by Mark Simos
Figure 7.8 is a melody in an “Over the Rainbow” vein—though not nearly as good. (Like its immortal precursor, it has an octave leap—but at the end of the section, and down instead of up!) It was written without an explicit, predetermined chord progression in mind.
FIG. 7.8. Stand-Alone Tonal Melody. Good melodic continuity.
This melody moves relatively freely in range and intervallic variety, showing both continuity and unity. It covers a good range for vocal melody: an octave plus diminished fifth, from the leading tone below the tonic to a fourth above the octave. It mixes steps, skips, and leaps, avoiding extended passages that are purely scalar or arpeggiated. The phrases cadence on structural tones that take on meaning and create contrast in terms of stable vs. unstable relation to the tonal center (F). The melody uses several motivic devices, including repetition, sequence, and extension. It has an arc, a narrative; it stands on its own and tells a story.
Partly as a result of these attributes, one could harmonize this melody in various ways. Two possibilities are shown in figure 7.9.
FIG. 7.9. Alternate Harmonizations of a Stand-Alone Tonal Melody
As these alternate harmonizations demonstrate, different harmonic settings can shift chord tone/non-chord tone relationships of given notes, e.g., the C(add9) chord against the D note in the last line of version 2. We can range from what a more restrained jazz “comper” might do, towards more extensive reharmonization—even shifting tonic, subdominant, or dominant harmonic functions on key notes (as in the fifth measures of the respective versions). Because of this freedom, progressions can follow their own voice-leading or bass-line logic, as in the line cliché in the second line of version 1, or the more unexpected linear bass line in the second line of version 2.
This harmonic flexibility can be used in developing or revising the song, or structurally—varying harmonizations of different repetitions of the same melodic figure or phrase within or across sections of the song. What’s remarkable is that harmonic progressions of such distinctive design can be created “after the fact,” set to melodies initially composed largely by reference to a tonal center and motivic principles. However, a key to this flexibility is that both melodies and harmony are, in the end, firmly rooted in the feel of major/minor tonality. This is evident from certain aspects of the example melody. The key is unambiguously suggested, by clues such as the leading-tone interval resolving back to the tonic, and the presence of tritone intervals (such as the E and BD in measure 6). Even the melody’s motivic construction is typical of techniques used for tonal melody. By relying heavily on sequence (both consecutively and at matched points in phrases) at the interval of the second, the melody combines figural integrity with the effects of recoloring chord tone/non-chord tone relations of constituent notes to the tonal center.
I believe these techniques lie at the heart of the elegance and sophistication of the Great American Songbook and related styles. While these vocal melodies characteristically moved in ways very different than vocal lines developed for
early polyphonic music, it could be argued that the melody/harmony relation-ships of this classic era of song repertoire express deep contrapuntal principles. Certainly, they represent an apotheosis of the progressive set of techniques explored so far in this chapter, for achieving independence of vocal melody with underlying harmony. These techniques allow creation of melodies with intrinsic interest in melodic contour, well suited to expressive styles that emphasize range and other aspects of vocal performance. The generations of interpreters, arrangers, and improvisers that have used this repertoire as a foundation have built their art on these inner principles of construction in the songs themselves.
For contemporary songwriters, these techniques also extend process flexibility, allowing us to work either melody first or chords first while still achieving a measure of independence in the resulting texture. A critical question remains. Can these techniques be applied and extended—or can comparable techniques be found—in other melodic and harmonic traditions?
Modal Melody/Harmony
Much of the world’s melodic music, and much contemporary popular music as well, is not driven primarily by functional harmonic progressions. I’d include in this category a vast territory of modal and pentatonic melody that can be found in music of many cultures and eras. (I won’t address purely atonal melody here; this would take us into domains of experimental composition beyond our scope.) One characteristic of such melodies is that they hold intrinsic interest without chords. Many such melodies were composed by musicians with no training in formal harmony, and in communities and styles where harmonic accompaniment was not part of traditional ensembles. This independence from a harmonic substrate can extend to ambiguity of where the tonal center is felt to be. For certain pentatonic melodies, even hearing them in the context of a full diatonic modal scale allows for alternate interpretations.
Such traditional melodies are beautiful and memorable in their own right, and in some traditions (such as Irish séan-nos or “old-style” unaccompanied ballads) are performed even today with no harmonic accompaniment. However, as traditional music styles have made the transition to contemporary audiences, one aspect of this evolution has been adventurous exploration of diverse approaches to harmonic accompaniment—for music originally conceived and performed in primarily melodic contexts. Accompaniment generally makes such music more accessible to modern harmonically acclimatized ears. Yet, the modal structure of the music allows for intriguingly open possibilities for harmonic interpretation, since many of the tunes were not originally constructed around preconceived harmonies. Contemporary traditional musicians explore these alternative harmonic possibilities in innovative arrangements and adaptations.
