Introduction

Author’s Preface

Analytic Toolkit

I write for all readers who would seek knowledge of music rhythm in Agbadza songs. I do not presume that readers have advanced knowledge of the theory and analysis of any of the world’s musics, whether Western Art Music or any of the world’s ethnic, folk, or traditional musics. Although I am enculturated into Western culture and have been schooled in Western institutions, I am largely self-taught in music theory and analysis.⁶ It may seem enigmatic, therefore, that my scholarly interest lies in transcription, analysis, and aesthetic criticism.

My analytic toolkit, so to speak, grew from direct engagement with Ewe performance arts. In trying to figure out “how the music works,” I have used a variety of notation systems and have explored diverse theoretical traditions. My writing aspires toward engaging the most sophisticated aspects of Agbadza’s music without either mystifying or condescending to the curious reader. I always try to make available audio files so that readers may also be listeners who do the hard work of bringing together the music itself with its representation in words and graphics.

The Musical Rhythm of Agbadza Songs

Our journey into the rhythm of Agbadza songs begins with the fundamentals of musical time in Ewe dance-drumming. The path begins with the bell part.¹⁰ In genres of Ewe dance-drumming music, the bell part sounds over and over as a recurrent temporal theme that gives to musical time a distinctive pattern or shape.

Selected Agbadza Songs

Let us now consider several songs. General rhythmic characteristics will emerge through discussion of these specific tunes and texts.¹⁸

In this discussion of musical rhythm in the twenty-five songs in the Alorwoyie Agbadza project, “Kaleworda” will represent a typical or average song. Its comparatively uncomplicated musical features are a good place to start.

Over the span of four bell cycles, song leader and singing group each sing the same two-sentence lyric about the lonely death of a strong warrior on a distant battlefield (see #7, Song Lyrics).¹⁹ The tune’s pitches array within an octave except for the upper A in the Leader’s opening motive (see #7, Lead Sheet). Leader works higher in the pitch set, while Group lowers the melody to its final note on the lower G. After opening the song with a dramatic, relatively wide upward leap (C to G), Leader moves in steps and modest leaps until another large downward leap (D to A) and final downward step to G. The group’s reply centers on C until it too descends to G with cadential leap-step motion (D–A–G). With the exception of B♭in m. 4, the tune uses five pitch classes.²⁰ To me, the song’s pitches move toward modal and temporal conclusion on the final G, but C also feels like another, complementary “tonicized pitch,” so to speak; this would mean that the song’s tonality is a pentatonic scale without half-steps in the modes G–A–C–D–F (2–3–5–6–1) and/or C–D–F–G–A (5–6–1–2–3). Both parts in the call-and-response are of equal duration – two “measures of bell,” so to speak – and set the text with the same time values as shown in Table 13.2.

The rhythmic design of its time values contributes to the musical personality of the melody. The words to the song are rendered in four nearly identical rhythmic figures, each spanning two four-feel beats (see Table 13.2). The idea stated in m. 1, that is, motion in eighth-note values between successive onbeats, establishes a pattern that is slightly modified in the three subsequent rhythmic patterns.

Subtle differences among these four rhythmic figures enable each variant to project its own quality to the flow of time within the span of one bell phrase and each has a particular relationship to notes in the bell phrase (Table 13.3).

Discussion of “Kaleworda” has introduced musical features common in most all Agbadza songs. Call-and-response between the leader and group parts is a foundational aspect of a song’s temporal design. The timing of the transfer in vocal action between Leader and Group parts, that is, the rhythm of call-and-response, and the consequent change in musical texture that results is an important component of Agbadza’s overall rhythm. Their exchange establishes a before-after temporal structure that provides an opportunity for antecedent-consequent musical logic, which may include aesthetic forces of tension-resolution. The timing of shifts in tonal centers within a pentatonic scale exerts yet another rather large-scale temporal effect. At a more fine-grained dimension, the rhythmic patterns of time values in the melody make polyrhythm with the bell phrase. As if it were another drum in the ensemble, the melodic rhythm may be heard to project musical forces toward other instruments, imparting nuances of accentuation on onbeat and offbeat timepoints to a listener’s interpretive experience of the polyphony.²¹ Finally, the song’s musical form, which is shaped by call-and-response design as well as by melodic factors of tunefulness, so to speak, has impact on a song’s rhythm through the comparative duration of its several sections.

Let us review the specific temporal features of this song that are characteristic of most songs among the twenty-five in the Alorwoyie collection. First, Leader was higher in the song’s range and had more tonal movement; Group quieted the rhythmic activity of the tune as it lowered the song’s pitches toward the finalis.²² Second, in a straightforward A1A2 form, Leader and Group both set the same text to identical time values; each part made a coherent melodic statement, but the two parts preceded and followed each other according to an Ewe musical logic of melodic gesture, pentatonic tonality, and rhythm governed by bell phrase and meter. Third, time values had a memorable theme – in this case, eighth-note motion through successive four-feel onbeats – that helped unify the tune.

Although I have proposed this song as being prototypical, every Agbadza song is unique. Overall, the genre has characteristic style, but each venerable song was intentionally crafted to convey particular meaning.

Like “Kaleworda,” this song spans four bell cycles and has two exchanges between Leader and Group (see #2, Lead Sheet). But the rhythmic design of “Miwua 'Gbo Mayi” is much more asymmetric and the relationship between Leader and Group much more intertwined.

The melody has three phrases with a rounded ABA form in the span of four bell cycles (see Table 13.5). Although the metric structure groups the ternary beats into sets of four (quadruple meter), the pattern of call-and-response confers an asymmetric design: 16 = (3+3) + 5 + 5 (see Table 13.4).

Leader and Group share in the song’s dramatic opening lyric, “Brave ones, open the gate. I will go” (see #2, Song Lyrics). Begun by Leader on four-feel beat three (m. 1), Phrase 1 requires a hand-off to Group on four-feel beat two (m. 2). Leader’s relatively long Phrase 2 fits neatly within one complete bell cycle: 1–2–3–4–1. As it did in Phrase 1, in Phrase 3 Group takes over the flow of four-feel beats from Leader on beat two (m. 4) with another five-beat gesture that extends through the next ONE: 2–3–4–1–2.

The tune adds more intricate melodic rhythm to this motion of metric units. The Leader begins the first phrase with upward and downward pendular leaps of a minor third interval (B–D–B) in a rhythm that aligns with the bell’s cadential motion over tones 5–6–7–1 (mm. 1–2).²³ Countering the structural tendency of the music to reach cadence on ONE, the Group quickly continues the melody’s rhythmic flow with an upward half-step on timepoint 2.2. Together, the melodic rhythm of the two sub-phrases in Phrase 1 articulates an important metric rhythm in Agbadza’s music: the oscillation within the span of one bell cycle between a half-measures “in three” and “in two” (see Table 13.6).²⁴

Each part is restricted to two pitches, but the Group part stands out for its long sustained note on C that sets the word with a key semantic image: brave Ewe warriors (see #2 Song Lyrics). Tonally and rhythmically, the melody creates a feeling of anticipation for phrase 2 (m. 3). Into this musical space, the Leader jumps boldly with a dramatic downward gesture that begins in polyrhythmic contrast to bell before aligning with its cadential tones to arrive at G on timepoint 1.1 (m. 4). In the lyric, this powerful melody establishes that the song is about struggle between the Ewes and their prototypical enemies, the Fon people of Dahomey. Although rhythmic motion of Phrase 2 achieves a sense of closure by aligning with bell’s cadence to ONE (m. 4), the Group again enters rather quickly (m. 4), this time with its own long phrase that arches upward to D before the final plunge to F♯ (m. 5), which to my ear leaves the whole song in an unresolved tonal condition. In a clever feature of the song’s text setting, the rhythm of the final word, “Dahomey,” imitates the two prior positions of “brave ones” (m. 2, m. 4). I especially enjoy the design of the rhythmic figures in this phrase, which suggest a palindrome: 3–2–1–2–3 (mm. 4–5) (see Table 13.7).

Some Agbadza songs feel especially drum-like (see Items #13 and #21): the sectional form moves quickly between Leader and Group, the melody reiterates only a few pitches, and the rhythms are repetitive and percussive. Compared to the tuneful setting of poetic text in songs like “Kaleworda,” these songs seem more like chants to “rally the troops,” so to speak. Because the singing functions like drumming, this song provides us with an opportunity to go deeper into the music of the drum ensemble.

The song lyric expresses quintessential warrior bravado: “The battlefield is for men. If I die, bury me there” (see #21, Song Lyrics). To enhance the feeling of urgency, Alorwoyie selected an extraordinarily intense composition for lead and response drums that sets the scene with the insistent statement, “On the battlefield,” and/or “The brave place” (see #21, Drum Language). Rhythmic intensity derives from the unusually short time span of the drum parts – only two four-feel beats. Two bounce tones from the response drum align precisely with a similar figure in the high-pitch support drum, thus joining the power of each instrument in a new synthesis (see #21, Full Score). One rhythmic consequence of the fusion of these two drumming parts is accentuation of the fast-moving eight-feel beats, which suggests a “double-time” feeling of tempo. (Compare to the quality of “cut-time” accentuation in “Ahor De Lia Gba 'Dzigo,” below.)

The song leader insistently intones the same lyric, “Battlefield-men’s place,” to a short descending motive (D–C–A) whose rhythm carries the feeling of metric closure – three–four–one motion of the four-feel beats – as well as the bell phrase’s cadence to ONE over strokes 5–6–7–1 (m. 1, m. 3, m. 5). The singing group responds with a sequence of two melodic phrases that end first on D (m. 3) and last on G (m. 5), which conveys a fleeting feeling of tonal and rhythmic stasis before the song’s next iteration.

The time values in Group’s part have an ingenious impact on the overall polyrhythmic texture. I enjoy hearing this rhythm as two successive occurrences of a four-note motive – ti ta ta ta – that is launched first from the pickup to four-feel beat two (timepoint 1.3) and then again from the onbeat of four-feel beat four (timepoint 4.1). The note with short time value (“ti”) functions like a temporal switch that toggles the melodic rhythm back and forth between the upbeats and the onbeats of the six-feel beats (see Table 13.8); the handclapping part gives phenomenal presence to this counter-metric field. The same toggling procedure happens within every cycle of the bell phrase: the short note on timepoint 2.2 shifts the bell’s long tones into unison with the flow of beats in the upbeat six-feel until the short note on timepoint 4.3 returns the long bell tones into unison with the flow of beats in the onbeat six feel. In this song, a similar procedure creates two identical rhythmic patterns that make very effective polyrhythmic interaction with bell.

Songs discussed thus far have illustrated rhythmic dynamism in Agbadza songs. Whether due to factors such as the duration of composed themes, formal design, metric accentuation, or the pattern of its time values, the melodic rhythm of these songs adds to the ever-changing quality of Agbadza’s overall musical temporality. The next two songs illustrate a different capacity: the steady and relatively unambiguous accentuation of one kind of metric field, that is, the flow of four-feel or six-feel metric beats. Although the musical rhythm of Agbadza will always be malleable to different interpretations, in these songs we hear and feel strong alignment between a song’s accentuation and the foundational time feels of Agbadza.

In many ways, “Ahor De Lia Gba 'Dzigo” is a classic Ewe song. The song lyric heralds a sneak attack on Adzigo, a legendary center for Ewe warriors, a message enhanced by the drum language’s command, “Put on your war belt” (see #17, Song Lyrics and Interview). Although more fully developed than “Kaleworda,” the design of the call-and-response, the melody’s shape, and the song’s form are typical for a dance-drumming song (see #17, Lead Sheet): an opening section (A1A1) in which Leader and Group twice exchange relatively long phrases (mm. 1–6); a middle section (B1B2) with faster call-and-response timing (mm. 6–10); and a reprise of Group’s phrase from the opening section (A2) (mm. 11–12).

Time values in Leader’s melody make a memorable rhythmic topography, so to speak (see Table 13.9, bold shading shows accentuation).²⁵

The prominent notes on every onbeat enable a listener to feel the melodic rhythm as conferring accent to the four-feel beats. With long notes initiated from timepoints 3.1 and 1.1 (mm. 3–4), Group’s reply reinforces this hard-driving onbeat rhythmic quality.²⁶ Because it continues with the same text and time values in its B and A2 section, the entire song has an “onbeat four” quality of rhythmic accentuation. This is not the full story, however, as will be discussed below after a brief detour into the theory of Ewe meter.

In Agbadza’s musical meter, four-feel beats with ternary subdivision (dotted quarter notes) always are balanced by six-feel beats with binary subdivision (quarter notes). The co-existence of two types of metric units imparts to the music a permanently ongoing three-with-two temporal ratio (3:2 over a half-measure; 6:4 within one bell cycle) that makes patterns in Agbadza’s music amenable to different rhythmic interpretations. The timing of the implicit four-feel beats will be so familiar to persons competent in Ewe music that the explicit iteration of the six-feel beats by the hand-clapping part in the Alorwoyie recordings likely makes for a pleasing counterpart. Just as some songs align to the four-feel beats, a song may also “be in six,” if I may put it that way.

“Dzogbe Milador” exhibits steady accentuation of the onbeat six-feel beats (see #12, Lead Sheet). Because the time values in the A section (mm. 1–5) tend toward uniformity in eighth notes, they do not suggest a particular accentual pattern in and of themselves. However, the syllabic division of words in the text and the choice of pitches in the tune bring out the “onbeat six-feel,” suggested by the bold shading in Table 13.10.

While this quality of rhythmic accentuation is unequivocally present in the Leader’s part, in the Group part (m. 2), the consecutive eighth notes on pitch A present a more rhythmically malleable situation that could be felt in sets of three, i.e., organized within ternary beats three and four.

In the B section (mm. 5–7), Leader and Group combine their incomplete melodic fragments to set one line of text to a full tuneful idea; the melodic rhythm continues to accentuate the onbeat six-feel beats (see Table 13.11, bold shading shows accentuation).

For the first time in our discussion, this song has a C section with important new information in the lyrics. In the A section, Leader and Group both conveyed the message “As warriors, we are prepared to die on the battlefield.” In the B section, the song belittled the effectiveness of the enemy’s weapons, “Your guns cannot shoot. Your knives cannot cut.” The confidence expressed in these lines is tempered in the C section: “Men will die in battle, while women await their own deaths back at home.” As if to give the turn in the song’s poetry a new musical setting, the melody’s pattern of steady accentuation changes dramatically from being “in six” to being “in four” (mm. 7–9). Melodic motion on B♭ and D confers the feeling of grouping within ternary beats onto the long set of nine eighth notes that lead to the onbeat dotted quarter note on G in m. 8, that is, the four-feel groove (see Table 13.12, bold shading shows accentuation).

Then at the midpoint of m. 8 (timepoint 3.1) comes a striking departure from the prior time values of eighths and quarters: a dotted figure followed by highly distinctive duplet motion through beat one of the next bell cycle (m. 9).²⁷ This is a clear instance of melody dramatizing the meaning of song text. In the closing return of section A (mm. 9–10), Group brings back its opening phrase, thereby ending the song with a return to its accentuation of the six-feel beats.

