1
JAZZ IMPROVISATION
Theory, Analysis, Context, and Process
Most technical writings on jazz focus on improvised lines and their underlying harmonic progressions. These writings often overlook the basic fact that when one listens to jazz, one almost never hears a single improvised line, but rather a texture, a musical fabric woven by several musicians in real time. While it is often pragmatic to single out an individual solo line (I will be doing the same), it is important at all times to remember that an improvised solo is but one thread in that fabric, and it is a thread supported by, responded to, and responsive of the parts being played by the other musicians in the group. This book will explore the process of player interaction in jazz, and the role this interaction plays in generating improvised music.
I'd like to begin by examining a recorded excerpt from an improvisation by saxophonist Cannonball Adderley. To some extent, recordings in jazz take the placeâand assume the same textual authorityâof scores in European music, and I will rely on them throughout this work. Furthermore, I will make extensive use of transcriptions in my analyses, and while the authority of transcription has been widely critiqued in ethnomusicological circles,1 I intend to circumvent that issue by viewing transcriptions as heuristics in my discussions of recorded performances. In other words, my transcriptions are meant to serve as visual aids to the recordings and to my commentaries on them. I do not assume that they capture all aspects of a recorded performance, and I will be routinely taking information into account that is not available in the transcription. I regard them as representations of performances rather than pieces, and at times those representations will be approximateâsome of the music I will be dealing with is extremely complicated by the standards of any music notational tradition. It is important to remember that in jazz, a score (whether a "lead sheet," arrangement, or transcription) occupies no privileged ontological position, a notion that I take as an article of faith; my use of them is entirely pragmatic.
The excerpt, transcribed in figure 1.1, comes from the end of Adderley's first improvised chorus on the jazz standard "Groovin' High."2 As a starting point, the analysis will first examine Adderley's improvised line by itself, before demonstrating that an examination of the musical context within which the improvisation took place can generate further insight. The analysis will introduce and briefly address a number of musical and analytical issues to be dealt with at length in later chapters.
Rests in mm. 27 and 28 suggest a preliminary parsing of the improvisation into three large gestures, which are bracketed in figure 1.1 and labeled A, B, and C. Each gesture is of a different length: A is nine beats long, beginning in m. 25 and extending through the downbeat of m. 27; B is approximately four beats long, beginning with a pick-up to beat 3 of m. 27 and extending to m. 28, beat 2; and C, the longest of the three, extends approximately fourteen beats from a pick-up to beat 4 of m. 28 through the downbeat of m. 32.
At first glance, comparing the scalar sixteenth-note improvised lines of gestures A and B with the somewhat less-scalar eighth-note line of gesture C suggests that this excerpt is made up of two quite different kinds of music. Hie following analysis supports that preliminary intuition, first describing the differences between gestures A-B and gesture C, and then suggesting possible reasons for these differences.
Besides sharing a preponderance of sixteenth notes, gestures A and B have many other similarities. As noted earlier, they are primarily scalarâthat is, they are almost exclusively composed of stepwise motionâand both gestures are highly chromatic. A and B are also similar in another, less obvious, way in that both the contour and the melodic pitches of the improvised lines relate to the underlying harmony. Comparing gestures A and B with the underlying harmony reveals some consistent harmonic processes. Figure 1.2 presents a "harmonic analysis" of these melodic gestures. The upper staff adds analytical beams to the transcription, the middle staff gives a crude, "close position" realization of the chord symbols; these chords are
Figure 1.1. Cannonball Adderley, "Groovin' High," saxophone solo, mm. 25-32.
Figure 1.2. Cannonball Adderley, "Groovin' High," saxophone solo, mm. 25-28.
presented over a simple realization of the chord progression's root movement on the lower staff. The chord symbols in this figure are standard in jazz performance practice, and are interpreted as follows: The superscript-7 denotes a minor seventh chord, constructed from a minor triad and minor 7th; the symbol C-7 would therefore designate the chord containing the pitches C, Eâ, G, and Bâ. The plain superscript 7 denotes a "dominant" seventh chord, made up of a major triad and minor 7th; C7 would therefore be C-E-G-Bâ. The superscript Maj7 designates a major seventh chord, containing a major triad and major 7th; CMaj7 would thus be spelled C-E-G-B. Finally, these chords may contain additional alterations: C-7(â5), for instance, would designate a C minor seventh chord with a flatted-5th, or C-EââGââBâ.