Expanding the Songbook
Even when working melody first, the Great American Songbook writers composed from an intuitive grounding in the language of tonal harmony. Like natural language, we absorb this vocabulary and “grammar” through acculturation and continual exposure to music in that harmonic style. Similarly, to learn to compose modal or pentatonic melodies, you need to gain some familiarity with the repertoire, idioms, and traditions where such melodic forms are native. It’s helpful to learn various modal and pentatonic scales, and to set out bravely to write in them. But to write convincingly in these styles, and more importantly, to make such resources part of your vocabulary as a writer, there’s no substitute for full immersion in the real stuff. I developed much of my own musical style through years of listening to diverse folk and world music repertoire, particularly Irish and American old-time ballads and fiddle tunes rooted in modal and pentatonic melodies.
To expand your work in the full 360° songwriting sense, I warmly recommend that you check out “the music of the rest of the planet.” But though I love these traditions, and hope you’ll find new (and old) music you’ll fall in love with as well, I also believe these resources can be applied broadly across genres and styles. Through my continuing education in contemporary music by my Berklee students (that is, teachers), I’m struck by the strongly modal and pentatonic aspects of melodies across a wide variety of current music—from neo-soul to topline pop writing (not to mention the resurgence of interest in roots music styles).
Counterpoint in Modal Melody/Harmony
Just as they’re conceived independently of chords, modal and pentatonic melodies also exhibit intriguing freedom in rhythm and phrasing. This rhythmic freedom stems in part from an intimate connection of song melodies to lyrics, although melodies and lyrics also intermix and recombine over time. Irregular time signatures and phrase lengths might exactly match the lyrics’ ebb and flow, or might sometimes mischievously fight the lyric with archaic mis-settings, breaking “conversational prosody” rules right and left.
Because many earlier traditional melodies were not composed based on a harmonic substrate, they depend for their musical interest and integrity, not just on a richer variety of tonalities to modern ears, but also on a remarkable dense web o
f motivic relationships within the music. This dense motivic approach can be applied to working chords first, with chords drawn from modal scales. In principle, this polarity between tonal and modal resources could be explored in a full 360° songwriting sense, working from directions of both melody and harmony. The resulting musical possibilities hold the promise of achieving a different kind of independence.
To bring these various musical worlds together, into a comprehensive “global school” of songwriting, is work for another day. (I admit it—there’s my “hole for a sequel, big enough to drive a truck through.”) But as a suggestive example, I offer, in figure 7.10, a melody composed without a preconceived single harmonic interpretation. The melody is mostly pentatonic, shifting to a modal feel at the end of the phrase. Yet, it is not strongly marked as being in any particular folk or ethnic style.
FIG. 7.10. Pentatonic/Modal Melody
This melody was neither generated in response to chords nor locked to an implicit set of chords. Though quite different in character from the tonal melody example in figure 7.8, like that melody it has characteristics well-suited to vocal melody. The melodic contour is shaped by a unique peak or high point (the G in measure 1), and valley or low point (the D in measure 5). It uses varied interval sizes and directional changes. The only arpeggiated chords occur across measure and implied subphrase boundaries (F D A across measures 2 and 3, and A F D across measures 4 and 5). Though using a range of rhythmic durations, the melody does not depend for interest solely on repeating rhythmic figures; it’s no more rhythmically than harmonically driven.
Rather, the sense and narrative of the melody derives largely from its motivic construction. The melodic contour elaborates an initial figural idea (D G F D) with a variety of motivic devices. Again, as with the tonal melody of figure 7.8, it uses sequence—but at the distance of a fourth. The result is a weakening of the sense of tonal center, with melodic contour taking more prominence. It also employs other devices such as isomelody (same melodic contour, varying rhythmic pattern), displacement (C A G A in measures 4 and across measures 6 and 7), and truncation.
As we sketch various harmonizations of this melody, differing chord choices change the chord tone vs. non-chord tone roles of various notes. Not only individual chords, but the overall apparent tonal center and mode (and therefore the palette of available diatonic chords) can shift. We might select from alternatives sketched during composition, or structure several alternatives directly into contrasting harmonizations of repeated melodic phrases.
Figure 7.11 shows four different “melody-led” harmonizations for the melody of figure 7.10. Harmonizations (a) and (b) interpret the melody in the tonal centers and modes of D Dorian and G Dorian respectively, made possible by the pentatonic figure and use of sequence. Harmonization (c) uses surprising juxtapositions of chords against the melodic line, as root motion in the progression takes on its own “melodic” and motivic aspect against the vocal melody. Harmonization (d) shows how the melody can preserve narrative flow, set against a suspended chordal drone.