Multistability is the normal condition of musical rhythm in Agbadza. The primacy of the four-feel beats notwithstanding, the design of tunes usually enables more than one way to interpret the song’s rhythmic accentuation and melodic grouping. I suggest that this very quality of temporal dynamism is a reason why traditional genres of music like Agbadza have been popular among Ewe people for centuries. The songs and drumming never will become stale as long as people listen creatively. We return to “Ahor De Lia Gba 'Dzigo” to illustrate.

Above, “Ahor De Lia Gba 'Dzigo” served to exemplify steady accentuation of the four-feel onbeats. Returning again to this song, we can observe how its melodic rhythm also conforms to the resultant rhythm of time values in 3:2 between quarter notes and dotted quarter notes – ta ti ti ta, ta ti ti ta, etc.²⁸ In this song, the four-note 3:2 pattern is phrased ti ti ta TA, that is, from offbeat pickup, through onbeat two, toward onbeat one, with the final “ta” aligning to the moment when the two timing streams come together in unison (bold shading and capitalization shows accentuation). From the temporal perspective of the “three side” of 3:2, the melodic rhythm in this song may be said to consistently align with the “ands” of six-feel beats, that is, the flow of upbeat six-feel beats (see Table 13.13, bold shading shows accentuation).

Ewe metric theory reveals another consequence of accentuation on the “upbeat six”: moments of unison between upbeat six-feel beats and the onbeat four-feel beats occur on four-feel beats two and four, not one and three. In other words, notes timed to flow of the upbeat six-feel beats tend to accentuate the backbeats, a well-established hallmark of music in the African Diaspora.²⁹

Summary

The foregoing discussion has familiarized us with the overall nature of musical temporality in Agbadza songs and provided opportunity to articulate many of its more sophisticated features of rhythm. Let us summarize.

The bell part establishes the conditions of musical time:

• ▪ ever recycling temporal condition

• ▪ duration of time span or measure

• ▪ distinct pattern of sounded time values and unsounded timepoints using two time values – long and short

• ▪ two grouping shapes of full theme: (A) ta ti ta ta ta ti ta, (B) ta ta ti ta ta ta ti

• ▪ segmentation into fragments: (A) ta ti ta + ta ta ti ta

• ▪ onbell and offbell timepoints

• ▪ cadential motion over onsets 5–6–7–1 toward fleeting moment of stasis (ONE)

• ▪ strokes 7 and 1 are onbeat in the four-feel

• ▪ toggling between onbeat six-feel beats (onsets 1, 2, 3) and upbeat six-feel beats (onsets 4, 5, 6, 7)

Meter establishes duration, subdivisions, and structural accents:

• ▪ twelve timepoints within one bell cycle

• ▪ ternary-quadruple time or “the four-feel beats” (four groups of three) is foundational

• ▪ binary-sextuple time or “the six-feel beats” (six groups of two) is a permanent complement

• ▪ three-with-two (3:2) is omnipresent

• ▪ the accentual force of four-feel beats ranges from most stabile to most motile as follows: 1–3–4–2, that is, downbeat, midpoint, backbeat, backbeat

• ▪ four-feel beats 1 and 4 are onbell; four-feel beats 2 and 3 are offbell

• ▪ six-feel beats 1–3 are onbell, six-feel beats 4–6 are offbeat

• ▪ three timepoints within one ternary beat: the onbeat timepoint (1.1), the afterbeat timepoint (1.2), and a third timepoint (1.3) that may function as either an unaccented pickup if it leads to a subsequent onbeat tone or an accented offbeat if no onset occurs on the subsequent onbeat

• ▪ two timepoints within one binary beat: onbeat and upbeat

• ▪ matrix conception: steady flow of onbeats, offbeats, and upbeats in 3:2 ratio at different durational values

Accentuation, heightened feeling of a particular musical moment, is made in several ways:

• ▪ structural: resulting from permanent nature of bell, meter, recurring themes of parts in drum ensemble, and scale/mode

• ▪ compositional: resulting from design of song and lead-response drum composition

• ▪ onbell and onbeat accents: structural

• ▪ offbell and offbeat accents: compositional

• ▪ agogic accent (relative time value or duration of a note)

• ▪ positional accent: first or last note in a group

• ▪ pentatonic scales and modes: multiple potential tonal centers

• ▪ song finali often is tonal center but not always

• ▪ recurrent cyclic nature of music time continuously refreshes accentual patterns of motility and stasis

Song design has impact on musical rhythm in many ways:

• ▪ overall duration: from relatively short to relatively long

• ▪ organization of motion through metric fields and bell phrase

• ▪ moments of beginning and ending on bell and within meter

• ▪ timing of transfer between Leader and Group; rhythm of call-and-response; each part may achieve melodic closure or, alternatively, the two parts may combine to make one phrase

• ▪ duration of Leader and Group parts: long Leader–short Group; short Leader–long Group; equal duration of Leader-Group

• ▪ rhythm of tonal motion: motion toward and arrival at tonal centers; timing of moments of tonal stasis on bell and in meter

• ▪ temporal features of musical form (design of melody considered together with design of call-and-response): A sections – tuneful, B-sections – percussive, C sections – tuneful but different and distinctive

• ▪ overall before-after temporal/tonal patterns: from temporally busy and high-pitched at a song’s beginning to temporally quiet and low-pitch at its end

Melodic rhythm, that is, the design of time values in a melody, projects temporal force just as do the musical instruments in the drum ensemble:

• ▪ Duet with bell and each instrument

• ▪ Composite rhythm with other parts

• ▪ Metric placement of onsets

• ▪ Surface pattern: variegated time values make a definite rhythmic shape; unvaried time values have neutral temporal shape and are susceptible to being shaped by the force of other parts (malleability)

• ▪ 3:2 as a pattern of time values (ta ti ti ta); melodic rhythm often phrased ti ti ta ta.

• ▪ consistent accentuation of a metric field, and/or a metric rhythm such as three–then–two, or three–four–ONE

• ▪ musical dramatization of the meaning of song lyrics by a shift in accentual pattern or other means

• ▪ non-isochronous timing of two-note, short-long figures when short first note is onbeat

• ▪ temporal motion toward accentuation at the end of phrases

• ▪ clever design: palindrome; short riff repeated with difference on bell or meter; alignment with instruments in ensemble

• ▪ internal references: motivic variation, melodic sequence, recurring rhythmic figures

Conclusion

The onbeat four-feel groove in duet with the seven-note bell theme provides the ultimate temporal logic of Agbadza, but perhaps because this foundation is so well established, a plethora of countervailing forces may be put in play without threatening the music’s groove. Agbadza’s melodic rhythm might be characterized as iridescent: it resists a one-way interpretation and may be perceived to change depending on its setting in musical context. In Agbadza, everything musical happens within an interactive network of mutual influences: instrumental parts in a multi-part ensemble, meter as a dynamic matrix, and songs with multistable temporal design. Songs are designed to fit with other parts in interesting and musically satisfying ways. Like the other components of Agbadza music, a song acquires its full nature only in relationship to things outside itself.

Tala

The remarkable facility in rhythmic play demonstrated by musicians and dancers throughout the Indian subcontinent is as impressive as it can be bewildering for the listener. From local and regional practices, through devotional and popular genres, to the heavily theorized concert traditions of the North (Hindustani music) and South (Karnatak music), rhythmic complexity abounds. A performance may begin without even a pulse, where melodies seem to float unpredictably in musical space. Yet increasing rhythmic regularity leads to the establishment of repetitive sequences of beats, both evenly and unevenly distributed, which provide the frameworks for elaborate melodic and rhythmic compositions, variations, and improvisations. The entrance of drums – also essentially melodic in their subtle manipulations of pitch, timbre, stress, and resonance – is invariably a moment of great visceral as well as intellectual excitement. Together, singers, dancers, instrumentalists, and drummers build their performances around the anchors provided by the beats; they subdivide these beats in myriad ways, playing with different rhythmic densities and syncopations. The thrilling, rapidly articulated sequences with their offbeat stresses can temporarily disorient the listener until all seems to resolve in a triumphant convergence of surface rhythm and target beat. The rhythmic system as a whole and the individual frameworks of beats that serve to organize rhythmic expression are known as tala.

The Sanskrit term tala (Hindi: tal; Tamil: talam) is an ancient concept described in treatises close to 2,000 years old, and still today the word carries the same essential meaning of a handclap. Any attempt to summarize what are arguably the world’s most complex and virtuoso rhythmic-metric practices must necessarily begin with a definition of tala, for it differs from Western meter in a fundamental way. Meter is implicit: it is a pattern that is abstracted from the surface rhythms of a piece, and consists of an underlying pulse that is organized into a recurring hierarchical sequence of strong and weak beats. On the other hand, tala is explicit: it is a recurring pattern of non-hierarchical beats manifested as hand gestures consisting of claps, silent waves, and finger counts, or as a relatively fixed sequence of drum strokes.

The repetitive beat patterns of a tala are often thought of as cyclic, and certain words describing the cycle (avartana, for instance) are based on the Sanskrit root vrt, meaning “turning” or “revolving.” The circular representation shown in Figure 14.1, taken from an Urdu book published in 1869, maps out tintal, Hindustani music’s most prevalent tala of four beats, with each beat lasting four counts for a total of sixteen: it contains quasi-onomatopoeic syllables for the drum strokes (dha, dhin, ta, tin) used to represent the tala. Conceptually, the cycle begins and ends on sama (Hindi: sam; Tamil: samam – here, at the top of the dial), which is the beat representing the most common point of melodic and rhythmic confluence.

14.1 Cyclic representation of tala

Throughout this chapter, readers will encounter many examples of clapped beat structures as well as syllables representing the strokes that articulate rhythms. All are encouraged to engage physically with these phenomena by performing the patterns of claps, waves, and finger counts, and by orally expressing the syllables. For it is through physicality and orality that the musical system is taught. Such an embodiment of tala is crucial not only for achieving rhythmic competence and engendering creativity as a performer but also for deriving enhanced aesthetic pleasure as an audience. Indeed, audience participation through gestures marking tala is prevalent in the concert traditions, especially in Karnatak music, and allows audiences to experience and appreciate more keenly the rhythmic architecture of performance.

Tala in Karnatak Music

As an abstract structure, tala finds its most canonical form in the concert tradition of southern India: Karnatak music. We begin with the example of adi tala: a series of claps, silent waves, and finger counts that provides the framework for roughly 80 percent of songs and other composed works in this repertoire. In Table 14.1, the eight-count sequence of hand gestures provides both visual and sonic markers that allow performers and listeners alike to know precisely where they are within the tala at any given moment. This pattern begins with a clap of the right hand down onto the upturned palm of the left hand on count 1, followed on counts 2, 3, and 4 by taps of the pinky, ring, and middle fingers of the right hand onto the left palm; it continues on count 5 with another clap, and on count 6 with a wave, which is where the right hand turns palm upward and effects either a light tap of the back of the right hand on the left palm or a bounce in the air above it; another clap and wave on counts 7 and 8 conclude the sequence, and the pattern cycles back to repeat from count 1. As stated earlier, the tala has no internal accent structure like Western meter, not even on the clap marking count 1. The gestured 8-count pattern functions to provide a solid temporal reference for complex surface musical activity.

A musician trained in a Western tradition might well approach the clapping pattern of adi tala by doing the hand gestures and counting out the eight counts. Yet this approach is rare in South Asia, where musicians tend to use syllabic sequences to mark time rather than numbers. This results in a qualitatively different way of experiencing one’s relationship to the tala. The syllabic sequences are based on solkattu: a rich vocabulary of drum strokes and sounds that are expressed as onomatopoeic syllables like ta, di, tom, and nam. Returning to Table 14.1, we note the presence of eight syllables that should now replace the numbers as one performs the gestures. As a basic exercise, one begins with density level 1, where each hand gesture is accompanied by one syllable. Density level 2 doubles the speed of articulation of the syllables, although one must remember to strictly maintain the original pace of the hand gestures so that now each one is accompanied by two syllables. Density level 3 doubles the speed of articulation yet again, so that four syllables accompany each gesture. These three density levels are known as trikala, the “three speeds,” and all students of Karnatak music, whether melodic or percussive in form, learn this fundamental technique of changing rhythmic densities while maintaining the original pulse. Students of vocal music, for example, proceed through defined sets of scalar exercises sung to the solfège names Sa Ri Ga Ma Pa Dha Ni Sa, all the while clapping adi tala and applying the “three speeds.” As can also be seen in Table 14.1, an additional rhythmic exercise arranges the same syllables in triple time.

Adi tala is fundamentally duple in character, and as such it fits into the first of five classes of rhythm. This first class, or jati, is known as caturasra, or “four sided,” and is commonly articulated with the solkattu sequence ta ka di mi (or ta ka jo nu). As shown in Table 14.2, there are four other jati of 3, 7, 5, and 9 (this is their traditional order), each with its own pattern. The jati are important in several ways: they show how the beat may be internally subdivided into quadruplets, triplets, septuplets, quintuplets, and nonuplets, respectively; they can form the basis for calculating larger units of rhythmic improvisation; and they can serve to modify tala structures. This last point necessitates a brief discussion of the suladi sapta tala system.

Appearing first in the late nineteenth century, the suladi sapta tala (seven primordial tala) system quantifies seven basic categories, each with a distinctive gestural structure. The three gestures are laghu (symbol I: a clap plus a variable number of finger counts), drutam (symbol O: a clap plus a wave), and anudrutam (symbol U: a single clap). Adi tala belongs to the triputa category, which comprises one laghu and two drutam. The length of the variable laghu is determined by one of the jati categories: in the case of adi tala, the clap is followed by three finger counts for a total of four counts, and thus the laghu is “four sided.” Another, more cumbersome name for adi tala is therefore caturasra jati triputa tala. As can be seen in Table 14.2, the combination of seven tala categories with the five jati results in thirty-five distinctive tala structures, from three counts up to twenty-nine. What is interesting is that adi tala is not the only structure comprising eight counts, and yet tisra jati matya tala (clap pinky ring clap wave clap pinky ring) and khanda jati jhampa tala (clap pinky ring middle index clap clap wave) differ markedly in their arrangements of gestures.

In truth, however, very few of the thirty-five structures have been employed in performance practice, though one does occasionally hear uncommon tala structures used for exercises and technically challenging showpieces called pallavi that are designed to demonstrate technical virtuosity. The vast majority of compositions, including those of the greatest composers from the Golden Age of Karnatak music in the late eighteenth and early nineteenth centuries – the so-called “Holy Trinity” of Tyagaraja, Diksitar, and Syama Sastri – are set in just four tala: adi tala plus three that do not even belong to the suladi sapta tala system. These are also given in Table 14.2 and comprise only claps and waves: rupaka (3 counts), misra capu (7 counts), and khanda capu (5 counts). These structures are very likely to have entered the concert tradition from local or regional practices. Rupaka is an interesting and somewhat confusing case, as it shares its name but not its structure (clap clap wave) with one of the seven suladi categories (comprising one drutam and one laghu), and it also appears to be a relatively modern substitute for the ancient tisra jati eka tala (clap pinky ring). In practice, rupaka and tisra jati eka tala are interchangeable, and musicians choose according to the teaching lineage through which they have acquired their knowledge.

