PRINCIPLES OF RHYTHMIC INTEGRATION IN AFRICAN DRUMMING.

Rhythm in African music has been a subject of much interest to ethnomusicologists (see Kauffman 1980). While it would be generally agreed that much has been accomplished in the common effort to explain "exactly what is the African treatment of rhythm" (Jones 1959), there is at present no consensus on how this should be done. Each individual has tried to interpret African rhythm from his or her background knowledge of Western music and theory as either a performer or an ethnographer (Kauffman 1980).

Of all these studies, the works of David Locke (1982), John Chernoff (1979), A. M. Jones (1954), and James Koetting (1970) are directly relevant to this discussion. Locke and Jones have focused primarily on Ewe dance drumming, Chernoff observed the Dagomba and Ewe,(1) while Koetting worked on Akan drumming, mainly to devise a notation system aimed at interpreting its holistic perception. Locke's background as a performer is reflected in his somewhat precise musical references and an approach that emphasizes descriptive as well as analytical methods. As a social scientist, Chernoff combines his learning experience with the aesthetics of his acculturative experience in Ghana. His approach is descriptive rather than analytical. Jones' transcriptions and theories have generated much discussion over the years. His experience with Tey, a native Ewe master drummer, was limited. He worked in a "foreign" environment where the drum ensemble was not readily available to him. Moreover, the use of his transcription machine with his informant limited his observations to one isolated part of the ensemble at a time. As a result, the vertical organization of his material is not adequately dealt with as he could not properly perceive the interrelationship of the whole ensemble. Nevertheless, the expositions of all these scholars show growing awareness of the intricacies of African drumming, partly because of the systematic approach adopted and partly because of their practical experience in African drumming.

The problem apparent in most of these studies stems from a general lack of a holistic approach. As Koetting points out:

   [T]o analyze the patterns of a drum ensemble piece individually is to miss
   the main characteristic of the music, which is the totality of sound
   produced by the interrelation of the various parts. This is particularly
   true in viewing the relation between the master drum and the rest of the
   ensemble.... What is needed is a comprehensive analysis that can encompass
   similarities and differences as components of the whole.... A deeper probe
   of the music--involving such detail as the precise beginnings of master
   drum patterns, possible verbal meanings in subgroup or individual
   supporting patterns, and dance associations--would have to be made before
   any trustworthy conclusions could be reached. (1970, 139)

I endorse the formulation of the problem by Koetting, since it confirms my own experience as a native scholar (Anku 1986; 1988; 1992; 1993; 1995a; 1995b). This study seeks to probe deeper into the internal structures of drumming by taking into account ethnographic evidence of "perception norms," which are often missed in analytical observations and in the formulation of theories on rhythm in African drumming.

Musical examples for this study are based on extracts from an Akan adowa drumming transcription provided as an appendix to this paper. The adowa ensemble (Akan funeral music) is made up of the following instrumentation: two boat-shaped bells (dawuro), two hourglass drums (donno), one hand drum (apentemma), one stick drum (petia), and a pair of talking drums (atumpan).

The most significant aspect of "multirhythm" perception in drumming is that the various composite patterns are heard in integration and not as isolated units. While in most other forms of music we have become accustomed to the tune concept (melody, voice leading) as distinct from harmonic and other accompanying elements (Seeger 1966), there appears to be no such clearly defined or comparable tune concept in drumming. It should be noted that even though some instruments tend to stand out loud and distinct (for instance, the adowa bell--dawuro--is the highest pitched, while the master drum--atumpan--is the most resonant), they do not in any way represent a melodic concept of the drum ensemble. In many multidrum cultures, however, drumming may simulate melody as the drummers verbalize a random selection of successive sound patterns, using nonsense syllables to represent what emerges from the sound palette, typical of the entire ensemble. What is usually verbalized is not played by any one particular instrument but is a result of the integration.

Evidence of this integration in the perception of the various instrumental patterns is demonstrated by Okyrema Asante, celebrated Akan master drummer, who is able to perform alone an entire popular band drumming type such as asiko or adowa, which is intended for a full drum ensemble (in a manner similar to the jazz drum technique). Even though this manner of performance is not typical in Akan drumming, the drummer is able to do so not by playing a succession of predetermined isolated patterns as traditionally prescribed but by performing what he perceives as the expectancies of the integration. One of the plates published in African Rhythm and African Sensibility features Gideon Folie Alorwoyie, an Anlo master drummer, performing in a similar manner (Chernoff 1979, 172). Such examples demonstrate the ultimate presence and recognition of emergent and resultant rhythms as legitimate conceptual and perceptual phenomenon in African drumming.

