Rhythm, meter, and cross-cultural time: Western notation, West African polyrhythm, and additive meters

Intuition Beginner

Every piece of music has a pulse — a regular beat, like a heartbeat. Most Western music groups beats into patterns of two, three, or four. The time signature tells you the grouping: 4/4 means four beats per measure (a march or a pop song), 3/4 means three (a waltz). This is the system behind most music you hear in the West, from a Beethoven symphony to a Spotify playlist.

But this is just one tradition. In West African drumming, musicians layer several different pulses on top of each other. One plays three notes in the same time another plays two — a polyrhythm. In Balkan and Turkish folk music the meter can be 5/8, 7/8, 9/8, or 11/8 — pulses added together rather than divided. In Indian classical music the rhythm follows a tala, a cyclic pattern of beats.

Teental, the most common North Indian tala, has sixteen beats and returns to beat one — the sam — after every cycle. These are different musical systems for organising time, each coherent on its own terms. Rhythm is the most universal element of music: every known culture has rhythm, but not all have melody or harmony. Understanding cross-cultural meter is the entry point to non-Eurocentric listening.

Visual Beginner

The picture shows four ways human cultures organise musical time. At the top, Western time signatures divide a measure into equal beats: 4/4 (four quarter-note beats), 3/4 (three, the waltz), and 6/8 (six eighth notes grouped as two compound beats of three). At lower left, the standard West African bell pattern places seven strikes on a twelve-pulse cycle, creating the polyrhythmic foundation of much sub-Saharan ensemble music. At lower right, the North Indian teental cycle unfolds across sixteen beats as four groups of four, returning to the sama after each iteration.

Worked example Beginner

In 1972 Steve Reich composed Clapping Music, a five-minute piece for two performers clapping a single rhythm. The rhythm is a version of the standard West African bell pattern in 12/8 time — twelve pulses per cycle, with strikes on a fixed subset of those pulses.

Step 1. Both performers clap the same twelve-pulse pattern in unison, bar after bar.

Step 2. After about twelve bars, one performer shifts the pattern by one pulse — what was the second note is now the first. The other performer keeps the original pattern unchanged.

Step 3. The two performers are now in syncopation: the same pattern, offset by one pulse, produces a new polyrhythm. The listener hears interleaved accents that neither performer plays alone.

Step 4. Every twelve bars the shifting performer advances by one more pulse. After twelve shifts the patterns realign, and the piece ends.

What this tells us: rhythm is universal, but the systems for organising it are culturally specific. Reich's 1970 trip to Ghana, where he studied Anlo-Ewe drumming, gave him the bell pattern that drives the piece. A single twelve-pulse cell, phased against itself, generates the entire composition.

Check your understanding Beginner

Formal definition Intermediate+

The formal vocabulary of Western meter distinguishes time signature, beat unit, and metrical hierarchy. A time signature $\frac{n}{d}$ specifies $n$ beats per measure, with the note value $1/d$ receiving one beat (where $d=4$ is the quarter note, $d=8$ the eighth). Simple meters — $\frac{2}{4}$, $\frac{3}{4}$, $\frac{4}{4}$ — divide each beat in two; compound meters — $\frac{6}{8}$, $\frac{9}{8}$, $\frac{12}{8}$ — divide each beat in three, with the dotted quarter as the beat unit. Hypermeter extends the hierarchy across measures: just as beats group into measures, measures group into hyperbeats, typically in two- or four-measure spans.

The Cooper-Meyer durational-pattern framework ^{[CooperMeyer1960]} analyses rhythm as a nested hierarchy of five levels — note, motive, phrase, period, movement — each carrying one of five rhythmic groupings drawn from Greek poetic meter (trochee, iamb, dactyl, anapaest, spondee). A single passage carries different rhythmic groupings at different hierarchical levels simultaneously.

The Lerdahl-Jackendoff A Generative Theory of Tonal Music ^{[LerdahlJackendoff1983]} formalises rhythmic cognition as four hierarchical structures: grouping (how notes form motives, phrases, sections), meter (the pattern of strong and weak beats), time-span reduction (the structural importance of events within a time span), and prolongational reduction (tonal-hierarchical importance). Each structure is governed by well-formedness rules — hard constraints that any analysis must satisfy — and preference rules — defaults that can conflict, requiring the analyst to choose among competing interpretations. Metrical well-formedness requires that every beat at level $k$ is also a beat at all lower levels.

