^{page 5 note 1} This statement, of course, does not apply to such norms of conduct as are imposed upon the community by a ruling power, in which case the process is exactly reversed.

^{page 5 note 2} A name given to them by the Uasin-Gishu Masai.

^{page 5 note 3} Census figures of 1932.

^{page 6 note 1} The expression which comes nearest to it is ave nyumba yange, ‘those of my house’.

^{page 6 note 2} i.e. those groups within a clan who trace their kinship by actual genealogies.

^{page 6 note 3} Until the establishment of intertribal peace the Kitosh lived in walled villages (dzingova) in protection against Masai raids.

^{page 6 note 4} The word Maragoli denotes the country of the Logoli.

^{page 6 note 5} A full description of the hut and its sociology is not intended here.

^{page 7 note 1} In Kitosh bananas do not grow so well and are, therefore, planted only in valley bottoms or on the fertile sites of abandoned cattle-kraals.

^{page 9 note 1} The notion that cattle-herding is below the dignity of grown-up men is of recent origin, as formerly the danger from wild animals and enemies made cattle-herding a job which required the attention of full-grown men.

^{page 9 note 2} Omusaza and omukali, and omwana and omukana respectively.

^{page 10 note 1} The wife is not, however, adopted into the husband's clan but remains a member of her own clan.

^{page 11 note 1} Provided that the husband is not impotent.

^{page 11 note 2} Such conduct is, however, considered adulterous only when engaged in with the intention of secrecy. A wife who openly leaves her husband to live with another man, with no intention to return, is considered divorced and merely the bride-wealth is returned; see also footnote i, p. 15.

^{page 11 note 3} i.e. by bringing the matter before the judicial council of clan elders (ekiruazo); see below, p. 27.

^{page 12 note 1} Some of my informants insist, however, that this is a new, although now general, development which sharply contrasts with conditions in pre-European days.

^{page 12 note 2} Such protection, however, is given to the wife within limits only. As a rule, the marriage cattle given for her have in the meantime been handed on as bride-wealth for another marriage and her brothers will, therefore, urge her to return to her husband.

^{page 13 note 1} Both hut and garden are only apportioned to the wife by the husband, who remains the owner (omwene) of both, i.e. he retains the full rights of control and disposal.

^{page 13 note 2} A husband may not climb upon the storage shelf, an attic-like arrangement above the cooking- and sleeping-partitions of the Logoli hut, nor may he look into the granaries on the yard, unless he wants to take stock of the supplies or sell part of the crops.

^{page 14 note 1} As distinct from her quality as a wife and member of the household.

^{page 14 note 2} As a rule one cow is deducted for each child she has borne.

^{page 14 note 3} Even in this case the deceased husband's brother or nearest classificatory brother legally steps into the father's place, and not the new husband. Thus in the case of daughters if they are raised at the mother's new husband's place (and not at the place of the husband's brother), he can claim one cow of the marriage cattle as compensation for the bringing up of the girl.

^{page 15 note 1} This is the traditional law, which in recent judicial practice is changing; see below, p. 46. According to customary law in Kitosh a woman carries the sole blame only when she commits adultery away from her hut, while her lover is also subject to a fine if he has entered her hut. In Maragoli the wife alone is considered the guilty party in all circumstances.

^{page 15 note 2} Omwivuli alikudzila kuvulavu, ‘the begetter will die on the public place’.

^{page 16 note 1} In Kitosh a man's obligation to pay marriage cattle is not limited to his own sons but extends to the ‘sons’ of his lineage group. The above argument applies, therefore, primarily to Maragoli.

^{page 16 note 2} i.e. the garden apportioned to his mother; cf. footnote i on p. 13.

^{page 17 note 1} As an elder put it in a Logoli text: ‘The first-born [son] is who is helping his father dividing the things to those smaller [children]. Because now he is as a brother to him [the father], and if he [the son] has his child [then] that son is a “husband” [i.e. a married man of full status]. He is saying things, then his father listens indeed and he says: “This my son he is saying things like a husband.”’