Rhythmic Play in Karnatak Music

The performance of a typical piece of Karnatak concert music begins with alapana, which is an exposition of the melodic motivic characteristics of a raga without tala. Although the melodies may seem to be free of any regular sense of rhythm, many musicians insist that they are mindful of an underlying pulse against which melodic expression is measurable. Alapana is often quite short, but nevertheless the expanding range of melodic motifs is complemented with increased surface rhythmic density.

Following the alapana is the composition, a text set to melody (or an instrumental rendition of one) that is always framed by a tala and thus always accompanied by a drum. The great percussion instrument of Karnatak music is the mridangam, a barrel drum with two heads made from layers of animal hide, laced together, and capable of being tuned by means of a permanent black compound applied as a low, circular mound in the center of the right head and by the application of a temporary ball of dough stuck and flattened onto the left. While the left head provides a deep, resonant bass, the right produces a variety of timbres depending on how and where the fingers and palm strike it. As the first strains of the melodic composition are delivered, the mridangam player must quickly identify the tala and the tempo, which then remain constant throughout the piece. The starting point for the melodic composition may occur anywhere in the cycle and could even begin on a half beat. Experienced drummers will likely know many compositions and may even play along with some of the prominent rhythmic signatures of the melody. As an accompanist, the drummer’s role is to support the melodic unfolding of the composition, mark the ends of sections of the exposition with rhythmic cadences, and contribute to the increasing energy and intensity of sections of improvisation that follow.

The rhythmic patterns played by the drummer fall into two categories: those that structurally maintain the flow of time, and those that disrupt it through rhythmic formulas that are calculated to terminate on a target beat within the tala, most commonly sama. The first category is known as sarvalaghu (from Sanskrit words implying “all short/easy”). Table 14.3 outlines a few simple examples of sarvalaghu, each of which the reader is encouraged to read out loud while clapping the structure of adi tala. These patterns have a tendency toward internal repetition that subdivides them into two halves and thus reinforces the repetitive groove resulting from the distinctive timbres and articulations. The groove takes on a particularly heavy swing in examples 4–7 (for instance, in example 5 one should sharpen the attack on TAka and exaggerate the weighty resonance of JOnu), and example 8 suggests greater surface rhythmic density, pointing toward patterns that become increasingly complex as pieces develop. Once one has gained familiarity with these patterns, one should double their speed to get a sense of how they sound in performance (yet maintain the original tempo of the clapping pattern – a metronome mark of roughly 84 counts per minute is a fairly typical performance tempo). A drummer will switch between many different sarvalaghu patterns according to the flow and rate of activity of the melodic exposition and its development.

The second category is known as kanakku, “calculation,” which is a vast and complex topic too large for anything but a cursory introduction. We shall briefly look at endings (mora), shapes (yati), and complex designs (korvai). All are configured in such a way as to create temporary uncertainty only to find familiar ground once again by directing our attention to a target beat. The simple examples given in Table 14.4 are borrowed from David Nelson’s exemplary Solkattu Manual.¹

A mora is a rhythmic cadence that ends a section of the music. In its simplest form, it is a sequence of strokes that is played three times: the reason for three statements of a given pattern is important. With just two statements, it would be difficult to anticipate the target beat, whereas with three, the pattern is not only more firmly established in the listener’s mind but also the temporal distance from the second to the third can be predicted to be the same as from the first to the second. The mora shown in Table 14.4 features the pattern ta ta kt tom tom ta tam, which covers 7 pulses (kt stands for kita and occupies 1 pulse). The body of the pattern is the 6-pulse statement ta ta kt tom tom ta, and tam is its end point. Tam may be followed by no gap at all, or more commonly with a gap of a variable number of pulses. In this case, tam is followed by a 2-pulse gap for a total of 3 pulses: tam – –. The mora, then, comprises (statement) + [gap] + (statement) + [gap] + (statement) for a total of 24 pulses. If the rate of rhythmic action in adi tala is 4 pulses per count, then the 8-count cycle will comprise 32 pulses. It follows, therefore, that in order to target the sama of the cycle on count 1 the mora should start after a gap of 8 pulses (in this case, those pulses are occupied by part of a sarvalaghu pattern from Table 14.3, ta – din – din – na –); in other words, the mora begins on the third count that is marked by the ring-finger gesture.

Yati refers to a series of operations that create shapes in the mind of the listener. The truly interesting ones among them are the cow’s tail (gopucca) and the river mouth (srotovaha), which represent narrowing and expanding operations, respectively. Retaining the same statement used for the first mora, we can see in Table 14.4 how elements are subtracted from the original phrase in gopucca yati, while the reverse is true in srotovaha yati. The gap in each instance is reduced to 2 pulses [tam –], and once again the total number of pulses for each sequence is 24. Therefore, these mora also begin on the third count of adi tala. By combining these two shapes, one can create two more: damaru yati (a small hourglass-shaped drum) with gopucca-srotovaha, and the barrel-shaped mridanga yati with srotovaha-gopucca.

In accompaniment, the mridangam player tends toward shorter, simpler mora structures. Yet often near the end of a concert piece the spotlight may shift over to the drummer for a solo that can run anywhere from two to ten minutes. This is the tani avartanam, and it marks a special moment of great concentration for the other performers on stage who attempt to maintain the clapping pattern of the tala as the only accompaniment to the sounds of the drum. Here the rhythmic designs are longer and more complex, and may involve changing the surface rhythmic density from duple to triple time, or even to quintuplets and septuplets. Compound mora structures are also increasingly likely, where a mora is repeated three times, thus prolonging the tension before a resolution on the sama of the cycle. But the tani avartanam must also have at least one grand, pre-composed structure: the korvai.

A korvai may feature all manner of clever rhythmic thinking, but at root it comprises a yati plus a mora. In the relatively simple example given in Table 14.4, there are four phrases of 10, 4, 4, and 3 pulses, respectively, which create the narrowing shape of gopucca yati. The composition repeats phrases abcd three times, then bcd, then b, and finally the mora statement and gap constructed from (c)+[d]+(c)+[d]+(c). A korvai may in fact be extensive, combining many sections, as long as it ends with a mora. They are often difficult to execute, and difficult to follow, but they represent the pinnacle of arithmetic thinking merged with musical aesthetics and technique, and they are quite thrilling to experience.

Finally, one may sometimes find more than one percussion instrument on stage in a Karnatak music concert, most commonly a ghatam (clay pot) and a khanjira (small tambourine). These are wielded with extraordinary technical skill, and are capable of replicating anything the mridangam can do.

A Brief Word on Local and Regional Rhythmic Traditions

There exists an extraordinary diversity of approaches to rhythm in South Asia, yet outside the concert traditions of Karnatak and Hindustani music there is relatively little detailed documentation or analysis of how precisely rhythm works. Certain scholars have unearthed evidence contradicting the notion that rhythm in South Asia is rigidly organized into isometric, that is, unchanging cycles of beats and counts. Jim Sykes² has described Sinhala Buddhist ritual music in Sri Lanka where drumming patterns can sound like unmetered speech, where long and short drum syllables set in lines of drum poetry often do not coincide with beats or pulses, and where sometimes the beats may even be stretched to match the durations of drum words. Without an insider’s understanding of the rhythmic logic of the drums in these ritual contexts, these irregular cycles and rhythms are difficult or impossible to count. Richard Widdess³ has noted how in many older repertoires of religious genres – such as Sikh gurubani hymns in the Punjab, Sufi devotional qawwali songs from Delhi, temple traditions from Lord Krishna’s heartland of Vrindaban, and religious and ceremonial music and dance among the Newar of the Kathmandu Valley in Nepal – heterometric rhythmic organization survives alongside isometric structures, and may have been (or indeed may still be) far more widespread that we realize. In a heterometric composition, the tala changes from section to section of the piece, unlike the concert traditions where the tala changes only if the composition does. In his work with Shi’a drumming groups active during Muharram (the annual period of mourning) in Muslim centers across India and Pakistan, Richard Wolf ⁴ also documented examples of heterometric structure. Additionally, in his study of Kota tribal drumming from South India,⁵ he introduced the important analytic idea of beats as anchor points that act as signposts that are especially important in coordinating group rhythmic practices. The spaces between beats can be flexible through non-uniform inflation, just as they are in the Karnatak system where the variable laghu can expand from 3 to 4, 5, 7, or 9 counts. Indeed, the rationale for many of the observations Wolf has made about a wide array of drumming practices focuses on the importance of the number of beats – both as a series of foundational anchors and also as stressed strokes in a surface rhythmic pattern – in the naming and identification of a tala. Of course, not all carry the name tala: though the term is widespread, others like cal (Hindi: “motion; gait”) or ati (Tamil: “beat”) are also found, and some traditions appear to have no word for tala at all.

One other very important analytic concept introduced by Richard Wolf is “stroke melody.” This resonates with what I wrote earlier about the extraordinary variety of different pitches, timbres, and articulations that drummers can produce on their instruments, either solo or in ensembles featuring several different kinds of membranophones and idiophones (small cymbals, for instance, that have always traditionally marked anchor points). Stroke melodies are prominent throughout South Asia, and are fixed patterns whose combinations of timbres and stresses set up what might best be described as a groove: a repetitive rhythm rooted in bodily movement that often involves offbeat stresses and that conveys a feeling of motion (compare the Hindi term cal). They often establish an underlying framework for other musical activity, and sometimes through variation, expansion, and changes in density they can be the focus of the musical performance itself. As we shall see, stroke melodies are also very important in the Hindustani concert tradition.

Hindustani Tala

The Hindustani tala system in fact harbors two systems that over the past two centuries have become enmeshed to such an extent that few would acknowledge any separation whatsoever. Yet extricating one from the other can prove instructive. The first lies within the domain of dhrupad: widely considered to be the oldest genre still performed, a dhrupad performance features a substantial unaccompanied alap (compare this with alapana), followed by one or two compositions set in tala and accompanied by a barrel drum called either pakhavaj or mridang (compare this with mridangam, to which it is structurally similar). The second pertains to all other types of concert music: vocal genres such as khayal, thumri, and so on; instrumental music of the sitar, sarod, and so on; and the dance form that during the twentieth century came to be known as kathak. All these genres are accompanied by the tabla, which along with the sitar has become a globally recognized symbol of Indian music.

Of the hundreds of tala structures that have been listed over the centuries in Sanskrit and Indo-Persian treatises, only four continue to appear with any regularity in the modern dhrupad repertoire. Of these, cautal and dhamar (12 and 14 counts, respectively) dominate slow-tempo compositions, and sultal and tivra (10 and 7 counts) frame those in fast tempo. Matra (etymologically linked to meter) is the commonly used word for a count: in the past, the matra corresponded to a healthy human pulse, but it is now conceived as a flexible unit dependent on tempo: laya. In the three categories, slow, medium, and fast (vilambit, madhya, drut), the matra can range from 12 per minute in the case of slow khayal compositions up to 720 in ever-accelerating instrumental climaxes.

Table 14.5 maps the beats of tala structures for dhrupad using only clap and wave gestures: unlike Karnatak tala, finger counts are not generally used, and certainly not systematically so. What dhrupad has in common with Karnatak practice, however, is the strict maintenance of the clapping pattern by performers and audience as an external representation of the tala in use, which in turn frees the pakhavaj player to support the melodic unfolding of the composition, mark the ends of its sections with rhythmic cadences, and contribute to the increasing energy and intensity of the performance – as is precisely the case with the mridangam player. Moreover, just as the mridangam player may choose from various sarvalaghu patterns to fill the tala cycle and contribute to rhythmic flow, the pakhavaj player too adopts repetitive, groove-like patterns. The first examples were notated in the early 1850s by Wajid Ali Shah, king of Awadh, a lavish patron and practitioner of music at his court in Lucknow. He called them theka, “accompaniment.” A theka is a fixed sequence of drum strokes that, when repeated relatively unchanged cycle after cycle, creates a recognizable representation of a tala – an aural symbol of it – and thus its presence largely obviates the need for the clapping pattern to mark time. However, Wajid Ali Shah’s theka for cautal would in subsequent years be interpreted merely as a kind of filler pattern akin to sarvalaghu, and was superseded in the late nineteenth century by another pattern that even today continues as the established theka representing cautal. The paradox is that in spite of the presence of these theka patterns, there is still a heavy reliance on external clapping patterns for dhrupad and pakhavaj performance. By contrast, clapping in non-dhrupad genres is rare. This raises three points: (1) that dhrupad and Karnatak performance are less removed from one another than is generally assumed; (2) that the pakhavaj accompanist spends very little time playing theka but rather quickly shifts gears into filler patterns and compositions, thus creating the need for an external set of markers for the tala; and (3) that theka is probably not native to the pakhavaj but instead owes its presence to the influence of the tabla. Theka is first linked to tabla in texts from the early nineteenth century.

Table 14.5 Tala structures for dhrupad

First appearing in the early eighteenth century, the tabla was, organologically speaking, a pakhavaj split into two parts and played with skins facing up. The substitution of a small hemispherical kettledrum for the bass gave drummers the ability to create extremely active pitch inflections of its resonant sonority, allowing it to replicate not only the rhythmic language of the pakhavaj but also that of the barrel drum dholak, widely used in local and regional musical genres as well as in the traditions of the Qawwals, who sang Muslim devotional genres as well as khayal, among other things. It was this flexibility that led to a growing preference for tabla in all genres other than dhrupad, but the drum owed its rapid spread throughout northern parts of the subcontinent to the popular songs and dances of female entertainers. Such performances were labelled “nautch” by the British (a corruption of nac, “dance”), and were very much the rage in the eighteenth and nineteenth centuries among Indians and foreigners alike.

Three rhythmic patterns dominated the nautch: a compound duple with hemiolic properties called dadra (3+3 / 2+2+2), a lilting ghazal (3+4), and a duple kaherva (4+4). These were articulated on tabla with short, fixed patterns – stroke melodies – that likely underwent embellishment and intensification without deviating greatly from the beat pattern other than to end a section of music or dance with a short cadential flourish. As tabla began to be used for other genres with longer beat patterns, the grooves largely remained intact, but were played twice in order to fill the extended tala structures. A case in point is ektal, whose clapping pattern derives from cautal for pakhavaj, but whose theka consists of a dadra groove played twice, as seen in Table 14.6. Yet in order to differentiate between the repetitions, and thus ensure that the theka marks the sam of the cycle, the repeat removes the bass resonance (voiced phonemes dhin dhin become unvoiced ones, kat tin). This highlights the timbral difference between the opposing polar axes of sam at the beginning of the cycle (clapped and marked with an X) and khali halfway through it (waved and marked with a 0). Khali, the “empty” beat, therefore becomes an important marker in a bipartite structure. The same process can also be seen in Table 14.6 with the most common of all tabla tala structures, titala, which in an older form was called dhima titala that was adapted from dhrupad-pakhavaj repertoire (but that is now rarely heard there). Dhima was only one form of titala, however, as other stroke melodies appropriate to different song genres were also in common usage. A few of these are shown in Table 14.6, but most are now obscure owing to the almost complete dominance of tintal (also represented in Figure 14.1), which was originally used in dance accompaniment. Nowadays, the vast majority of vocal and instrumental genres of concert music accompanied by tabla use tintal, jhumra, dipcandi, ektal, jhaptal, and rupak (this last, a structural anomaly, is representative of an iambic lilt influenced by regional and popular forms of 3+4). These patterns are also given in Table 14.6.