Additive meter — common in Balkan, Bulgarian, Turkish, and some Indian and Middle Eastern traditions — builds a measure as a sum of unequal beat-lengths rather than as the division of a single beat. The Bulgarian rachenitsa is $\frac{7}{8}$ notated as $2+2+3$; paidushko is $\frac{5}{8}$ as $2+3$; the Macedonian lesnoto is $\frac{9}{8}$ as $2+2+2+3$; $\frac{11}{8}$ often appears as $2+2+3+2+2$. Bela Bartok's ethnomusicological transcriptions of Bulgarian folk music (1903-1918) introduced these meters into the twentieth-century art-music vocabulary; Stravinsky's Rite of Spring (1913) alternates mixed meters within a single movement, with measure-to-measure changes such as $\frac{2}{4}-\frac{3}{4}-\frac{2}{4}-\frac{3}{4}-\frac{3}{4}$.

West African polyrhythm layers several simultaneous rhythmic streams, each with its own pattern, against a shared underlying pulse. The Anlo-Ewe of Ghana use a twelve-pulse reference cycle on which the gankogui (iron bell) strikes a standard seven-note pattern — the same pattern that Steve Reich cites as the source for Clapping Music. The Afro-Cuban son clave places five strokes on a sixteen-pulse cycle as $3+2$ (or $2+3$ in rumba clave). Chernoff's ethnography ^{[Chernoff1979]} emphasises the multidimensional, polymetric character of Ewe and Dagomba drumming: different instruments are heard in different meters simultaneously, and no single metric interpretation is privileged by the ensemble.

Indian tala is cyclic: a fixed number of beats (matras) is grouped into sections (vibhags), and the cycle (avartan) returns to the first beat (sama) after each iteration. Teental has sixteen beats as $4+4+4+4$; eka is a single-beat cycle; rupaka has seven beats as $3+2+2$ or $3+4$; jhaptala has ten as $2+3+2+3$. The tabla articulates the structure through bols — mnemonic syllables (e.g., dha, dhin, ta, tin, ga) each corresponding to a specific stroke on the drums. A tihai is a thrice-repeated melodic-rhythmic figure whose final note lands on the sama, cadentially closing the cycle; the design of tihais is a central compositional art-form in Hindustani music.

Syncopation is the deliberate displacement of accents onto metrically weak positions. Groove — the felt sense of rhythmic flow in jazz, funk, and Afro-Cuban music — emerges from the interaction of syncopation with systematic microtiming (Keil 1995; Iyer 2002): the placement of notes slightly before or after the strict beat, at offsets of 10-30 milliseconds, that the listener perceives as a characteristic "feel" rather than as error.

Counterexamples to common slips

• Slip: "African music has no meter." It has sophisticated cyclic and polymetric organisation, distinct from but equal in complexity to Western linear meter. The Ewe bell pattern is as strictly structured as a Bach fugue.

• Slip: "Polyrhythms are random." They are highly structured. The same Ewe bell pattern recurs across hundreds of performances and can be transcribed with millisecond-accurate precision.

• Slip: "Bartok invented additive meter." He documented it from Bulgarian, Romanian, and Hungarian folk traditions starting in 1903. The meters themselves predate his compositions by centuries.

• Slip: "Western meter is the most sophisticated." Indian tala cycles (some extending to 108 beats in Carnatic music) and Central African polyrhythm (Arom 1991) achieve equal or greater complexity.

• Slip: "Syncopation is mistakes." It is the deliberate displacement of accents that produces groove in jazz, funk, and Afro-Cuban music. A "straight" performance of a funk bassline feels wrong precisely because the groove requires microtiming.

• Slip: "Meter is universal in music." Gregorian chant, Japanese gagaku, and free jazz all employ non-metric rhythm. Meter is one option among several for the temporal organisation of sound.

Key theorem with argument Intermediate+

Theorem (Meter as attentional entrainment). Meter is not a property of the acoustic signal. It is a perceptual organisation constructed by the listener, who attends to the signal by entraining to a periodic pulse and inferring a hierarchy of strong and weak beats. Two listeners can hear different meters in the same acoustic signal, and the accents of metrical experience need not be present as accents in the waveform.