The emergent rhythm may be defined as a random and aesthetic selection of a continuum of "peaks of prominence" of sound patterns, from a palette of integrated rhythms. The resultant rhythm, however, is a more definite outcome of an integration, conceived monolithically (such as resulting between two bell rhythms in Akan adowa or between two aburukwa support drums in Akan kete drumming). The intrinsic perception of time synchronization of the various composite parts of the ensemble is to a great extent embedded in the performers' awareness and expectancies of the emergent and resultant rhythms.

The subjective approach adopted in the present study is crucial to understanding the problem in focus. It is a legitimate consideration in part because it is acknowledged by the various African traditions and because their musical references are ultimately expressed from individual or group perception. The norms of rhythm perception of a people, therefore, should provide a premise for such analytical discussion of procedures of rhythm, especially when examining music communicated entirely by oral tradition.

The intensity factor in drumming, for instance, should not be overlooked in the discussion of drumming procedures. In a particular performance, for example, one may observe the peculiar use of certain popular rhythms, or the enactment of certain popular movements in dance, or the significant use of certain rhythmic phrases in vogue, as for example in Akan or in Ewe drumming (see Exs. 1 and 2). Such utterances, which are either chanted, sounded on drum, or sounded by any other means available (e.g., the use of the bugle in Ewe borborbor drumming) may generate a renewed energy in the performance participation. This behavior is often initiated by a single participant. If it is timely, it draws the focus of the other participants. The response, of course, is instantaneous and spontaneous and in turn heightens the excitement level. The success of a performance may be measured in part by the degree of intensity (as may be expressed in terms of the depth of emotion or the extent of gaiety) of the performance. In the drum ensemble itself, articulations of certain rhythmic patterns may tend to highlight well-known rhythmic passages, which may also give rise to heightened intensity.

[Examples 1-2 ILLUSTRATIONS OMITTED]

Given this allowance of individual expression, the author's personal experience and perspective as a drummer may provide crucial information for this analysis.

Structural Aspects of the Integration

In drumming, rhythms are articulated into groups of various structural units. These rhythmic groups may be perceived on various levels as either (a) a span of two patterns (on a third synthesis level); (b) the pattern span itself (on a second synthesis level); (c) in groups of half patterns (on a first synthesis level); and (d) in four equidistant divisions of the pattern as beat units. This structural stratification bears some resemblance to the principle of grading (Nketia 1974, 133) and may be represented graphically as shown in Figure 1.

[Figure 1 ILLUSTRATION OMITTED]

These structures are manipulated in various ways, both linearly, in terms of each individual part, and vertically, within the rhythmic units of composite parts of the ensemble. Vertically, they are perceived in integration to achieve the same structural proportions that equate fundamentally to the pattern span. These are then compounded, in the main, in double pattern (antecedent/consequent) relationships on the third synthesis level.

In all these, the span of the "time-line" pattern (also referred to as the time span) is the main structural referent. In Example 3, each composite part is articulated within a framework of twelve equidistant time points (twelve eighth notes) constituting the pattern span. Spans of twelve and sixteen time point (sixteen sixteenth notes) patterns are typical in African drumming. Each composite span establishes an orientation or relationship with the regulative beat (marked with an arrow in Ex. 3).

[Example 3 ILLUSTRATION OMITTED]

Vertical consideration of composite patterns establishes three possible integral relationships with one another by the following procedures: (a) overlapping; (b) interlocking; and (c) adjacency and alternation. The resultant rhythm from such combinations constitutes a monolithic rhythmic concept.

Overlapping refers to composite patterns that are heard in juxtaposition, thus in somewhat staggered relationship. Overlapping occurs between two patterns of different relationships to the regulative beat (see Fig. 2a). An interlocking relationship occurs when two composite rhythms are superimposed and thus share the same orientation with the regulative beat (see Fig. 2b). Adjacency and alternation is perhaps the most frequently observed of the three relationships. It refers, as the name implies, to rhythms that are in adjacent (antecedent/consequent or call-and-response) relationship to one another (see Fig. 2c). These appear on all levels of structure.

[Figure 2 ILLUSTRATIONS OMITTED]

Call-and-response and hocket techniques are common in African music. They are not limited to vocal music but permeate all forms of performance practices. As with emergent and resultant rhythms, the performers' ability to maintain the overall timing organization of the piece rests in part on their perception of the various relationships of call-and-response and hocket patterns. These techniques find their highest expression in drumming. Call-and-response structures are easily recognizable between rhythmic phrases of wider structural spans, while shorter units interact in hocket fashion.

