² Bohannan, P., African homicide and suicide, Princeton University Press, 1960.Google Scholar I also found valuable his more detailed study of traditional procedures among the Tiv, although there was little in that book about homicide: Bohannan, P., Judgement and justice among the Tiv, O.U.P., 1957.Google Scholar

^{page 58 note 1} See, e.g., Appendix V, Crimes of Violence, Staff report to the National Commission on the Causes and Prevention of Violence, prepared by Mulvihill, Tumin and Curtis, Vol. 11, U.S. Govt. Printing Office, Washington, D.C., 1969 at 153. Homicide rates for American cities range from a high of 28.3/100,000 to a low of 2.5/100,000 per annum. The rate for Western Nigeria is about the same as that for Minneapolis, Minn., and Long Beach, Calif. One should be wary, however, of international comparisons.

^{page 58 note 2} The 1973 census had just begun when this research was completed. The figures in this study refer to the 1963 census, which has been severely criticized. However, the figures for Western State are thought to be accurate and were all that were available at time of writing.

^{page 59 note 1} Demographic estimates of the population of Ibadan run from a high of 2 million to a low of 1 million. It is impossible to know exactly how many people live in the city when migration is so heavy both in and out of the city, on a seasonal and annual basis. Currently, Lagos with over 2.5 million is thought to have usurped the title of largest city in Black Africa. For a detailed and scholarly examination of the phenomenon, unique in traditional Africa, of Yoruba urbanization, see Mabogunje, A., Urbanization in Nigeria, University of London Press, 1968.Google Scholar

^{page 59 note 2} Statistical information throughout, unless otherwise indicated, is from Western State of Nigeria Statistical Abstract, Ministry of Economic Planning and Reconstruction, Statistics Division, Vol. XII, Ibadan, 1971. The figure for Ibadan urban density is from Mabogunje, op. cit.

^{page 60 note 1} The absence of records and statistics on a year-to-year basis is a serious difficulty for anyone doing research. Annual police statistics have not been published since 1966. Eight different series of law reports have published Western State cases since 1955. Not one of those series publishes all cases in a given year or on a year-to-year basis. There are no statistics available, even in unpublished form, on judicial administration. Federal Supreme Court judgments are just now beginning to catch up from a two-year backlog in reporting. In 1967 a decision was made to include all High Court judgments in the official files at the capital. However, it is not clear that all cases were sent in for the early years. The judgments varied greatly in the amount of detail they included. In one case the name of the accused was nowhere to be found.

^{page 60 note 2} When a driver is prosecuted for “manslaughter-dangerous driving”, he will appear on the police manslaughter files as an accused. However, if ten people were killed in the accident, the manslaughter figure would only include the one driver being prosecuted. Thus the manslaughter figure is somewhat magnified by the inclusion of a few cases of prosecutions against criminally negligent drivers. However, the manslaughter figure is not a figure for death due to road accidents. The number of victims of road accidents annually averaged 91 in Ibadan police district alone. A total of 135 deaths were recorded in Ibadan in 1972. Western Nigerians, like Americans, are more likely to be killed by motor vehicles than by relatives, or in riots, or by any other unnatural cause.

^{page 60 note 3} Ralph Tanner used something like the sudden and unnatural death figure in calculating the incidence of homicide in Uganda. His calculations are based upon a raw figure of 1267 homicides, a total which was reduced by police investigation to almost half, or 711 murders. Tanner computes homicide rates by district, and the figures range from a high of 195/100,000 per annum to a low of 3/100,000 per annum. A later commentator takes the figure of homicides reported to police and arrives at a figure of 17.8/100,000 per annum for Uganda as a whole, with highs of 43.9 and 33.3 for Karamoja and Kampala districts respectively, and a low of 5.4 for Kigezi District. The number of homicides in Uganda has always been thought to be high for Africa, and such regional discrepancies point out the hazards of quoting any figure for a country as a whole. See Tanner, R., Homicide in Uganda, 1964, Scandinavian Institute of African Studies, Uppsala, Sweden, 1970, 10ff. The later research, which uses data from the years 1965,1966,1967 and 1968, was done by M. Mushanga, Dept. of Sociology, Makerere and reported in M. Mushanga, “Criminal homicide in Western Uganda”, Makerere University Monograph, April, 1972, 3.Google Scholar