Argument. The signal carries events at various time intervals; the listener selects a subset as the beat and groups the beats into a metrical hierarchy. The selection is underdetermined by the signal itself.

(1) Metric ambiguity in minimal stimuli. Reich's Clapping Music presents the same twelve-pulse bell pattern offset against itself; listeners report hearing the downbeat migrate as the phase shift progresses, even though the acoustic signal changes only by a one-pulse displacement per cycle. The signal does not dictate where the downbeat lies.

(2) Cross-cultural metric reinterpretation. The same twelve-pulse bell pattern is heard as a single twelve-cycle by Anlo-Ewe listeners and as four groups of three (a 4/4 with triplets) by Western listeners trained in divisive meter. London's Hearing in Time ^[London2004] formalises this as meter-as-entrainment: the listener's attention forms a periodic expectation whose period is the beat, and the metrical hierarchy is the pattern of stronger and weaker expectations, not a feature of the waveform.

(3) Lerdahl-Jackendoff preference rules. The GTTM metrical structure is generated from the signal by preference rules — regularity (spans of equal duration), length (longer events attract stronger beats), local stress (louder or higher-pitched events attract beats), harmonic change (points of harmonic shift attract beats) ^{[LerdahlJackendoff1983]}. These rules can conflict; when they do, the listener's interpretation is a resolution, not a reading-off. Two listeners resolving the conflict differently hear two different meters.

(4) Syncopation and groove. Syncopation — accenting an off-beat — only makes sense relative to a metric grid. The grid is in the listener; the off-beat accent is in the signal. The groove of jazz, funk, and Afro-Cuban music emerges from the interaction of the listener's metric entrainment with the performer's systematic microtiming: the groove is not in the signal alone, and it is not in the listener alone, but in the relation between them.

(5) The limitation. Some metrical cues are unambiguously in the signal: a series of equally spaced loud attacks strongly suggests a beat at that periodicity, and harmonic changes in Western tonal music typically align with strong beats. The claim is not that the signal is irrelevant but that it underdetermines the meter. The listener does the rest.

$\square$

Bridge. The meter-as-entrainment thesis builds toward 34.02.04, where Gregorian chant — unmetered by design — shows the limit case of a repertory in which the listener's entrainment is deliberately not solicited, and the rhythm follows the prose text instead. The foundational reason is that metrical entrainment is a cognitive act distinct from acoustic event-detection, and this is exactly the central insight that identifies meter with attentional structure rather than with signal property. Putting these together, the bridge is between the Western mensural system (which prescribes a meter for the performer to reproduce) and the West African ensemble (which leaves the meter to the listener's inference from the polyrhythmic streams); the pattern recurs throughout the advanced results that follow, where each interpretive debate turns on how much of rhythm is in the signal versus in the listener.

Exercises Intermediate+

Interpretive debates Master

This unit's contested terrain is organised around five interpretive debates, each anchored in primary scholarship and each turning on how much of rhythm is in the signal, in the listener, or in the cultural tradition.

Debate 1 — Is "African rhythm" a Western projection? Agawu 1995 ^[Agawu1995] argues in African Rhythm: A Northern Ewe Perspective that the category "African rhythm" — supposedly more complex, more polyrhythmic, more embodied than Western rhythm — is a construction of Western ethnomusicology (Chernoff 1979; Arom 1991) that African listeners do not recognise as a description of their own experience. The Northern Ewe, Agawu contends, hear the bell pattern and the song melody as a single integrated texture, not as a polyrhythm requiring special apparatus. The counter-argument (Arom 1991 ^[Arom1991]; Toussaint 2013 ^{[Toussaint2013]}) is that the structural complexity of the patterns is real and measurable: the seven-on-twelve bell pattern is a mathematical object comparable to the five-on-sixteen son clave, and these comparisons are substantive.

