^{page 250 note 1} My data on the Kaguru were obtained during field research in Tanganyika in 1957–8. This was made possible by a research assistantship awarded by the Department of Sociology and Anthropology, University of Illinois. For further information on the Kaguru, see Beidelman, T., ‘ Hyena and Rabbit: a Kaguru Representation of Matrilineal Relations ’, Africa, xxxi (1961), pp. 61–74.CrossRefGoogle Scholar

^{page 250 note 2} Needham, R., ‘ The Left Hand of the Mugwe: an Analytical Note on the Structure of Meru Symbolism ’, Africa, xxx (1960), pp. 28–33.Google Scholar

^{page 250 note 3} Bernardi, B., The Mugwe; a Failing Prophet, London: Oxford University Press, 1959.Google Scholar

^{page 250 note 4} Hertz, R., ‘ The Pre-eminence of the Right Hand: a Study in Religious Polarity ’, in Death and the Right Hand (trans. R., & Needham, C.). London: Cohen & West, 1960.Google Scholar

^{page 250 note 5} Hocart, A., Kings and Councillors. Cairo: Printing Office, 1936, pp. 161–2, 260.Google Scholar For a more general discussion of dualistic symbolic classification, see Hocart, A., Social Origins, London: Watts & Co., 1954, pp. 86–96.Google Scholar

^{page 250 note 6} Werner, A., ‘ Note on the Terms Used for “ Right Hand ” and “ Left Hand ” in the Bantu Languages ’, Journal of the African Society, xiii (Oct. 1904), pp. 112–16.Google Scholar

^{page 250 note 7} Evans-Pritchard, E. E., Nuer Religion. Oxford: Clarendon Press, 1956, pp. 142, 184, 211, 212, 233–6, 245, 296Google Scholar; ‘ Nuer Spear Symbolism ’, Anthropological Quarterly, xxvi (1953), pp. 1–19Google Scholar. Despite its title, Torday's early article provides very little information on the problems with which this paper is concerned, Torday, E., ‘ Dualism in Western Bantu Religion and Social Organization ’, Journal of the Roy. Anth. Inst. lviii (1928), pp. 225–46.Google Scholar

^{page 251 note 1} For example, the Fulani and the Masai. For the Fulani, see Needham, op. cit., p. 30. Some indication of the Masai data may be found in H. Fosbrooke, ‘ An Administrative Survey of the Masai Social System’, Tanganyika Notes and Records, no. 26 (1948), pp. 1–50. The Baraguyu would also merit such analysis. Unfortunately, my data on these people is too incomplete at present to make this possible, see Beidelman, T., ‘ The Baraguyu ’, Tanganyika Notes and Records, no. 55 (1960), pp. 260, 263, 270.Google Scholar

A recent article by Cory reports an especially fine example of dualistic symbolism of social organization. Unfortunately, Cory does not discuss its relation to the society in which it occurs. See Cory, H., ‘ Religious Beliefs and Practices of the Sukuma/ Nyamwesi Tribal Groups ’, Tanganyika Notes and Records, no. 54 (1960), pp. 14–26Google Scholar. Such concepts also seem to be especially important to the Tonga, see Colson, E., Marriage and the Family among the Plateau Tonga, Manchester University Press, 1958, pp. 130, 166, 217, 272Google Scholar. The Nguru, whose land borders that of the Kaguru, also appear to have a rather elaborate (but at present poorly described) system of dual symbolic classification. McVicar, T., ‘ Wanguru Religion ’, Primitive Man, xiv (1941), pp. 18–22Google Scholar.