Rhythmic Play in Hindustani Music

In dhrupad, a melodic composition is stated several times before undergoing many kinds of rhythmic transformation of text and tune, with the beats being subject to increasingly denser subdivisions. The pakhavaj accompanist tries to match these, and can draw on a variety of filler patterns, as noted earlier, or on compositions known broadly as paran. Table 14.7 outlines two simple paran compositions: the first is known as sath paran, sath meaning “with; together,” suggesting its use in accompaniment; the second is a mohra that incorporates a threefold repetition generally called tihai in Hindustani music (compare this with the Karnatak mora). In these, the typically unbroken stream of stroke phrases in the sath paran contrasts with the broken pattern initiating the mohra before the tihai (ghege tite kata gadi gena dha –) directs our attention to the sam.

There is no equivalent of Karnatak music’s tani avartanam in dhrupad, but instead the pakhavaj may be heard as a discrete solo item in a concert. Here, drummers embark on sets of longer, varied paran structures, including the chakradar paran that will comprise the threefold repetition of a paran plus a mohra, calculated to end on sam. Pakhavaj players have resurrected many older, obscure tala frameworks as the basis for solo performances, many possessing names of Hindu gods: Brahm, Rudra, Lakshmi, and so forth. They also recite and play compositions that blend drum syllables – bol – with lines of verse praising gods: the elephant-headed Ganesh, Remover of Obstacles, is a popular subject for an opening piece, an invocation for an auspicious blessing.

The role of the tabla as accompaniment to vocal genres differs from pakhavaj, because it is confined to a far greater extent to maintaining the theka, with very few opportunities for solo flourishes. In instrumental music, the modern trend has moved increasingly toward a collaborative performance where the accompanist is given several opportunities to perform solo, during which time the melodist maintains the raga composition as an aural marker of the tala. Theka and melodic composition become important frames of reference for the tala structure in the absence of the clapping gestures of Karnatak music and dhrupad. This is true also of lahra, a tune specifically designed to accompany the discrete genre of tabla solo.

Many different types of composition are available to the tabla player, but once again all fall into one of two categories: those that maintain the structure of the cycle and those that are calculated to end on a target beat, the sam. As far as the latter category is concerned, tabla borrows heavily from the structures of the pakhavaj: tukra is the equivalent of mohra, ending with a threefold tihai, and a chakradar-tukra repeats that structure three times. A gat is a specialized composition of mostly tabla material and is prized for its specialized techniques: some end with a tihai (gat-tukra), and others blend tabla material with pakhavaj phrases (gat-paran).

What truly sets tabla apart is the manner in which pieces that maintain the cycle are structured and performed. Peshkar (“presentation”) and bant (“division”) are slower, introductory compositions, qaida (“base; rule”) is the primary vehicle for developing variations on a theme, and rela (“torrent”) presents a stream of rapidly articulated phrases. There is considerable evidence to suggest that these compositional types emerged from theka and its embellishments and variations, particularly those for tintal. Crucially, all are subject to transformations dependent on the khali of the cycle. Take the popular late nineteenth-century Delhi qaida shown in Table 14.8: the 8-count theme (dhati tedha tite dhadha tite dhage teena kena) occupies the first half of the cycle, and is then repeated in the second half (tati teta tite dhadha tite dhage dheena gena). The right-hand, treble strokes remain the same, but the left-hand bass strokes change from open, resonant sounds to damped ones. As noted earlier with ektal theka, the transformation is represented by a phonemic change from voiced syllables (dha, ge) to unvoiced equivalents (ta, ke) as the theme approaches the khali, and then by the return of voiced syllables as the repeat returns toward the sam. Typically, the qaida is then played at twice the rhythmic density, though the theme continues to be subject to the bipartite division of the tala into sam and khali halves. Variations (vistar, “spreading”) are built from the components of the original theme by repeating, permutating, expanding, and compressing its phrases. Dohra (“double”), for example, is a common method of repeating the opening phrase three times. The qaida ends with a tihai based on the original theme or one of its variations. This Delhi qaida with a short sequence of variations and concluding tihai (bracketed) can be seen in Table 14.8.

Rhythmic Diversity or Unity?

In a region of the world so obviously socio-culturally diverse, different approaches to musical rhythm are to be expected. Yet as this necessarily brief introduction to rhythmic thought and practice in the Indian subcontinent has tried to show, there is much more that unites the region than divides it, in spite of the tendency of many musicians and scholars to maintain distance between Karnatak and Hindustani music systems, or between elite/concert and local/regional traditions. The fundamental orality of rhythm is ubiquitous, as is the explicitness of tala either as gestured, quantitative structures based on arrangements of beats or as qualitative stroke melodies that articulate and represent them. Indeed, it is clear that both Karnatak and Hindustani rhythm combine these quantitative and qualitative approaches to tala, and that the rhythmic strategies in their respective performance contexts are really not so different. The all-important beats of a tala are anchors that organize the flow of time, frame composition, and coordinate creativity: we have seen that this flow can be maintained with stroke melodies like sarvalaghu and theka, and through various rhythmic compositions that are bound to and reflective of the structure of the tala cycle; and we have noted how the flow may be disrupted by rhythmic patterns and compositions calculated to target a specific beat, most commonly sama/sam, the principal marker of creative confluence. One does not need to understand complex theory to sense the sheer excitement of rhythmic performance in South Asia, but an awareness of its beat structures and patterns of maintenance and disruption will most certainly enhance enjoyment of what is one of the world’s most thrilling systems of rhythm.

Setting the Scene

Just outside the temple gates, a string of hawkers sells grilled saté, peanuts, and suckling pig, while nearby, the fast cyclic melodies of the bamboo gamelan joged accompany a flirtatious dance. Passing into the first temple courtyard, I’m immediately enveloped by the bright bronze clangor of the gamelan gong kebyar, whose high energy, syncopated rhythms, and tight coordination give the twenty-five-piece ensemble of gongs and metallophones its name: kebyar – explosion. Thirty paces away, the members of a small gamelan geguntangan accompany a cast of singer-dancers in an arja theatrical performance, the intricate interlocking of drums and cycle-marking gongs driving the action beneath long ornamented melodies of voices and flutes. There are other ensembles in the next courtyard, of iron and wood and bamboo, singers and dancers and puppets, short trance-like cycles and slow winding melodies, each sounding independently yet in concert: a truly Ives-ian rhythmic experience.

Unraveling Balinese Rhythm: Cyclicity

An understanding of Balinese rhythm begins with an understanding of Balinese cyclicity. Not all Balinese music is cyclic to be sure. The flashy introductions of gamelan gong kebyar works are almost exclusively through-composed, as are diverse practices of sung poetry and much contemporary composition. The extended melodies of the classical instrumental form lelambatan, while cyclic, are often so long and elaborate they’re unlikely to be perceived as such by the uninitiated listener. And while most gamelan works do contain cyclic melodies, their successions accumulate to large-scale forms conveying a sense of forward motion and change.

That said, most of the rhythmic features for which Balinese music is internationally celebrated are made possible through cyclic construction. Whether the brisk 2-beat vamp of an arja fight scene or the slow 128-beat pengawak melody of a refined legong dance, much Balinese music is built to repeat without ending, the final note of one cycle melting into the first note of its repetition. Important moments in a cycle are punctuated by strokes on a collection of hanging gongs creating a hierarchy of more and less structurally significant tones in a melody. The final note of each cycle is generally marked by the largest hanging gong, often the low-voiced bronze gong ageng, or “great gong,” while smaller gongs punctuate other important points, typically evenly distributed throughout the cycle.

A common 8-beat gong structure called Bapang uses the medium-sized kempur (often abbreviated “P”) and small high-pitched klentong (“t”) in a symmetric pattern with gong ageng (“G”). Together they create the cyclic structure (G) – P – t – P – G, while the horizontal gong kempli strikes a steady beat. A contrasting 8-beat structure called Gilak uses just the lower gong ageng and kempur in an unusual asymmetric pattern of (G) – – – G P – P G. Its dense second half gives Gilak the strong feeling that Balinese composers prefer for dance characters like warriors, for instance. Thus does rhythm create feeling – rasa – in Balinese music.

While mid-twentieth-century Western composers didn’t show particular interest in or understanding of Balinese rasa, many were drawn to Balinese cyclic construction. In his opera Death in Venice, premiered in 1973, Benjamin Britten employed cyclic melodies and even mimicked cycle-marking gongs. An excerpt from the Act I, Scene 5 beach scene, shown in Example 15.1, sees the tamtam and double bass playing the role of gong ageng, demarcating a 9-beat cycle. The dry quarter-note strikes of the tuned drum and repetitive cello pizzicato act as kempli beat-keepers. Above these structure-marking instruments, the xylophone plays a zigzagging melody that repeats with variation every 9 beats, while the glockenspiel cycles a gently varied melody every 4 or 8 beats, in cross-rhythm to the cyclic structure.⁶ Here and elsewhere, only the instruments discussed are transcribed.

15.1 Cyclic structure in Death in Venice

Later in the same scene, Britten employs two tamtams of contrasting sizes as well as octave leaps in the cello, bass, and harp to outline a 9-beat cyclic structure with gongs of different pitches, similar in feel to Gilak.

Colin McPhee likewise made use of cyclic structures in his celebrated gamelan-inspired piece, Tabuh-Tabuhan: Toccata for Orchestra and Two Pianos, which premiered in Mexico in 1936. In Example 15.2, excerpted from Rehearsal F of the first movement, McPhee perfectly parallels a Bapang structure: low bass and cello pizzicati as well as octave Gs in Piano II function as gong ageng, while bowed cello and harp fill the role of the first kempur stroke. The high klentong at cycle’s midpoint is marked by a higher chord in Piano II, doubled on viola, and the harp sounds the final kempur. Above these cycle markers, the flutes and clarinets perform an 8-beat call-and-response that, while not strictly cyclic, maintains some consistency of pitch and contour.

15.2 Cyclic structure in Tabuh-Tabuhan

A Different Metric Conception: End-Weightedness

One of the most distinguishing features of Balinese cyclicity, yet that does not translate to these Western compositions, is “end-weightedness” – a system of metric organization where the stress is felt not at the beginning of each cycle but at its end. Instead of counting an 8-beat gong pattern with the strongest beat at the start (1-2-3-4-5-6-7-8-1), Balinese musicians feel the gong ageng at cycle’s end (8-1-2-3-4-5-6-7-8). In the preceding discussion, the first “G” of each gong pattern appears in parentheses because it actually “belongs” to the previous cycle; Bapang is heard as (G) – P – t – P – G, not G – P – t – P – (G). While this may seem a mere semantic argument, the distinction between front-weighted and end-weighted meter is vital to an understanding of Balinese melodic and rhythmic construction. Melodies are composed to lead toward strong beats, their grouping structure boundaries ending rather than beginning at these points. Melodic contours and rhythmic patterns often feature increased levels of activity or motion leading up to a gong stroke with more stasis or stability in its wake. Known as ngegongin, “leading to the gong,” this practice creates an underlying rhythm – somewhat akin to a harmonic rhythm – that densifies over the course of a cycle. End-weighted thinking can thus generate waves of stasis and motion, ngubeng and majalan, as cycles repeat again and again.⁷ At a yet more foundational level, it dictates textural relationships among musical strands in a composition, resulting in a uniquely Balinese-style heterophony.

Heterophony and Stratified Polyphony

Like many traditions across Southeast Asia, most Balinese music could be considered heterophonic: different instruments or musical strands simultaneously perform a single melody, each in their own way. Yet, unlike the relatively un-systematized heterophony of much Indian or Arabic classical music, where diverse instrument idioms and personal improvisatory styles shape subtleties of timing and ornamentation, Southeast Asian heterophony is often highly systematized, characterized rather by differing rhythmic densities across an ensemble. Thus, it is perhaps more accurate to think of such music as rhythmically stratified polyphony.

“No voice in the gamelan is without its rhythmic function,” McPhee wrote.⁹ In the internationally renowned gamelan gong kebyar, the main melody is generally played by the 10-keyed metallophone ugal, whose player uses grace notes, note doublings, and the limited timing flexibility reserved for solo instruments to ornament a simple quarter-note rhythm. Loosely doubling the ugal may be a group of bamboo flutes – suling – playing the same melody with flexible timing, ornamentation, and pitch shading. A pair of 5-keyed metallophones called calung (“cha-loong”) play a sparser abstraction of the melody compressed into a one-octave range. They track the ugal's contour largely at the half-note density, aligning most importantly with the gong ageng at cycle's end. The calung’s spare melody is often called the pokok or “core melody” and, as we will see, forms the melodic basis for all other strands in the texture, acting as McPhee’s “stem.” At half or one quarter the density of the calung, and one octave lower, is a pair of metallophones called jegogan, playing the sparsest version of the melody at the whole (or double whole) note density. These various instruments comprises the second, fourth, and fifth staves in Example 15.3: a 16-beat melody from the dance piece Oleg Tumulilingan. Vertical pitch class convergences are circled, and the first note on each staff is parenthesized to indicate end-weightedness.

15.3 Balinese stratified polyphony

The middle staff in Example 15.3 begins to offer insight into the impact of end-weighted thinking on melody-making. A pair of metallophones one octave above the calung, and playing at twice their density, the penyacah derive their melody from the calung’s, anticipating each new calung arrival with the tone either directly above or directly below it in the gong kebyar’s 5-tone scale (here notated C♯-D-E-G♯-A).¹⁰ Only beats 6–7 show the composer (or player) exercising increased artistic license, temporarily misaligning penyacah from calung to create a smoother contour in the penyacah’s melody. The top staff in Example 15.3 shows an elaborating melody played by an octet of metallophones called gangsa. These musicians, too, heterophonically track the calung’s tones at the half-note density, in this case filling in the other seven sixteenth notes in an idiomatic elaboration style called norot. Characterized by neighbor-note oscillation,¹¹ norot also reflects an end-weighted conception. The gangsa’s C♯-C♯-D-C♯ contour, for instance, anticipates the calung’s beat-4 C♯; this melodic segment ends on the beat with calung. The gangsa’s next segment is felt to begin on the weak second subdivision, leading to the following calung tone A through the same pitch anticipation technique.

