Man Vol. 17, Number 3, pgs 546-548

Man Vol. 17, Number 3, pgs 546-548 - September 1982

The authority of ancestors

1. Kopytoff's criticism of Calhoun (Man (N.S.) 16, 125-6), in which he also criticises my article (Africa 43, 122-33) challenging his original assertion, conveniently ignores one of my most important points and one which should be of interest to him in connexion with semantic accuracy. Kopytoff's argument with which I took issue was:
Insistence on the conceptual primacy of this division between the living and the dead ... is an ethnocentric distortion of the African world view, a distortion that prevents our understanding of what we have persisted in calling 'ancestor cults' and 'ancestor worship' (1971: 136).
And further on his insistence that because Africans do not semantically differentiate dead ancestors and living elders, they do not perceive a difference between the two categories.

As I noted, it is quite true that there may be no word that stands precisely for 'ancestor' and therefore one finds the use of the same words commonly accepted to stand for, gloss if you will, 'elders', 'grandfathers', 'great ones' (though it is noteworthy that while all these terms may be used to describe ancestors in a general way, the terms themselves are not interchangeable among the living persons in these categories). We often do something comparable when we refer to our 'forefathers'

or 'forebears'. The real point, though, is that the word used to translate 'ancestral spirt' (rather than 'ancestor'), which in Bantu languages is most frequently a variant on the root -zim-, is assigned to a noun class different from that used for living persons or animals. Kopytoff attempts to hoist me with my own petard when he claims that I admit to the several possible meanings of this term, but the fact remains that while living being classes can be and are used to gloss the words we translate as 'elder' etc. to render 'ancestor', the reverse is not true. An elder among the Luguru may possess an mtsimu (pl. mitsimu); he cannot be one unless he is dead. As I also noted before, it is significant that in Swahili the only other words used to describe things which are assigned to this class (commonly used for trees, plants, rivers, mountains, the moon) are those we gloss (I would say 'translate') as god, mungu, pl. miungu and prophet, mtume, pl. mitume.

Kopytoff also chooses to disregard my other point: that in Ndebele the word used for ancestral spirit has the root -loz- (-loy- -log- and -loj- are cognates and connote witchcraft/sorcery). From this I developed the hypothesis I have discussed widely with African colleagues and with which they concur: namely, that there exists a spiritual power which can be tapped licitly or illicitly. The elder can tap it by way of the ancestral spirits to uphold his authority. The witch can tap it by his/her evil powers. But the source of the spiritual power tapped is the same.

As I made clear in my original article, I found myself in general agreement with much of what Kopytoff said, and am grateful to him for giving me the stimulus to carry out further research.

J.L. Brain
State University of New York
New Paltz
Brain, James L. 1973. Ancestors as elders in Africa: further thoughts. Africa 43, 122-33.
Kopytoff, Igor 1971. Ancestors as elders in Africa. Africa 41, 129-42.

Followers of the controversy that Igor Kopytoff's paper 'Ancestors as elders in Africa' has stimulated may be forgiven a certain bemusement, for whilst Kopytoff is claiming a different methodology ('cultural semantics') it is not clear that he is saying anything positively different from his adversaries. He is of course making a number of negative, polemical points, largely to do with the dangers of false conceptualisation, but these points (whose problematical substantiation has been demonstrated in recent correspondence) do not clear the way to any new interpretation of the role of ancestors, or of elders. I wonder whether the trouble does not lie with the view, which all

parties seem to have accepted (and which is reflected in the heading to this correspondence), that what most characterises African ancestors, or elders, is their authority.

Let us return to the common Bantu terms that Kopytoff (I think very profitably) focuses upon: the nominal or adjectival stem -kulu (Guthrie's CSS*-kúdù) that can be associated with the verbal radical -kula (Guthrie's *-kúd-). The latter is general to a very large number of Bantu languages and has the virtually constant meaning 'to grow up', 'to grow (instransive)', 'to mature'. Where we have detailed accounts of the contexts of its use (e.g. Bemba or Ndembu) it is clear that when the word is used of persons the growth so denoted is both physical and social, and that it does not happen all at once but is gradual and relative (cf Turner 1968: 83; Richards 1956: 121 sqq. and Appendix A). The nominal and adjectival form, -kulu, often carries the sense of 'adult', 'grown-up' or 'mature' but it also appears frequently in kinship terms and other words denoting relative status, meaning 'elder', 'senior' or simply 'big' in the sense of socially important. Kopytoff's own suggested gloss is the French grand- an imaginative translation since this term did in fact come into our own kinship terminology to distinguish very much the same categories of kin (grandparents, grandchildren) that are distinguished by the Bantu -kulu (e.g. Bemba shikulu, grandfather, mwishikulu, grandchild). Returning, however, to the Bantu word, the point to make is that -kulu, even when its sense is simplest and most matter of fact (as in the Swazi 'great hut' of a homestead, indlunkulu, or Luganda mukulu meaning 'headman' or 'boss'), its reference is still intrinsically social and relative. Hence its widespread use in kinship terms - a feature that Kopytoff largely ignores. It is used to distinguish older from younger siblings (as in Ganda, Luba, Kikuyu, Kaguru) but even more commonly it distinguishes generations - the senior or 'grown' parents from those who are more immediately parents, the extended or 'grown' children from those who are more immediately a person's children. The former distinction, that is -kulu in grand-parental terms, is variable and exists alongside other terms for grandparents (e.g. the Swazi alternatives of babe'mkulu or gogo for grandfather) but the word for grandchild in eastern and southern Bantu languages is extraordinarily constant; Nyoro mwijukuru, Gisu umwitzukhulu, Kuria mocokoro, Ambo musikulu, Gogo mwizukulu, Nyanja mdzukulu, Ndembu mwijikulu, Zulu mzikulu, and so on. In all these terms -kulu (or -koro) is clearly present, associated with an only slightly less constant middle element -iju/-itzu-/-co-/-iji-/-zi-. How one should derive this term etymologically as a series in common Bantu is open to doubt: my

