Djembe & Mande Music Page
(last revision 02/26/99)

Rainer Polak, Bayreuth

Jenbe Music in Bamako
 Microtiming as Formal Model and Performance Practice1

[this is a slightly changed and shortened version of an article published 1998 in Iwalewa Forum 2 (ISSN 1435-7356; ISBN 3-89654-241-X): pages 23-42. With the kind permission of the African Studies Centre of the University of Bayreuth, Germany.]


Playing with microtiming is a remarkable feature of musical rhythm in the Western Sudan. Both African and Western musicians and musicologists have time and again wondered at the capability of Malian, Senegambian or Guinean drummers to excitingly place notes on, next to, apart from, or in between expected points of time. This musical practice results from a complex interaction of different factors. Yet, in my view, the so-called "systematic variations in duration" (Bengtsson 1974) would form its basis. This concept holds that the equal pulse base underlying African and other dance-oriented music is often inflected according to regular and constant patterns. Jenbe drummers, in particular, make extensive use of a limited number of sophisticated inflection patterns to enrich their playing in a specific manner. This paper addresses the drummers' microtiming both as a system and as a practice in the context of professional festival music in Bamako. 

Jenbe drummers in Bamako perceive drum patterns to be wrong when there are slight deviations of timbre sequence or position relative to the ensemble's polyphony. This is in accordance with what is regarded as defining the "pattern" in West African drum ensemble music (Koetting 1970: 120; Pantaleoni 1972:59). On the other hand, the drummers allow extraordinary variability concerning the microrhythmic realization of their rhythmic figures. The timing of impulses may fluctuate and swing to an extent that no equal pulse base can be established underlying the rhythmic figure. This contradicts the definition of West African drum patterns (Pantaleoni 1972: 58; Koetting 1970: 122). And it made me experience frustration since I have first started field research among professional jenbe drummers in Mali.2 Lacking the feel of a pulse base, recognition and mimesis of drum patterns for me often was difficult to attain. Notation sometimes appeared to be a lottery without prize. In some cases abstracting two pulse bases from one drummed pattern, both of which approximately explaining what was actually played, seemed to be the most truthful approach. The recognition of a rhythmic Gestalt in relation to more than one perception frame helped me practicing it afterwards. Consequently I started to transcribe many patterns in two versions. The following illustration is an example. Version 1 shows the dunun (accompanying bass drum) pattern of rhythm3 menjani in a ternary feeling. But the same pattern can be perceived to be a binary rhythm, too, as transcribed in version 2.

Figure 0 - different realizations of the dunun pattern in the rhythm menjani:

"1, 2, 3, 4" represents the beat or meter. "." represents a single pulse of an ongoing pulsation. "O" represents full strokes producing open sounds on the dunun. " -" represents stopped strokes producing muted sounds on the dunun.

The aim of the first part of this article is to come to an explanation of the jenbe drummers' microtiming from a formal, etic point of view.

Pulsation Theory I

The ethnomusicologist Chris Waterman once experienced a similar problem during a jam session. Everything went fine with himself as a bass player encountering a Yoruba dundun drummer. But when a "fine" Wolof sabar drummer joined in, Waterman was confused by the latter's microtiming "of more-or-less triplets placed so far behind the pulse, like a water skier leaning way back on his ropes, that the groove almost capsized". The Yoruba musician put it more drastically in asking: "Is this drumming? Is he a drummer?" (Waterman 1995: 93). Obviously the Senegalese was disregarding rules that both the American and the Nigerian were regarding as defining musical rhythm as such. Microtiming from the Western Sudan does not only raise a notational problem but one of musical practice and analysis, too.

The extraordinary and sometimes eccentric ways of timing in the Western Sudan have attracted musicians' and musicologists' attention before.4 But none of them systematically considered its obviously ambivalent, if not contradictory relation to the theory of elementary pulsation underlying all danceable African music. Engaging in this fundamental theoretical concept, however, seems to be a precondition for a sufficient explanation of our phenomenon in terms of ethnomusicological analysis.

