^{page 168 note 2} Recent literature on the Nyoro provides a notable exception to this, cf. Beattie, J. H. M., ‘Informal Judicial Activity in Bunyoro ’, Journal of African Administration, ix. 4, Oct. 1957, 188–95Google Scholar, and Bunyoro, an African Kingdom, 1960. There is also interesting material in Cory, H., The Indigenous Political System of the Sukuma, 1954Google Scholar; Cory, H. and Hartnoll, M. M., The Customary Law of the Haya Tribe, 1945Google Scholar; Gulliver, P. H., Social Control in an African Society, 1963Google Scholar; Huntingford, G. W. B., The Nandi of Kenya, 1953Google Scholar; Marr, L. P., An African People in the Twentieth Century, 1934Google Scholar; and Tanner, R. E. S., ‘Law Enforcement by Communal Action in Sukumaland, Tanganyika Territory ’, Journal of African Administration, vii. 4, Oct. 1955, 159–65Google Scholar. Monica Wilson's Good Company, 1951, though essentially dealing with communities of coevals, is also most instructive and of great comparative interest in this context.

^{page 168 note 3} Fallers, L. A., Bantu Bureaucracy, 1956, p. 115.Google Scholar

^{page 168 note 4} Taylor, B. K., The Western Lacustrine Bantu, Ethnographic Survey of Africa, 1962, pp. 49–50, 58, and 69.Google Scholar

^{page 169 note 1} J. Scherer (Ha) and A. Trouwborst (Rundi) in Les Anciens Royaumes de la zone interlacustre méridionale, Ethnographic Survey of Africa, pp. 195 and 147.

^{page 169 note 2} There is a good bibliography of works referring to the Nyamwezi in the East African volume of the Africa Bibliography series of the International African Institute, 1960.

^{page 169 note 3} Apart from work in the Kahama area, most of which was carried out in Busangi chiefdom, I also worked in two chiefdoms, Uyui and Ibili, which are near Tabora. The situation which I found there appears to be typical of central and southern Unyamwezi, though some variation may no doubt be found from place to place. I shall say a little about this Tabora area at the end of this paper.

^{page 169 note 4} I refer to the rural African population here. Since some of the things which I describe have been changed since Tanganyika became independent in 1961, I should stress that I write in the ‘ethno graphic ’ present throughout.

^{page 171 note 1} I have borrowed the concept ‘large mesh ’ here from Barnes, J. A., ‘Class and Committees in a Norwegian Island Parish ’, Human Relations, vii. 1, 1954, p. 44.Google Scholar

^{page 171 note 2} The division of labour between the sexes in fact pervades most areas of the social organization of Unyamwezi. For ease of presentation I have confined specific discussion of it to some of the more detailed parts of my account. It should, however, be realized that for most activities co-operative groups consist of people of the same sex.

^{page 171 note 3} The best material on these rites has been obtained by Cory, who actually joined two of the societies. See his ‘The Buyeye, a Secret Society of Snake-charmers in Sukumaland, Tanganyika Territory’, Africa, xvi. 3, July 1946, 160–78Google Scholar, and ‘The Buswezi’, American Anthropologist, lvii. 5, Oct. 1955, 923–52.Google Scholar

^{page 172 note 1} I refer to the ceremonies for the death of a male commoner in northern Unyamwezi. The ceremonies for women and children and also those for chiefs and headmen differ from these in some respects. There is also some regional variation.

^{page 173 note 1} More than one of these forms of relation between different sets of ties may, however, be present in some situations.

^{page 176 note 1} The title ntemi basically refers to the chief of a chiefdom. Its adoption by co-operative groups and secret societies is found all over Unyamwezi.

^{page 177 note 1} They are also traditional in Sukumaland; cf. Cory, 1954, p. 67.

^{page 178 note 1} Though some people prefer the word ‘moot’ for such gatherings the word ‘court’ has very many uses in English besides its formal legal one. Beattie, 1960, p. 67, and Cory, loc. cit., use the term court to describe institutions closely comparable to those under discussion.

^{page 178 note 2} This is not to say that outsiders may not join in the consumption of fines. It is only an offender's neighbours, however, who are specifically informed of the date of a feast.

