• MTO Home
  • Current Issue
  • Previous Issues
    By Author By Volume       
  • Submit
    Submission Guidelines Submit Online
  • Jobs
    Current Job Listings Submit Job Listing
  • Dissertations
    All Dissertations New Dissertations List Your Dissertation
  • Events
  • Webinars
  • About
  • Links
  • SMT

Volume 18, Number 3, September 2012
Copyright © 2012 Society for Music Theory

Schenker and the Tonal Jazz Repertory: A Response to Martin

Mark McFarland


KEYWORDS: Jazz, Schenker, Henry Martin, Steve Larson

ABSTRACT: Henry Martin recently introduces new forms of the Urlinie other than Schenker’s 3-, 5-, and 8-lines to provide superior readings of works from the tonal jazz repertory. This study questions the superiority of these new Urlinien; Schenker’s three forms of the Ursatz are preferred instead. Martin’s dissatisfaction with orthodox Schenkerian theory is traced to his belief in the inviolability of the original melody, which excludes the use of implied tones and forces the Urlinie to appear in the upper voice, thus eliminating the Deckton.

Received July 2012

PDF version
 

[1] If one were asked to summarize Steve Larson’s approach to the analysis of tonal jazz, the first thing that would likely come to mind would be his belief in the applicability of Schenkerian analysis to this repertory. In fact, Larson strove as much as possible to approach jazz using as orthodox a form of Schenkerian theory as possible.(1) In his review of monographs by Forte 1995, Gilbert 1995, and Martin 1996, Larson carefully explained how deviations from Schenkerian orthodoxy by these authors were, in his own words, “more apparent than real” (Larson 1999, 114–17). More specifically, Larson’s analytic methodology brought three modifications to Schenkerian theory into orthodoxy: (1) the frequent lack of spacer in Gilbert’s Urlinien and in Martin’s “Common Bop Background Forms,” (2) Martin’s concept of “prolongation by arrival” (one replaced by the concept of the auxiliary cadence), and (3) the unresolved or unexplained appearance of dissonances by all three authors.

[2] Martin’s recent review of Larson’s Analyzing Jazz: A Schenkerian Approach (Martin 2011a) along with the former’s keynote address at the Dutch-Flemish Society for Music Theory conference (Martin 2011b), serve as a response to Larson’s use of an orthodox Schenkerian methodology. Indeed, Martin’s keynote ends with a list of modifications to Schenkerian theory that are implied by his analyses, these modifications consisting primarily of new forms of the Urlinie. His goal is to show that since tonal jazz works “can be found in which a conventional Schenkerian interpretation may be unconvincing,” he hopes to show that “background paradigms differing from Schenker’s provide superior readings” (Martin 2011b, 1).

[3] From the publication history of these two men, there is clearly a difference of opinion between them on whether to use an orthodox or modified Schenkerian approach. This disagreement is not limited to these two scholars; a recent event has suggested that there is widespread disagreement between jazz scholars. At the recent Steve Larson Memorial Conference held at the University of Oregon at Eugene in March 2012, the penultimate paper was read by David Heyer, a former graduate student of Steve Larson during the latter’s final years. In the question and answer session that followed the final three papers,(2) Dr. Heyer asked the audience, one composed almost exclusively of jazz theorists, how many of them recognized that sevenths and ninths were dissonant intervals in jazz; not a single hand was raised. It seems that one of the main tenets supporting Larson’s use of orthodox Schenkerian theory is not widely accepted.

[4] Thus, there is a schism in current tonal jazz research dividing those who employ orthodox or modified Schenkerian theory, with the majority of scholars—at least among those attending this conference—choosing the latter approach. Because Larson’s voice has been silenced, this study is a response to Martin’s call for the adoption of a modified form of Schenkerian theory, one labeled neo-Schenkerian in one of his more extensive graphs (Martin 2011a, 129). Rather than simply going over the same analytic ground that separated Larson and Martin, this study will focus on a revealing statement in Martin’s keynote to which Larson ran out of time to respond in print:

Typically, Schenkerian analysis privileges the three original prototypes to such an extent that they serve as models for the inference of implied tones, i.e., those missing from the music, but enhancing the coherence of an analytic level. I suggest that the analyst try to infer any missing notes from the harmonic and voice-leading events of the original melody (Martin 2011b, 17).