The idea is that one infinite series of equidistant smallest time units forms an unconscious unilinear flow somewhere in or behind the minds of African musicians during their performance. This pulsation would screen or break down different aspects of organization of time like ensemble synchronization, pattern composition or overall rhythmic structure. The concept was elaborated as "Nennwerte" (Dauer 1966), "density referent" (Hood 114), "fastest pulse" (Koetting 1970), "valeurs opérationelles minimales" (Arom 1984) or "Elementarpulsation" (Kubik 1988).5  Note that the different concept of beat (which bundles two, three or four pulses) was confusingly termed "gross pulse" by Koetting (1970) and was misleadingly termed "pulsation" by Arom (1984). Conversely, Kauffman speaks of "common fast beat" referring to pulsation. This terminological diversity seems impractical and unjustified as all are definitely talking about the same thing. I simply will speak of "pulses" or "pulsation" referring to the phenomenon of smallest time units and of "beat" referring to a higher analytic level of meter.

After its implementation the concept of pulsation has quickly been accepted as a basic assumption in scholarly thinking about African rhythm (Merriam 1982: 445). There have been reflections about its questionable accordance with emic conceptualization, but as an analytical tool its validity has been unquestioned ever since:

It seems almost undebatable that most African rhythms can be related to fast regular pulse. (...) Musical scholars are probably more aware of the density referent than are performers of the music, but the absolute accuracy of rhythmic relationships in performance seems to attest to at least an unconscious measuring of fast, evenly spaced units. (Kauffman 1980: 407).
Deviations of equidistance between pulses were considered unimportant or natural by-products of the inexactitude immanent in any human motional activity. "The listener's and the the player's time perception 'corrects' this physical irregularity and unconsciously bends the beats into a regular series" (Kubik 1972: 33; see Merriam 1982: 451). Yet deviations which could neither be explained physically nor by the psychology of perception still have not been addressed by the means of musicological analysis. Referring to the musical cultures of the Western Sudan (Niger, Burkina Faso, Mali, Senegal), Gerhard Kubik spoke of completely irrational anticipations and retardations (Kubik 1988: 77). The present paper instead tries to take into account Dane Harwood's comment on recent research in microtiming: "Human cultures not only invent rules, they also invent rules how to play with the rules" (Harwood 1995: 77).

Swedish musicologist Ingmar Bengtsson considers microrhythmical shifting to be an essential requirement of any dance music. He elaborated the concept of "systematic variation in duration". By this he meant that an equal pulse base is regularly and constantly inflected according to specific patterns, in contrast to "agogic", i.e. only local or arbitrary variation in microtiming (Bengtsson 1974: 202). Jenbe drummers extensively use a limited number of sophisticated patterns to inflect the pulse base underlying their music. I have so far been able to detect four basic systems of pulse inflection practiced in Bamako.

Two Models

In the following I try to outline two models [all four models are covered in the print version] for the specific forms of pulse inflection as applied by jenbe drummers from Bamako. A musical example for each of the four models is given in the appendix.

1. The first form is based on a binary pulsation. This means that two pulses are bundled in one beat. The binary pulsation is inflected by retarding every second pulse. Thus, pulses come in pairs of two. This is a very simple structure practiced in a wide range of musical genres.6 If the grade of inflection is sufficiently strong, the larger distance between the bundles of two pulses allows the feel of an extra pulse: a ternary target pulsation comes into existence.

Figure 1 - inflection of pulsation model 1:

The gray areas mark the time span a retarded pulse may be placed in. Thin additional lines indicate the wider space between the pairs of two contracted pulses.

2. The following form sounds stranger to listeners socialized elsewhere than the Western Sudan. The model in Figure 2 shows a quarternary pulsation. It is inflected since every second, every third and every fourth pulse is pulled forward. The second pulse might vanish by being fused into the first pulse as double-stroke. Thus, a pulse is swallowed, and the virtual target pulsation has only three steps per beat instead of four.