^{page 61 note 1} The relevant statutes on homicide are contained in ss. 306–329 of the Criminal Code, cap. 27. Excusable homicide includes killing in the course of suppressing a riot, as well as the usual excusable homicide of self defence. Joint and several responsibility is set out for murder and manslaughter where it is not clear by whose hand life was taken. Killing by fighting may be murder, manslaughter, or homicide in self defence, according to the circumstances. The relevant law is clearly set out in , Brett and , McClean, The Criminal law and procedure of Lagos, Eastern Nigeria and Western Nigeria, London, Sweet and Maxwell, 1963. A revised edition has been prepared.Google Scholar

^{page 61 note 2} A difference between the Nigerian and pre-1957 English law is in the unlawful purpose doctrine. To be murder in England, killing in prosecution of an unlawful purpose must be done in the act of committing a felony involving violence. In Nigeria it is sufficient if death is caused by an act done in prosecution of any unlawful purpose, not necessarily a felony. However, in Nigeria (and this appears to contradict the first distinction) the unlawful purpose must have caused an act of such a nature as to be likely to endanger human life. This last proviso was not necessary in England: See Brett and McClean, op. cit 683. s. 1840. There is also a distinction between English and Nigerian law with regard to killing in defence of property. English law distinguishes between defence of property against forcible felony and defence with respect to a misdemeanour. In Nigeria the distinction is silillar to that in the U.S., namely, between defending a dwelling house and other property. Only in defending a home is harm to the trespasser justified.

^{page 61 note 2} A clear account of the difficulties presented in the various defences offered in killing involving allegations of witchcraft is set out in R. B. Seidman, “Witch murder and ???rea: a problem of society under radical social change”, (1955) 28 M.L.R. 46.The author shows how each defence (self-defence, insanity, mistake, provocation) fails, leaving the courts enforcing a norm which is at odds with what the greater part of traditionally oriented society believes.

^{page 62 note 1} The Queensland Code was the model for the Penal Code of East and Central Africa also. The Penal Code of the North (of Nigeria) was drafted from the Sudan Penal Code and differs from the Criminal Code Act in several important aspects. The Western State Criminal Code is identical, word for word, with the Federal Criminal Code, which is in effect everywhere where the Criminal Code of the North does not operate. Parallel citation of the Western State Code is a formality; the State Attorney-General can technically only prosecute underc the Western State Code.

^{page 62 note 2} S.28 of the Criminal Code sets out the law on criminal insanity: ”A person is not criminally responsible for an act or omission if at the time of doing the act or making the omission he is in such a state of mental disease or natural mental infirmity as to deprive him of the capacity to understand what he is doing, or of capacity to control his actions, or of capacity to know that he ought not to do the act or make the omission”. The legal point most often at issue in the judgments was whether or not the accused could be judged insane at the time of the act, given that the expert testimony usually referred to some time after the crime. Delays or the total absence of expert testimony made it impossible to distinguish between killing by an insane person, insanity caused by or after killing, and insanity which developed in custody after the crime. Nigerian judges were also reluctant to ante-date evidence of insanity after the crime.

^{page 62 note 3} In Wolfgang’s study of homicide in Philadelphia the proportion of offenders declared criminally insane was very small, i.e. 2.7%. This compares with about 33% of homicide offenders declared criminally insane in Britain. Wolfgang also notes that of those available for trial, typically around 4% are reported insane. The figure of 16 acquittals out of 205 offenders, or 8%, is high in comparison with other reported figures he has noted. See Wolfgang, M., Patterns in criminal homicide, University of Pennsylvania Press, Philadelphia, 1958, 312 ff. for U.S. figures and reported British figures.CrossRefGoogle Scholar

^{page 63 note 1} A detailed analysis of 44 cases involving accused who showed signs of mental illness will be presented separately.

