ng’s review of Peter H. Smith’s Expressive Forms in Brahms’s Instrumental Music: Structure and Meaning in His Werther Quartet,” Music Theory Online 14.1.

KEYWORDS: Brahms, Schenkerian analysis

ABSTRACT: This commentary discusses the Schenkerian interpretation of the opening of Brahms’s Piano Quartet No. 3, earlier addressed by Samuel Ng and Eric Wen in this journal. It argues that the central question at issue—the location of the structural IV–V motion—is clarified by factors such as rhythm, meter, register, instrumentation, and tonal expectations. Aspects of complexity are also identified, including Brahms’s way of giving divergent structural meanings to an inherently ambiguous passage by setting it in different contexts.

Received April 2008

[1] In the previous issue of Music Theory Online, Eric Wen attacks Samuel Ng’s reading of the opening of Brahms’s Piano Quartet no. 3, presented in Ng’s review of Peter Smith’s monograph of this work. While Wen calls Ng’s interpretation “surely incorrect,” I find his discussion lacking in arguments that would justify such surety. Since Wen’s discussion raises important questions about analytical justification and analytical rhetoric, I believe a comment on Wen’s comment may be worthwhile.

[2] The main difference between Ng’s and Wen’s views concerns the location of the structural dominant in the opening large progression (mm. 1–32); see Example 1 for an annotated score of mm. 1–27. Ng criticizes Smith’s reading of the G major chord in m. 21 as the dominant, suggesting that the dominant only appears in m. 27. Wen, on the other hand, agrees with Smith in this issue, but criticizes both Smith and Ng of overlooking “an important Classical tonal procedure” in their analyses. This procedure relates to the restatement of the opening idea in B minor in m. 11 ff. Example 2a (combining relevant parts of Wen’s Examples 9 and 10), illustrates his conception of the B minor chord as part of a descending chromatic 5–6 progression leading to a IV6. The basic progression is similar to that at the opening of Beethoven’s “Waldstein” Sonata, one of the classical precedents that Wen discusses in his commentary.

[3] While Wen’s discussion of classical precedents is interesting, it by no means settles the issue of the dominant’s location in the Brahms passage. From Ng’s voice-leading graph (his Example 7) we may infer that his conception of the underlying harmonic framework is, in fact, essentially similar to Wen’s. Example 2b presents a reading whose bass-line framework is extracted from Ng’s graph (his Example 7); the counterpointing upper voices are supplied by the present author (Ng’s graph is less clear in this respect). In both Example 2a and 2b, the B minor chord (VII) functions as an intermediate element between I and IV, which then goes on to V. (The G-major and F-major chords of Example 2a are replaced by root-position chords in Example 2b, but this does not affect the basic meaning of this framework.)(1)

[4] The primary difference between Ng’s and Wen’s readings thus does not concern the underlying harmonic framework but the timing of its elements. According to Wen, the structural IV–V motion occurs mm. 20–21; according to Ng it occurs in mm. 25–27. Example 1 illustrates the alternative views of harmonic hierarchy by note names, boxes, and parentheses beneath the score. The invocation of classical precedents is clearly insufficient for assessing the merits of these two readings. Instead, we have to study the compositional features in the Brahms score to see what they suggest with respect to this issue.


Example 1. Brahms, Piano Quartet No. 3, mm. 1–27,
three conceptions of bass line hierarchy

(click to enlarge and see the rest)


Example 2. Brahms, Piano Quartet No. 3,
three readings of harmonic structure
(click to enlarge)

[5] I submit that such features hardly warrant contending that the first F–G bass motion (mm. 20–21) is “surely” the decisive one. In view of the great rhetorical weight of the opening C and B (mm. 1–2 and 11–12), this motion seems too incongruously fleeting to function convincingly in the harmonic framework, overriding the B in structural priority. To be more precise, factors of rhythm, register, and instrumentation ally to support a perception of the later F–G motion (mm. 25–27) as more decisive, in accordance with Ng’s reading. The F–G motion is rhythmically more pronounced in mm. 25–27, and the long note values associate with the opening C and B, as does the resumption of the original low register and the return of the piano.

[6] While these factors support Ng’s locating the IV–V motion in mm. 25–27, they also suggest modifying one aspect in his reading. Ng indicates the F bass as prolonged from the end of m. 20 to m. 24 through its neighbors G and G. The structural status of the first F (m. 20), however, fails to be supported by rhythm, register, or instrumentation. On the basis of these factors, this F would be more logically interepreted as a local dominant back-related to the preceding B. This reading is illustrated in Example 1 (lowest line), Example 2c, and, in more detail, Example 3 (which shows notes in their actual registers).(2)


Example 3. Brahms, Piano Quartet No. 3, mm. 1–32, voice-leading sketch

(click to enlarge and see the rest)