Figure 2 - inflection of pulsation model 2:

The gray areas mark the time span an anticipated pulse may be placed in.

Jenbe drummers in Bamako seem to apply systematic variations in duration according to a common principle: A basic pulsation is inflected in such a way that the extended or reduced space between pulses gives rise to the impression of an additional or a lost pulse. Specifically resulting from this strategy, a second, virtual target pulsation is brought on. The–now two–pulsations set the limiting values for the microrhythmic tolerance in placing the impulses of a rhythm pattern.

At this point I turn to integrating my argument into discussions of rhythm in the Western Sudan and Africa in general.

Pulsation Theory II

Gerhard Kubik's interpretation of microtiming in this region as completely irrational proved insufficient. Jenbe drummer's at least, and probably other musicians in the Western Sudan, too,7 systematically and intentionally bend time by inflecting the pulsation of smallest time units in sophisticated ways that conflict with standard pulsation theory. Yet theoretical assumptions have to be evaluated on the grounds of empirical findings, not the other way round. In order to reconcile pulsation theory with empirical timing practice as modeled above, it has to be reconsidered or softened in two respects. First, equidistance of pulsation is to be enriched with the possibility of non-equal but nonetheless regular time spans between the single units. Second, pulsation is not unilinear by definition. A single pulsation is not in every case binding upon all members of an ensemble playing at one time. On the contrary, either an alternating or a parallel relation to two different pulsations seems to be a characteristic possibility of Malian and neighboring music.

Roderic Knight established the phenomenon quite distinctly in his comprehensive dissertation on the music of Manding Griots in Senegambia. He stated the "tendency for the lilting |x . x x . x| rhythm so common in the kumbengo [basic kòra-pattern used to accompany song] to become a series of evenly measured notes in the donkilo [vocal pattern] [...] while the triple-pulse rhythms of the kumbengo are maintained. The intent does not appear to be to create a carefully measured 2/3 between the voice and the instrument. It is more like the rhythmic counterpart of the familiar melodic technique of heterophony in which each part is playing the same melody, but in its own idiomatic way" (Knight 1973: 270-271; inserts by the author).

Knight's statement emphasizes the element of tolerance and playfulness in performance practice. This is important. It reminds us that the bilinear pulsation I proposed as defining the musicians' limiting values in pattern realization are important mainly in etic models. In practice the ways of acting in between is what makes the groove go round. But I think we have to go further in looking for accurate analogies to gain a better understanding of the phenomenon. To do so I have to discuss another theoretical assumption of African musicology at an even higher level of abstraction than before.

There have been some projects in ethnomusicology trying to evaluate and condense literature in order to specifically characterize "African" rhythm. The most ambitious efforts in my view focused on features like cross rhythm or multipart relationships between patterns and multiple or non-unilinear organization of time (Kaufmann 1980; Stone 1982; Kubik 1988). Until now it is generally accepted that this element of "multiplicity" or ambiguity can predominate only on one level of time organization, and that on at least one other level intersubjective ground has to be established unequivocally in order to enable pattern composition, individual orientation and ensemble synchronicity. This was held to be the function of pulsation, most sharply in contrast to the ambiguous level of meter or Gestalt perception (Kauffman 1980: 393). The character of clear pulsation upon which polymeter is built even is expressed in the terms of "elementary pulsation" and "density referent". It is paradox that the search to understand ambiguous or multi-perceivable time in African music should change the hypothesis of unilinear pulsation into an a priori pre-condition. Yet exactly this was taking place. "It is precisely with the recognition of unequal beats composed of equal pulses that ethnomusicology begins to break out of a linear perspective" (Stone 1985: 140). But what if pulsation itself is unequal, multilinear and ambiguous?