Debate 2 — Is meter in the signal or in the listener? The Lerdahl-Jackendoff 1983 framework ^{[LerdahlJackendoff1983]} treats meter as a cognitive construction by the listener, generated from the signal by preference rules. Yeston 1976 ^[Yeston1976], in The Stratification of Musical Rhythm, had earlier argued that meter is a stratified phenomenon with multiple simultaneous levels, anticipating the cognitive view. London 2004 ^[London2004] gives the modern psychological formulation in Hearing in Time: meter is a form of attentional entrainment, a periodic structure in the listener's attention that need not correspond point-for-point to the acoustic signal. The opposing view — that meter is in the signal as a pattern of accents — persists in acoustic-musicological circles but cannot account for the metric ambiguity of Reich's phase pieces or the cross-cultural reinterpretation of identical signals.

Debate 3 — Is Western divisive meter "natural"? Western music theory since Rameau 1722 ^[Rameau1722] has treated meter as divisive: a whole is divided into equal beats, each divided into subdivisions. Hauptmann 1853 ^{[Hauptmann1853]} gave the systematic Hegelian defence of divisive meter as the only natural form. Ethnomusicology has dismantled the normative claim — additive Balkan, Turkish, and Indian meters are no less natural — while preserving the structural distinction for analytical purposes. The debate has both a descriptive dimension (divisive and additive meters are structurally different) and a normative dimension (neither is more natural than the other), and only the descriptive claim survives.

Debate 4 — Cooper-Meyer versus Lerdahl-Jackendoff. Cooper and Meyer 1960 ^{[CooperMeyer1960]} introduced the durational-pattern analysis of rhythm using Greek poetic feet as basic units at five hierarchical levels. Lerdahl and Jackendoff 1983 ^{[LerdahlJackendoff1983]} preserved the hierarchical insight but recast the units as cognitively generated grouping and meter structures, with preference rules replacing the prosodic vocabulary. The debate is whether rhythmic analysis should be formal-prosodic (Cooper-Meyer) or cognitive-generative (Lerdahl-Jackendoff). The cognitive programme has dominated music theory since 1983, but the formal-prosodic programme retains adherents in ethnomusicology and performance studies where the description of specific rhythmic shapes in prosodic terms remains useful.

Debate 5 — The geometry of rhythm. Toussaint 2013 ^{[Toussaint2013]}, in The Geometry of Musical Rhythm, represents rhythmic patterns as points on a circle and studies their mathematical properties: symmetry, evenness, inter-onset-interval vectors, and closeness to the maximally even distribution. The seven-on-twelve West African bell pattern and the five-on-sixteen son clave emerge as mathematically distinguished objects, both near-maximally-even distributions of their strokes on their cycles. The debate is whether such mathematical properties explain the cross-cultural prevalence of these patterns (the strong Toussaint claim) or merely describe them after the fact (the weak claim). The cross-cultural empirical evidence — that similar patterns recur in West African, Afro-Cuban, Brazilian, and Middle Eastern traditions without documented historical contact — supports the strong claim more than is sometimes conceded.

Synthesis. The five debates fit together as a sequence in which each stage both builds toward the next and appears again in every subsequent one. The foundational reason is that rhythm and meter are simultaneously acoustic, cognitive, and cultural objects, and each debate foregrounds one of these dimensions at the expense of the others: Agawu foregrounds culture, Lerdahl-Jackendoff foreground cognition, Toussaint foreground the acoustic-mathematical. The central insight is that no single level exhausts the phenomenon — this is exactly what identifies a performance of the Ewe bell pattern with both a mathematical seven-on-twelve distribution and a culturally situated Ewe ritual object, and the bridge is between the formal-geometric and the ethnographic analyses of the same musical event. Putting these together with the Cooper-Meyer-to-Lerdahl-Jackendoff trajectory, the pattern recurs throughout this unit's connections: meter is the listener's cognitive construction, and that construction is shaped jointly by the acoustic signal, by the cultural tradition in which the listener was trained, and by the mathematical regularities of cyclic patterns that recur across cultures without contact.

Full argument set Master

Proposition (Clapping Music realignment). In Reich's Clapping Music (1972), after twelve phase shifts of one pulse each, the moving part returns to its original alignment with the stationary part, and no smaller positive number of shifts produces a realignment.