The analytic problems associated with treating the original melody of the lead sheet as inviolate when it does not serve as the basis of improvisation will be explored throughout this paper, in conjunction with a careful reading of points made in Martin’s article and the musical examples that support them.(3) These analyses will be performed in order to explore the question of whether Schenkerian theory requires modification when applied to the anomalous works from the tonal jazz repertory Martin cites in his study,(4) and if so, what is the best method to derive these modifications.

[5] Miles Davis’ “Four” is, in some ways, the least controversial graph in Martin’s study as it ultimately reduces to the ascending Urlinie spacer spacer spacer spacer that was first introduced in Neumeyer 1987a. Rothstein (1991) noted that Neumeyer’s ascending Urlinie follows what he refers to as the “Gestalt Nature” of Schenkerian thought: it employs Schenker’s concept of melodic fluency (stepwise motion), the Gestalt principle of “good continuation” (reference to a single scale in a single direction), and the principle of the imaginary continuo (a set of pitch classes that belong to successive chords from which the foreground voice-leading is derived). For these reasons, he concedes that Neumeyer’s ascending Urlinie can span an entire work and should at least be assigned to a deep level of the middleground. While I have no problem with the ascending Urlinie in Martin’s reading of “Four,” I question the scale degrees at which his Urlinie is interrupted, the number of appearances of interruption within this opening chorus, and the radically different reading of the two halves of this antecedent-consequent period.

[6] In his article on the ascending Urlinie, Neumeyer does not provide any examples containing interruption. It is possible, however, to determine on which scale degree this would occur. In Neumeyer’s contemporaneous article on the 3-part Ursatz (Neumeyer 1987b), the traditional combination of the Urlinie and the Bassbrechung is joined by a second Urlinie, one that moves in combination with these other two parts. This second Urlinie is confined to stepwise motion, although it does not exclusively descend; in fact, a traditional 5-line can be counterpointed against the second Urlinie spacer spacer spacer spacer spacer , or a traditional 3-line against spacer spacer spacer . Neumeyer does not provide an example of a three-part Ursatz with his ascending Urlinie from spacer to spacer . From his other examples, however, it is clear that if there was interruption in this ascending Urlinie, it would occur at spacer , which would be counterpointed against spacer in the second Urlinie and spacer in the Bassbrechung, and would obviously coincide with dominant harmony.

[7] Martin’s reading of “Four” does not place the interruption within the final section at spacer , but rather at spacer , yielding spacer spacer ||spacer spacer spacer spacer .(5) This placement would seem to point to the work of Day-O’Connell, which Martin cites as an influence (Martin 2011a, 127 and Martin 2011b, 1). Day-O’Connell’s rationale for the placement of interruption at spacer derives from his work on the pentatonic,(6) while Martin explains the placement of his interruption due to the subordination of the leading tone to spacer in this passage.(7) The emphasis given spacer here is more easily explained as an upper neighbor figure to spacer .(8) If this note does not assume this role, then fundamental tenets of Schenkerian theory must be modified, including the harmony as well as bass and treble scale degrees at which interruption appears. If this new type of interruption represents a standard modification of this Schenkerian principle in the tonal jazz repertory, it would represent a significant advancement in our understanding of these works. Without this supporting evidence, neo-Schenkerian theory represents a radical departure from orthodoxy; indeed, Martin mentioned after his paper at the Eugene conference that perhaps the word “Schenkerian” should be eliminated from his analytic approach.