[7] Tonal expectations constitute another significant factor. Here Ng’s and Wen’s perspectives differ sharply. On the basis of the local tonicization of B minor in mm. 11–20, Ng finds the subsequent G-major chord in m. 21 unexpected, “defying normative harmonic logic.” Wen, by contrast, calls it the “long-expected dominant,” thus invoking the global perspective of C minor. Both the local B-minor and the global C-minor perspectives are, I would suggest, pertinent for the perceptive listener, as illustrated beneath the graph in Example 3. In terms of modulation technique, the underlying idea is the use of the F-major chord as a pivot, the V of B minor turning into the IV of C minor. The F7–G motion in mm. 20–21 (Wen’s IV–V) suggests such a reinterpretation, but a more decisive occurrence of such an event takes place in mm. 25–27, after the prominent re-establishment of F as the dominant of B minor (VI–V in mm. 23–25). In addition to rhythm, register, and instrumentation, such an impression is supported by tonal expectations. As Ng correctly observes, the first G-major chord (m. 21) is locally prepared by no elements pointing outside the key area of B minor. By contrast, the second G-major chord (m. 27) is prepared by a chromatic passing chord—an “inverted Italian 6th”—that unmistakably signals the upcoming dominant.

[8] Citing Marianne Kielian-Gilbert’s expression, one might describe these events as an “oscillation” between B-minor and C-minor perspectives.(3) As discussed above, however, various compositional factors suggest a hierarchy for such an oscillation: the local establishment of C minor in mm. 20–21 functions within a larger B minor context, which, of course, functions within a yet larger C minor context (see Example 3). Despite the difference in hierarchical level, the G-major chord in m. 21 bears significant associational relationships with those in mm. 9–10 and m. 27, recalling the former and foreshadowing the latter. Such associations are especially important for explaining how an unlikely instance of double mixture such as VI—“defying normative harmonic logic”—manages to make a meaningful effect within the B minor context.(4) As shown by brackets in Example 3, the top-voice motion D–D (mm. 21–23) echoes the implicit top-voice motion linking the first two phrases; this chromatic relationship is also reflected in other details.

[9] Wen should be credited for one substantial observation in his Brahms analysis, concerning a parallelism between the faux bourdon progressions in each of the two opening phrases. Wen’s graph illustrating this feature (from his Example 9) is reproduced here as Example 4a. While a study of this parallelism reveals interesting compositional aspects, it does not support Wen’s reading of structural relationships but rather highlights features that speak against it.

[10] In contrast to a classical model such as Beethoven’s “Waldstein”—in which the parallelism between the I–IV6 and VII–IV6 progressions in mm. 1–8 is based on literal repetition—Brahms has composed the latter part of this parallelism so as to weaken considerably its “goal” F-major . Example 4b illustrates how the two -chord progressions correspond with the actual music. The goal chord of the first progression, G major (m. 9–10), is emphasized by its strong metrical position, long duration, and root position, but the corresponding F-major (m. 20) is deprived of all these features. The treatment of meter is especially interesting: the F-major occurs at the weak third beat in a hemiola. Such an attenuation of the F-major is at odds with Wen’s notion of this chord as not only equivalent to the G-major chord of the first phrase but structurally surpassing all elements that have occurred since the opening I.


Example 4
. Brahms, Piano Quartet No. 3,
rhythmic treatment of “parallelistic” progressions

(click to enlarge and see the rest)

[11] While the -chord progression of the second phrase may have been modeled on that of the first, it is thus transformed in a way that undermines the goal status of the F-major chord in m. 20 and necessitates the more pronounced approach to this harmony in mm. 22–24. As indicated in Example 4 (uppermost stave), the VI–V progression (G–F) may be understood as a kind of compensation for the chords weakened by the hemiola rhythm in 19–20. This idea is supported by the way in which the chromatic parallel minor thirds in mm. 21–25 repeat those in the hemiola passage (see circles in Example 4b).(5) Moreover, the rhythm of the bass figure in mm. 21–22 and 23–24—three quarter notes followed by a long note—suggests a correspondence with the rhythm leading to the closing V chord of the first phrase (mm. 8–9), accentuating the G and F basses (mm. 22 and 24), and supporting the perception of the latter as the true goal of the VII–IV motion (= I–V in B minor).(6)

[12] All in all, while Wen bases his discussion on classical models such as the “Waldstein,” his Brahms reading fails to follow such a model in one significant respect. To cite Carl Schachter, “it is as much a part of the composer’s art as it is of the sculptor’s or painter’s to be able to create clear and distinct shapes.”(7) At the opening of the “Waldstein,” a “clear and distinct shape” of structural connections is secured by features such as the unified chromatically descending bass, the lucid parallelism between the I–V6 (mm. 1–4) and VII–IV6 (mm. 4–8) progressions, the strong emphasis on the dominant (mm. 9–13), and the effect of the preparatory A–A inflection (m. 7–8) in dispelling the impression of preceding local tonicizations. While the structural framework in Wen’s Brahms analysis is formed by elements similar to those in the “Waldstein” theme, one seeks in vain for equivalent clarifying factors that would secure the connection between these elements and the hierarchy among them. As suggested by the above discussion, however, the fault does not lie in Brahms’s art—as if he were a bad emulator of Beethoven—but in the analysis. A closer attention to factors such as rhythm, register, instrumentation, and tonal expectations leads us to discover a shape that is as “clear and distinct” as Beethoven’s—albeit more complex.