As far as I know, Hans Kroier was the first to relate cases of different quantizations of one pattern to both the concept of systematic variations of duration and to pulsation theory.8 In his enlightening magisterial thesis at the Free University Berlin he termed the ambiguity of reference to two virtually parallel pulsations in Cuban rumba as "metrical ambivalence" (Kroier 1992: 6). Concerning structure, bilinearity of pulsation indeed closely resembles another fundamental feature of African rhythm on the level of beat or meter: polymetrics, the combination of more than one meter referred to at the same time. Both polymetrics and dual pulsation aim at the same objectives with analogous devices: Polymetrics rests upon unilinear pulsation but produces ambiguity in meter. Dual pulsation reinforces a non-inflected, metrical beat but produces ambiguity in pulsation. In both cases ambiguity aims at splitting perception by parallel reference systems coordinated by the numerical proportions of 2 against 3 or 3 against 4. In this perspective, polymetrics and (analogously termed) polyquantization have more in common than contrasting features. Moreover, it is no problem to realize both structures at the same time. Some rhythms in jenbe music of Bamako actually contain patterns realizing both structures at the same time. This results in very complex relations if seen mathematically, but still fits into the groove without sounding overly construed.9

I now leave the formal aspects of systems of variation in duration and turn to its practice in performance briefly viewed from the perspectives of repertoire, interaction and cultural change.

Timing as Practice

The repertoire of jenbe music consists of drum patterns which are performed as ensemble rhythms. The patterns of most rhythms are inflected according to only one systematic variation in duration. The Maninka villagers of Sagele10 practice about twelve individual jenbe rhythms. All of the repertoire is explicitly termed "Maninka" and emically viewed as cultural property specific to the region and the ethnic group. On the other hand, parts of the repertoire originally stem from as far as the Mauritanian border and Senegal.11 Obviously there are processes of incorporating foreign rhythms into the practice seen as the Maninka's "own tradition", even though their own tradition is partly defined in contrast to those other regions and groups where the incorporated material originally comes from.

In Bamako, moreover, the repertoire comprises about 60 rhythms from all over the country and beyond. Rhythms of very different instrumental ensembles and musical genres are transposed into jenbe music, and new rhythms come forward almost every year. Urban professionals have to come to terms with this rapid transformation of their repertoire. To keep it practicable, they combine those individual rhythms which are marked by structural similarities. The rhythms thus being treated as one larger family share similar drum patterns and one form of pulsation and inflection. Generally, the same accompaniment patterns and signal phrases are applied to each of them. Thus e.g. two jenbe accompanying patterns12 are applied to more than 30 rhythms in Bamako. The specific accompanying patterns which existed for most rhythms get out of use. Lead-drum patterns become exchangeable to a certain degree, too. Some rhythms are merged into one piece and finally are fused into a new super-rhythm. E.g. maraka, the rhythm most often played at festivals in Bamako, nowadays includes elements of variation which formerly represented and were performed as independent rhythms, like take, sogolo and manenka-mòri. That is to say, specifically inflected pulsation is a basic structure that–together with pattern and rhythm–constitute character and finally identity in the repertoire of jenbe music.

I will now try to investigate the role microtiming plays in the context of performance interaction. In Bamako, the jenbe drum ensemble performs to the occasion of certain festivities.13 Its main objective is to frame the interaction between singers, dancers and the encircling crowd of participants. The lead drummer coordinates the repetitive course of interaction. First he challenges anybody from the audience circle to come in for a dance solo. If somebody follows his call, he immediately pushes the tension towards a maximum level. The climax goes on for a very short period of time until the sequence is finally cut off by a signal phrase which at the same time gives way to the next cycle. The hot phase when the lead-drummer encounters a dancer is called "golobali", that is "precipitation", in Bamana.14 Among others, two musical means are available to make it work: one might increase tempo and/or loudness; and one plays dense patterns that continually fill each pulse with a sounding impulse.