^{page 178 note 3} These words kusumula and masumule are also used in some extra-neighbourhood contexts. A person who offends a kinsman or affine may occasionally agree to pay some compensation without a neighbourhood hearing and the offence and payment are described by these words. Again, I have heard the term masumule used for a fine which the ritual elders of a chiefdom demanded from someone who mislaid some regalia which only they could replace.

^{page 179 note 1} For accounts of ostracism in Sukumaland see Cory, op. cit., p. 85, and Tanner, 1955, pp. 161–5.

^{page 179 note 2} Tanner, op. cit., p. 164, gives an interesting case where a particularly rich and powerful Sukuma individual was apparently capable of withstanding ostracism. Northern Unyamwezi is a poorer area than Sukumaland and I did not come across such cases there.

^{page 179 note 3} In the conflict between neighbourhood and chiefdom organization, which I discuss later, the situation was temporarily altered when the neigh bourhood courts started to hear cases which were not normally considered within their jurisdiction.

^{page 180 note 1} These figures do not appear to be unusual. Thus Kahama District as a whole, which had a population of 111,736 in 1957, the courts dealt with 2,303 cases in 1957 and 2,086 cases in 1958.

^{page 180 note 2} Unlike the official distinction between ‘crimi nal’ and ‘civil’ cases, that made here between offences directly against the State and those against persons is my own.

^{page 182 note 1} What follows is an elaboration rather than an explanation of the situation. It may be noted that Beattie, 1957, p. 191, reports a greater degree of overlap in the sorts of cases dealt with by Nyoro neighbourhood and chiefdom courts.

^{page 182 note 2} Beattie, op. cit., pp. 194–5. Gibbs, J. L. in a discussion of ‘The Kpelle Moot: A Therapeutic Model for the Informal Settlement of Disputes’, Africa, xxxiii. i, January 1963, pp. I – IIGoogle Scholar, has paid particular attention to the psychological implications of comparable procedures.

It should, however, be mentioned that the Nyamwezi system of fining and feasting is something more than a mechanism for repairing breaches in village harmony. Reasons for imposing new fines are eagerly sought and disputed at feasts and to this extent the system may also be considered as a self-perpetuating one for the distribution and consumption of wealth within a village.

^{page 182 note 3} I use ‘right in rem ’ as defined by Brown, Radcliffe in Structure and Function in Primitive Society, 1952, p. 33Google Scholar. The people themselves consider adultery as a type of theft. A similar attitude is recorded for the Soga by Fallers, 1956, p. 118.

^{page 183 note 1} It is of interest here that chiefdom courts usually refuse to hear divorce cases during the rainy season. One of the reasons for this is that their settlement demands the presence of witnesses who would have to leave off their agricultural work for the case.

^{page 183 note 2} Cory, 1954, p. 85, reports a similar situation in Sukumaland. What may be in doubt, however, is whether the offender had just cause for acting as he did.

^{page 184 note 1} Beattie, op. cit., p. 194, reports similar behaviour among the Nyoro.

^{page 184 note 2} Gulliver, 1963, part iv, presents a very full and elegant account of the complementarity of Chiefdom Courts and different types of informal courts, including parish assemblies, among the Arusha. Neighbour-hood relations as such are an important feature of Arusha parishes but, particularly in the mountain area of the country, the age-group system which divides the parish population into a number of interdependent groupings provides the main basis for internal parish organization (loc. cit., pp. 19, 24, 34, and passim). Both Beattie, 1957, p. 195, and Gibbs, 1963, p. 10, also stress the complementary nature of informal and chiefdom courts, and there is comparable material in Bohannan's Justice and Judgment among the Tiv, 1957, ch. ix and passim.

^{page 184 note 3} Cory, 1954, p. 94, shrewdly foresaw the possibility of such a connexion for Sukumaland.

^{page 184 note 4} The smallest official branches usually covered one or two chiefdoms.

^{page 185 note 1} One aspect of this was the way in which T.A.N.U. cleverly encouraged the use of the titles D.C. and P.C. Tanu to refer to their District and Provincial Chairmen.