Argument. Let the rhythmic cell be a function $p:\mathbb{Z}/12\mathbb{Z}\to \{0,1\}$ where $p(i)=1$ means a clap on pulse $i$ and $p(i)=0$ means a rest. A one-pulse phase shift of the moving part is the index map $i\mapsto i+1(\mathrm{mod}12)$, so after $k$ shifts the moving part claps on pulse $i$ whenever the stationary part claps on pulse $i-k(\mathrm{mod}12)$. Realignment occurs when the two parts clap on the same pulses, that is, when the set $S=\{i:p(i)=1\}$ satisfies $S=S+k(\mathrm{mod}12)$.

The shift generates the cyclic group $\mathbb{Z}/12\mathbb{Z}$ of order twelve, so $S+12=S$ for any set $S$. For $1\le k<12$, the set $S+k\ne S$ whenever $S$ has no nontrivial rotational symmetry. The standard bell pattern has seven strikes and five rests on a twelve-cycle, and no rotation by $k\in \{1,2,\dots ,11\}$ leaves it invariant: the pattern's inter-onset-interval sequence is not periodic with any period dividing twelve other than twelve itself. Therefore $S+k\ne S$ for $1\le k<12$, and the realignment occurs at exactly $k=12$. $\square$

Proposition (Three-against-two polyrhythm has period six). The three-against-two polyrhythm, in which one stream strikes at times $\{0,2,4\}$ within a six-pulse cycle and the other at times $\{0,3\}$, has realignment period exactly six pulses.

Argument. The three-strike stream has period two (the spacing between its strikes); the two-strike stream has period three. The two streams realign at pulses that are multiples of both periods, that is, at multiples of $\text{lcm}(2,3)=6$. Within one six-pulse cycle, the coincidence set is $\{0,2,4\}\cap \{0,3\}=\{0\}$, so the streams coincide only at the start of each six-pulse cycle and not within it. Since no positive multiple of six smaller than six exists, the realignment period is exactly six. The listener perceives the polyrhythm as a single six-pulse cycle with a characteristic internal accent pattern, not as two independent cycles of two and three. $\square$

Connections Master

• Music fundamentals: rhythm, melody, and harmony 34.01.01. The survey unit 34.01.01 introduces rhythm, melody, and harmony in a single overview; this unit is the depth specialisation of the rhythmic component, extending the survey's brief mention of polyrhythm into a full cross-cultural treatment. Builds toward the depth treatments of harmony in 34.01.02 pending and musical form in 34.01.03 pending, both of which assume the metrical vocabulary established here.

• Gregorian chant and Notre Dame polyphony 34.02.04. Gregorian chant is the limit case of unmetered rhythm in the Western tradition: the prose-driven free rhythm of the chant repertory is intelligible only against the metrical framework that the Western tradition later constructed for itself. The Notre Dame rhythmic modes (c. 1170-1230) are the first measured Western meter and the direct ancestors of the time-signature system analysed in this unit's formal definition.

• Cinema verite and direct cinema 34.05.02. Both this unit and 34.05.02 analyse cross-cultural art forms against Eurocentric-canonical narratives: just as Western meter is one tradition among many rather than a universal, so the Hollywood narrative is one cinema tradition among many rather than a universal. The methodological parallel — treating the dominant Western form as one option rather than as the norm — is the same in both units.

• Aesthetics: beauty, art, and judgment 20.04.01. Rhythm and meter raise the central aesthetic question of 20.04.01: is the beauty of a rhythmic pattern a property of the pattern itself (formalism) or of the listener's response to it (response theory)? The meter-as-entrainment thesis defended in this unit's Key theorem aligns with the response-theoretic answer — the pattern's metrical beauty is partly constructed by the attending listener, not wholly present in the acoustic signal.

Historical & philosophical context Master

The systematic Western theory of rhythm begins with Rameau, whose Traite de l'harmonie reduite a ses principes naturels (1722) ^[Rameau1722] subordinated rhythm to harmony in the framework that would dominate French theory for two centuries. Hauptmann's Die Natur der Harmonik und der Metrik (1853) ^{[Hauptmann1853]} gave the Hegelian defence of divisive meter as the only "natural" metrical form — a position ethnomusicology has since dismantled but which set the terms of the nineteenth-century debate. The Cooper-Meyer The Rhythmic Structure of Music (1960) ^{[CooperMeyer1960]} introduced the five-level hierarchical analysis using Greek poetic feet, recasting rhythm as a nested structure rather than a flat sequence of durations; it remains the point of departure for all subsequent rhythmic theory.