The more one relies on increase in tempo and the higher the absolute tempo, the less important becomes the inflection of pulsation. Slow tempo gives fuller scope to playful microtiming and its perception. Yet even at extremely fast tempi a dexterous musician has room to move in microtiming. Rather than only being caused by the physical limitations of drumming and listening, the inversely proportional relation of tempo and inflection represents a function of style. There are numerous aspects that differentiate style in jenbe playing, among others regional, ethnic and personal identity. The stylistic differences between the old and the young stand out the most.

It was not before the 1950s that jenbe drumming became a full-time occupation in Bamako. In the rural regions where the urban musicians originally stem from a drum ensemble consisted of at least three and up to six musicians. In the course of urban professionalization the cast was reduced to only two persons for commercial reasons. Performing with only one jenbe and one accompanying bass drum became typical of the Bamako tradition from the 1960s until well into the 1980s. It is specifically exacting to play by twos. You are less free to improvise.15 Yet more important to our concern is that you are less free to play loud and fast and to make extensive use of rolls. Tension has to be created and maintained in more economic ways (especially the phase of precipitation mentioned above means that the drummers have to meet an increased demand for tension by their performance partners). Systematic variation in duration supplies a most deliberate means to do so. Especially when restlessly realizing all pulses with accentuated strokes during échauffements, the inflection of pulsation becomes clearly audible and thus gains full efficacy. Playing with inflection of pulsation was an essential feature of the jenbe style of Bamako from the 1960s well into the 1980s.

During the same decades, however, festive performances in Bamako have tended to become part of a increasingly differentiated social sphere, that is "secular" entertainment culture. Among other things, this causes a decrease in age of the audiences. This might be reinforced by other social and demographic developments that favor the influence of the youth on the expressive culture in Bamako. The young, however, love to dance at a faster pace than the old. Apart from that, the "traditional" festival culture has been influenced by "folkloristic" ballet groups and recently by international percussion-ensemble concert music.16Both of these prefer large ensembles and fast dancing.

In 1995 I recorded a CD featuring three different soloists playing the very same jenbe-drum but producing very different sounds and styles (Polak [ed.] 1996). One was the then 79 years old Bani Dunbia, who–together with Daba Keita17– became the godfather of professional jenbe music in Bamako during the 1960s and 1970s. His playing is not only slower in absolute tempo but manages with less marked relative increase in tempo, too.18 But his ways of pulse inflection sometimes swing spectacularly. When I played the recordings to his musical children and grandchildren, they showed a mixture of respect and amusement: Even in the very 'status-laden' context of an industrially produced CD they found their former master's style so old-fashioned as to shake their heads and smile. The style that presently gets drummers the most engagements–faster, louder and more notes–places diminishing emphasis on inflection of pulsation.

These rough examples of lines of change were intended to show that microtiming as a feature of musical style is rooted in the wider social and cultural context of musical practice.

Searching for Groove

Let me conclude with some comments on the theory of "participant discrepancies", as it was brought forward in a special issue of Ethnomusicology in 1995. Its essence is that all played rhythm naturally deviates from the exactitude of abstract time; and that thus all ensemble music represents a socializing interplay between the participants' individual microtimings. Charles Keil, the protagonist of Participatory Discrepancies, shapes the identity of his concept mainly in dichotomically opposing it to other concepts. He puts "good" groove against "bad" musical syntax, he claims practice to be more important than context, and playfulness to be more human than patterned structure. Formerly, irregularities in microtiming were supposed to be insignificant. Keil (1995: 4) asserts that to be out of time and out of tune is the key to making good music in a social context.

