Oral Literature in Africa (15 page)

A few additional features should be mentioned briefly. One is the sound system. There is the ‘dominantly vocalic quality of the Bantu sound-system,
the absence of neutral and indeterminate vowels, and the general avoidance of consonant-combinations’ (Lestrade 1937: 302) Bantu languages differ in the use they make of this system. Shona, for instance, is definitely staccato, Swahili to some extent so, and the tonal systems also vary. Some use a regular long syllable, as in the Nguni languages of the south (including Zulu and Xhosa):

Strongly-marked dynamic stresses, occurring in more or less regular positions in all words of the same language, and the fairly regular incidence of long syllables also usually in the same positions, give to Bantu utterance a rhythmic quality and a measured and balanced flow not met with in languages with irregular stresses and more staccato delivery

(Lestrade 1937: 303).

The particular genius of each language gives rise to various possibilities in the structure of verse. The type of ‘prosody’ often used exploits the grammatical and syntactical possibilities of the language, which is not, as in English, bound by a fixed word order. Alliterative parallelism is easily achieved. Thus in the Zulu proverb

Kuhlwile / phambili // kusile / emuva

It is dark / in front // it is light / behind (‘it is easy to be wise after the event’).
¹⁴

(quoted Lestrade 1937: 307)

there is perfect parallelism, idea contrasting to idea in corresponding position, identical parts of speech paralleling each other (verb for verb, and adverb for adverb), and, finally, number of syllables and dynamic stress exactly matching each other. Similar effects are produced by ‘cross-parallelism’ (chiasmus) where the correspondence is to be found crosswise and not directly, and by ‘linking’, the repetition of a prominent word or phrase in a previous line in the first half of the next one. The kind of balance may even extend to correspondence in intonation and, though very different from more familiar ‘metrical’ forms, is felt to provide perfect balance and rhythm by native speakers of the language (Lestrade 1937: 307–8).

To these linguistic resources on which the Bantu speaker can draw we must also add the whole literary tradition that lies behind his speech. There is, for one thing, the interest in oratory and in the potentialities of the language which is typical of many Bantu peoples. ‘They have the germ of literary criticism in their very blood’, writes Doke, and discussions of words and idioms, and plays on tone, word, and syllable length ‘all provide hours of entertainment around the hearth or camp fire in Central Africa’ (Doke 1948: 284). There is the rich fund of proverbs so often used to
ornament both everyday and literary expression with their figurative and elliptical forms. There are the ‘praise names’ that occur so commonly in Bantu languages, forming on the one hand part of the figurative resources of Bantu vocabulary and word-building, and on the other a form of literary expression in its own right, often elaborated in the south into full praise verses of complex praise poetry (see Ch. 5). To all these literary resources we must, finally, add the formal genres of Bantu literature—prose narration, proverb, riddle, song, praise poetry. In each of these the artist can choose to express himself, drawing both on the resources of the language and on the set forms and styles placed at his disposal

From the artless discursiveness and unaffected imagery of the folktales to the stark economy of phrasing and the elaborate figures of speech in the ritual chants, from the transparent simplicity and highly-charged emotion of the dramatic songs to the crabbed allusiveness and sophisticated calm of the proverb, and from the quiet humour and modest didacticism of the riddle to the high seriousness and ambitious rhetorical flight of the praise-poem

(Lestrade 1937: 305).

I have written at some length about the basis for oral literature in the single Bantu group in order to illustrate from one well-documented example the kind of resources which may be available in an African language. Other languages and language groups in Africa have other potentialities—some in common with Bantu, some very different—but a similar kind of analysis could no doubt be made in each case. There is no reason, in short, to accept the once common supposition that African languages, unlike those of Europe, could provide only an inadequate vehicle for the development of literature. This point is made here in general terms and will not be repeated constantly later, but it is necessary, in the case of each analysis of a single literary form, to remember the kind of literary and linguistic resources that, though unmentioned, are likely to lie behind it.
¹⁵

III

It is necessary to examine briefly the general relevance of certain other elements, particularly tone, metre and other prosodic forms, and music.