^{page 19 note 1} Taxation lists from Maragoli and Kitosh show the following data on polygyny:

This compilation requires two corrections, (a) The number of polygynous families is larger than indicated as plural wives are often not registered, for purposes of tax evasion, (b) The lists being compiled on the basis of local headmanships, a man with wives in two different headmanships is listed in each location separately, hence the number of people with more than two wives appears lower than it actually is.

^{page 19 note 2} Avahalikwa, from okuhalika, to marry a second wife.

^{page 20 note 1} The newly married wife who lives with her husband in the homestead of her father-in-law cannot sleep there as she has to avoid her father-in-law.

^{page 20 note 2} Co-wives often cultivate their gardens jointly, beginning in the garden of the great wife.

^{page 20 note 3} Also linguistically they are distinguished from full siblings: amitu (plur. avamitu) is the half-sibling of the same sex, i.e. a man's half-brother or a woman's half-sister, and mbozua (plur. avavozua) is the half-sibling of opposite sex, i.e. a man's half-sister or a woman's half-brother.

^{page 21 note 1} If they are many they are kept in a cattle enclosure that adjoins the hut of the great wife.

^{page 22 note 2} In a sense which will be discussed below, pp. 28 sq.

^{page 22 note 3} The marriage of a senior wife by the son of a junior wife is not permitted, even if the difference in their age would not be too great to rule out their marriage under ordinary circumstances.

^{page 23 note 1} Cf., however, footnote 2 on p. 29.

^{page 23 note 2} This may, perhaps, be due to the fact that until recently cattle were an insecure possession owing to the frequent Masai raids.

^{page 24 note 1} The following argument is put into the past tense as recent changes apply to the whole tribe and not merely the progressive section.

^{page 24 note 2} The same argument applies to money, only more so as money has merely exchange value and no consumption value.

^{page 24 note 3} See below, pp. 28 sq.

^{page 25 note 1} Cf. below, p. 27.

^{page 25 note 2} A full discussion of these points will be undertaken in a subsequent paper.

^{page 26 note 1} Such cattle will accrue to them in the meantime partly through return gifts which their wives' kin are supposed to make from the increase of the marriage cattle and partly from the increase of the inherited cattle.

^{page 26 note 2} A person will contribute to his brother's bride-wealth, even if his brother owns several times as many cattle as he needs for marriage.

^{page 26 note 3} See the above summary, points 3, 6, 7, 8.

^{page 27 note 1} Called omukulundu among the Logoli, omugasa among the Vugusu, and eligutu among the Wanga.

^{page 27 note 2} Called asai in Maragoli and wele in Kitosh.

^{page 28 note 1} With the exception of the Wanga tribe.

^{page 28 note 2} Avakana vatula kuhiri javo kudzia handi, ‘daughters leave their clan to go elsewhere’. Individually, however, they retain membership of their father's clan; cf. p. 10.

^{page 28 note 3} Actually, this explanation was given to me repeatedly when I pointed out that the married daughter as such was not lost to her family.

^{page 29 note 1} e.g. through raiding or in exchange for crops or services (of the magician or witch-doctor), but not as compensation for criminal offences or the violation of a taboo.

^{page 29 note 2} Actually, a maternal uncle and a sister's son are entitled to one bull out of the legacy. But this share in the legacy is balanced by strictly reciprocal gifts—consciously regarded as such—at circumcision and marriage. Besides, these relatives can never claim more than one bull, regardless of how big or small the legacy is.

^{page 29 note 3} In Kitosh.

^{page 29 note 4} To quote from a text: ‘If you have your friend and he is [too] poor in cattle to take a wife, and you have cattle you cannot give him … even if you like him overy much you cannot do so.’ The only exception to this is when the friend is a virongo, a ‘circumcision friend’, whose functions are similar to those of a blood brother elsewhere.