[13] Finally, despite arguing for a “clear and distinct” prolongational hierarchy in the Brahms passage, I would like to point out an aspect of ambiguity in one of its constituent parts.(8) While there is evidence that the G-major chord in m. 21 does not yet represent the structural dominant, this evidence is based entirely on the impact of the preceding events. If one considers the passage of mm. 21–27 in isolation, there is nothing to contradict Wen’s notion of it as embodying a prolongation of V. Its design is reasonably unified, and the steadily descending long notes in the top voice could well be heard as forming a third-progression moving from the fifth to the third of the G-major triad. Hence this passage is inherently ambiguous: it can fulfill divergent prolongational functions, depending on the context. Brahms, rather characteristically, takes advantage of such ambiguity. While the G-major chord in m. 21 is too weakly prepared and expressed to establish the structural dominant, this passage reappears in the recapitulation preceded by a strongly established dominant (mm. 217 ff.). In such a context it actually participates in a dominant prolongation.

[14] The main purpose of this commentary is not to offer, at last, a “correct” reading of the Brahms passage. Its main purpose is to give an idea of the kind of musical factors that should be allowed for in approaching a “correct” or descriptive reading. Apart from its observations of parallelism, Wen’s discussion shows little concern for such factors. Its critique of Ng’s reading is based on presumptuous rhetoric (“surely incorrect”) rather than rational arguments about the music. Such a manner of discussion is not without precedents in Schenkerian literature, but it is an unfortunate tinge in an otherwise valuable tradition. A more solid attention to analytical justification would certainly benefit the Schenkerian cause and its contribution to musical understanding.

Olli Väisälä
Sibelius Academy
Department of Composition and Music Theory
PL 86
00251 Helsinki


1.  According to Wen’s analysis, the G-major chord is represented by the root-position chord at the end of the first phrase (mm. 9–10), but the F-major chord is expressed as a real chord (m. 20, followed immediately by a root-position F7). The idea of two parallel phrases is enhanced, however, if the F-major chord also occurs in root position, as in Ng’s analysis. The substitution of chords for chords in this framework also occurs in other pieces, such as Beethoven’s Op. 31/1, one of Wen’s examples.
Return to text

2. The precision of register and instrumentation in structural clarification also applies to the relationship between the G and G bass notes, the latter of which is interpreted as structurally superior in Example 3 (in accordance with Ng’s discussion).
Return to text

3. Marianne Kielian-Gilbert, “Interpreting Schenkerian Prolongation,” Music Analysis 22/1–2 (2003): 51–104.
Return to text

4. Cf. the discussion of the preparation required for chromatic chords in Edward Aldwell and Carl Schachter, Harmony & Voice Leading, 3rd edition (Belmont: Thomson–Schirmer, 2003), pp. 546–547.
Return to text

5. As indicated in Example 4, there are enharmonic implications in both instances: the applied leading-tone chord to the G-major triad in m. 19 “should” have E, and the relationship in mm. 22-23 also implies an enharmonic reinterpretation.
Return to text

6. The interpretation of hypermeter in mm. 21–25 is not completely straightforward. The surrounding events would imply that mm. 21, 23, 25 are strong in relation to 22 and 24, but the accentuation of mm. 22 and 24 together with the ensuing relationships produce a hint, at least, of the reverse relationships.
Return to text

7. Carl Schachter, Unfoldings (New York: Oxford University Press, 1999), 124.
Return to text

8. Several authors have discussed ambiguity in Brahms from various viewpoints. For a recent example, see Peter Smith, “You Reap What You Sow: Some Instances of Rhythmic and Harmonic Ambiguity in Brahms” (Music Theory Spectrum 28/1 [2006]: 57–97), which, among other things, discusses small-scale metric ambiguity in the passage discussed here.
Return to text

Copyright Statement

Copyright © 2008 by the Society for Music Theory. All rights reserved.

[1] Copyrights for individual items published in Music Theory Online (MTO) are held by their authors. Items appearing in MTO may be saved and stored in electronic or paper form, and may be shared among individuals for purposes of scholarly research or discussion, but may not be republished in any form, electronic or print, without prior, written permission from the author(s), and advance notification of the editors of MTO.

[2] Any redistributed form of items published in MTO must include the following information in a form appropriate to the medium in which the items are to appear:

This item appeared in Music Theory Online in [VOLUME #, ISSUE #] on [DAY/MONTH/YEAR]. It was authored by [FULL NAME, EMAIL ADDRESS], with whose written permission it is reprinted here.

[3] Libraries may archive issues of MTO in electronic or paper form for public access so long as each issue is stored in its entirety, and no access fee is charged. Exceptions to these requirements must be approved in writing by the editors of MTO, who will act in accordance with the decisions of the Society for Music Theory.

This document and all portions thereof are protected by U.S. and international copyright laws. Material contained herein may be copied and/or distributed for research purposes only.

Prepared by Brent Yorgason, Managing Editor
Updated 09 June 2008

Number of visits: