DEFINITION AND GENERAL CHARACTERISTICS

Definition

By polyphonic instruments, we mean instruments designed for melodic music, i.e., provided with pitch levels from a predetermined scale, and played in such a way that the musician's hands simultaneously perform two parts which differ in both melody and rhythm.

General characteristics

The three polyphonic instruments found in the Central African Republic are the xylophone, the sanza, and the harp. They are only rarely used independently, their primary function being to provide support for a song which is invariably sung by the musician himself, but may also involve a response from a choir, or simply from a partner. This instrumental support does several things. It defines the periodic and modal framework for the song, and in so doing, furnishes a set of metric, rhythmic, and melodic reference points. The regularity of the invariant periodicity and the stability of the reference points help the musician to develop a nearly automatic motor behaviour, which frees his mind from attention to manual activity. When he has reached this stage, he can devote his attention entirely to melodic and verbal improvisation in his song. Far from confining itself to an organisational function, the instrumental formula provides a sort of ‘launching pad’ for the musician's inspiration by suggesting melodic associations.

Functionally speaking, the xylophone may be set apart from the harp and the sanza, as it is used primarily to accompany group dances. Accordingly, it is habitually a part of larger ensembles, including percussion instruments (particularly drums). The harp and the sanza, however, are used on more intimate occasions, althouth this is not incompatible with the adjunction of one or more percussion instruments, such as double bells or rhythm sticks.

WHAT IS RELEVANCE?

Having developed a method and shown it to be well-founded, we now have the technique we need to transcribe and analyse polyphony. We can record unwritten polyphonic music part by part, and our recordings enable us to see how the parts combine and interlock. They thus provide the raw material for further research; however, analysis as such cannot begin without a transcription.

We now come to a fundamental methodological question. Do we want a rational and meaningful transcription based on the way the people who use the music perceive and understand it, i.e., do we really want to follow through with our stated purpose of setting aside whatever Western notions we might be tempted to project onto African music? If so, before we actually start to transcribe, we must develop another device to enable us to determine what we can and ought to extract from our recorded material. Referring only to criteria present in the musical tradition under study, the investigator should be able to decide which data in the raw material are meaningful, and which are not. We therefore require a device like a sequence of filters, which can separate out whatever is meaningful. This device must operate on the basis of the notion of relevance.

The musicologist has, in fact, a choice of two alternatives. He may remain as close to the recorded material as he can and try to note every detail with maximal accuracy. This approach, while apparently convincing from the acoustic standpoint, nevertheless suffers from the fact that there is no limit of ultimate accuracy, as transcriptions like Bartók's have shown. The other possibility would be to try to determine beforehand what the members of the community consider to be meaningful in their own music.

INTRODUCTION

Our study of African rhythm and discovery of what we believe to be its basic features suggest there is a common denominator which can provide a link between African rhythm and the principles underlying the kind of rhythm practiced in the West. A critical examination of Western terminology has revealed a number of inherent contradictions but has also shown where they come from. We have seen that they are connected with the development of the idea of rhythm in the West. A short history of this development has helped us to understand this more clearly. We have found that, in Emmanuel's words, ‘the development of rhythmics is related to the development of language’(1926: III).

We have seen that the Indoeuropean languages which gave rise to Vedic poetry and later to Greek poetry had pitch accents and no stress. Their versification, the ‘verbal rhythmics’ which ‘stylises the rhythm of language’ (Emmanuel 1926: 106) and is therefore an intermediate stage between the unmeasured flow of speech and strictly proportional musical rhythm, was based only on the quantitative contrast of long and short syllables and not on dynamic intensity.

We should here recall that most African languages are tone languages and also have no stress. In such languages (cf. Book I, 2.2.5), the pitch of a syllable is a distinctive feature; the same syllable spoken on different pitches can change the meaning of the word containing it. Tones are to these languages as pitch accent was to Indoeuropean languages. In either one, stress may have an expressive role, but is unrelated to phonology, i.e., does not distinguish the meaning of words.

The analysis of any cultural phenomenon is reductive by definition. This follows from the fact that, in the terms of Gilles-Gaston Granger (1967:2), ‘any attempt to understand any facet of man must start with a reduction of our experience to a set of correlative marks’. On this principle, if we want an adequate description of how Central African polyphonies operate as systems, we must find a suitable set of correlative marks, and say why we consider them to be basic. A further need will be to examine the various facets and levels in our data where we find these marks at work.

The first step in our analysis will be to reduce this music to writing by transcription. This involves a set of preliminary operations enabling pitch (degrees on a scale) to be discriminated, and temporal organisation (metrics and rhythmics) to be grasped. It will therefore also be necessary to give a principled explanation of the scores appearing in the remainder of this volume.

The main purpose of the present chapter is to provide an explicit statement, and in so doing to justify the choice, of the criteria we apply to the analysis of Central African types of polyphony and polyrhythmics. Two important questions must, however, be dealt with beforehand. The first concerns all types of Central African music, while the second relates only to music of the polyphonic type.

Can the term mode be applied to the arrangement of strictly pentatonic melodic material?

Can the notion of harmony be applied to the vertical structures engendered by this same material?

The importance of the answers we provide to these questions for our choice of analytical criteria and our results is immediately evident.

The method described can be used to decompose polyphony into a coherent set of elementary monodies. It is therefore the first method to be devised which is capable of providing students of musicology, acoustics, and organology with a recording of the individual parts in any piece. It thereby makes it possible to conduct laboratory studies which were previously restricted to monodic music: pitch analysis, decomposition of the frequency spectrum by real-time analysis, sonographic studies of harmonics and overtones, and melographic recordings of melodic curves. It also constitutes a new tool for ethnomusicological research into the musical psychology of peoples with orally transmitted polyphonic traditions, particularly, into how far the musicians consciously develop their polyphony and what individual limits are placed on their musical ‘productivity’. A few of these possibilities will now be illustrated by specific examples.

We wanted to find out to what extent each of the singers in a group of Pygmies was conscious of the polyphonic structure of a specific piece, i.e., of its metric, rhythmic, modal, and vertical, i.e., ‘harmonic’, organisation. After recording the first singer in the traditional order of entry, we asked each of the remaining performers to develop a counterpoint to the melody of this initial ‘cantus firmus’. This was done by about ten people. We found that, despite (or perhaps because of) the experimental conditions, all were perfectly clear about how the metre, rhythm, mode, and vertical organisation of their parts should be directly related to the ‘cantus firmus’.

Conversely, it is impossible, by merely listening, to determine how Pygmy vocal polyphony is organised, despite the obviously coherent relationships among the undeterminable number of voices.

Conducting research in this way under experimental conditions calls for extreme caution. That is why checks and controls must be included in our procedure and applied throughout the experimental process. Brief reference has already been made to one kind of check, namely, the fact that headsets are provided to enable various people to monitor the recording operation: the musicians, the head musician (where the function of leader exists), the masters, and the elders, all of whom have an intimate knowledge of the music from having been involved in it. They listen through their headphones to all the individual performers being recorded. The investigator must maintain constant visual contact with them so that, if they object to any musical event, he will see their opinion reflected with such speed and clarity on their faces that he will immediately know there has been a mistake. If the investigator himself thinks he has heard something wrong during the performance, he can direct a questioning glance at them, and they will either confirm his impression or let him know that everything is as it should be. Clearly, this constant questioning of the accuracy of the output in no way relies on the investigator's theoretical assumptions, but is based entirely on the proper of improper performance of musical material within a given group, i.e., on culturally relevant criteria.

There is a second sort of checking procedure built into the method. If a musician makes a mistake which gets by even the kind of monitoring described above, yet is serious enough to affect the coherent musical organisation of the piece, the next musician to record, who uses the erroneous part as a reference, will obviously be bothered by it in his own performance.

We are now about to describe a new method which has enabled us to obtain the results presented in this work. It came, not as a brainstorm, but as a result of the kind of long and sometimes tortuous processes, as is often the case. The reader may find it helpful if we outline the main stages in the process which led to its development; we therefore hope he or she will excuse the personal nature of some aspects of the following presentation.

In 1967, we were working on the monodic song-tales of the Ngbaka people in the Central African Republic. These pieces have no regular accentuation and are performed a cappella, i.e., they contain no materialised metric or rhythmic point of reference. We were faced with the problem of finding out how to determine the periodicity of these songs, which meant discovering the key to their temporal organisation. It occurred to us that an answer to this question might be found if we first recorded the song and then asked the singer to record the same piece again, this time clapping his hands. We naturally avoided telling him exactly how this should be done. The Ngbaka performer would then straightforwardly provide us with the basic pulsation for the piece involved. We hoped to obtain a new recording in this way, containing enough metric reference points to enable us to transcribe the durations within the song. The singer's handclaps must be intimately connected with the music, and could therefore leave no doubt as to their accuracy, and consequently, their validity as reference points.

When I got off the plane in Bangui, the capital of the Central African Republic, on 30 November 1963, it was the first time I had ever set foot on the African continent. I had no idea that this date was to mark a turning point in my career.

The country was celebrating the anniversary of its independence on the following day, 1 December. Groups from several provinces were to put on shows of traditional music during the festivities. Within twenty-four hours, I was thus able to make myself acquainted with a variety of the types and techniques of the country's music.

At the time I was a performing musician and played the French horn in a symphony orchestra. For me, as for many Europeans in the 1960s, there was no ‘real’ music outside Western music. I was conditioned by a cultural background too narrow to allow easy comprehension of anything as different and as disconcerting (in the etymological sense of the word) as African music. I thus found myself suddenly plunged into a universe of sound which was as strange to me as it was unexpected: there were groups of percussion instruments playing tightly interlocked rhythms, orchestras of up to twenty wind instruments in which each musician would play only a single note in an extremely strict and precise polyphonic latticework, choirs singing incredibly complex and yet perfectly coherent contrapuntal music, and many other surprises. Everything was performed from memory with no conductor to provide coordination. The extent to which the performers were apparently unconstrained was equally bewildering.

The result of this first encounter was ‘love at first sight’.

In autumn 1982 a former student of mine, the Puerto Rican composer Roberto Sierra, brought to my attention a collection of instrumental and vocal ensemble music of the Banda-Linda tribe from the Central African Republic, recorded by Simha Arom. The record ‘Banda Polyphonies’, then several years old, was no longer available so I re-recorded it on to a cassette and made a photocopy of Arom's introductory text. Having never before heard anything quite like it, I listened to it repeatedly and was then, as I still am, deeply impressed by this marvellous polyphonic, polyrhythmic music with its astonishing complexity.

For many years I have been fascinated by the musical epoch from Vitry and Machault to Ciconia, and since my acquaintance with his work in 1980, by the music of Conlon Nancarrow. Undoubtedly my interest in the music Arom has recorded stems also from the proximity I feel exists between it and my own way of thinking with regards to composition: that is, the creation of structures which are both remarkably simple and highly complex. The formal simplicity of sub-Saharan African music with its unchanging repetition of periods of equal length, like the uniform pearls of a necklace, is in sharp contrast to the inner structure of these periods which, because of simultaneous superpositioning of different rhythmic patterns, possesses an extraordinary degree of complexity. Gradually, through repeated listening, I became aware of this music's paradoxical nature: the patterns performed by the individual musicians are quite different from those which result from their combination.

INTRODUCTION

Of the principles that may be said to underlie Central African polyphony, some are current in a much wider geographical area, extending from Ethiopia to South Africa.

That is why I have chosen in this summary of previous studies, rather than restricting it to Central African polyphony (which has hitherto been little studied), to devote it to all of sub-Saharan Africa, so as to compare some of the techniques observed or studied elsewhere, both in earlier times and today, with analogous procedures now flourishing in Central Africa. As the reader will see, the documentation concerning this country is very sparse; indeed, out of the hundred or so accounts that I have succeeded in locating, only seven touch upon Central African polyphony. Let me emphasise, nevertheless, that this summary is not exhaustive but is a presentation of the most significant texts that throw light on the subject.