Generally, we should be grateful to Keil for having focussed interest on variability and tolerance in realizing musical structures in rhythm practice. The articles of Prögler and Alen that follow Keil's introduction show that microtiming in African and Afro-American musical rhythm is a complex result of at least three different factors. Prögler (1995: 37) describes a so-called "intermediate beat swing" of a jazz drummer's ride tap. This swing is produced by the shifting of some pulses in relation to a non-inflected beat. It corresponds to what Bengtsson called systematic variations in duration, and jenbe polyquantization described in model three is only a specific if sophisticated form thereof. Prögler's findings moreover state the strategy that an ensemble might play in synchronicity but out of phase (1995: 38, 44-46). This means that the musicians' parts proceed in parallel at the same tempo, but with one musician's phase slightly shifted relative to another musician's phase. Alen instead measures the shifting of pulses in relation to specific drumming patterns. He finds that these might depend on body techniques related to the patterns and to spatio-motor conditions of instruments (Alen 1995: 69). To approach a project as ambitious as to search for groove in African/derived music, the relation of these at least three factors would be a major research problem.

Many of the responses to Keil (published in the same issue of Ethnomusicology) tended to question his monopolizing and isolating approach to human quality in music. I agree with this criticism. While the micro-level of African and African-derived music is still waiting to be taken seriously in scholarly discourse, it is no good to use it as an ingredient in an ideological and moralizing attack on Western musicology or music pedagogy.

One of the ideals of ethnomusicology is to consider formal aspects of musical action within different contexts, like style, performance, society or culture. Participatory Discrepancies construct a contradiction between musical syntax and other macro-structures and timing, the latter strictly being reduced to the sub-syntactic level (Keil 1995: 2). I have tried to describe a microtiming practice in a model. This practice, however, does not seem to form an aesthetically or otherwise independent domain. It rather relates to musical form as well as performance context. As for jenbe music from the Western Sudan, at least, inflection of pulsation does not represent a mystical essence of making groove together. It forms an element of style intentionally applied or left out in performance depending on both individual decisions and culturally determined conditions.

Appendix: Musical examples

All musical examples show drum patterns. Western African drum patterns are cyclical in character, often starting, changing and ending elsewhere than on a definite beat 1. In order to meet graphical needs, however, the patterns are shown as starting from a (in some cases arbitrary) beat 1 and accordingly extending to another beat 1 which would represent the starting point of the next cycle of the same pattern.

Example 1:

Example 1 refers to model 1 as shown in Figure 1 in the text. It shows a possible realization of the jenbe accompanying pattern in rhythms dansa, sanja, madan, tansole and many others. "S" represents slaps, "T" represents open tones.

Example 2:

Example 2 refers to model 2 as shown in Figure 2 in the text. It shows a possible realization of the jenbe accompanying pattern in rhythms sunun and others. "S" represents slaps, "T" represents open tones, "B" represents bass tones.

References cited

Alén, Olavo, 1995: Rhythm as duration of sounds in TumbaFrancesa, in: Ethnomusicology 39/1: 55-72.

Arom, Simha, 1984: Structuration du temps dans les musiques d'Afrique Central, in: Revue de Musicologie 70/1: 5-36.

Bailleul, Père Charles, 1996: Dictionnaire Bambara-Français. Bamako: Editions Donniya (I. ed. 1981).

Beer, Johannes, 1991: Trommelrhythmen der Malinke-Hamana/Guinea, in: Simon, A. (ed.), Rhythmen der Malinke. Berlin: Liner notes to CD 18, Museum Collection.

Bengtsson, Ingmar, 1974: Empirische Rhythmusforschung in Uppsala, in: Hamburger Jahrbuch für Musikwissenschaft 1: 195-219.

Berger, Harris M., 1997: The Practice of Perception: Multifunctionality and Time in the Musical Experience of a Heavy Metal Drummer, in: Ethnomusicology 41/3: 464-488.

Drame, Adama und Arlette, 1992: Jeliya - être griot et musicien aujourd'hui. Paris.

Harwood, Dane, 1995: Response to Keil, in: Ethnomusicology 39/1: 74-77.

Kaufmann, Robert, 1980: African Rhythm: A Reassessment, in: Ethnomusicology 24/3: 393-415.

Keil, Charles, 1995: The Theory of Participatory Discrepancies: A Progress Report, in: Ethnomusicology 39/1: 1-20.