The modern cognitive theory crystallised with Lerdahl and Jackendoff's A Generative Theory of Tonal Music (1983) ^{[LerdahlJackendoff1983]}, which applied the generative-grammar programme of Chomsky to tonal music and produced the four-structure analysis — grouping, meter, time-span reduction, prolongational reduction — that has dominated music theory since. Yeston's The Stratification of Musical Rhythm (1976) ^[Yeston1976] anticipated the cognitive view by treating meter as a stratified phenomenon, and London's Hearing in Time (2004) ^[London2004] gave it the modern psychological formulation as attentional entrainment.

The ethnomusicological counter-tradition runs from Chernoff's African Rhythm and African Sensibility (1979) ^{[Chernoff1979]} through Arom's African Polyphony and Polyrhythm (1991) ^[Arom1991] to Agawu's African Rhythm: A Northern Ewe Perspective (1995) ^[Agawu1995], which contests the very category its predecessors constructed. The mathematical-comparative thread is represented by Toussaint's The Geometry of Musical Rhythm (2013) ^{[Toussaint2013]}. Clayton's Time in Indian Music (2000) ^{[Clayton2000]} provides the canonical treatment of tala as cyclic meter.

Bibliography Master

Primary sources and treatises

• Rameau, Jean-Philippe. Traite de l'harmonie reduite a ses principes naturels. Paris: Jean-Baptiste-Christophe Ballard, 1722. Translated by Philip Gossett as Treatise on Harmony. New York: Dover, 1971.

• Hauptmann, Moritz. Die Natur der Harmonik und der Metrik. Leipzig: Breitkopf und Hartel, 1853. Translated by W. E. Heathcote as The Nature of Harmony and Metre. London: Swan Sonnenschein, 1888.

• Cooper, Grosvenor W., and Leonard B. Meyer. The Rhythmic Structure of Music. Chicago: University of Chicago Press, 1960.

• Yeston, Maury. The Stratification of Musical Rhythm. New Haven, CT: Yale University Press, 1976.

• Chernoff, John Miller. African Rhythm and African Sensibility: Aesthetics and Social Action in African Musical Idioms. Chicago: University of Chicago Press, 1979.

• Lerdahl, Fred, and Ray Jackendoff. A Generative Theory of Tonal Music. Cambridge, MA: MIT Press, 1983.

• Arom, Simha. African Polyphony and Polyrhythm: Musical Structure and Methodology. Cambridge: Cambridge University Press, 1991. Originally published as Polyphonies et polyrythmies instrumentales d'Afrique centrale (Paris: SELAF, 1985).

• Agawu, Kofi. African Rhythm: A Northern Ewe Perspective. Cambridge: Cambridge University Press, 1995.

• London, Justin. Hearing in Time: Psychological Aspects of Musical Meter. Oxford: Oxford University Press, 2004. 2nd edition 2012.

• Clayton, Martin. Time in Indian Music: Rhythm, Metre, and Form in North Indian Rag Performance. Oxford: Oxford University Press, 2000.

• Toussaint, Godfried T. The Geometry of Musical Rhythm: What Makes a "Good" Rhythm Good? Boca Raton, FL: CRC Press, 2013.

Musical works cited

• Reich, Steve. Clapping Music. 1972. In Works: 1965-1995. Nonesuch Records, 1997.

• Bartok, Bela. Mikrokosmos, Vol. 6, nos. 148-153 ("Six Dances in Bulgarian Rhythm"). 1939. Budapest: Editio Musica.

• Stravinsky, Igor. The Rite of Spring. 1913. London: Boosey & Hawkes.

Secondary scholarship

• Keil, Charles. "The Theory of Participatory Discrepancies: A Progress Report." Ethnomusicology 39, no. 1 (1995): 1-19.

• Iyer, Vijay. "Microstructures of Feel, Macrostructures of Sound: Embodied Cognition in West African and African-American Music." Ph.D. diss., University of California, Berkeley, 2002.

• Honing, Henkjan. Musical Cognition: A Science of Listening. New Brunswick, NJ: Transaction Publishers, 2013.

• Titon, Jeff Todd, ed. Worlds of Music: An Introduction to the Music of the World's Peoples. 6th ed. Belmont, CA: Schirmer Cengage Learning, 2016.