Bali’s stratified heterophonic polyphony influenced mid-twentieth-century Western composers in two key ways. First, the simple concept of heterophony allowed composers wishing to divorce themselves from the constraints of functional harmony a fresh organizing principle. Britten, for instance, had a “strong interest in the vertical conflation of linear material” and found kinship in Balinese music where “any ‘harmonic’ element is a by-product of, and directly related to, the melody.”¹² In the orchestral Prelude to Act II of Death in Venice, an excerpt of which is shown in Example 15.4, Britten employs the timing flexibility of solo instruments like ugal and suling to create a close heterophonic texture in the strings, while sustained tones in bassoon, horn, and bass might be said to mimic various gongs.

15.4 Flexible heterophony in Death in Venice

Even more widespread, however, was the adoption of a rhythmically stratified polyphony. At times, Western composers employed this device without the vertical coincidence of common tones so central to Balinese stratification. In the Prologue to Britten’s 1940 operetta Paul Bunyan, a simple E-F♯-G-F♯ contour is layered at varying densities: the piccolo cycles it at the sixteenth-note density, the oboe in triplets, the clarinets in eighths, the horns in quarter notes, the trombone in half notes, and so on. As Cooke observes: “There can be no doubt that the polyphonic stratification of this passage, in which rhythmic activity increases as the register rises, was influenced by Balinese music.”¹³ Somewhat truer to a Balinese conception of density stratification is the passage in Example 15.5, excerpted from Act III, Scene 2 of Britten’s 1957 ballet The Prince of the Pagodas. Two gongs, a larger and a smaller, outline a cyclic structure doubled by Piano II, harp, and low strings (not shown here). Piano I, piccolo, and xylophone perform a stratified polyphonic melody, vertically aligning almost every eighth note and filling in the spaces in their own densities and idioms. The trombones play a more rhythmically diverse melody, only very occasionally stating the same pitch class as the other instruments. Here, Britten may be imitating the freer idioms of ugal and suling, but is as likely breaking from the Balinese palette altogether.

15.5 Stratified polyphony in The Prince of the Pagodas

McPhee, too, used stratified polyphonies in his work. In the first excerpt from Example 15.6, Rehearsal M of Tabuh-Tabuhan’s opening movement, a 3-note core melody (pokok) in Piano II accompanies shifting sixteenth-note ostinati in Piano I. The ostinati’s C-G-A♭ downbeats almost perfectly parallel the C-G-A♮ pitch collection of the pokok, allowing vertical alignment at the beginning of most measures. The second excerpt shows just the piano and bassoon parts from Rehearsal G of the piece's third movement, where vertical pitch class convergences outline a 3+3+2 rhythm.

15.6. Stratified polyphony in Tabuh-Tabuhan

The ostinati in both these excerpts likely grew out of what McPhee saw as the gamelan’s “blossoms”: fast-moving melodic elaboration styles used on gangsa and other instruments. Their distinctive contours, and the playing styles that generate them, bring us to one of Bali’s most idiosyncratic musical techniques: interlocking polyphony.

Interlocking Polyphonies

Interweaving polyphonic traditions, from Central African horn and vocal musics to West African drum ensembles, often feature widely divergent ranges in pitch and timbre among performers, allowing listeners to follow individual musical strands while also taking in the polyphonic whole. A Balinese performance is an altogether different sort of experience. Many non-Balinese musicians are first drawn to Balinese music (and continue to love and play it for years) because of a distinctly Balinese brand of interlocking, where a melody or rhythm is seamlessly shared between two or more performers such that the resultant composite is a single, smooth strand of music. One often cannot discern which musician has performed which note in an interlocking passage; the perceptual effect is of a group of musicians each playing the entire passage, in perfect synchrony, much faster than humanly possible. As a performer, the embodied experience of playing some notes of a melody while feeling as though one has played them all is equally exhilarating! Whether that melody is blindingly fast and flashy or slower and more subtly constructed, the act of meticulously fitting together jagged rhythmic puzzle pieces to form a perfectly smooth surface texture with one’s musical partners is deeply satisfying. And while similar interlocking techniques exist in other Indonesian practices (the Central Javanese saron’s imbal technique, for instance), in Bali they dominate musical textures.

The 3+3+2 rhythmic patterns in Example 15.6 exemplify a familiar division of eight pulses in a duple metric framework that crosses cultures and continents. One of the most basic Balinese forms of this rhythmic device occurs in the vocal genre kecak, often called “the monkey chant” by Western observers due to its incorporation into a twentieth-century simian dramatization of the Ramayana Hindu epic. In it, a chorus of dozens or even hundreds of men chant the syllable “cak” (“chak”) in a thick polyphony of syncopated ostinati, their short, dry shouts grounded by a vocalized kempli beat-keeper. Each simultaneous strand is built to rhythmically complement the others and their tight interlocking creates a steady, uninterrupted stream of notes. There are many different patterns in a kecak performer’s toolkit. One of the most common is a group of three interlocking rhythms together known as cak telu (“three”), so named because each rhythmic strand contains three notes; cak telu performers chant 3+3+2 rhythms, each offset from his partners by a single pulse. The top left transcription of Figure 15.1 shows cak telu’s interlocking texture in end-weighted notation; on the right, the three individual rhythms are transcribed in Western notation, their different rotations circled.

15.1 3+3+2 rhythms in polyphony

A similar group of patterns is used in the interlocking crash cymbal (ceng-ceng) playing of the high-energy processional ensemble gamelan beleganjur. Here again, a stream of uninterrupted notes is created by three concurrent 3+3+2 rhythms in different metric rotations, performed against a beatkeeper as in the middle transcriptions in Figure 15.1. Called kilitan telu, the slightly denser rhythms of these interlocking patterns ensure no performer has two consecutive rests.¹⁵

Short, interlocking ostinati such as these were particularly compelling for Steve Reich, whose interest in minimalism favored “a dense, tactile, repetitive, and driving music.”¹⁶ As such, he was drawn to the layered polyphonies and “precise rhythmic blending of the ensemble”¹⁷ that he found in Balinese music as well as in drumming practices from Ghana. The middle section of Reich’s 1973 Music for Pieces of Wood is kin to the telu patterns just discussed.¹⁸ The final transcriptions in Figure 15.1 show three adjacent rotations of a closely related 3+2+3 rhythm, played on claves in Reich’s piece, with one additional player keeping a steady beat. While the 3+3+2 rhythmic trope is widely used, “phasing [it] to interlock with an audible regulating pulse as Reich does nevertheless points to Kecak [and kilitan telu] in its specific arrangement.”¹⁹ It seems particularly noteworthy that, while kilitan telu patterns fill in extra notes just before the main 3+3+2 onsets, Reich fills in the notes directly following them. I see this as a perfect reflection of end-weighted versus front-weighted metric thinking, with the extra kilitan telu notes leading to strong onsets in the 3+3+2 construction while, in Reich, they lead away from them.

Balinese melodic interlocking generally requires two musicians rather than three (though sometimes four are needed), but the principle remains the same: multiple strands of syncopated patterns, frequently rhythmic rotations of one another, together create an uninterrupted melody that can be played very quickly if necessary. Often generically termed kotekan, meaning “interlocking parts,” this Balinese approach to melody-making comes in a variety of techniques. In each two-person technique, a pair of musicians playing the same instrument type – two gangsa metallophones, for instance – shares a melody between them through carefully prescribed roles and rhythmic formulae. One musician in an interlocking pair is the polos, “basic” player, whose pattern generally falls more “on the beat” and whose pitches more closely track the slow pokok core melody; her partner, the sangsih or “complementary” player, performs a related rhythm, often with more off-beat onsets.

When the gangsa’s norot melody from the top staff of Example 15.3 is played sufficiently fast, no one player is capable of performing it alone. The melody is then divided between polos and sangsih partners per Example 15.7.²⁰

15.7 Interlocking gangsa norot

Here again, interlocking partners play identical rhythms offset by one subdivision. The polos, notated stems down, strikes mostly on the beat and sounds just the core melody’s tones; the sangsih plays largely off the beat and often strikes non-core melody upper-neighbor tones. Each note sounds in at least one of the two strands, resulting in an uninterrupted stream of onsets. Pairs of Balinese hand-drums play comparable interlocking patterns of bass strokes, slap strokes, and various ringing strokes, which find partial reflection in the alternating pitch ranges of Britten’s xylophone melody from Example 15.1.

Related to the 3+3+2 rhythms of kecak and kilitan telu are two styles of melodic elaboration frequently referenced by McPhee. In both, elaborating instruments like gangsa anticipate the arrival of new pokok core melody tones through repeated three-note ascending or descending gestures at the sixteenth-note density, and here the impact of end-weightedness on grouping structure is particularly salient. If “x” is the current pokok tone (irrelevant to the contour of these elaborations), and “1” the upcoming tone (very relevant in this end-weighted music), gangsa players in these figuration styles approach a note in a descending pokok line with descending contours: (x)-2-1-3-2-1-3-2-1; they likewise anticipate a pokok tone in an ascending passage (in this case tone 3) with ascending contours: (x)-2-3-1-2-3-1-2-3. Both these are considered majalan, “kinetic” or “moving” contours. When the pokok tone repeats – what Balinese musicians call ngubeng or “static” – composers have many more contour options.²¹

There are two quite distinct versions of this three-note elaborating process. In ubit telu (“three”), when partners interlock, both polos and sangsih play the middle of the three tones. Polos also plays the upcoming pokok tone, either higher or lower depending on the current pokok contour, and sangsih plays the non-pokok tone. Two majalan contours – ascending and descending – and two ngubeng options for ubit telu are shown in Example 15.8. The top staff shows the basic contour; in the second, the two voices are divided, with polos notated stems down. The bottom two staves of the example show eight beats of ubit telu for a core melody combining motion with stasis.²² As in norot figuration, the melodic idiom itself generates pairs of interlocking rhythms that are identical (or similar) but metrically offset, together creating an uninterrupted stream of notes.

15.8 Interlocking ubit telu

Comparing these contours to McPhee’s from the second excerpt in Example 15.6 again draws into sharp relief the distinction between front- and end-weighted metric conception. In ubit telu, the 2+3+3 rhythm beginning just after the beat leads the melody smoothly to the following downbeat; in McPhee’s writing, a 3+3+2 rhythm starting on the downbeat simply repeats on the next downbeat, without the sense of arrival that end-weighted ubit telu engenders.

While polos and sangsih share the middle of three tones in ubit telu, in ubit empat (“four”), which is otherwise identically constructed, a higher fourth tone is added instead, always coinciding with the lowest and thus generating an irregular accent pattern. Because minimizing motion across the instrument is prioritized in ubit empat, polos and sangsih interchangeably align with the core melody for this technique. This necessitates a much more varied rhythmic palette for each individual player than does ubit telu. Four possible ubit empat patterns are notated in the top two staves of Example 15.9; the bottom two staves show ubit empat for the first eight beats of a melody from the dance piece Teruna Jaya.

15.9 Interlocking ubit empat

The snaking contours and shifting rhythmic accents of ubit empat are reflected in the top excerpt from Example 15.6, where McPhee alternates between a 2+3+3 and 3+3+2 accent scheme using virtually identical contours and pitch collections to the empat patterns in Example 15.9. Reich uses a similar technique in his 1973 piece Six Pianos, where various rotations of a low-middle-high collection of tones seem to mirror ubit empat construction, and alternating right and left hands loosely parallel polos and sangsih interlocking, per Example 15.10.²³

15.10 Interlocking in Six Pianos (top four voices)

The preceding examples show composers largely referencing the composite contours of Balinese interlocking melodies, with only Reich taking advantage of the syncopated rhythms of their individual strands. Because Balinese interlocking is designed to result in a seamless composite, it’s not surprising that contour, for these composers, was often more central to the aesthetic than the rhythmic makeup of individual parts. That said, McPhee did sometimes reference discrete polos and sangsih rhythms in his compositions. Example 15.11 shows an excerpt from Tabuh-Tabuhan where two flutes (doubled on clarinet and violin) play an interlocking passage of syncopated rhythms with every sixteenth-note subdivision filled. Other instruments not notated here play at slower densities of eighth, quarter, half, and whole notes, together creating a full stratified polyphony.

15.11 Interlocking in Tabuh-Tabuhan

Complicating the Picture: Balinese Linearity and Malleability

While each of these composers drew inspiration from Balinese stratified and interlocking polyphonies, their music is only a partial reflection of Balinese rhythmic aesthetics, because they were not focused on Balinese feeling, rasa. Each broke Bali’s rhythmic idioms into two- and four-beat cells of music to be repeated, manipulated, and repeated once more, bypassing the circuitous melodies and expansive structures that make up of so many of Bali’s greatest compositions. It is telling that, when played Reich’s 1971 composition Drumming, the famous Balinese musician Wayan Tembres “smiled and listened politely to part 1 but was unmoved and at last asked: ‘Is that all it does? Doesn’t it go anywhere?’”²⁴ For Reich, repeating a “static or slowly evolving cyclic pattern at a steady tempo” was “designed to focus listener attention on compositional process. This led him to draw from [Ghanaian] Ewe and Balinese music as though they were canvases of patterns unfolding on neatly ruled tempo grids stretching to infinity.”²⁵ On paper his approach, like Britten’s and McPhee’s, may seem Balinese; in reality, Balinese rhythm is far more malleable and unpredictable.

Music in Bali is very often cyclic, based on formulaic ostinati. Yet within that system, and guiding it, are deep-seated principles of forward motion, variety, and change. The complex, virtuosic patterns of gong kebyar introductions, where an ensemble-wide rhythmic unison of fast, unmetered passages embodies the genre’s characteristic “explosive” aesthetic, are entirely through-composed. Yet we don’t even need to go that far to find a Balinese practice more developed and varied than Britten, McPhee, and Reich present. A pokok will as often be a carefully shaped 16-, 64-, or 256-beat melody as a 2- or 4-beat one, and each section in a composition will be linked to the next with contrasting introductory, transitional, and concluding material. Formal organization, then, often centers on progression, not repetition. In ubit telu and norot, as we’ve seen, melodic contours move across the instrument as they follow the pokok, different for static and kinetic moments; in ubit empat, melodic motion in the pokok requires ever-shifting shades of 3+2+3 and 2+3+3, ascending and descending, with now polos and now sangsih on the beat. What’s more, each internally cyclic melody is varied through both subtle and extreme shifts in tempo, dynamics, and melodic elaboration to create a forward-directed listening experience. This “temporal and gestural flexibility … is obscured once captured within Western staff notation,”²⁶ reducing a dynamic temporal experience to strict repetition.