The significance of tone in literary forms has been most fully explored in West African languages, though it is not confined to them. In these
languages tone is significant for grammatical form and for lexical meaning. In, for instance, Yoruba, Ibo, or Ewe, the meaning of words with exactly the same phonetic form in other respects may be completely different according to the tone used—it becomes a different word in fact. The tense of a verb, case of a noun, even the difference between affirmative and negative can also sometimes depend on tonal differentiation. Altogether, tone is something of which speakers of such languages are very aware, and it has even been said of Yoruba that a native speaker finds it easier to understand someone who gets the sounds wrong than someone speaking with incorrect tones.

This awareness of tone can be exploited to aesthetic effect. Not only is there the potential appreciation of unformalized tonal patterns and the interplay of the tones of speech and of music in sung verse (see Ch. 9), but tones also form the foundation of the special literary form in which words are transmitted through drums (Ch. 17). In addition tone is apparently sometimes used as a formal element in the structure of certain types of orally delivered art. In Yoruba not only do tonal associations play a part in conveying overtones and adding to the effectiveness of literary expression, but the tonal patterning is also part of the formal structure of a poem. One light poem, for instance, is based on the tonal pattern of high, mid, low, mid, with its reduplicate of low, low, mid, low:

Jo bata—bata o gb’ona abata (2)

Ojo bata—bata (2)

Opa b’o ti mo jo lailai
.

Dancing with irregular steps you are heading for the marsh (2),

If you will always dance with those irregular steps,

you [? will] never be a good dancer.

(Lasebikan 1956: 48)

Another, in more serious vein, gives a vivid description of a great battle, adding a note of authenticity with the author’s claim to have been an eye-witness:

Ija kan, ija kan ti nwon ja l’Ofa nko—

Oju tal’o to die mbe ?

Gbogbo igi t’o s’oju e l’o wo’we,

Gbogbo ikan t’ó s’oju e l’o w’ewu ęję

Ogọro agbọnrin t’ò s’oju e l’o hu’wo l’oju ode;

Sugbọn o s’oju mi pa kete n’ile we nibi nwọn bi mi l’omo;

Agba ni ng o ti da, mo kuro l’omode agbekorun r’oko
.

What about a great fight that was fought at Ofa—

Is there anyone here who witnessed a bit of it?

Although the trees that saw it here all shed their leaves,

And the shrubs that saw it were all steeped with blood, And the very stags

that saw it grew fresh horns while the hunters looked on,

Yet I saw every bit of it, for it was fought where I was born.

I do not claim to be old, but I’m no more a child that must be carried to the farm

(Lasebikan 1955: 35–6)

Lasebikan comments on the tonal structure of this poem. It falls into four distinct divisions:

How are these divisions marked out ? Not by means of rhymes as in English poems, but by the tone of the last syllable of the division’. He shows how the actual words used are carefully chosen to fit this tonal structure, for possible alternatives with the same meaning and syllable number have tonal compositions that ‘would spoil the cadence of the poem’

(Lasebikan 1955: 36).

A similar but more detailed analysis has been made by Babalola of the way tonal patterning is a characteristic feature of the structure of Yoruba hunting poetry (
ijala
)
. The musical and rhythmic effect of this poetry arises partly from tonal assonance—specific short patterns of syllabic tones repeated at irregular intervals—or, alternatively, from tonal contrast which ‘seems to … increase the richness of the music of the ijala lines by adding to the element of variety in successive rhythm-segments’ (Babalola 1965: 64–5, 1966 (Appendix A
passim
)).

In other forms and areas too we sometimes see tonal correspondence. There is sometimes tonal parallelism between question and answer in ‘tone riddles’ or within the balanced phrases of some proverbs (see e.g. Simmons 1958 (Efik); Van Avermaet 1955 (Luba)). The use of tone correspondence in some poetry is so striking as to have been called a species of ‘rhyme’ (e.g. the ‘tonal rhyme’ of Efik, Ganda, and possibly Luba poetry (Simmons 1960a; Morris: 1964: 39; Van Avermaet 1955: 5; Stappers 1952
a
).

Some of the detailed analyses of the significance of tone in literature are controversial, and little enough work has as yet been done on this formal aspect. But as linguists increasingly stress the general importance of tone in African languages throughout the continent, so we can expect many more studies of this aspect of literature.