^{page 30 note 1} Only in exceptional cases—if a man cannot find land in his own clan, if he has quarrelled with his clansmen, or if he has been haunted by disease or adverse fortune—marriage may be temporarily or even permanently matrilocal.

^{page 30 note 2} Cf. above, pp. 8 sq. and 13.

^{page 31 note 1} Cf. above, p. 4.

^{page 31 note 2} i.e. exported from reserve, not necessarily from colony.

^{page 31 note 3} According to a 1932 estimate (Agricultural Memorandum, Carter Report evidence, vol. iii) the value of crops consumed annually in the district amounts t o £750,000 or ca. 190s. per household of 4·5 persons.

^{page 32 note 1} According to the labour returns of 31 October 1937 a total of 79,897 natives of Nyanza Province were in employment, of whom 30,796 or 38 per cent, were Bantu of North Kavirondo. The above figure of 124,260 is based on the assumption of an even distribution of wages over the whole province.

^{page 32 note 2} The value, at local market prices, of goods annually consumed by a family of 4·5 members amounts to approximately 390s.: 130s. for the items listed above, 190s. for cereals, pulses, &c, 50s. for fowls, eggs, milk, IOJ. for firewood, 10. for upkeep of homestead and replacements of native-made household articles. A fuller discussion of these figures will be found in my forthcoming book, The Bantu of Kavirondo.

^{page 33 note 1} A fuller analysis of these changes is impossible within the scope of the present paper, but where they affect the family the relevant data will be given in the proper context.

^{page 35 note 1} The inquiry was made in the location of the Yonga clan, South Maragoli, an area which is characterized by a number of features making for contact developments: a sector school which until 1926 was an out-station of the Friends' African Mission with a missionary in charge, a District Commissioner's camp, a dispensary, a recruiting agency and camp of a Kericho tea company, and proximity to a main motor-road with a distance of 16 miles to Kisumu, the rail terminus and provincial centre, of 15 miles to the centre of the Kakamega gold-fields, and of 22 miles to the Kakamega district station.

^{page 35 note 2} The actual amounts I have found extremely difficult to assess, as the money is seldom sent directly to the wife but reaches the household through indirect channels only. The following is a typical case. The husband who has saved 30s. gives 20s. to a friend, who borrows 12s. to pay his own tax, hands 3s. to the husband's wife for buying clothes, &c, and pays 5s. to a neighbour who helps the wife in the garden. Of the remaining IOJ. the husband brings 5s. along himself when he comes home and sends 5s. to a brother who uses this amount together with his own money to buy a young heifer in which the husband then becomes a part-owner.

^{page 36 note 1} See pp. 41, 43, and 49.

^{page 36 note 2} The term squatter in Kenya refers to individuals or families who have entered into a contract with a European settler whereby they are given the right to reside, cultivate gardens for their own food-supply, and graze their stock on the settler's land, in return for which rights they have to work a fixed number of days or hours per day on the settler's farm. According to the 1930 labour returns 2,540 male squatters from the Nyanza Province lived on neighbouring European farms, mainly in the Trans-Nzoia and Usain-Gishu districts of the Rift Valley Province.

^{page 38 note 1} The wages paid are about 20 per cent, lower than those paid by Europeans (5s. to 7s. as compared with 8s. to 10s. per month) and no food is given, except i n the case of permanent servants, but the work is considered less exacting.

^{page 38 note 2} The conditions for large-scale farming are more favourable in North Kitosh than elsewhere in the district owing to the ampleness of land, large stretches of flat country, and the proximity of the railway.

^{page 39 note 1} At least in the predominantly agricultural areas.

^{page 39 note 2} Under the present hut- and poll-tax ordinance the hut-tax levied from every plural wife equals the hut- or poll-tax of 12s. per annum levied from every adult male. Thus a man with three wives has to pay 36s.