Example 1. Miles Davis, “Four” from Blue Haze, opening chorus

spacer

(click to enlarge and see the rest)

Example 2. Schenkerian Graph of Davis’ “Four,” opening chorus

spacer

(click to enlarge and see the rest)

[8] Example 1 presents a transcription of the opening chorus of “Four” from the album Blue Haze,(9) followed by a more traditional graph of this work (Example 2), one influenced by Neumeyer’s 3-part Ursatz. As Neumeyer points out, “The forms of the three-part Ursatz provide musically satisfying solutions which hold close to the theory’s limits and should be considered as alternative responses to ‘Is it spacer or spacer ?’” (Neumeyer 1987b, 28). Instead, there are two linear progressions in “Four” that work in tandem; in this reading after the interruption, there is an ascending fourth progression from spacer up to spacer (measures 29–31) and a descending third progression from spacer down to spacer (measures 25–31). I have nevertheless heeded Rothstein’s comment on this topic: “it is probably best to assign such a three-part counterpoint to a deep layer of the middleground rather than to the background, and to choose one of the linear progressions as superior in status to the other” (, 306). With this clarification, an upper neighbor supported by supertonic harmony embellishes the opening note of the upper linear progression (spacer /I–I6spacer /II), while both spacer of the Urlinie and spacer of the ascending linear progression are supported by dominant harmony. When interruption occurs at the end of the first half of the work, it coincides with the initial arrival of dominant harmony. Davis’ final note at the half cadence is spacer , the thirteenth of the chord, or a sixth above the bass that substitutes for the consonant fifth, spacer . The leading tone is, admittedly, an implied tone. Yet the motion from spacer -spacer spacer spacer in the upper linear progression, with its arrival on dominant harmony at the interruption, seems to imply the addition of this note in the upper voice. This is especially true since there are two appearances of this ascending motion, the first of which does not lead to interruption or dominant harmony, thus preparing the listener for these two events on the immediate repetition of this material.

[9] My reading violates the original melody most obviously in its use of the implied note spacer just discussed, but also in the replacement at the middleground level of Davis’ foreground dissonant spacer with the consonant spacer . Each of these violations requires a change of a note in the original melody, which is anathema to Martin. This explains why interruption at the end of the B section in Martin’s middleground description of “Four” arrives on spacer rather than spacer ; although spacer is not a member of dominant harmony and therefore represents an unresolved dissonance, spacer is used at the point of interruption instead of the required spacer since the latter pitch does not appear in the original melody.

[10] While the importance of the original melody cannot be overstated—it is the one thing that remains constant among all performances of the same tune—the treatment of the original melody as inviolate requires modifications to some of the most fundamental tenets in Schenkerian theory. In “Four,” these changes forbid a traditional interruption on spacer since this note is not heard in the melody, although, pace Larson, the final spacer as chordal thirteenth represents spacer at the middleground. Further, an unchangeable original melody requires the appearance of two interruptions in “Four,” neither of which coincides with spacer . An inviolate original melody that must contain the Urlinie also undermines the concepts in Schenkerian theory that feature the interplay between contrapuntal voices, most obviously the cover tone. This explains why, although only the cadences of the antecedent and consequent phrases differ in the opening statement of “Four,” the former passage employs a background descent from spacer while the latter ascends from there to spacer , ignoring the simultaneous descent in an inner voice.(10) The reader is now invited to review the opening statement of “Four” (Example 1), keeping in mind both Martin’s middleground descriptive analysis found above and my reading (Example 2), in order to determine whether the latter’s neo-Schenkerian Urlinie provides a superior reading.

[11] After Martin presents a graph of “Sentimental Journey” that conforms to a traditional 3-line with interruption (Martin 2011b, 3–5), he follows this with graphs of Buster and Bennie Moten’s “Moten Swing” and Sy Oliver’s “Opus One,” which both use unorthodox Ursätze, spacer spacer spacer and spacer spacer , respectively (Martin 2011b, 6–9). The Ursatz spacer spacer is gapped and therefore breaks Schenker’s rule of melodic fluency mentioned above and to be discussed more fully below. On a less controversial note, “Moten Swing” seems to use the note spacer spacer as its Kopfton, although this note is never supported by tonic harmony and always appears as an incomplete neighbor to spacer . It is for this reason that “Moten Swing” uses a neighbor note Urlinie, a concept also to be discussed at more length below.(11)