As the added beams in figure 1.2 demonstrate, the contour of the improvised line emphasizes harmonically significant pitches: after the initial Bâ moves down to A (the 3rd of F7), the line ascends to its high point, C (the 5th), before descending chromatically to Eâ (the 7th). Eâ is even more strongly emphasized than the previous harmonic pitches A and C for several reasons: (1) as an eighth note, Eâ receives an agogic accent after the long string of sixteenths; (2) it also receives a metric accent as a result of its falling on the beat; and (3) it is approached by double chromatic neighborsâFâ above and D⎠below. As the two brackets below the upper staff in figure 1.2 point out, Adderley employs this double chromatic neighbor figure motivically, transposing it just after the downbeat of m. 27 to embellish C (the 5th of F7) at the end of the first gesture and again in m. 28 to embellish Aâ (the 7th of Bâ7) at the end of gesture B. These double neighbors, when combined with the following rests, bring about a sense of melodic closure, and help define the endings of these gestures. Gesture B not only ends with the same closing motive as gesture A, but also like A emphasizes harmonically significant pitches through its contour, beginning on Aâ (the 3rd of F-7), ascending to an embellished F (the root of F-7), passing through Bâ (the root of Bâ7), before finally descending to Aâ (the 7th of Bâ7).
Another similarity between gestures A and B worth noting is the rhythmic placement of these harmonically significant pitches. Without exception, all of these pitches occur in relatively weak rhythmic positionsâeither off the beat (as in the first two beamed pitches of each gesture) or, if on the beat, on the metrically weak second beat of the measure (as in the third beamed pitch of each gesture). This rhythmic displacement of harmonic pitches keeps the music off balance and pushes it forward, making the improvised melody seem as if it floats over the meter, rather than being constrained by it.
Figure 1.3. Eâ minor pentatonic scale.
Figure 1.4. Cannonball Adderley, "Groovin' High," saxophone solo, mm. 29-32.
Gesture C contrasts markedly with gestures A and B. Immediately noticeable is the change from sixteenth to eighth notes. Also noticeable is the fact that the melody of gesture C is not as consistently stepwise as the melodies in A and B, but contains a greater variety of melodic intervals. This contrast is, in part, a result of the melodic resources used in each gesture: A and B are constructed from a combination of chromatic and diatonic scales that Adderley uses to fill in gaps between chord tones, while C is constructed primarily from an Eâ minor pentatonic scale, written out in figure 1.3. Die minor pentatonic scale is frequently used by improvisers to give the music a "bluesy" or "funky" feel. This bluesy feel derives from the use of "blue" notesâprimarily the flatted 3rd Gâ and 7th Dâ, which in figure 1.4 clash colorfully with the G⎠and D⎠of the underlying EâMaj7.
Figure 1.4 shows not only the clashes that result from playing the Eâ minor pentatonic scale over EâMaj7, but also that gesture C differs from A and B in the rhythmic placement of its harmonic pitches. Whereas in figure 1.2 we saw that chromatic neighbors displaced harmonic pitches to relatively weak rhythmic positions throughout A and B, harmonic pitches in C regularly occur on the beat, frequently right on the first beat of the measure. This has the effect of "rooting" the melody to the beat, binding it to the meter, in contrast to the way the melody breaks free and hovers above the meter in gestures A and B.
Figure 1.5. Comparison of gestures A-B and gesture C.
In addition to the differences discussed so far, two more very subtle differences exist between gestures A-B and gesture C. These differencesâwhich are difficult if not impossible to capture in standard musical notationâhave to do with subtle manipulations of pitch and rhythm. In gesture C, pitches tend to be inflectedâthat is, they may go in and out of tune, or rather "slid up" to, "squeezed," or "smeared," to use common jazz lingoâin contrast to the precise, in-tune pitches of gestures A and B. Subdivisions of the beat in A and B are also more precise than that of C: in A and B, the melody divides the beat into four very even sixteenths, while in C, the beat is divided into two uneven notes, the first longer and the second shorter. In jazz slang, the sixteenths in gestures A and B are played "straight," while Adderley "swings" the eighths in C. The long/short pattern of the swing eighths further emphasizes the beat in gesture C, increasin...