Knight, Roderic C., 1973: Mandinka Jaliya. Ph.D. thesis, Los Angeles.

Koetting, James, 1970: The Analysis and Notation of West African Drum Ensemble Music, in: Selected Reports in Ethnomusicology 1/3: 116-146.

Kroier, Hans, 1992: Die Rumba in Matanzas (West Kuba). Magistral Thesis at Freie Universität Berlin: Unpublished manuscript.

Kubik, Gerhard, 1988: Einige Grundbegriffe und-konzepte der afrikanischen Musikforschung, in: Kubik, Zum Verstehen Afrikanischer Musik: 52-113. Leipzig.

Merriam, Allen P., 1982: African Musical Rhythm and Time Reckoning, in: Merriam, African Music in Perspective: 443-461. New York.

Pantaleoni, Hewitt, 1972: Three Principles of Timing in Anlo Drumming, in: African Music Society Journal 5/2: 50-62.

Polak, Rainer, 1996: Das Spiel der Jenbe-Trommel. Musiker, Feste und Musik in Bamako. Magistral Thesis at Universität Bayreuth: Unpublished Manuscript.

Polak, Rainer, 1997: Zeit, Bewegung und Pulsation. Theorierelevante Aspekte der Jenbe-Musik, in: Jahrbuch für musikalische Volks- und Völkerkunde 16: 59-69.

Prögler, J.A., 1995: Searching for Swing: Participatory Discrepancies in the Jazz Rhythm Section, in: Ethnomusicology 39/1: 21-54.

Stone, Ruth, 1985: In Search of Time in African Music, in: Music Theory Spectrum 7:139-148.

Waterman, Christopher A., 1995: Response to Keil, in Ethnomusicology 39/1: 92-94.

Selected Discography

Kanté, Mamadou and various artists, 1993: Les Tambours du Mali. Boulogne: CD PS 65132, Playasound.

Polak, Rainer (ed.), 1996: The Mali Tradition: The Art of Jenbe Drumming. Munich: CD BL-P 001, Bandaloop Production.

Polak, Rainer (ed.), 1997: Dònkili - Call to Dance. Festival Music from Mali. Leiden: CD PAN 2060, Pan Records.

Simon, A. (ed.), 1991: Rhythmen der Malinke. Berlin: CD 18, Museum Collection.


 [1] The present text is partly based on two preliminary ones: Polak, Rainer, 1997: Zeit, Bewegung und Pulsation. Theorierelevante Aspekte der Jenbe-Musik, in: Jahrbuch für musikalische Volks- und Völkerkunde 16: 59-69; and a paper held at the 42th Annual Conference of the Society for Ethnomusicology in Pittsburgh in October 1997 entitled "Some Relations between Micro-Timing, Meter and Performance in Jenbe Drum Ensemble Playing".

I am grateful to the Graduate Program 'Intercultural Relationships in Africa' and the 'Universitätsverein Bayreuth' for financing the trip to Pittsburgh, to Dr. Eric Charry for reading and encouragement and to Stefan Röhrich who kindly corrected my English. And I thank Madu Jakite and all the drummers of Badialan and their families for receiving me again and again with increasing respect and warmth.

Note on orthography: What I choose to spell "jenbe" has been spelled "jembe" or "djembe" or "djembé" by many others. My spelling is in accordance with dictionaries, e.g. Bailleul 1996, and the DNAFLA (Direction Nationale pour l'Alphabetisation Fondamentale en Langues autochtones) in Bamako.

  [2] Since 1991 I have been carrying out a case study among a group of professional festival drummers in Bamako. So far five periods of three to five months each have been accomplished. My research mainly consists of participant observation including apprenticeship and, in the meantime, performance practice as a jenbe player.