Ideophones and other forms of sound association are so important in non-Bantu as well as Bantu languages that they are worth mentioning again at this point. Thus there are the important sound associations in Yoruba which connect, for instance, a high toned nasal vowel with smallness, or low toned plosives with huge size, unwieldiness, or slow movement, often intensified by reduplication (Lasebikan 1956: 44) the connection in Ewe and Gbeya with the vowel /i/ and a lateral resonant consonant in ideophones for ‘sweet’,
or the common use of back rounded vowels for ideophones indicating ‘dark, dim, obscure, foggy’, etc (Samarin 1965
a
: 120). In many cases ideophones constitute a high proportion of the lexical resources of the language. In Gbeya, in the Central African Republic, about 1,500 ideophones have been recorded. A number of African languages are said to have twenty, thirty, or even forty ideophones just to describe different kinds of ‘walking’ (Samarin 1965: 117, 118). Indeed it has been suggested by one authority that ‘ideophones by count constitute, next to nouns and verbs, a major part of the total lexicon of African languages’ (Samarin 1965: 121).

Many different instances of these graphic ideophones could be added to the earlier (Bantu) examples we have cited. We have, for instance, the Yoruba representation of water draining out drop by drop (
to to to
)
,
a lady in high heels (
ko ko ko ko, ko
)
,
or a stalwart in heavy boots (
ko ko ko
) (Lasebikan 1956: 44). The Zande ideophones
digbidigbi
(boggy, oozy, slushy),
guzuguzu
(fragrantly, sweet smelling),
gangbugangbu
(listlessly),
degeredegere
(swaggeringly), and
gbaraga-gbaraga
(quarrelsome-ly).
¹⁶
The effectiveness of such forms for descriptive and dramatic formulations and the poetic quality they can add to ordinary language need not be re-emphasized except to say that this point constantly reappears in detailed analyses of African oral art.

Then there is the topic of prosodic systems in African verse. This is a difficult question which, apart from one classic article by Greenberg (largely devoted to Arabic influence) (Greenberg 1960; also 1947; 1949), has received relatively little attention. It is clear that many different forms are possible which can only be followed up in detailed accounts of particular literary genres. Broadly, however, one can say that six main factors can be involved: rhyme, alliteration, syllable count, quantity, stress, and tone (here and throughout this section I draw mainly on Greenberg 1960). Of these, the last three can only be used when there is a suitable linguistic basis. Tone, for instance, can only perform a prosodic function in tonal languages; and this influence of language on form can be seen in the way Hausa has been able to keep to the quantitative form of its Arabic verse models, whereas in Swahili, equally or more dominated by the Arabic tradition, the nature of the language has precluded the use of quantitative metres and instead turned interest to rhyme and syllable count (Greenberg 1947; 1949; 1960: 927, 935).

The Arabic influences on African prosody have been summarized by Greenberg and need not be repeated in detail here. In spite of various difficulties and uncertainties the Arabic-based styles are somewhat easier to analyse than some-other forms of African poetry, partly because, for one familiar with the Arabic forms, the parallel African ones are ultimately recognizable partly because they appear in local
written
forms in which the evidence tends to be somewhat more plentiful and accurate than for oral African poetry. Classical Arabic prosody is a source of verse forms in several African languages, particularly the
qasidah
(ode) based on quantity and rhyme. This occurs, for example, in learned Fulani poetry, and in learned (and sometimes in popular and oral) Hausa poetry; in both cases there is the retention of quantitative features made feasible by the forms of these two languages. The post-classical
tasmit
has also influenced African forms. Again it uses quantity and rhyme, but with more stress on rhyme. It occurs in several West African languages (Hausa, Kanuri), but has reached its highest development in Swahili where ‘it is by far the most common form in both learned and popular poetry, whether sung or recited’ (Greenberg 1960: 935). Since Swahili does not possess vowel quantitative distinctions, this principle has been replaced by that of syllable count accompanied by rhyme. Its most popular form is the four-line stanza (each line containing eight syllables), with the basic rhyme scheme aaab/cccb/dddb/…, but five-line stanzas (
takhmis
) also sometimes occur (Greenberg 1960: 934–6, 1947; also Harries 1962: 9 ff., Hichens 1962–3).