^{page 39 note 3} These opinions were obtained by asking a considerable number of monogamous but well-to-do persons what they thought of polygamy.

^{page 39 note 4} Cf. above, p. 28.

^{page 39 note 5} Cattle sold for export from the district are valued at £3,000 per annum (Agricultural Officer's annual report, 1937, for North Kavirondo). The cattle traded for cash within the district or sold for slaughtering must represent a value many times as high, as the value of exported ox hides alone amounts to £28,524 per annum. There is a big demand for cattle for meat markets, which nowadays supply most of the meat consumed, as ritual killings and other killings in fulfilment of kinship obligations are getting rare.

^{page 40 note 1} Annual fees in mission boarding-schools range from 45s. to 70s. for tuition and boarding.

^{page 40 note 2} Several ‘letters to the editor’ from former native Christians of Kavirondo have recently been published in the East African Standard, denying that a Christian life is incompatible with polygamy.

^{page 40 note 3} This state of affairs has led to so many complications, especially lawsuits of expelled Christians against the mission for compensation or for the return of individually owned lands granted for school and church purposes, that the setting aside of further Christian villages has recently been prohibited by a Local Native Council resolution (Resolution 7 of 193 3, North Kavirondo Local Native Council).

^{page 41 note 1} Fragmentation of holdings is said to have served the purpose of minimizing the risk of a wholesale destruction of the family gardens through hailstorms, locusts, or evil magic.

^{page 42 note 1} Out of 80 married Christians of one church community (Vihiga, South Maragoli) for whose expulsion from the church the motives and circumstances have been ascertained by me, only 6 left for the exclusive reason of marrying a second wife, for 8 it was one of several reasons, and 15 left for other reasons but married a second wife a year or more after they had left the church.

^{page 45 note 1} At the Church of God Mission, Bunyore, the total attendance in the first year was 13 pupils, as against 100 boarders and 33 day pupils in 1938.

^{page 45 note 2} Besides Bunyore at Mumias, and Kakamega (Roman Catholic), Butere (Church Missionary Society), and Kaimosi (Friends' African Mission).

^{page 45 note 3} Present at the debate were 52 teachers representing all missions in the area as well as the Government School. The reasons given in support of girls'education were: to enable women to instruct their children in hygiene (which would lower child mortality), to make homes more comfortable and ‘as nice as those of Europeans’, to give the wives self-respect and thus make them respected by their husbands (which would decrease wife-beating), to break down the barriers of (ritual) shame between the sexes, to enable wives to read cookery books and cook more varied food than the monotonous gruel or porridge three times a day (as the resources of a fuller diet were all there), to increase the wives' knowledge of agriculture, to make husbands and wives have more interests in common.

^{page 46 note 1} The main reasons are: clan jurisdiction has been superseded by tribal jurisdiction; clans no longer form military units for raids and defence; land, when purchased for money, can be held with almost the same security outside the territory of one's own clan; Christian communities cut across clan groups; common sacrifices to clan ancestors have become erratic and half-hearted. A fuller analysis of these changes and the evidence for them will be given in my forthcoming book.

^{page 47 note 1} The statement that the hope for economic returns is the chief motive for the education of sons and daughters is based on numerous talks with native fathers who had children in school; it is also confirmed by all missionaries with whom I have discussed the subject.

^{page 47 note 2} Owing to the difference in education, the replacement of clan jurisdiction by tribal courts in which seniority is of little significance, the decreasing belief in the ancestor cult and in the power of curses, and the decreasing economic importance of the father's legacy.

^{page 48 note 1} The Kericho tea plantations, for instance, employ hundreds of children from Bunyore, the most densely populated location of North Kavirondo.

^{page 50 note 1} Cf. B. Malinowski, ‘Culture’, in Encyclopedia of Social Sciences.

^{page 51 note 1} Cf. Oberg, K., ‘Kinship Organization of the Banyankole’, Africa, vol. xi, no. 2, pp. 151 sqGoogle Scholar.