Call-and-response techniques are essentially a learned behavior deeply rooted in oral traditions of African peoples. Chorus responses of all kinds may be observed--between a teacher and his students, between a priest and the congregation of worshipers, chorus responses to ordinary greetings, and at sociopolitical gatherings--all of which form an integral part of day-to-day behavior. It is, ultimately, a compositional device found to facilitate communal music participation.

Specific Analytical Considerations

There are two important analytical perspectives under focus: (a) the drummer's "internal" holistic perspective and (b) the "external" holistic perspective.

The "Internal" Holistic Perspective

The main considerations in understanding the internal holistic perspective are the descriptions of how the performers find their cues and maintain their relationship with other composite parts of the ensemble: in other words, how the performer perceives his entries. One may distinguish between primary and secondary relationships.

Primary relationships are those that depend directly on the time line (for instance, donno and bell, apentemma and bell, atumpan and bell). Each performer perceives the time line in integration with his assigned pattern as a way to facilitate the cue entry process. Since there are no external concepts of timing gestures, such as those of the conductor in the Western orchestra, this built-in device is a crucial one.

Secondary relationships, on the other hand, refer to instances when the performer establishes multiple integration with patterns other than the time line. At certain intensity levels, the relationship between petia support drum and atumpan, or petia and apentemma, becomes so strong that one or both performers will momentarily disassociate from the time line (without losing the flow of the basic timing relationship of the piece) to indulge in some sort of dialogue, with one usually serving as a pacemaker for the other.

These kinds of momentary shifts in focus occur frequently throughout the piece. They may also arise because certain rhythms integrate more easily than others, or because they help to facilitate timing perception more easily. That is, there is a constant search for a pacemaker, thus, an easily integrated rhythm. As the relationships during the piece unfold in intensity and complexity, there is often a need to make frequent reference to the time line for confirmation of entries. This is especially important at changes in pattern orientation and with the introduction of complex staggered rhythms such as those referred to by the Akan as akita (literally, "to hold back").

Traditionally, an Akan drummer may make ironic reference to his pacemaker as his dawuro (bell): Wo na woye me dawuro, literally, "you are my bell." This is because the pacemaker provides the same referential role as the time line even though he may be playing on a composite instrument other than the bell.

At the height of intensity, the perception of the holistic composition becomes inseparable into composite rhythms. The performer no longer feels his rhythms as composites but as part of the monolith. It is the desire of every performer to sustain this perceptual unity, arising from the integration, for as long as possible.

One may also observe how each instrument relates within the holistic configuration of the ensemble from the internal perspective. In adowa drumming one may easily observe a dual distribution of instruments on the basis of assigned rhythmic patterns. There are two bells and two donno parts, the apentemma and petia rhythms related closely, while, at the same time, petia provides interesting rhythmic interchanges with the atumpan. The same is true between the low and high pitch relationships of the atumpan drum pair.

The two bell patterns have an interlocking relationship. Traditionally, one of the two bells is referred to as the interlocking bell, ntwamu. This interlocking effect confirms the close association between the rhythms of these two parts. Together they produce the resultant shown in Example 4. The second bell pattern easily integrates in the following relationship, as the two patterns share the same rhythmic motive at a certain point (see Ex. 5). To maintain the timing relationships correctly between the two instruments, the perception of the resultant along with the individual bell pattern is a helpful guide (see Ex. 6).

[Examples 4-6 ILLUSTRATIONS OMITTED]

Similarly, the two donno rhythms are perceived in integration with the regulative beat of the time line serving as their main point of reference. The "male" donno externalizes the beat for the ensemble, which also defines the dancer's basic walking steps. Together, they produce the resultant shown in Example 7. Both patterns interlock and are best perceived as a monolithic rhythm.

[Example 7 ILLUSTRATION OMITTED]

The donno resultant is heard in adjacent relationship to that produced by the bells (see Ex. 8). This adjacent relationship between the bells and the donno drums in turn produces another resultant rhythm, which Akans verbalize as shown in Example 9. Kyankyena kyankyena are nonsense syllables; opanin didi oyi ayiwa means "if the elder eats all the food, he must as well wash his own dish." The bells, donno, and resultant, verbalized rhythm combine as shown in Example 10.

[Examples 8-10 ILLUSTRATIONS OMITTED]

The apentemma and petia play various patterns that correspond to theme changes played by the master drummer. These themes in turn define structural relationships with the time span (see Example 11). In themes and variations A, B, and C (To me ku me, Adampom, and Otwe bedi nprem), the overlapping gives rise to the perception illustrated in Example 12. Once this relationship is established, one can enter a number of possible secondary relationships, such as with the male donno, which provides a steady referent (see Ex. 13).