[12] Rather than arguing in favor of a traditional Schenkerian reading for these two works, I would simply like to point out that the melodic ambitus of the two tunes is a perfect fourth and a tritone, respectively, while both tunes contain exactly four distinct melodic pitches. Under such tight melodic constraints, it is not surprising that a traditional Ursatz cannot be applied. These pieces seem similar to anomalous works from the Classical repertory such as Chopin’s Prélude op. 28, no. 4 in E minor. The structure of this work is described in the instructor’s manual to Forte and Gilbert’s Introduction to Schenkerian Analysis:

The text instructions describe this piece as having an interruption on spacer , in contradiction to Schenker’s stipulation that interruptions, no matter what the primary tone, can occur only on scale degree spacer . Regardless of what one calls it, it is clear that there is a caesura in m. 12, and that the melodic line that begins the piece is retraced beginning in m. 13. Reading the melody of m. 12, we see that G functions as an accented passing note within the consonant skip A–Fspacer , hence our contention that A (spacer ) is the prolonged note in that measure (Forte and Gilbert 1982b, 98).(12)

In the textbook itself, Forte and Gilbert state unequivocally that this work “has as its primary melodic note spacer , but no fundamental line.” They then cite other works that are similarly anomalous, including the Scherzo of Beethoven’s Piano Sonata op. 28 and Brahms’ Intermezzo op. 76, no. 4 (Forte and Gilbert 1982a, 388). Martin describes “Moten Swing” as similar to and “Opus One” as a true example of a “riff-like tune.” Such pieces favor “repetition and rhythmic interplay over voice-leading motion through descending linear progressions” (Martin 2011b, 6–7). Martin has communicated to the present author that there are many “riff-like tunes” similar to “Opus One” or “Moten Swing,” including Garland’s “In the Mood,” Parker’s “Cool Blues,” and Coltrane’s “Blue Train” and “Cousin Mary.” Until there is a thorough study of these works, however, it remains unclear whether a significant subset of the tonal jazz repertory feature melodies constrained to the extent that the appearance of a traditional Urlinie is impossible, or whether “Moten Swing” and “Opus One” are simply anomalies like the works by Chopin, Beethoven, and Brahms mentioned above.

Example 3a. Martin’s neo-Schenkerian graph of Noble’s “The Touch of Your Lips”

spacer

(click to enlarge and see the rest)

Example 3b. Larson’s graph of Noble’s “The Touch of Your Lips”

spacer

(click to enlarge and see the rest)

Example 4a. Leadsheet Notation of Noble’s “The Touch of Your Lips”

spacer

(click to enlarge)

[13] Martin’s graph of “The Touch of Your Lips” (Example 3a) introduces another new type of Ursatz, this one consisting of a double-neighbor figure, spacer spacer || spacer spacer spacer (Martin 2011a, 128, and Martin 2011b, 10). This reading of the work was prompted by the one in Larson 1998 that uses a traditional three-line with interruption (Example 3b). Both these graphs are based on three sources: the lead sheet for “The Touch of Your Lips” (Example 4a), as well as Larson’s transcription of Bill Evans’ plan for (Example 4b) and improvisation on (Example 4c) “The Touch of Your Lips.” Martin specifies the disagreement he has with Larson’s reading, one that relates to the bass arpeggiation of the tonic triad in the first half of the work:

Example 4b. Bill Evans’ Plan for
“The Touch of Your Lips”

spacer

(click to enlarge and see the rest)

Example 4c. Bill Evans’ Improvisation on
“The Touch of Your Lips”

spacer

(click to enlarge and see the rest)

the interesting wrinkle . . . is the III is a major rather than minor triad, which means that at quite an upper level of structure, the diatonicism of C major is abandoned. The significant assertion of IIIspacer 3 is a reason that I decided that the original Schenkerian spacer spacer ||spacer spacer spacer interrupted form worked less well with the song: IIIspacer 3 does not support spacer , while B4 and B3 in mm. 13–16 are prominently featured in the melody (Martin 2011b, 10).