  [3] In Bamana, the local language of Bamako and lingua franca in Mali, there is no abstract term for the individual elements of festival music repertoire. The specific identities which represent the repertoire are referred to in three ways: either as mere proper names (e.g. menjani, suku, dansa, dununba etc.); or as names composed of occasions or groups or institutions with the suffix "-fòli" which very generally means "playing" music or "saying" parole (e.g. denbafòli, jinafòli, kòmafòli, furasifòli etc.); or as names composed with the suffix "-dòn" which means dance (e.g. wolosodòn). Talking about drum ensemble music, I refer to these identities as "rhythms". I hope that the context in each case this term is used makes clear whether I speak of rhythm in the sense of repertoire or of rhythm as general aspect of music.

  [4] E.g. Knight 1973, Beer 1991. The jenbe virtuoso Adama Drame states that "Le kenkeni [specific term for dunun, accompanying drum] de base et le djembe solo font complètement des rhythmes décalés. Il y a une espèce de paralleles rhythmes" (Drame und Senn-Borloz 1991:258; insert by the author).

  [5] For general integration of the concept into African musicology see Kauffman 1980: 396, 407-414 and Kubik 1988: 71-87. A more systematic discussion spinning around the emic/etic dichotomy is Merriam 1982: 444-461.

  [6] It is used in other Malian musical genres as well as in Nigerian dundun (personal comment on the Pittsburgh paper's draft by Prof. Kofi Agawu). Jazz, Swing (cf. Prögler 1995: 37) and sampling Hip Hop musicians also make their rhythm swing by this form of inflection. I guess it is found in various parts of the world beyond Western Africa, too: E.g. some tala (rhythm modes) as played on tabla or dholak in Pakistan and Northern India, like quaval or kharva, contain the same system of variation in duration.

  [7] There is some evidence for this generalization. Knight's (1973) examples are of Griot music (kòra playing and song) in Senegal. Drame and Senn-Borloz (1992) as well as Beer (1991) give examples of similar structures for jenbe music from Guinea, Mali, Burkina Faso and Côte d'Ivoire. Dr. Eric Charry's (personal comment) and my impressions of listening to diverse Malian musical genres and preliminary research in Bamana bala (xylophone) music point to the same direction.

  [8] The term "quantization" also was used by Kroier. He is an ethnomusicologist and composer of sampling music of high rhythmic qualities.

  [9] See appendix: musical example 5 in original article.

  [10] This is a village of about 1500 inhabitants in the Cercle of Siby. It is situated in the so-called Monts Mandingues about 80 km to the south-west of Bamako towards the Guinean border.

  [11] One of these rhythms is "soron" (elsewhere called "sunun") from Kaarta, a region between Kita and Nioro where Kagòòrò-Bamana formed a kingdom in rivalry to Segu in the period from the 17th to the 19th century. Another is "sabar" of the Wolof and Mandinka from Senegal, which originally is not played on jenbe but on the drum ensemble of the same name. Another interregional rhythm practiced in the Mande Mountains is "kisa" which is related to "kassa" from Northern Guinea (Simon 1991) and to "dansa" from Xhasso (and since the 1960s: Bamako), a region between Bafulabe and Kayes near the Senegalese border in Western Mali.

  [12] These are shown in the appendix: example 1 and  example 3 in original article.

  [13] These are mainly marriages and, to a lesser extent, circumcisions, spirit possession cult initiations, baptisms, engagements and singular political or public events.

  [14] This is called (fr.) "échauffement" in the international percussionists' jargon.

  [15] When Bamako drummers compare their style with Guinean jenbe playing, they put it clearly: "You do not play solo [French/international musician's jargon in the original] in our places".

  [16] This is an interesting process of folklore feed-back as both of them represent institutionalized, arranged and transformed versions of the festival performance of the streets.

  [17] This Guinean born drummer formerly had been jenbe player in Fodeba Keita's legendary Ballets Africaines.

  [18] For more detailed descriptions and full transcriptions of this CD see Polak 1996.

Send a comment to the author, Rainer Polak <>