^{page 321 note 2} I use the term ‘covenant’ after Hocart, whose two brief papers are among the most useful written on this problem: A. M. Hocart, ‘Blood-brotherhood’, pp. 185–9; ‘Covenants’, pp. 190–4; The Life-Giving Myth, Methuen, London, 1952.Google Scholar

^{page 321 note 3} See Tegnaeus, H., Blood-brothers, Ethnographical Museum of Sweden, Stockholm, 1952Google Scholar. This is the broadest comparative study available, but it makes little attempt at a detailed analysis of the various social and ideological factors involved.

^{page 321 note 4} Evans-Pritchard, E. E., ‘Zande Blood-Brother-hood’, Africa, vi (1933), pp. 369–401.CrossRefGoogle Scholar

^{page 322 note 1} Anyone familiar with Evans-Pritchard's paper will immediately recognize how much my article owes to his. However, most of the data for this article were collected before I had read his.

^{page 322 note 2} Beattie, J., ‘The Blood Pact in Bunyoro’, African Studies, xvii (1958), pp. 198–203.CrossRefGoogle Scholar

^{page 322 note 3} Beattie, J., ‘Nyoro Kinship and Affinity’, International African Institute Memorandum xxviii (1958).Google Scholar

^{page 322 note 4} See Beidelman, T. O., ‘Hyena and Rabbit …’, Africa, xxxi (1961), pp. 61–74CrossRefGoogle Scholar; ‘Right and Left Hand …’, Africa, xxxi (1961), pp. 250–7.Google Scholar

^{page 322 note 5} In 1881 Last reported a case of blood covenant between a coastal African and another, who appears to have been either a Kaguru or Ngulu, although he is not completely clear concerning this. His account is very different from the accounts I collected from present-day Kaguru and Ngulu informants: ‘In the afternoon, whilst I was in the village, I witnessed the ceremony of making brotherhood between a coast-man and a Native. A mat about eight feet long was placed outside on the ground, and the two applicants for brotherhood were seated one at each end, each having his man behind him. A small puncture was then made in the arm of each, a few drops of blood extracted; these were mixed and partaken of by both men. A sword and knife were then produced; the man who stood behind the coast-man took these in his hand, placed the bare sword on the coast-man's head, and began scraping the knife up and down the blade; at the same time he asked the Native candidate for brotherhood a number of questions, in the answers to which the Native pledged himself to help the coast-man by every means in his power, and under all circumstances to treat him as a brother, before all others. This side of the questioning lasted about twenty minutes. Then the Native who stood behind the village applicant took the sword and knife, and having placed the sword on the villager's head and begun scraping, he went through the same questioning with the coast-man, who in his turn pledged himself to help the Native by every means in his power, and to advance his interest in every way he could. This promise he kept, probably, until he saw a good chance of benefiting himself, and then his so-called brother might go to the wall’, Last, J. T., ‘The Usagara Mission: Mamboia’, The Church Missionary Intelligencer (1881), pp. 556–7.Google Scholar

Another account states that the German adventurer Peters made a covenant (Blutsbruderschaft) with people in the Kaguru area by sharing milk with them, see p. 124, Müller, F. F., Deutschland-Zanzibar-Ostafrika, Rütten und Loerning, Berlin, 1959.Google Scholar

^{page 322 note 6} Beidelman, T. O., ‘Further Adventures of Hyena and Rabbit …’, Africa, xxxiii (1963), pp. 54–69.CrossRefGoogle Scholar

^{page 323 note 1} Few kinship data are at present published for the Kaguru. I hope to publish several detailed analyses of Kaguru kinship at a later date.

^{page 323 note 2} The Kaguru text merely states weja wakikola umbuya (they had taken hold of comradeship) the phrase which follows, idyene disikudaha kubidulwa (which could not be changed) makes it clear that a blood covenant is meant.

^{page 323 note 3} Afi seye cha ndugu (Now we are kin [with one another]).

^{page 323 note 4} mwiko, forbidden: I discuss this term in the next section of this paper.

^{page 323 note 5} Monitor-lizard considers Eagle should be bound to his (Monitor-lizard's) children quite as much as to himself.

^{page 323 note 6} The best feathers for arrows are thought to come from eagles.

^{page 323 note 7} Njaho: a term of great familiarity used to elders with whom one stands in such a relation because they are members of an alternate generation from one's own. Such conduct would not be possible between matrilineal kin of the same or adjacent generations.

^{page 323 note 8} Monitor-lizard means his own children but says ‘your’ to indicate that Eagle should treat these children as though they were his own.

^{page 324 note 1} Literally, ‘We ate [one another's] blood’, aseye chidiya sakami.

^{page 324 note 2} Eagle does not willingly give his feathers but has them taken from him. This is emphasized by the use of the passive, dyaputulwa ‘he was stripped’.

^{page 324 note 3} The heads of drums used at dances to celebrate initiation are usually made from monitor-lizard hide.

^{page 324 note 4} In writing of Kaguru comments about the blood covenant I use the present tense, since Kaguru still preserve the vocabulary and some of the beliefs of blood covenant. In writing of Kaguru blood-covenant practices I use the past tense, since these customs are no longer carried out.

^{page 325 note 1} Evans-Pritchard states: ‘The blood itself is the “medicine”, the material element of the magical complex, and it becomes such through association with the spell and rite’ (op. cit., p. 401). Elsewhere he notes: ‘For the spell has no virtue in itself. It can act only through the blood’ (ibid., pp. 384–5).

^{page 325 note 2} This may well be a modern and less precise use of a term which traditionally had a far narrower meaning restricted to blood comrades.

^{page 325 note 3} I could not find out whether, in the case of blood comrades living far apart, witnesses from both men's kin and neighbourhoods were required. However, this seems unlikely.

^{page 325 note 4} The liver is considered by Kaguru to be one of the tastiest portions of an animal. I did not find that the Kaguru attribute any special qualities to the liver which would account for this. However, the liver may in some way be associated with the closest kin (matrilineal relatives), for when a Kaguru slays an animal (wild or domestic) and holds a feast, the animal's liver should be presented to a mother's brother (bulai), that is, to a man of the first ascending generation within one's own matriclan.

^{page 325 note 5} To Kaguru the heart is the centre of thought and feeling.

^{page 325 note 6} Most Kaguru form the plural of ndugu (kinsman) as wandugu, even though this is contrary to Swahili in which the plural and singular forms of the word ndugu are the same.

^{page 326 note 1} Evans-Pritchard makes a similar observation for the Zande, but interprets it far differently. He describes Zande blood-covenant relations as well defined and Zande kin relations as diffuse and based on sentiment. In contrast, Kaguru blood comrades had quid pro quo relations, it is true, but their terms were undefined and depended upon the inherent limiting nature of reciprocal relations themselves. Kaguru kin relations, in contrast, are formal and fairly clearly defined, at least those involving only men.

^{page 326 note 2} I have not described Kaguru mlongo groups in any previous publication. It suffices here to note that at present the only function of such a group is to proscribe some food or other item from its members during a certain period of their lives.

^{page 327 note 1} Such confusion, apparently based on superficial ethnography, may be found even in recent studies. In a recent monograph, Winans writes of ‘quasikinship ties produced by blood-brotherhood’. Winans implies that blood covenants are still made by the Shambala, and it is regretable that he tells us so little of this important institution (Winans, E., Shambala, Kegan Paul, London, 1962, pp. 70, 74, 94, 123).Google Scholar

^{page 327 note 2} E. E. Evans-Pritchard, op. cit., pp. 370, 398.

^{page 327 note 3} J. Beattie, ‘The Blood Pact…’, loc. cit., p. 200.

^{page 327 note 4} E. E. Evans-Pritchard, op. cit., p. 400.

^{page 327 note 5} E. E. Evans-Pritchard, op. cit., p. 397.

^{page 328 note 1} T. O. Beidelman, ‘Right and Left Hand …’, loc. cit., p. 253.

^{page 328 note 2} I am aware that for Kaguru menstrual blood (nhamu, tamu or nhume), labiadectomy blood (nhtunga, eel), blood of childbirth (fula, rain), have different names from ordinary blood (sakami or sakame), but Kaguru agree that all these are aspects of blood in general (sakami).

^{page 329 note 1} Cory describes the practice of putting substances used to heal circumcision wounds into contact with women's vaginas or urine. Unfortunately, he provides no explanation for this and, indeed, does not even specify exactly which tribes do it, other than that they are from eastern Tanganyika. However, it may well be that this has a basis similar to the Kaguru and Ngulu practice which I describe, see Cory, H., ‘Jando’, J.R.A.I, xlvii (1947), p. 161Google Scholar. Similar emphasis upon blood and its essentially matrilineal nature is made by the Kaguru's matrilineal neighbours to the east, the Ngulu. They also practise blood covenant (soga), or rather did so in the past but not today. Blood (damu or mpomi) was eaten on liver (ini) in the same manner as by the Kaguru. In the past, Kaguru and Ngulu are said to have made blood covenants with one another.