In performance, the Oleg Tumuliligan melody from Example 15.3 cycles twenty-seven times with virtually no repetition among them.²⁷ The composition alternates between a higher and lower pokok, and transitional high-to-low and low-to-high versions; it moves among different elaboration techniques on the gangsa in different cycles; and it features extreme tempo changes that see players shifting between breakneck interlocking polyphonies and slow, languorous figuration where polos and sangsih perform in parallel contours rather than interlocking parts. On top of that are starts and stops, short inserted ugal solos, and accented rhythmic breaks. Such devices are central to Balinese gamelan performance. A cyclic melody will frequently be interrupted by an angsel: an increase in volume leading to a brief syncopated rhythmic pattern and followed by a sudden stop, similar to the end of McPhee’s passage in Example 15.11. This approach to cyclicity shapes waves – ombak – of volume and intensity shifts in even the most insistently repetitive passage. More variable and dynamic still are the ocak-ocakan rhythms of the reyong gong-chime, an elaborating instrument that shifts between melodic and rhythmic roles, alternating interlocking polyphonies with tightly coordinated shots of cluster chords and rhythmic accents. Britten was particularly drawn to these syncopated tone clusters, using them, for instance, to represent the Prince-as-Salamander character in The Prince of the Pagodas.²⁸

While a Balinese composition may look repetitive on paper, then, experientially it is anything but. Each of these elements brings variability and goal-oriented motion to Balinese music, obvious even to the unseasoned listener. Many expert listeners, though, find their greatest joy in the yet subtler manipulations of collective Balinese improvisation. The works and words of mid-twentieth-century composers cast Bali as a culture of precision and repetition, fostering music that was virtuosic and complex but allowed performing Balinese minimal agency in its creation. Reich unequivocally claimed that “Balinese mallet playing is composed and allows no improvisation.”²⁹ Of course, as Tenzer argues, “improvisation was not Reich’s interest. … He was not predisposed to envision how improvisation could be integrated into a collectivity, or ponder the possibility that it is not only in ‘microvariations’ that ensemble musicians ‘give life to the music.’”³⁰ McPhee did acknowledge restricted improvisation on certain instruments, as in the flexible timing and ornamentation of the ugal and suling, and composers like Britten took advantage of that aesthetic as we saw in Example 15.4. Tadzio’s theme in Death in Venice, too, has a rhythmic freedom that mimics the Balinese solo instrument trompong.³¹ But McPhee also maintained that “other than in solo parts there can be no place for spontaneous improvisation [in Balinese music. …] Unison in the different parts must prevail or utter confusion results.”³²

These and other Western composers, musicians, and thinkers missed what, to many Balinese, makes their music come alive: the constantly shifting hues of four reyong players twisting in and around the norot contour, making flexible what on gangsa is utterly formulaic; the two drummers of an arja theatrical performance simultaneously improvising fast patterns of low bass and high ringing strokes, interlocking with the perfect imperfection that only decades of partnership can bring; the solo drummer stretching her idiom, inventing new patterns in response to a dancer’s movements.³³ All this, too, is essential to Balinese rhythm: unpredictable, flexible, creative, constantly evolving. True linearity within cyclicity.

Embracing the Contradictions

Every perspective, Tenzer notes of intercultural fusion and hybrid composition, “is a betrayal of some other perspective.”³⁴ One might wonder whether McPhee, Britten, and Reich, in imitating Balinese rhythm, actually overlooked its most crucial qualities. Improvisation, flexibility, end-weighted thinking, and linear aspects of structure, tempo, dynamics, and figuration are, after all, truly fundamental features. Yet, artistically speaking, these composers never hoped to express themselves in a “Balinese” way; faithful imitation never was the point. Their mishearings instead suggested new creative options, much as Bix Beiderbecke’s self-taught fingerings, while technically “wrong,” facilitated the jazz trumpeter’s innovative musical style.³⁵ Unlike composer and Bali scholar Michael Tenzer in his 2003 gamelan composition Puser Belah, or Evan Ziporyn a decade before in his fusion work Aneh Tapi Nyata,³⁶ earlier composers like McPhee weren’t looking to problematize tensions between contrasting systems. Rather, they worked like Balinese experimentalist composer Dewa Ketut Alit, who unapologetically draws inspiration from John Adams and György Ligeti, Michael Tenzer and various modern jazz performers, taking their diverse styles wherever his sensibilities lead him.³⁷ Alit’s recent forays into Western idioms, from his chamber works Ameriki (2018) and Simalakama (2019) to his Open My Door (2015) for orchestra, likewise borrow from Western techniques in a piecemeal way, turning the tables to reveal an analogous internationalizing process, but in an opposite direction.³⁸

Of course, colonial history and power dynamics necessarily complicate this picture, and we should be wary of too blithely comparing the two perspectives. Western musicians, producers, and listeners have a long history of “put[ting our] Others in (small) boxes”³⁹ – presuming to delineate “authenticity” for our non-Western counterparts, and to expect it from them, while simultaneously seeing their musics as “a kind of natural resource that is available for the taking.”⁴⁰ This ideological imbalance reinforces stereotypes shaped by Western discourses, often denying musicians from the Global South the agency to define themselves and their musics – and, by extension, the freedom to truly experiment – but giving Western musicians compositional carte blanche.⁴¹ It may now be incumbent upon non-Balinese composers drawing from Balinese techniques to work toward decolonizing that exchange.⁴² There is still plenty of space within such deliberate interactions for radical acts of imagination. Of the inherent tension between tradition and experimentalism, Alit opines: “I think … I need both. You need both. … It’s both good because, for me it gives us inspiration. So traditional thing is good because it’s strong root. But new is good because it gives you the future.”⁴³ As Lou Harrison once said of the compositional process, the best results come from “just playing in the sandbox.”⁴⁴ It was in this creative spirit that Bali’s cyclic melodies, specially stratified heterophonies, and rich interlocking polyphonies motivated new approaches to formal structure, vertical relationships, and rhythm for McPhee and his musical descendants. Each borrowed just the things he needed to “create a music based on his fantasies.”⁴⁵

Hybrid compositions, Pete Steele argues, “confound simplistic binary relationships, and forge an indeterminate ‘third space’”⁴⁶ where creative agency balances artistic borrowing. At its worst, such a space embraces only the tropes, reinforcing stereotypes, perpetuating musical and cultural misrepresentations, and ultimately, as Lipsitz warns, “creat[ing] new sources of misunderstanding, misreading, and misappropriation.”⁴⁷ At its best, though, Steele contends this third space could “embody the gradual destabilization of European and North American hegemony, and serve as [an] exemplary symbo[l] of a more integrated post-colonial world.”⁴⁸ It may be that McPhee, Reich, and Britten’s adoption and creative reinterpretations of Balinese rhythm amount to appropriation, at least by twenty-first-century standards. Unlike Tenzer and Ziporyn, these earlier composers never truly sought to steep themselves in Balinese culture, nor to consciously engage with the experiences of Balinese musicians, yet nevertheless felt entitled to borrow from them without reservation. The act of borrowing, though, came from a place of honest appreciation, curiosity, and respect; their work initiated an exchange, setting the stage for future generations to go deeper. And “without minimizing the very real dangers of cross-cultural appropriations and misunderstandings,” as Lipsitz contends, “we must nonetheless be open to the kinds of knowing hidden within some ‘incorrect’ perceptions.”⁴⁹ Walking a tightrope between Western and Balinese perspectives, Colin McPhee, Benjamin Britten, and Steve Reich did what innovators have always done: they took elements from disparate ideas and modes of perception and, through their fusion, arrived somewhere new. For them, as for Alit, Tenzer, Ziporyn, and many others, the fusion of Balinese and Western musical elements was and is, unabashedly and experimentally, about embracing the complexity of the encounter and leaping into the sandbox with both feet.

Neo-African Genres

Most of the best-known Latin American and Caribbean music genres, such as salsa, reggae, and reggaeton, are products of a syncretic process in which musical elements derived from Africa and Europe were creatively combined and reworked over generations, or even centuries, to produce distinctively new entities. However, some of the region’s most powerful and rhythmically rich musics are much closer to the African idioms brought by the several millions of slaves transported to the Americas from the early 1500s to the 1870s. Such genres that survive today are best seen as neo-African rather than African per se, in the sense that they have changed and evolved since taking root in the Americas, but along overwhelmingly African-derived aesthetic lines, without any overt European influence or inspiration.

Such musical traditions are to be found primarily in association with African-derived religions, especially since religious musics often tend to be conservative and practitioners are likely to assert the importance of maintaining ancestral continuity. For various reasons, such music genres and their associated religions are strongest in Cuba and Brazil. Spanish and Brazilian colonists, unlike their British counterparts in North America, allowed significant numbers of slaves to purchase or otherwise gain their freedom (in a process called “manumission”), such that they and their descendants could form urban societies (in Cuba: cabildos) where they were able to perpetuate, in however modified forms, their African-derived religions and associated songs and dances. Even more important to the survival of these traditions was the fact that the importation of enslaved Africans to Cuba and Brazil continued and even intensified in the 1800s, lasting until the 1870s, unlike in the United States and British colonies, where most slaves were brought in the 1700s. Hence, many black Cubans and Brazilians know the ethnic ancestry of their forebears, and several Afro-Cuban religious songs are still sung across the Atlantic in Nigeria and elsewhere in West Africa.

Perhaps the richest and most vigorous traditions of neo-African drumming are those associated with the Afro-Cuban religion known as Santería, or more properly, Regla de ocha. Santería is a sort of streamlined consolidation of Yoruba-derived practices and beliefs, centering on lively ceremonies in which music – especially call-and-response songs accompanied by a trio of batá drums – plays a central role, often inducing spirit possession by devotees. For somewhat more festive ceremonies called bembé, a set of two or three conga-shaped drums (cachimbo, mula, and caja) are used, together with a cowbell.

Many batá and bembé rhythms are based on a feature common in much Western and Central African music, namely, polyrhythm, meaning a composite rhythm with two (or possibly more) basic pulses occurring at once. This quintessentially takes the form of a twelve-beat ostinato, with a cowbell providing a syncopated seven-stroke “time-line,” while other percussion parts divide the cycle into groups of twos and threes. In several cases, it may be the vocal line and/or the dance moves – typically involving stepping in twos or threes – which set up or reinforce polyrhythms in the vocal and drummed parts. Example 16.1 shows a common bembé rhythm, consisting of an ostinato in which the bell plays the standard West African time-line, the high-pitched cachimbo plays a rhythm essentially in threes (as in $\begin{array}{c}6\\ 8\end{array}$ or $\begin{array}{c}12\\ 8\end{array}$), while the mid-pitched mula drum pattern is in twos (suggesting $\begin{array}{c}6\\ 4\end{array}$).¹ Over this basic polyrhythm, a lead drummer, playing a conga-like drum with a stick in one hand and the bare palm of his other, improvises freely, often alternating between patterns that suggest either the $\begin{array}{c}6\\ 4\end{array}$ or $\begin{array}{c}6\\ 8\end{array}$ feel.

16.1 Polyrhythmic bembé ostinato

Aside from Afro-Cuban religious music and dance, such polyrhythms are common in Haitian Vodou drumming, including such basic dance rhythms as yanvalou. They are also found in the music of candomblé, the Yoruba-derived religion that thrives today in Brazil. In general, the $\begin{array}{c}12\\ 8\end{array}$ polyrhythms are most prominent in genres that remain strongly African in character, are associated with African-derived religions, and are typically performed by ensembles of percussion instruments with singing, usually without any melodic instruments such as guitar.

A broader category of percussion-dominated neo-African genres, and especially those that may be associated with secular entertainment dance rather than religion, use predominantly duple meters, which could be notated in $\begin{array}{c}4\\ 4\end{array}$ or cut time. In the Caribbean, as well as elsewhere, some of these rhythms, as shown in Example 16.2, are based on particular cells, such as those which, adapting Cuban terminology, could be referred to as the tresillo (pronounced “tray-see-yo,” three-three-two), the “habanera” rhythm (whose related incarnations will be discussed further below), the cinquillo (“seen-key-o,” a sort of decorated version of the tresillo), and the “amphibrach,” which could be seen as a displaced version of the cinquillo, and/or a slightly elaborated version of the habanera cell.

16.2 Rhythmic cells: tresillo, habanera, cinquillo, amphibrach, clave (3+2 and 2+3)

Thus, for example, in the Afro-Puerto Rican bomba genre, in which an individual dances in front of a trio of drummers, while others sing call-and-response songs, the belén rhythm is based on the habanera pattern, the sicá and bambulé are based on the tresillo, and the gracimá is based on the amphibrach. These patterns also abound in various other predominantly duple-metered Afro-Caribbean genres, such as Jamaican kumina drumming, Trinidadian Orisha (or Shango) music (derived, like Santería, from Yoruba roots), Dominican palo drumming (which may be either secular or religious in context), and the entertainment-oriented belé and gwo ka of Martinique and Guadeloupe, respectively.

Another particularly important pattern in “Latin” music – which connotes Afro-Cuban music and its various forms, derivatives, and relatives – is clave (clah-vey). In Spanish, clave means “key” in a metaphorical sense, as in the key that unlocks a rhythm, but as a musical term it derives more directly from the hard-wooden pegs called clavija that could be used in building ships, or to beat the pattern when dockworkers were having informal rumba sessions. Clave, as shown in Example 16.2, is usually understood as implying a two-bar pattern of $\begin{array}{c}4\\ 4\end{array}$ in either “three-two” or “two-three” form, with the “three” side being relatively syncopated and the “two” side not. (The rhythmic jingle “SHAVE and A hairCUT – TWO BITS” coheres nicely with three-two clave.)

In some Afro-Cuban music (as in that associated with the abakuá brotherhoods), the $\begin{array}{c}12\\ 8\end{array}$ standard time-line shown in Example 16.1 is often “abbreviated” by omission of the third and final bell strokes, affording a five-stroke “two-three” configuration which itself is referred to as clave, and can be seen as a generative source for the familiar $\begin{array}{c}4\\ 4\end{array}$ clave of rumba and popular music.

The $\begin{array}{c}4\\ 4\end{array}$ clave pattern seems to have arisen in connection with Afro-Cuban rumba (and was notated as such in an 1850 composition by Louis Moreau Gottschalk). While the term rumba has been loosely used to denote various creole Cuban and Latin genres, properly speaking it refers to an Afro-Cuban dance-music genre in which three conga drums (or cajón boxes), a pair of hardwood clave sticks, and two other lighter sticks (palitos, tapped on some surface, such as the side of a conga) accompany singing and dancing, either by a couple, or else, in the case of colombia, by a solo male. The most common rumba subgenre is the guaguancó (wah-wahn-CO), whose basic ostinato, with a typical vocal refrain, is schematized in Example 16.3. (Note that in rumba, the final stroke of the “three” side falls an eighth note later than in the more familiar son clave.)

16.3 Rumba montuno “Consuélate”

Over this basic pattern, a solo conga (or quinto) drummer provides lively improvisations, while the singing takes a two-part form, consisting of an initial section of verses sung by a solo vocalist, followed by a montuno section of call-and-response singing, during which dancing occurs. The pre-composed choral response (coro) coheres with the clave pattern, such that a singer or clave player who renders it reversed will generate a jumbled-sounding “crossed” (cruzado) effect and inspire glares, scowls, and eye-rolling among the other musicians and knowledgeable listeners. Example 16.3 shows the coro of a familiar rumba (“Consuélate”), with clave (in this case, three-two, with the phrase emphasis beginning on the second measure shown), a very schematic rendering of the two-conga ostinato, and the pattern of the palitos. Note how the main accents in the coro coincide with the clave strokes.