The information available is scattered across some six hundred years. The very first mention of an African musical instrument dates from the fourteenth century, whereas the first description of a performance that would seem to have been polyphonic in character dates from 1497. These descriptions derive from a wide range of sources. They include travellers' tales, as well as the accounts of missionaries, geographers, explorers, ethnographers avant la lettre, members of the colonial administration, stationed in various countries. These witnesses, whether they had any musical knowledge or not, always had an opinion about the music that they happened to hear, and they always gave their aesthetic judgement about it.

What all these witnesses have in common is their attempt to give an account, with varying degrees of astuteness or competence, of performances they had attended, and things they had seen and heard.

INTRODUCTION

Although each human group has its own musical idiom, the musics of the peoples living in the Central African Republic have many common characteristics. The general principles upon which these different musics are based show a certain degree of uniformity. I shall summarise these general principles as a way of introducing an analysis of the polyphonic and polyrhythmic musics which are the subject of this study.

But before briefly surveying the general features of traditional Central African music, let me remind the reader that, as in most African societies, this music is popular, oral, anonymous, undatable, collective, and for internal use.

1. – Popular music as opposed to ‘art’ music because it does not require knowledge of any formal theory on the part of those who inherit it and perform it. The African musician does not verbalise the abstract principles upon which his music is based.

2. – Oral music, because it has no notation and is transmitted from generation to generation by word of mouth and memorisation only.

3. – Anonymous and undatable music: its possessors cannot say either who composed the pieces they have inherited, nor when it was first performed.Collective music, because the whole community, inasmuch as it is a cultural entity, is responsible for its preservation and subsequent transmission.

4. – Music for internal use, because, produced in the society in question, by and for its own members, it constitutes a means of expression and communication – even of communion – that is peculiar to the society, just as much as its natural language is.

VOCAL AND INSTRUMENTAL MUSIC

The Central African peoples, like so many others in Black Africa, do not generally distinguish vocal from instrumental music.

The method of polyphonic analysis described above is experimental. The question must now be raised whether the use of this kind of experimentation in ethnomusicology is acceptable. Before proceeding, we may remark that the procedure receives initial support from the fact that a number of experiments involving the same object have yielded, perhaps not identical, but at least extremely similar and equally revealing results. Paradoxically, the theoretical assumptions of the method are confirmed in its application. We may summarise these assumptions as follows: each part in polyphonic music is itself a coherent entity; these parts also fit together in a coherent way; and the way they fit together two by two remains unchanged, however many times we record them. We furthermore assume that the combination of all the parts is equivalent to the set of the individual parts and to the set of all possible combinations of these parts. In applying the method, we can confirm each and every one of these hypotheses.

The procedure consists essentially of obtaining a diachronic reproduction of a synchronic event. The fundamental principle of the successive entry of voices or instruments, which characterises the polyphony and polyrhythmics of this part of Africa, is nevertheless respected, even while the music is being ‘disassembled’. In some cases, entries take place in indeterminate order; in others, they are strictly ordered in connection with the structure of the piece, and can be indicated by the musicians. The musicologist has no part in this breakdown, and is a mere spectator to the musical performance. The method follows this order step by step. The procedure has the further advantage of being applicable in the field, i.e., in the musicians' own cultural setting.

In our discussion of earlier works on the subject, we noted that little attention has been devoted to African polyphony. Unlike many other aspects of music transmitted by oral tradition, reference transcriptions of African polyphony are rarely available, and when provided are only fragmentary at best. This seems to be mainly attributable to the difficulty of transcribing this kind of music. Several experts, among them Bruno Nettl (1964:84), have pointed out the problems involved:

Gerhard Kubik (1961: 200) is equally sceptical about transcribing African instrumental music: ‘Transcription of African instrumental music from tape recordings poses enormous problems. In most cases, it will lead to incorrect results.’ In a subsequent publication on xylophone music (1964a), Kubik reaffirms that transcription from tape recordings is impossible. He suggests instead that an attempt be made to learn to play an African instrument before transcription is undertaken. We will return below to the effectiveness of this approach.