FIG. 7.11. Harmonizations of Melody from Figure 7.10
Modal Palettes and Mosaics
While these various harmonic settings reveal something of the flexibility of the source melody, they are not necessarily distinct, in sound or process of generation, from tonal harmonizations showed earlier. If you build modal progressions seeking only the same kinds of “narrative progressions” we associate with tonal harmony, the resources can appear limited, and even confining. However, there is a different approach that can produce strikingly different textures. If, in an independent tonal melody, we melodize the key, not the chord, perhaps “independent modal harmony” involves harmonizing the mode, not the tune.
Where the link between independent tonal melody and harmony depends on a shared tonal framework, the link between modal/pentatonic melody and harmony can be thought of as a kind of “canvas”—painted or colored by chords drawn from various “palettes,” each defined by a given mode. Since melodic phrases can move between modes or, with pentatonic materials, allow for alternate modal interpretations, sometimes multiple palettes are available.
You can draw chord sequences from these palettes that create their effect more strongly by motivic principles—a kind of chord “mosaic” as it were. Such mosaics may involve rapid changes of chords compared to more conventional progressions, though the chords themselves are largely diatonic to a mode (at least in any adjacent passage). In contrapuntal terms, the chord sequence “thread” is brought to more prominence relative to the melody than in typical melody/harmony textures. (In a songwriting context, of course, these denser contrapuntal melody/harmony textures need to be reconciled with an appropriate focus on the lyric.)
In figure 7.12 we show some additional harmonizations for the melody of figure 7.10, in this “mosaic” textural style. For ease of comparison, I’ve kept harmonic pace consistent across the versions. From a process perspective, these chord mosaics can be generated in a more independent fashion, following their own design rules and implicit “root tone melody” relationships. Thus they may interact with the vocal melody in complex and surprising ways. Unusual non-chord tone relationships can be smoothed by the linear context of the moving chords, while even chord tone matches can be surprising in context. Shifts to alternate chordal palettes can “presage” or “echo” characteristic tones in melody (or vice versa). Note the use of motivic devices such as retrograde, sequence in root tone motion, and even rotational displacement—used not only within a given “mosaic” progression, but across the alternatives as well.
FIG. 7.12. Alternate Mosaic Harmonizations of Melody from Figure 7.10
Whether setting chords to melody or melody to chords, the contours of modal and pentatonic melodies seem to lend themselves to the slowly unfolding “melodies” of these flexible harmonic settings—moving in simple cycles, narrative progressions, or the chord “mosaics” discussed above. Achieving this independence in its fullest sense depends on understanding the contour principles discussed in the “Melody” chapter, the qualities of root tone motions discussed in the “Harmony” chapter, and the progression of contrapuntal strategies laid out in this chapter. I believe this is new territory to explore, potentially comparable in expressiveness to melody/harmony relationships in the tonal vocabulary of the Great American Songbook. Call it the Great Global Songbook—still to be written.
CHAPTER 8
Structure
In each facet we’ve toured—rhythm, lyrics, melody, harmony—we’ve engaged with structural aspects. Yet I’ve held off detailed discussion of structure as a whole until now. We tend to associate different kinds of structure with different facets: rhyme schemes for lyrics, phrase structures in melody, etc. But structure in songs is not the province of any one facet; rather, structural patterns are expressed in layered ways in multiple facets. Through structure, we unfold or “grow” seeds in any facet into matching question/answer phrases, up to an entire section. We also use structure in a top-down way, architecting songs from well-worked forms and using seed material to fulfill those structural plans.
Structure in the Song
In this chapter, we’ll touch on three aspects of structure in the context of
the song:
Phrase structure. Melody, lyrics, and chords weave together in a structural phrase: a duration (e.g., a number of bars if the phrase is metrically regular, though it need not be) marking a unit of musical time. The boundaries of a structural phrase are generally conveyed to the listener’s ear by some marking event in one or more facets—the first syllable of a line, the attack of a chord. But the overall structural phrase is not determined by the phrasing of material in any one facet. At times, it might only be expectations based on prior phrases, anticipated symmetry or regularity in structure, or genre conventions that tell the listener where to hear phrase boundaries. This allows us to counterpoint structure in any facet, not only against material in other facets, but against perceived bo
undaries of the phrase itself. This can be seen particularly clearly in lyric settings that shift the start of the lyric line in front of or behind the musical downbeat. The start of that next phrase might be cued with a chord change, but might be implied. As we’ll see later in this chapter, a similar principle is active when chords change on weak metric beats. Harmonic expectations may be set against the expectations established for the phrase length.
Motivic structure. In each facet, material is structured into patterns of similarity, contrast, and heard transformations. We’ve alluded to this as motivic structure, and seen its importance for creating compelling and memorable melodies and chord progressions. As motivic structure involves a great deal of repetition, it’s important to note that it appears strongly in the lyric facet as well. We tolerate and are moved by a remarkable amount of repetition in lyrics—repetition that would feel marked and mannered if heard in ordinary speech or even poetry. This is motivic structure “peeking out at us” through the lyrics.