There is evidence of elaborate dual classification among such West African peoples as the Fon and Ashanti, though this does not seem to have been analysed in any systematic fashion. See also Morton-Williams, P., ‘ The Yoruba Ogboni Cult in Oyo ’, Africa, xxx, no. 4, p. 372.Google Scholar

^{page 251 note 2} Jeffreys, M., ‘ Dual Organisation in Africa ’, African Studies, v (1946), pp. 82–104, 157–76CrossRefGoogle Scholar. Less helpful discussions prompted by Jeffreys's paper were later presented by Haekel, J., ‘ Die Dualsysteme in Afrika ’, Anthropos, xlv (1950), pp. 13–24Google Scholar, and by Jensen, Ad., ‘ Dual-Systeme in Nordost-Afrika ’, Anthropos, xlviii (1953), pp. 737–59Google Scholar. I was unable to secure a recent article on dualism: Straube, H., ‘ Das Dualsystem und die Halaka Verfassung der Dorse…’, Paideuma, vi (1957), pp. 342–53.Google Scholar

^{page 252 note 1} Needham, op. cit., p. 26.

^{page 252 note 2} Of course, it may be that more complete data would reveal that a fuller set of terms cannot be subsumed under the categories of ‘ right ’ and ‘ left ’. Unfortunately, I was not aware of this problem in 1957–8 when I did fieldwork among the Kaguru. Consequently, my information on such classification is undoubtedly far less than it otherwise might be. However, such a classification of Kaguru concepts about their society is sufficiently important to appear even from such incomplete data.

^{page 252 note 3} The term kulume derives from -ume or -lume which is a masculine root, e.g. mwanalume, man (Swahili, mwanamume); mulume, husband (Swahili, mume). The term kumoso appears also to have a relation to words in other Bantu languages: umotho, moso, Needham, op. cit., footnote, p. 23. The Kaguru terms I cite are slightly different from those used in Last's study of this language. Last gives: mkono wa kulumi (right hand) and mkono wa kumono (left hand), pp. 123, 133. Last, J. T., Grammar of the Kaguru Lan- guage. London: Society for Promoting Christian Knowledge, 1886.Google Scholar

Many Kaguru use Swahili words for right and left rather than the Kaguru words. The Kaguru language is being rapidly replaced by Swahili through the schools, courts, and trade. These Swahili words are often linguistically very similar to Kaguru. They often seem to have the same attributes which Kaguru words for ‘ right ’ and ‘ left ’ have: the right hand is called mkono wa kulia, hand for eating; mkono wa kuume, hand of the male. The left hand is called mkono wa kuke or mkono wa kike, both meaning ‘hand of the female’; and mkono wa kushoto, ‘left hand’. I did not hear Arabic Swahili forms such as yamini, right (twaa yamini, to take an oath), and shemali, left. Some Swahili dictionaries list an additional expression for right hand which I did not encounter: mkono wa mlio. The meaning of this is not clear. Some authorities derive this expression from the verb kulia, to cry; mlio, noise. They then would say this expression means ‘the hand for making noise’, for drumming or playing a musical instrument. However, L. Harries informs me that this expression may derive from kula, to eat. He states that the term mlo, food, and the term mlio, eating vessel, are used at Lamu in the northern portion of the Swahili-speaking area.

^{page 253 note 1} This interpretation of a female's use of her right hand in intercourse was obtained from Kaguru males.

^{page 253 note 2} For a discussion of some of the implications of this principle of classification, see Lévi-Strauss, C., Les Structures éleméntaires de la parenté. Paris: Presses Universitaires de France, 1949, pp. 486–502.Google Scholar

^{page 253 note 3} It is not clear how Kaguru would resolve any possible contradiction between this and their other theory of relationship by right and left which I previously presented. Presumably, Kaguru may not see any contradiction in these two principles.

^{page 254 note 1} This seems to conflict with Hertz's interpretation of the symbolism of right and left. Most of the terms in the Kaguru classification are mentioned in Hertz's paper and are assigned to the same respective categories of right and left as he mentions. But Hertz's analysis presents the left as invariably subordinate to the right. He tries to discount as a ‘ secondary development ’, op. cit., p. 109, the Zuñi example which conflicts with this interpretation, The Kaguru classification of kin groups unquestionably gives eminence to the left (female) side. It would be interesting to learn if this is true of other matrilineal societies with dualistic classification.

^{page 254 note 2} See T. Beidelman, ‘ Hyena and Rabbit ’, supra cit.