[Examples 11-13 ILLUSTRATIONS OMITTED]

It is extremely difficult to sustain the off-beat character of the handdrum rhythms without a pacemaker. Together with petia, they produce different resultants in each of the theme changes mentioned above. For instance, the resultant that emerges in To me ku me from the integration is shown in Example 14. In the Adampom and Nsupa themes, a different resultant is heard (see Ex. 15).

[Examples 14-15 ILLUSTRATIONS OMITTED]

Like apentemma, petia patterns change in response to the various theme changes of the master drum. In Adampom, the pattern is in adjacent relationship to the time-line pattern while in Otwe be di nprem, it overlaps (see Ex. 16). In its secondary relationships, petia combines especially well with the donno group, apentemma, and with atumpan. The donna I provides the strong beats with the petia "off-beat" patterns in Adampom. Together with the apentemma they produce definite resultants in each theme relationship, while providing response rhythms that interchange with the master drum (see Ex. 17).

[Examples 16-17 ILLUSTRATIONS OMITTED]

The master drummer's themes and variations are presented as a succession of patterns that establish various orientations with the regulative beat. The other performers look to him as a driving force behind their changing perceptions, while they in turn provide the main ingredients of the background ostinato along which many possible integrations are established with each theme orientation.

In its primary relationships, the atumpan's To me ku me theme is heard in interlocking relationship with the time line. Both patterns are in the same orientation to the regulative beat (see Ex. 18).

[Example 18 ILLUSTRATION OMITTED]

The Adampom theme overlaps with the time-line pattern (see Ex. 19). In identifying its entry cue, the bell pattern is heard in integration with the Adampom theme as shown in Example 20. The pattern for the Otwe bedi nprem theme would be perceived as a complement to the time-line pattern (see Exs. 21 and 22).

[Examples 19-22 ILLUSTRATIONS OMITTED]

The Yenhu no nso theme is an overlapping relationship (see Ex. 23) and would be perceived as indicated in Example 24.

[Examples 23-24 ILLUSTRATIONS OMITTED]

Example 25 shows how Nsupa overlaps with the time line. It would be perceived as indicated in Example 26.

[Examples 25-26 ILLUSTRATIONS OMITTED]

The complex network of the entire background provides a cumulative resultant that in turn becomes a steady referent for the atumpan themes and variations.

The atumpan has an imposing timbre and resonance that allows several secondary relationships to be established. The various associating timbres of the background appear to be fragmented in between the "windows" of the atumpan rhythms. The petia and donno I rhythms, however, may be observed to be the closest associates or pacemakers of the atumpan variable rhythms.

This monolithic perception of the ostinato background of the rest of the ensemble places the atumpan drummer in a responsible and challenging position from which to steer the rhythmic floor of the performance.

The "External" Holistic Perspective

The external perspective represents a holistic consideration of the ensemble's rhythmic configuration. From this perspective, patterns do not just coexist or integrate on a unilateral level with one another; they are also subjected to different transformations due to a number of factors.

While rhythmic patterns are often isolated and discussed in their own terms, they are not normally heard with such clarity in actual performance context. That is to say, they are not easily recognizable with such clarity and independence as they are often assumed in theory. They are affected by various playing techniques such as open and muted tones of the bells and petia; low and high tones of the donno and atumpan; and open smack and palm techniques of the apentemma hand drum. These playing modes vary considerably in intensity.

It may be observed that references to tonal and timbral contrasts also imply a concept of dynamic (loud and soft) contrasts. These are complicated by overlapping, interlocking, and adjacent alternations of the patterns mentioned earlier. The staggered dynamic distribution, coupled with the complicated network of pattern integration (superimposition of soft and loud sections onto integrated patterns), naturally obscures the audibility of certain portions of individual patterns. This phenomenon is further complicated by the listener's proximity or location in the performance arena. It is often said that certain soft rhythmic nuances that are played but that otherwise are not heard in performance are meant for the drummers' personal edification, enjoyment, or as timing devises and may not be intended for the listener.

The procedure whereby resonant events tend to eclipse less resonant ones is termed masking (Josephs 1967, 45-47). For instance, Example 27 shows a petia rhythm in which the mutes (indicated by the smaller notes in Ex. 27a) are eclipsed and the pattern is transformed (see Ex. 27b). The same principle applies in the donno patterns, in which the high pitches tend to be more prominently heard than the low pitches.