INTRODUCTION

This Book deals with the temporal structure of African music, particularly African polyphony. All polyphonic music requires a temporal reference unit to provide a common denominator for its parts. We will therefore be discussing measured music, i.e., music comprised of durations with proportional values.

Let us recall that the distinction between measured and unmeasured music has a long history. It existed in the cultured music of the ancient Greeks, and in medieval musical theory, in the contrast between the cantus mensuratus (measured chant) and cantus planus (plain chant). Closer to our own times, it can be found in classical opera, where arias or measured pieces alternate with unmeasured recitativo secco.

Measured music, sometimes referred to by the Italian expression, tempo giusto, is thus defined by contrast with unmeasured music. The latter is not governed by fixed quantities, i.e., the values of durations are not strictly proportional. In unmeasured music, a note is only meaningful in its position with respect to preceding and following notes, i.e., in a melodic pattern. In measured music, however, all durations are strictly proportional.

According to Rousseau's (1768: 283) dictionary, the term measured ‘corresponds to the Italian a tempo or a batuta’. Proportional durations must be based on a reference unit. Different epochs and civilisations have had different ideas of such a unit. Ancient Greek music was based on a chronos protōs, the smallest indivisible unit of duration, whose multiples formed the foot and the metre. In the West, this unit is the beat provided by the conductor, whence the term batuta in Rousseau's definition of measured.

In recent years a number of studies have been published of early European and Asian notation for traditions which were primarily or partially oral. This paper examines the relationship between notation and performance in a primarily oral tradition that continues to the present, namely that of the shakuhachi honkyoku tradition. In this tradition, the relationship between score and sound has been determined largely within the context of oral transmission, which remains dominant within the honkyoku tradition despite the existence of notation for at least a century. The ways in which orality influences the relationship between notation and performance will be illustrated by examining in detail the honkyoku Kokū, a piece considered representative of the tradition. Certain observations made in the course of this study have clear implications for the study of early records of primarily orally transmitted music.

The shakuhachi, Japan's end-blown bamboo flute, is believed to have been first introduced into Japan from China in the eighth century as part of the ensemble of the Chinese court music, gagaku. By the eleventh century it was no longer used in gagaku performances.

In my studies of the use of the Japanese shakuhachi in the Fuke sect of Zen Buddhism, I have been continually puzzled by the tradition which traces the playing of shakuhachi as a religious practice back to the T'ang period Chinese Zen (Chinese: Ch'an) figure P'u-hua (read Fuke in Japanese). This tradition, recorded in the middle Edo period source Kyotaku Denki but certainly much older, does not attempt to ascribe to Fuke either the playing of shakuhachi or any knowledge of it whatsoever. Indeed Fuke's only musical connection was his ringing of a bell as he wandered the streets, the sound of which is said to have been captured after his death by a shakuhachi-playing disciple, Chang Po, and turned into the first piece of the Fuke repertoire, Kyotaku. It is now generally accepted that this history was largely fabricated in order to legitimize the Fuke sect, by establishing an elaborate genealogy leading back to the very source of the Zen tradition in Tang period China and to a figure (Fuke) traditionally associated with Rinzai Gigen (Chinese: Lin Chi I-hsüan), one of the great figures of early Zen.

Dunhuang, located in the western part of China's Gansu Province, was an ancient holy seat of Buddhism and a strategic place on the Silk Road, the ancient trade route along which eastern and western culture was transmitted. Approximately forty kilometres southeast at Dunhuang is a steep cliff with an average height of seventeen metres where, from the beginning of the fourth century, people dug Buddhist caves (mogao ku). A stone inscription made by Li Huairang in 698 states that at that time there were already more than one thousand caves, and for this reason, the Mogao caves have also come to be known as the ‘Thousand Buddha caves’ (qianfodong).

At the beginning of this century, the caretaker of the Mogao caves, a Daoist monk called Wang, unexpectedly discovered another cave through a crack which had opened up. In this cave, which had long been sealed off, were preserved many Buddhist scriptures and other documents, the latest of which date from 1002.