^{page 330 note 1} Hocart makes these points in his two articles, loc. cit.

^{page 330 note 2} Obviously, self-interest also served to maintain such a relationship.

^{page 330 note 3} In these terms, the Kaguru rule against making blood covenants with kin makes sense—as well as preventing redundant alliances.

^{page 331 note 1} The use of the term ‘twin’ is interesting, since Kaguru kill twins, which they regard as dangerous. Twins (as well as certain other unusual persons) are thought to have special abilities which they possess to the corresponding disability of their kinsmen. For Kaguru the closest possible relationship between persons of the same age and sex is to have been born of one womb (and blood) and therefore the closest and most equal relationship of all would be twins which are born of the same womb and yet are not separated by seniority. It is only this egalitarian and consanguineal aspect of twins that Kaguru appear to have in mind by this comparison.

For further discussion of twins, see T. O. Beidelman, ‘A Kaguru Version of the Story of the Sons of Noah …’, Cahiers d'études africaines, forthcoming, and ‘Kaguru Omens: an East African People's Concept of the Unusual, Unnatural and Supernormal’, Anthrop. Quart., 36 (1963), pp. 43–59.

^{page 331 note 2} Evans-Pritchard notes a somewhat similar observance among the Zande. However, according to him, a Zande theoretically owes the same obligations to all members of his blood comrade's clan, but that theory diverges from practice, for ‘The form of obligation is extended, but its content becomes progressively weaker the wider the extension’ (op. cit., p. 392). Kaguru theory and practice appear to have been more congruent, since one acted towards one's comrade's kin in the same way that he himself did—and thus progressively weaker relations on one's own part corresponded to those which occurred within his matrilineage and clan.

^{page 331 note 3} It is important to note that such behaviour was probably chiefly in terms of verbal deference and etiquette, rather than in terms of any far-reaching economic and political assistance. These relations can hardly have been very similar, in actual practice, to relations between matrilineal kin, since such relations did not affect rules of exogamy. Yet such rules are fundamental in preventing confusion and conflict between persons seeking to maintain the theoretically close and harmonious ties such as are ideally associated with members of a descent group.

^{page 332 note 1} Just as the term njaho is used by Monitor-lizard to Eagle when he makes his unjust request.

^{page 332 note 2} Siblings (viz. persons of the same generation and of the same matriclan or with fathers of the same matriclan) of the same sex as ego are differentiated by ego: mkulu- (elder than ego); mfuna-to (younger than ego). Siblings of the opposite sex from ego are not usually differentiated in this way (although on occasions they may be) but are usually simply termed lumbu-.

^{page 333 note 1} I deal in some detail with the problem of reciprocity among matrilineal kin in another article, ‘Further Adventures …’, Africa, loc. cit. This problem contrasts sharply with Hocart's comparison of blood comradeship and marriage as two forms of covenants.

^{page 333 note 2} Evans-Pritchard makes the same point for the Zande:‘The reciprocal nature of blood-brotherhood thus provides an integral system of sanctions by the very mode in which it functions. Social systems invariably generate their own sanctions by their mechanism of mutuality’ (op. cit., p. 390).

^{page 333 note 3} See T. O. Beidelman, ‘Witchcraft in Ukaguru’, pp. 57–98, in Middleton, J. & Winter, E., (ed.), Witchcraft and Sorcery in East Africa, Routledge, London, 1963.Google Scholar

^{page 334 note 1} Evans-Pritchard reports the same thing for the Zande, op. cit., p. 372.

^{page 334 note 2} In this sense, Kaguru blood covenants resemble those of the Nyoro: see J. Beattie, ‘The Blood Pact … ’, loc. cit., p. 200. The alliance aspect of the blood covenant is strongly emphasized by Hocart, loc. cit., who also rightly observes that, in this respect, blood covenant may be compared with marriage.

^{page 334 note 3} See Beidelman, T. O., ‘Beer Drinking …’, American Anthropologist, Ixiii (1961), pp. 534–49CrossRefGoogle Scholar; ‘History of Ukaguru …’, Tanganyika Notes & Records, nos. 58 and 59 (1962), pp. 11–39.

^{page 334 note 4} Some Kaguru refer to blood covenant between Kaguru and Baraguyu; others refer to a covenant of milk. I could not obtain consistent accounts from Kaguru on this. However, Baraguyu informants deny either practice. Among Baraguyu, a covenant (ol-mumai) is made by two men swearing an oath of loyalty to one another and then sharing beer and meat of the chest of a cow. No blood is exchanged. A person who broke such a covenant was thought to be in danger of illness or death owing to the oath which was broken and owing to some excitation of the meat he had eaten and which was thought, in some sense, to remain in his stomach.