own suggestion would identify this middle element with the common Bantu term of address for father (Guthrie's CSS* -yìcé/-yìcí/-yìcó) which would give the construction 'person +father + senior' i.e. person (child or descendant) of the grandfather.

How does all this relate to the use of the same stem -kulu or -koro in words for elders or ancestors in Bantu languages? - Essentially by recognising that the quality so denoted (we may call it 'social growth') is acquired over time through a person's relationships with others, notably his family standing, and that the 'big people', les grands, abakulu, are so, whether elders or ancestors, because they are the forebears , the grown persons, whose identity is preserved as much by their successors as by themselves.

This certainly would be the case I would argue for Kuria, who have no verb *-kora (the normal word 'to grow' is -kina), but who do have a highly significant set of words based on the element -koro, viz. omonto mokoro, a mature (grown-up) person, abakoro, ancestors (the commemorated dead), irikora a generation (formalised as a named class), sokoro grandfather, isakoro - grandfather of - nyakoro - grandmother of -. The linguistic point to make is that Kuria have no difficulty in distinguishing between an elder as a fully mature or grown man (the phrase omonto mokoro is used precisely in this sense) and an ancestor, omokoro, as a forebear, a remembered dead relative, but this does not imply that the two terms, linguistically cognate, lack a semantic overlap . Clearly they do derive their meaning from a common element. But this is not authority: rather it is the quality I have tried to define as social growth. Both stand at the other end of a relationship, or set of relationships, that secures their social identity. Whether as grandparents or forebears they are equally 'grown persons'.

Were this interpretation to be accepted - and I would be inclined to link it in the first place specifically to eastern and southern Bantu cultures - something of both Kopytoff's cultural semantics and of Forte's sociological analysis would be preserved. For what is central to Forte's account is that the ancestor as forebear images the relationships of the living: his identity both transcends and yet is dependent upon the living. Complementarily, Kopytoff points to what must be seen as a continuity in conception, that both senior fathers and ancestors as commemorated kinsmen owe their status to their juniors, although in different ways. The senior fathers or grown persons (abaisokor or abanto bakoro in Kuria) have a family status that is strictly relative to their living junior kin; the commemorated dead (abakoro in Kuria) have a less relative, more absolute status, but one that for that very reason is more dependent upon the good will and remembrance of the living. Both kinds of relationship may carry authority, but they do not necessarily. Indeed, formal authority of the adjacent generation type is often notably absent from such grandparental relationships (cf Snagree 1974: 66).

Malcolm Ruel
University of Cambridge
Guthrie, M. 1970. Comparative Bantu, vols 3 & 4. Farnborough: Gregg International.
Richards, A.I. 1956. Chisungu. London: Faber & Faber.
Sangree, W.H. 1974. Youths as elders and infants as ancestors: the complementarity of alternate generations, both living and dead, in Tiriki, Kenya, and Irigwe, Nigeria. Africa 44.
Turner, V.W. 1968. The drums of affliction. Oxford: Clarendon Press.

Malcolm Ruel's communication demonstrates incisively how much is to be gained from comparative semantic analysis of key terms in the vocabularies of culture-historically cognate societies. It can provide insights and questions that concentration on a single society can miss. The approach has yielded a rich harvest of understanding in Indo-European studies and should prove to be no less enlightening in African studies, as Ruel's analysis shows.

About James Brain's comment, I shall restrict myself to two points. First, I have nowhere offered the startling and simplistic proposition that 'Africans do not semantically differentiate dead ancestors and living elders' and that 'they do not perceive a difference between the two categories'. I can only suggest that he re-read my original article and my comment on Calhoun's argument. Second, when Brain states that the stem -zim- is 'used to translate "ancestral spirit"', one can only ask: used by whom? Clearly, by the anthropologist who begins with the English words 'ancestral spirit' on his mind. But the whole point of my argument is that we need to know what -zim- stands for in these Bantu languages, and not whether the English concept of 'ancestral spirit' can be rendered by the Bantu term -zim-. the time to start worrying about the latter problem will be when we write ethnographies of the English-speaking people in a Bantu language.

Igor Kopytoff
University of Pennsylvania

Go to the main Ancestors in Africa page Go to the main Experience Rich Anthropology page Go to the CSAC Anthropology Pages - online