While we will return to clave in discussing son and salsa, at this point it is relevant to point out another common feature of duple-metered neo-African and related creole rhythms. Genres like the guaguancó and the aforementioned bomba styles, being essentially in $\begin{array}{c}4\\ 4\end{array}$, do not feature classic $\begin{array}{c}12\\ 8\end{array}$ polyrhythms built into their basic structure. However, the improvisations played by lead drummers (as in rumba) often include syncopated passages that can be seen as temporary introductions of polyrhythms. The most common techniques of achieving this effect involve what could be seen as ternary phrasing of binary beat subdivisions or, conversely, binary phrasing of triplet subdivisions. Example 16.4 shows some examples of these two techniques in the form of high- and lower-pitched drum strokes, such as could be rendered on two congas, or a bongo. In measures two and three, a four-stroke phrase (suggesting duple time) is rendered in triplet quarter notes (i.e., ternary subdivision), while in measures four and five, a six-stroke phrase (suggesting ternary phrasing) is played in duple eighth notes (i.e., duple subdivision). Any transcription of a conga, bongo, or even piano solo in a salsa performance is going to reveal several instances of this device.

16.4 Rhythmic patterns suggesting polyrhythms

Hemiola-Based Genres in $\begin{array}{c}6\\ 8\end{array}$–$\begin{array}{c}3\\ 4\end{array}$

If $\begin{array}{c}12\\ 8\end{array}$ polyrhythms are characteristic of some of the most distinctively neo-African, percussion-dominated musics, related sorts of rhythmic patterns also undergird a vast category of creole or “mestizo” Latin American genres, in which hemiola patterns are typically rendered on stringed instruments rather than drums. As many musicians know, hemiola implies either the simultaneous or sequential combination of $\begin{array}{c}3\\ 4\end{array}$ and $\begin{array}{c}6\\ 8\end{array}$ rhythms. The sequential form quintessentially corresponds to the familiar “I like to live in Ame-ri-ca” pattern (from Leonard Bernstein’s West Side Story), which suggests a bar of $\begin{array}{c}6\\ 8\end{array}$ followed by one in $\begin{array}{c}3\\ 4\end{array}$ (with a constant eighth-note pulse). Vertical hemiola, by contrast, might feature one instrument playing a $\begin{array}{c}6\\ 8\end{array}$ pattern while another simultaneously plays a $\begin{array}{c}3\\ 4\end{array}$ pattern.

Scholars and pseudo-scholars have disagreed about the origins of this sort of rhythm in Latin American music, with some musicologists arguing for a Spanish or even Arab derivation. What seems abundantly clear, however, is that these rhythms came into vogue and spread most extensively not in Spain, but in the Caribbean Basin in the sixteenth and seventeenth centuries, as a syncretic product of the interaction of Spanish colonists and sub-Saharan African slaves and their descendants. The hemiola rhythms were of particularly early and vigorous appearance in Mexico (New Spain), where free or enslaved persons of African origin outnumbered Spaniards in the latter 1500s. For their part, many, if not most, of the Spaniards in the region consisted of Andalusians whose own music culture had been shaped by centuries of Moorish rule, which may well have contributed to a fondness for $\begin{array}{c}6\\ 8\end{array}$ meter, which continues to pervade current-day Moroccan music. It is easy to imagine Afro-Mexicans getting their hands on vihuelas or guitars and coming up with chordal and rhythmic ostinatos in which simple progressions (e.g., D minor–A) would be repeated with quasi-polyrhythmic strumming patterns. Such genres, under names like guinea, zarabanda, and cumbé, were taken up by local Spanish composers and made their way (complete with colloquial Afro-Latin mispronunciations of Spanish) back to Spain, where the “I like to live in Ame-ri-ca” hemiola became a stereotypical icon of Latin American music.

From what may have been the cradle of such rhythms – the area of Veracruz and Mexico City – these sonorities traveled by trade routes to rural Venezuela and beyond, extending to regions, such as highland Colombia, where there were few black people. Despite their evident Afro-Latin origins, these rhythms came to be played by string-based ensembles (especially using variants of guitars and vihuelas) and associated, ironically, with Spanish rather than African heritage. As these infectious rhythms disseminated, they came to animate a vast and diverse set of folk genres stretching from Mexico and Cuba down to the southern cone of Chile and Argentina.

Rhythms in hemiola-based genres (which some Latin American musicologists refer to as the cancionero ternario, i.e., the “ternary repertoire”) can take different forms. As mentioned, often the hemiola is a sequential alternation of $\begin{array}{c}6\\ 8\end{array}$ and $\begin{array}{c}3\\ 4\end{array}$ measures (as in “I like to…” etc.), which is generally known in Spanish as sesquialtera – from Latin, “six that alters.” Thus, for example, in the typical style of Cuban punto, as performed by predominantly white Cuban farmers in the central and western part of the island, ten-line décima verses sung in free rhythm alternate with instrumental interludes in which guitars and bandurrias repeat patterns roughly as shown in Example 16.5.

16.5 Cuban punto ostinato

In vertical hemiola, the $\begin{array}{c}3\\ 4\end{array}$ and $\begin{array}{c}6\\ 8\end{array}$ patterns occur simultaneously, constituting a polyrhythm. Such rhythms are structural ostinatos in genres such as the Colombian bambuco and are also common in the música llanera (plains-region music) of Venezuela and Colombia. Example 16.6 shows a typical ostinato that could occur in a Venezuelan joropo, in which the bass pattern is in $\begin{array}{c}3\\ 4\end{array}$ and the harp and ukulele-like cuatro play in $\begin{array}{c}6\\ 8\end{array}$.

16.6 Venezuelan joropo ostinato

In such música llanera, the three instruments typically reiterate the chordal ostinato while jumping back and forth – whether collectively or individually – between patterns that suggest either $\begin{array}{c}6\\ 8\end{array}$ or $\begin{array}{c}3\\ 4\end{array}$. The result is as intricate, exciting, and richly polyrhythmic as any West African drum ensemble, even though the original fusion with Afro-Latin musicians may have taken place in different regions and as long ago as the sixteenth century.

Genres in the cancionero ternario category are too vast to enumerate here, but aside from those mentioned would include the Mexican jarabe, son jalisciense (together forming the core mariachi repertoire), and huapango (in the broader son huasteco and son jarocho categories), the Ecuadoran pasillo, the Peruvian marinera, the Chilean cueca, and the Argentine chacarera. (Oddly enough, the “I like to live in Ame-ri-ca” pattern is conspicuously rare in Puerto Rican music, which Leonard Bernstein meant to evoke in his catchy tune.) Most of these genres today have a folkloric status, with the realm of urban commercial popular music being dominated by duple-metered genres, to which we may now turn.

Urban Binary Genres

For the last century, the vast majority of commercial popular music styles that dominate airwaves and dance clubs throughout Latin America and the Caribbean have been duple-metered, with the cancionero ternario genres receding into the realm of traditional folk music, however still vital and valued as national patrimony. This change has occurred on such a broad, hemispheric scale that it begs some grand explanation,² although the result has certainly brought the region’s popular musics in line with most of the rest of the world.

The Habanera and Cinquillo Complexes

The popularization of syncopated duple-metered genres can be documented in Cuba from around 1800, when the contradanza – a creolized version of the European contredanse (contra dance/country dance) – became the most popular genre in the innumerable dance halls of Havana and other towns. One factor distinguishing the Cuban contradanza from its European ancestors was the use of the habanera bass ostinato (which Cubans musicologists have called the ritmo de tango, or tango beat) shown in Example 16.2, which, depending on the nature of the ensemble, could be played on a string bass, on a tuba, by a pianist’s left hand, or on the lower strings of a guitar, or possibly on a drum like the timpani. Contradanzas were typically instrumental, but in Cuba in the 1840s, a languid, sung version of the contradanza called the “habanera” (not “habañera”) emerged and became internationally popular, such that outside Cuba its characteristic rhythm came to be known as the “habanera rhythm” – a designation we employ here.

Although simple, the habanera rhythm is in its way quite flexible and protean. If the sixteenth note is stressed, the rhythm shades into the three-three-two tresillo, while the addition of an eighth note after the first stroke affords the amphibrach pattern (also shown in Example 16.2). Such mutability may partly explain why the habanera beat and its variants have been adopted in diverse forms as the rhythmic cells of a wide variety of genres, from Afro-Cuban iyesá drumming to modern reggaeton. The amphibrach, for example, had become a familiar Caribbean creole rhythm as early as the 1750s, and in the latter 1800s went on to undergird the Brazilian maxixe and lundú. For its part, the habanera beat subsequently went on to become the underlying ostinato in the early Argentine tango. By the 1920s, the tango had acquired a quite different feel, but the rhythm persisted in the direct descendant of the habanera, namely, the Cuban bolero – like its predecessor, a slow, danceable, sentimental song, often featuring a guitar-based trio format, from the 1940s. The habanera pattern also constituted the standard bass rhythm of the 1950s chachachá. In the 1980s, Dominicans effected their own original adaptation of the bolero, and its habanera-pattern bass, in the form of bachata; all these basic genre patterns are schematized in Example 16.7. Most remarkable is the resurfacing of the pattern, with a strong three-three-two syncopation, in both up-tempo Trinidadian soca, in the 1970s, and medium-tempo reggaeton, which took its rhythm from Shabba Ranks’s 1991 dance-hall song “Dem Bow.” While it might not be accurate to say that these genres derive their rhythm from the nineteenth-century contradanza, the habanera beat is certainly a common creole Caribbean rhythm that has kept resurfacing over the last two centuries in diverse forms and genres. Example 16.7 also shows the typical rhythm of the cumbia, which, originating in Colombia, has become one of the most popular urban dance genres everywhere from Mexico to Argentina.

16.7 Basic patterns of bolero, chachachá, bachata; cumbia

As mentioned above, another basic creole Caribbean rhythm has been the five-stroke pattern that Cubans call the cinquillo. This pattern may have its origins in neo-African and especially Afro-Haitian drumming, as it features prominently in Haitian Vodou rhythms (such as banda), as well as Santería batá drumming, Martinican belé, and the siká and cuembé styles of Afro-Puerto Rican bomba. From the 1870s the cinquillo – in a two-bar form in which it is followed by a measure of even quarter notes – became the basic ostinato pattern of the Cuban danzón and the related Puerto Rican danza. The cinquillo also abounds in Trinidadian calypsos, Martinican zouk, Haitian konpa and popular songs such as “Chouconne” (better known as “Yellow Bird”), and other creole Caribbean genres animated by the colonial-era [Afro-]“French connection.” As with the habanera beat, the renderings of the cinquillo can vary dramatically in style, ranging from the thunderous pounding of Vodou drumming to the languid rubato of a Chopinesque Puerto Rican piano danza by Manuel Tavarez (1843–83). Example 16.8 shows the main theme of the 1893 danza “Mis amores,” by Puerto Rican composer Simón Madera (1875–1957), in which the left hand plays a characteristically “elastic” version of the two-bar cinquillo pattern.

16.8 Excerpt of danza, “Mis amores”

“Latin” Rhythm: Son, Mambo, and Salsa

Among the most dynamic, distinctive, and rich forms of rhythm in the Americas is that of “Latin music,” which, as mentioned, loosely denotes Afro-Cuban-derived popular dance music rather than “Latin American” in the broader sense. What is understood as Latin rhythm can be seen to have taken its modern shape in the 1940s and 1950s, especially in the contemporary Cuban son and big-band mambo, and perpetuated in salsa, the direct stylistic descendant of those genres. The essence of Latin rhythm is a composite ostinato structure created by a set of interlocking conventional patterns played on an ensemble of percussion and melodic instruments. These are typified in Example 16.9, which approximates some of the parts in the basic repeating pattern from the montuno section of the 1975 salsa song “María Luisa,” sung by Ismael Miranda. (Like traditional rumba and the creole Cuban son, salsa songs follow a two-part verses-montuno form.)

16.9 Salsa montuno, “Maria Luisa”

Several features are noteworthy here. One is the “anticipated bass” pattern, in which the bass, rather than emphasizing the downbeat of each measure, glides over it, sounding the root note of each measure’s chord on the quarter-note beat before it. Another feature is that while this montuno pattern consists, in this case, of a four-bar chordal and vocal refrain, the basic rhythmic ostinato is a two-bar entity cohering with clave – in this case, two-three clave, in which the “two” side is somewhat less syncopated than the “three” side. Even though the clave pattern itself is not being sounded by the hardwood sticks (as it would be in traditional rumba), the pattern still operates as an underlying, implicit rhythm that is ever-present in the performers’ heads. Thus, the capitalized syllables in the coro coincide with clave beats; conversely, if one were to clap the “incorrect” three-two clave in this song, the vocal and clave beats would miss each other and the sound would be jumbled and cruzado. The piano pattern (itself called guajeo or montuno) follows a two-bar rhythmic pattern that hits the downbeat of each odd-numbered measure (the less syncopated “two” side of the clave), and then creates tension by jumping off the rails, as it were, onto offbeats, and skipping the next downbeat (of the even-numbered measures, on the syncopated “three” side), before returning to the downbeat of the third measure. Meanwhile, the lead singer, or an instrumentalist, who is improvising “call” phrases (sonejos) in between the coro refrains need not slavishly stress the correct clave pattern, but must certainly be careful not to perform phrases that would suggest the wrong (in this case, three-two) clave. To do so would be regarded by many salsa musicians as a glaring error that muddies and jumbles the otherwise tight and coordinated composite rhythm. In fact, proper rendering of clave became a rather cliquish fetish among some salsa musicians, to the point that in recent decades a few prominent salseros have blasphemously defied the “clave police” in declaring that they no longer feel bound by such fussy rules.

Another dynamic Latin rhythm is that of the Dominican merengue, whose kinetic drive derives less from salsa-style syncopation than from its insistent, frenetic, four-on-the-floor pounding, articulated by the crisp, tight (apreta’o) timbres of the güira scraper and the tambora drum. In traditional merengue, the essence of the tambora pattern is its sixteenth-note roll leading up to the downbeat of each measure – a feature that may derive from Spanish military band music more than any African precedent.

Miscellaneous Rhythms

The categories outlined thus far could collectively be said to cover most of the characteristic rhythms for which Latin American and Caribbean music are known. They do not, however, completely exhaust the vast region’s soundscape, such that mention should be made of a few noteworthy genres not accommodated into this taxonomy. The reader, for example, may be wondering by now, what about reggae? The “boom-a-CHUCK-a-boom-a-CHUCK-a” so-called “skank” rhythm of roots reggae is indeed one of the hemisphere’s most characteristic and infectious rhythms, and has little structural relationship to the categories presented so far here. The evolution of this rhythm is itself enigmatic, as it came into vogue quite rapidly around 1968, perhaps deriving to some extent from the guitar or banjo strumming pattern of folk mento, and perhaps reinforced by Afro-Jamaican burru drumming, and quickly replacing the rhythm ’n’ blues shuffle beat of mid-’60s ska.