In the case of Kaguru, their report of a milk covenant is at odds with other information. Most Kaguru believe that adults should not taste human milk. Were a man, for example, to try to drink his wife's milk during love-play, she might tell him that he is a witch—although some Kaguru men appear to consider the idea of a woman's milk being drunk a pleasant temptation. Of course, what Kaguru actually do together by themselves remains their secret.

^{page 335 note 1} The Church Missionary Society, a Church of England Mission, which adopts certain somewhat fundamentalis attitudes.

^{page 335 note 2} Evans-Pritchard accounts in the same manner for the decline in the Zande blood covenant, op. cit., pp. 394–5.

^{page 335 note 3} The reader must rely upon my brief publications (already cited) and my present statements to support my remarks on Kaguru kinship. I hope to publish detailed information on Kaguru kinship and marriage at a later date.

^{page 335 note 4} A Kaguru parent might curse a child, but such a curse resulted not in loss of kinship so much as in expulsion from Kaguru society. It ended in ostracism by the matrilineage and thus, in the past, in probable death through neglect or in enslavement and deportation. Brothers or other relatives could not sever kin ties in this way. See T. O. Beidelman, ‘A Kaguru Version …’, loc. cit.

^{page 336 note 1} Members of alternate generations are not linked in this way.

^{page 336 note 2} See T. O. Beidelman, ‘Hyena and Rabbit …’, loc. cit.

^{page 336 note 3} I could secure no indication of any ritual obligain tions between blood comrades such as Hocart suggests and Evans-Pritchard reports, op. cit., pp. 390–1, 398.

^{page 336 note 4} This marriage pattern is now in decline.

^{page 337 note 1} This relaxation is true of all relations between alternate generations of kin, but it is more pronounced among classificatory kin than among those with whom one can trace close genealogical relationship.

^{page 337 note 2} Kaguru follow the practice of often naming grandchildren after grandparents or of having grandparents bestow several names on a child. These names may already be held by other living or dead kinsmen or they may not, but names are considered perpetual in the same sense that the matriclan and matrilineage are. Every Kaguru has many alternate names which he may use, some relating in some way to grandparents and some not.

The persons of alternate generations are also those nearest the spirit world (kusimu) which is inhabited by the ancestral spirits (wasimu), since they are those about to return to that world, viz. die, and those who have only recently emerged from it, viz. been born. Although this is not a currently popular belief among the Christian Kaguru, there is a good indication that traditionally many Kaguru believed in perpetual interchange of forces between the world of the living and that of the ancestors.

For an interesting exposition of this idea, see P. Tempels, Bantu Philosophy, Présence africaine, Paris, 1959, pp. 73–74.I hope to deal with these problems in some detail in a later essay,

^{page 337 note 3} Evans-Pritchard, op. cit., pp. 388–9.

^{page 338 note 1} I have already indicated elsewhere the dangers in such translations, Beidelman, T. O., ‘Kaguru Justice…’, Journal of African Law, v (1961), pp. 5–20.CrossRefGoogle Scholar

^{page 339 note 1} The word umoto, custom, clearly derives from moto, fire. Unfortunately, I could not find any ritual attached to fire-making among the Kaguru today, although I am well aware that fire has considerable significance for some other African peoples and that it sometimes has sexual associations.

It is said that in the past only a local dominant matriclan member could kindle a ritual fire at the start of each agricultural year. This is not, to my knowledge, done anywhere in Ukaguru today. Fire may be kindled by anyone of either sex. Kaguru say that because kindling a fire is hard work, men usually are sought to do it. However, a woman may and sometimes does kindle a fire herself.

^{page 340 note 1} A standard greeting among equals.

^{page 340 note 2} A term of great familiarity. Customarily, this is used to seniors of the alternate generation. Kaguru have very intimate, even ribald relations between kin of alternate generations; they may speak to them about anything.

^{page 340 note 3} A Kaguru idiom: ‘I didn't leave the donkey's home for nothing.’ This means: ‘I didn't come on this trip for nothing.’

^{page 340 note 4} Monitor-lizard means his own children, but he calls them ‘your children’, to indicate that his comrade should consider them the same as his [Eagle's] own.

^{page 341 note 1} A small hourglass-shaped drum used in initiation dances.