[Example 27 ILLUSTRATIONS OMITTED]

There are three types of masking. Type a occurs where the pattern is weakest, as in Example 27. Type b occurs physically between acoustically more resonant and less resonant instruments. Type c occurs by the choice of the listener's focus of perception. As Nazir Jairazbhoy has stated:

   There is good reason to believe that selectivity is also practiced within
   the musical sound range.... Most musicologists have probably experienced
   instances of unconscious selection, as for instance, when "hearing"
   (actually becoming aware of) something new in an often-heard piece of
   music. Even an identically repeated musical pattern need not necessarily
   sound monotonous if the listener's awareness shifts from one musical
   feature to another. (1977, 267)

For instance, the combination shown in Example 28a may be masked as shown, resulting in the "heard" pattern in Example 28b.

[Examples 28 ILLUSTRATIONS OMITTED]

This is where emergent rhythms, defined above as an aesthetic selection of peaks of prominence, from a palette of integrated rhythms, actually come into play. For instance, from the combination shown in Example 29a, the listener may actually hear the resultant indicated in Example 29b. In perceiving this resultant, the unmarked portions of Example 29a become subordinated. Of course, the subordinated rhythms are not altogether inaudible. They create an added dimension that may surface between more audible sound structures and also with the focus, proximity, depth, and intensity of the listener's perception.

[Examples 29 ILLUSTRATIONS OMITTED]

In this final example an attempt is made to delineate further illustrations of pattern integration as they may be heard from the listener's external perspective. The illustration isolates four different possible emergent patterns that may give rise to a diversity of perceptions (see Ex. 30).

[Example 30 ILLUSTRATION OMITTED]

The "typical treatment of African rhythm" involves much more than mere coexistence of absolute rhythmic patterns. Jairazbhoy observes that "students of ethnomusicology are delving so deeply into the music of other cultures that many are now bi-musical.... They are beginning to approach one of the goals of ethnomusicology--the ability to identify not only those structures `familiar to us in the notation of the accidental art' (which is in itself a legitimate aspect of the field), but to identify and be able to isolate structures recognized by the native musicians themselves" (1977, 270). In this same light, the role of bi-musical native contributors in meeting the challenges of explaining and expressing their own native arts adequately in "global" terms would continue to be a welcome contribution to the field.

(1.) [For more information on Ewe rhythmic practices, readers are referred to Agawu (1995), which was published after the completion of this article.--Ed.]

REFERENCES

Agawu, V. Kofi. 1995. African rhythm: A northern Ewe perspective. New York: Cambridge University Press.

Anku, Willie. 1986. Rhythmic procedures in Akan Adowa drumming. M.A. thesis, University of Pittsburgh.

--. 1988. Procedures in African drumming: A case of Akan/Ewe traditions and African drumming in Pittsburgh. Ph.D. diss., University of Pittsburgh.

Anku, Willie. 1992. Structural set analysis of African music. Vol. 1. Legon, Ghana: Adowa Soundstage Production.

--. 1993. Structural set analysis of African music. Vol. 2. Legon, Ghana: Bawa Soundstage Production.

--. 1995a. Towards a cross-cultural theory of rhythm in African drumming. In Intercultural Music, edited by Cynthia Tse Kimberlin and Akin Euba, 167-202. Vol. 1. Bayreuth, Germany: E. Breitinger.

--. 1995b. Creative processes in African-American drumming in Pittsburgh and its relationship to jazz. International Jazz Archives Journal 1, no. 3: 4-27.

Chernoff, John. 1979. African rhythm and African sensibility. Chicago: University of Chicago Press.

Jairazbhoy, Nazir A. 1977. The "objective" and subjective view in music transcription. Ethnomusicology 21, no. 2:263-273.

Jones, A. M. 1954. African rhythm. Africa 24:26--47.

Josephs, Jess. J. 1967. The physics of musical sound. Princeton, N.J.: D. Van Nostrand.

Kauffman, Robert. 1980. African rhythm: A reassessment. Ethnomusicology 24, no. 3:393-415.

Koetting, James. 1970. Analysis and notation of West African drum ensemble music. Selected Reports 1, no. 3:115-146.

Locke, David. 1982. Principles of offbeat timing and cross-rhythm in southern Ewe dance drumming. Ethnomusicology 26, no. 2:217-246.

Nketia, J. H. Kwabena. 1974. The music of Africa. New York: W. W. Norton.

Seeger, Charles. 1966. Versions and variants of the tunes of "Barbara Allen." Selected Reports 1, no. 1:120-167.

APPENDIX

[ILLUSTRATION OMITTED]

WILLIE ANKU is chair of the music department at the University of Ghana, Legon. His research interest is in theory and structural analysis of African traditional music. He was a visiting professor at California State University, San Marcos. He is the author of Structural Set Analysis of African Music, vols. 1 and 2 (1992; 1993).