Other popular genres are relatively independent of the African influences that have directly or indirectly enriched most Latin and Caribbean music styles. An especially substantial and musically fertile region is northern Mexico, together with the Mexican American communities in neighboring border states (and more far-flung but still substantial communities in such sites as Chicago). Here the main musical influences have been those of Germans who settled in the U.S. Southwest and elsewhere in the nineteenth century, bringing with them the keyboard accordion and the favored dances of the polka and waltz. These were adopted as the predominant rhythms of the distinctive Mexican and Mexican-American musics of the region. Most characteristically, the regional polka and waltz have been played on a conjunto ensemble traditionally comprising accordion, guitar, and the guitar-like bajo sexto, nowadays commonly supplemented by electric bass and drum set. This is the standard ensemble of norteño music (música norteña) and its Texas-Mexican counterpart, which might be generically called conjunto, música tejana, or “Tex-Mex.” Equally popular in recent decades has been the brass band format called banda, or tecnobanda if amplified vocals and perhaps keyboard and electric bass are added. Much of the repertoire of these groups, set to polka or waltz rhythm, might be categorized (e.g., on CD labels) as ranchera, which coheres with the dandified cowboy spirit reflected in preferred attire. If the songs are narrative and text driven, they may be referred to as corridos or, in the case of songs glorifying drug dons, narcocorridos. The term ranchera also comprises slow ballads (such as Vicente Fernández’s evergreen “Volver, volver”), typically employing mariachi instrumentation (with its characteristic violins and trumpets), set to what is essentially a greatly decelerated polka rhythm.

Another musical region little touched by Afro-Latin influences is the Andes, home especially to the Quechua- and Aymara-speaking populations of highland Peru, Bolivia, and, to some extent, Ecuador. The single most popular genre of this region is the huayno, a medium-tempo song for solo voice and, typically, stringed instruments such as charango, harp, and guitar. A typical huayno rhythm may outline a familiar quarter-eighth-eighth ostinato, in which case it can easily be accommodated into a cumbia beat, as in urban chicha music. Often, however, the first beat of this pattern is slightly shortened, affording a limping “long-short-short” ostinato that is not easily notated. Another distinctive feature of many huaynos (such as those by northern Peruvian songstress Dina Paucar) is their irregular and often uneven-numbered phrase lengths, which contrasts with the four-bar phrasings otherwise pervading Euro-American and Latin music.

Finally, another rhythmically distinctive set of genres is that associated with a different set of Indians – not Native Americans, but descendants of the over 400,000 people who came to the Caribbean from India as indentured workers in the period 1845–1917. These Indo-Caribbeans now constitute roughly half the populations of Trinidad, Guyana, and Suriname, and have cultivated their own dynamic music culture. This variously comprises traditional folk genres brought from India, creolized pop “chutney-soca,” and a few idioms that, though originally deriving from fragmented genres transplanted by the original immigrants, have over the generations developed – along wholly Indian rather than African lines – into thoroughly distinctive and unique entities. Particularly dynamic is the Indo-Trinidadian genre of tassa drumming, as performed at Hindu weddings, drum competitions, and other festivities by ensembles each consisting of two stick-played, shallow kettledrums (called “tassa”), a large bass drum, and a pair of cymbals (jhanjh, jhal).

Tassa drumming has evolved into a highly sophisticated, complex, and dynamic art form, provoking animated dancing and sustaining the interest of connoisseurs who prize virtuosity and creativity as well as adherence to established norms. The latter consist primarily of the repertoire of a dozen or so common “hands,” each of which comprises a composite ensemble rhythm (or sequence thereof), and a set of conventional cadences and riff types, which, though standardized, allow room for flashy improvisation (“cutting”) by the lead drummer. Tassa drumming is at once disciplined and thunderously loud, exciting, and as “hot” as any African or Afro-Latin music.

Unfolding Sound and Movement over Time

Concepts of rhythm are deeply interwoven with ideas of time, which are not universal between Indigenous communities. Each community has distinct ways of experiencing and marking time and the completion of life-sustaining cycles; for instance, Native American peoples in the Pacific Northwest through California complete community-specific ceremonies that mark the return of the salmon. Similarly, corn-cultivating Indigenous pueblos in Mesoamerica through South America complete community-specific ceremonies at different stages of the agricultural cycle. In many Indigenous communities, ceremonies with music and dance play a vital role in sustaining the delicate balance of these cycles, and humans and their music and dances play an important part in maintaining the natural environment. For this reason, many Indigenous communities only perform songs and dances during specific seasons and ceremonies, following the larger cyclical rhythms that have underpinned the lives of many Indigenous communities in the Americas for millennia.

Music and dance play a central role in the structuring of ritual time in many Native American ceremonies, and repeated rhythms can suspend time and create an experience outside the framework of normal space-time.² Layers of intricate repetition and variation in song and dance are integral in achieving this effect. Through song and dance, communities can express complex understandings of the cosmos and humanity’s role within it. Glossing these ideas as “religion,” however, would mischaracterize the full significance of many Indigenous music and dance traditions; instead, they are widely understood as ways of being in which the boundaries between sacred, social, and life-sustaining activities are kaleidoscopic and permeable.

Language and Rhythm

Most Indigenous American languages do not include a word that translates directly as “music”; for instance, in Nahuatl, one of the many Indigenous languages found in central Mexico, the word cuica means to sing whereas the word pitza means to play the flute or trumpet: There is no single term in Nahuatl that encompasses both of these ideas or translates to the European concept of “music.” Studies of the words used to describe the sounds of dance, instruments, and songs suggests that there is fluidity between language, music, and dance; for instance, in some languages, such as in several Native Alaskan languages, music and dance are synonymous.³

Similarly, many Indigenous languages do not appear to have an equivalent to the English word rhythm, and instead use other terms to express related ideas. When asked, Indigenous Maya scholar and musician Juan Francisco Cristobal shared with me that in Q’anjob’al Maya the word tx’olilal b’it is used to indicate the way sounds are ordered. In Kichwa, the Indigenous language spoken in Otavalo, Ecuador, Jessie Vallejo and Indigenous Kichwa scholar and musician Patricio Maldonado related to me that the word pacha, which translates to “time” in English, is the primary word used for rhythm. In addition to pacha, Vallejo and Maldonado shared that Kichwa uses emotive words to describe rhythms, including kushi for rhythms that are quicker and perceived as happier, and llaki for rhythms that are slow and perceived as sad.

Despite some broad similarities in how music, dance, and rhythm are conceptualized across Indigenous cultures in the Americas, each Indigenous nation and community developed its own distinct music and dance traditions. As of the 2000 census, in Mexico alone there were more than sixty distinct living Indigenous languages. When Europeans first arrived in North America, the geographic area comprising Canada and the United States boasted more than three hundred Indigenous languages, and to this day, South America alone is home to more than a quarter of the total language families in the world.⁴

Instruments

Indigenous American music features a wide range of melodic and percussive instruments: from the fiddle of the Métis people of Canada to the marimbas of the Maya and the many varieties of flute in South America, melody and rhythm go hand in hand. Although percussive instruments (particularly drums and rattles) are perhaps among the most widespread and iconic instruments, melodic instruments, including flutes and the human voice, are of equal importance in Indigenous American music. Furthermore, colonizers introduced new instruments that have become fully Indigenized and integrated into Indigenous traditions; for instance, Indigenous communities in South America and Mexico play stringed instruments that are modeled on European instruments but that are distinctly Indigenous. Examples include the Andean charango and the Mexican concha that use armadillo shells as the bodies for small lute-like instruments. These instruments provide rhythmic accompaniment to songs while simultaneously supplying harmony to support the melodic line.

Music and Dance Form

Form is integral to the structure of Indigenous songs, dances, and rituals. Using musically structured time and space, music and dance can depict expansive ideas about the cosmos and humanity’s place within it. In many cases, Indigenous musicians and ritual specialists carefully structure rituals, and songs and dances must be performed in a specific order. In these instances, form includes not only the rhythmic and melodic patterns of the songs, but also the broader ways in which songs are conceptualized in ceremonial settings – or even over a longer period, such as a full agricultural cycle.

Indigenous songs vary greatly in their structure, ranging from through-composed, meaning that there are no repetitions, to a strophic structure where the verses repeat the same rhythm and melody. Even if there are repetitions in the music, the dance can vary substantially, as is the case of the traditions of the Yurok, Karok, and Hupa people of northwestern California.¹⁴ In a number of cases, the broader repetitions of rhythms and choreography are determined by culturally significant numbers or by the context of the performance.

Conchero – a syncretic Indigenous dance tradition from Central Mexico – is one example of how culturally significant numbers can inform the rhythmic and melodic structure of music and dance. Although conchero is Catholic, it retains elements of pre-Hispanic music aesthetics and blends Catholic and Indigenous cosmologies; for example, it emphasizes the four directions that are each associated with sacred sites while also corresponding with the four directions of the cross. In conchero, these four cardinal directions are a critical part of both conchero choreography and rhythm. The number four is symbolically represented through the duality of the music and steps: Every step that is completed in one direction must be completed in the other. As a result, the structure of traditional conchero rhythms and dances typically follows an AABB format. This form reflects these two overlapping cosmovisions, and music and dance become vehicles for experiencing these broader connections. Through their movements, the dancers travel in the four directions and maintain the balance as they present their gestures and movements as an offering.

Similarly, in the Indigenous Andes, yanantin – a vision of the cosmos that views the universe as paired parts that, without both, would be incomplete – shapes musical form, rhythmic repetitions, and dance gestures.¹⁵ Yanantin is significant in rituals, including the celebration of the summer solstice that is celebrated across the former territories of the Inca Empire that reached across contemporary Peru, Bolivia, and Ecuador. In the Andean cosmovision, gender figures centrally in these pairings, and seasons, instruments, and geographic features are often ascribed symbolic genders; for instance, the rainy season, which fuels the growth of crops, is considered feminine, while the dry season is considered masculine.¹⁶

In Otavalo, Ecuador, the music and dance of the summer solstice festival, known as Hatun Puncha-Inti Raymi, serve a central role in allowing humans to communicate with the earth, spiritual beings, and the cosmos.¹⁷ The choreography of the dances in the festival varies between the dancers moving in straight lines or spiraling around musicians while they play short pieces on guitars, conch shells, harmonicas, melodicas, and transverse cane flutes that are made in pairs of male and female instruments. In the spiraling dances, the instruments play repeating phrases to accompany the chanting that uses a call-and-response structure: Each call-and-response pattern occurs twice, while the main melody is repeated three times by the instrumentalists. The stomps of the dancers connect their gestures to the earth and create a steady beat under the instruments and call-and-response vocals.¹⁸ The form of the music and dance movements for the Inti Raymi festival in Otavalo highlight how musical elements of time and collective movement can forge links between humans and their surroundings.

Meter and Beat Patterns

The rhythms of Indigenous music and dances are as diverse as the hundreds of languages found across North and South America, and communities have their own local or regional rhythmic styles that distinguish them from their neighbors. To try to understand relationships between the songs and dances of different communities, music researchers, particularly in the early twentieth to mid-twentieth century, sought to measure and quantify differences in Native North American songs. In an attempt to quantify these differences, researchers compared the number of rhythmic units commonly used in songs to create cross-cultural comparisons. For example, researchers have suggested that the songs of Mescalero and San Carlos Apache typically feature two durational values: one that is half as long as the other.¹⁹ In contrast, Bruno Nettl, who undertook comparative studies in the 1950s, concluded from his analyses that rhythms among the Paiute typically draw on three or four durational values, while Hopi and Zuni songs typically use approximately five or six different durational values.²⁰ While such categorization has been tantalizing for predominantly European music researchers, there are always exceptions and any broad statements warrant careful consideration.

Furthermore, recent research indicates that European and Indigenous perception can differ in how durational values are heard. Henry Stobart, an ethnomusicologist who has conducted extensive research in the Bolivian Andes, and Ian Cross, a music psychology and cognition researcher, have found that European listeners are likely to mishear the ratios in Indigenous songs. In particular, they found that Europeans are likely to hear ratios of 2:3 as 1:2 in Indigenous Bolivian music; furthermore, they noted that Europeans were likely to hear upbeats at the beginnings of songs where Bolivian performers heard downbeats.²¹ These findings point to the challenge of transcribing and analyzing rhythms in ways that do not misconstrue their fundamental qualities.

While many of the musical rhythms across North and South America use duple or triple meters, there are also plenty of examples that use complex meters, changing meters within songs, or added beats. These choices are related to both cultural and aesthetic preferences; for example, in the cases of conchero and the Inti Raymi festival mentioned above, symmetry and balance are culturally and musically significant, leading to predominantly symmetric melodies and choreography – and, by extension, symmetric meters. In other communities, asymmetry is a preferred aesthetic, such as in the case of the Yurok, Hupa, and Karok peoples of Northern California: as a result, the meters are often irregular or variations are added to create rhythmic asymmetry.²²

Across many Indigenous cultures, an underlying consistent rhythm provides the heartbeat for songs and dances, tethering instruments, vocal parts, and dance gestures to each other. Instead of conceptualizing singing and the voice as the musical focal point, the rattles, drums, stomping of feet, or other rhythmic actions provide the framework within which instruments, voices, and dance movements are structured. The songs of the Kwakiutl First Nation underscore this relationship: singing first begins after the drumbeat has been established, since the beat is understood to be of central importance.²³ Similarly, pow-wow songs begin with the drumbeat before the head singer introduces the song, allowing the singer to fit the melody into the grooves of the drumbeat.²⁴ When the melody moves between the strokes of the drum, it becomes highly syncopated with the drum and the movements of dancers. The composite rhythm of the drum, dance gestures, sounding regalia, and vocal line creates a rich rhythmic texture that cannot be fully grasped through the analysis of any single part.

Transcription

Although Indigenous American music is predominantly learned and preserved through oral tradition, Indigenous peoples in the Americas have long employed pictographs, symbols, and mnemonic devices to assist singers in recalling songs, rituals, and community history. For example, the Anishnaabe-Ojibwe have long used birch-bark scrolls to aid singers in recalling songs and histories.²⁵

European and Indigenous researchers have used transcriptions since the early twentieth century to codify the intricacies of Indigenous music and dance rhythms. Researchers transcribing Indigenous music have typically prioritized their research interests, often ignoring components of the performance they deemed extraneous, which in many cases included the sounds of dance steps or the nuances of the performance of instruments like rattles and drums.²⁶ In recent years, researchers have collaborated with Indigenous musicians to create transcriptions that more accurately reflect performance practices.²⁷ Since many Indigenous traditions use repetition with subtle variations, many transcriptions mark repeats in lieu of writing out the many iterations and variations of songs and dances. These transcriptions cannot fully capture the nuanced variations that occur in each reiteration. Ultimately, it is worth critically considering the value of analyzing Indigenous music with traditional Western analytic methods, especially in cases where analysis becomes removed from cultural contexts.