There are two points that need clarification in this crucial statement in defense of the superiority of this unorthodox Urlinie.

Example 5. Schenker Figure 15-1b from Der freie Satz

spacer

(click to enlarge)

Example 6. Graph of Schubert, “Die Sterne” D. 939

spacer

(click to enlarge and see the rest)

[14] First, when referring to the abandonment of C major with the appearance of IIIspacer at an “upper level of structure,” Martin is clearly alluding to Schenker’s dictum that the background remain strictly diatonic.(13) However, as Brown 1986 has pointed out, Schenker allowed chromaticism at the level of the deep middleground through the use of three techniques: prolongation of the fundamental line, mixture in the bass arpeggiation, and tonicization of the bass Stufen. In fact, Figure 15-1b in Der freie Satz reproduces the background structure of Larson’s reading, with spacer supported first by tonic and then major mediant harmony (Example 5). If Martin’s hesitation in the use of IIIspacer and the abandonment of diatonic C major relates to the relative scale of “The Touch of Your Lips,” this is also unproblematic, as demonstrated by Schubert’s strophic Lied “Die Sterne” (D. 939). This Lied is in Espacer major, with the third line of each four-line strophe tonicizing a different chromatic mediant: VIspacer , VIspacer , IIIspacer , and VIspacer (only IIIspacer is left unexplored). Each of these key areas are used to prolong the fundamental line: both VIspacer and IIIspacer support the Kopfton spacer , while VIspacer does—and the putative IIIspacer would—prolong spacer spacer (Example 6).

[15] My second point concerning Martin’s reading of “The Touch of Your Lips” stems from the statement “IIIspacer 3 does not support spacer , while B4 and B3 in mm. 13–16 are prominently featured in the melody” (Martin 2011b, 10). The first clause of this sentence is not problematic. The note B3 on the downbeat of measure 16 in the leadsheet, however, is changed by Evans in his plan to the note E4 (Example 4b).(14) It is therefore no surprise that in Larson’s graph of this work (Example 3b), spacer is supported by IIIspacer in an inner voice (measures 13–15) while spacer remains active as Kopfton above it. The leading tone scale degree is then held as a common tone with the melodic arrival of spacer (to be discussed below) along with dominant harmony and interruption at measure 16.1. In Martin’s reading (Example 3a), the arrival of spacer over IIIspacer harmony is prolonged, in combination with a register transfer down an octave, through the arrival of dominant harmony and interruption. Martin’s reading here is an accurate reflection of the tune’s leadsheet (Example 4a). It does not, however, reflect the melodic changes made in either Evans’ plan (Example 4b) or improvisation (Example 4c) on the tune. A detailed comparison between the graphs of Larson and Martin is therefore not possible, at least at this crucial spot in the score. This difference does highlight the role of the note B in this passage, one given special emphasis by Martin.

[16] His comment on the prominent appearance of spacer in the melody—Martin goes so far as to comment that “the arrival on spacer /IIIspacer 3 at m. 13 is among the most salient aspects of the piece (Martin 2011a, 127)”—requires more extensive discussion. As mentioned above, in both Evans’ plan and improvisation, spacer is melodically prominent in measure 13—lasting until measure 15 in Evans’ improvisation—while this note is changed to spacer in the final measure of both phrases. Although spacer does not appear in the original melody, with Evans’ melodic modifications, spacer is represented by spacer (a chordal thirteenth) in measure 16.1 in his plan for “The Touch of Your Lips.” The supertonic scale degree literally appears in measure 16.4 of Evans’ improvisation as the resolution of the chordal thirteenth on the previous beat, both pitches appearing within dominant harmony.

Example 7. Schenker Figure 461 from Der freie Satz

spacer

(click to enlarge)

[17] Yet even if Evans’ plan was the only source considered, spacer appearing at a cad

gipoco.com is neither affiliated with the authors of this page nor responsible for its contents. This is a safe-cache copy of the original web site.