^{page 63 note 2} A valid confession can be orally translated and transcribed by a police officer and then read over and consented to by the accused. If, however, the prisoner makes his statement through an interpreter, the interpreter must then testify in court before the confession will be admitted into evidence. The police officer taking the confession must also sign it. For Nigerian law on confessions see: The Evidence Act, ss. 27–32; also R. v. Omokaro 7 W.A.C.A. 146; R. v. Prater (1960) 1 All E.R. 298 (need for corroboration); and R. v. Kanu (1952) 14 W.A.C.A. 30.

^{page 64 note 1} Supreme Court Act No. 12 of 1960, s. 31 (4).

^{page 65 note 1} T. Asuni, “Homicide in Western Nigeria”, (1969), 115 British J. of Psychiatry. 1105–13.

^{page 65 note 2} A. Emovon and T. Lambo, “A survey of criminal homicide in Nigeria”, undated mimeograph, Behavioural Science Research Unit, University of Ibadan, Ibadan, Nigeria. See also T. Lambo, in 108 (1962), Journal of Mental Science, 256.

^{page 65 note 3} Bohannan, African homicide and suicide.

^{page 65 note 4} C. F. Lobban, “Law and anthropology in the Sudan: an analysis of homicide cases in the Sudan”, African Studies Seminar Series No. 13, Sudan Research Unit, University of Khartoum, February, 1972 (Ph.D. Thesis in Anthropology, Northwestern University, 1973) R. E. S. Tanner Homicide in Uganda 1964, Scandinavian Institute for African Studies, Uppsala, Sweden, 1970. M. Mushanga, “Criminal homicide in Western Uganda”, M.A. Thesis, Makerere University, Kampala, Uganda, 1972.

^{page 66 note 1} During the period studied new High Courts were established. Thus the number of High Courts sitting in 1966 would not be the same as the number of High Courts sitting in 1972. Comparisons between judicial divisions should be read cautiously. Courts were sitting in Ibadan, Abeokuta, Akure, and Oshogbo throughout the period. In addition to the 8 High Courts included in this study, two new High Courts, at Ife and Ilesha, have recently been established. Judgments from these two courts are not included, as they were not on file on January 1, 1973, although those courts had already handed down decisions at that date. The situation is made more confusing by the fact that judicial divisions do not coincide with administrative districts, nor do police districts coincide with either. The 8 judicial divisions represented here are: Akure, Ibadan, Oshogbo, Abeokuta, Ijebu-Ode, Ado-Ekiti, Oyo, and Ondo.

^{page 66 note 2} For questions concerning judicial administration the figures for 1968, 1969 and 1970 are the most reliable. A delay of several months, or even a year, can occur before the typescript of the judgment arrives at the Ibadan High Court Library. The number of cases for the earlier years is small and includes some cases charged prior to 1966. 1966 was when all judgments began to be put on permanent file. Previous to 1966 only selected judgments were kept on file.

^{page 66 note 3} The number of appeals decided should be qualified. Previous to 1967 all cases were appealed to the F.S.C., as there was no Western State Court of Appeal. The F.S.C. has also recently changed its method of reporting, and there was a backlog of about two years in reported F.S.C. judgments. Thus some Court of Appeal judgments would have also been decided by the F.S.C., but not all W.S.C.A. decisions would have been decided by the F.S.C. Also, some F.S.C. cases would have been the subject of appeals directly to that court.

^{page 67 note 1} Mushanga reports that men account for 89% of all accused. See M. Mushanga, “Criminal homicide in Western Uganda”, op. cit. at p. 2. Lobban reports that 58 of 360 homicide accused in her group were women (16%). Tanner does not report the sex of accused, only of victims. Wolfgang reports 82% men in his sample of 621 offenders.

^{page 68 note 1} Tanner's study indicates that women are overwhelmingly likely to be killed by relatives and family, and particularly by members of their immediate families. Wolfgang's figures agree with this. Men, Tanner reports, are more likely to be killed by friends and acquaintances as sexual rivals and thieves. Lobban comments that homicides concerning sexual jealousy, a common pattern in the North of the Sudan, account for many of the killings of acquaintances in her study. See Tanner, op. cit. Table 40 and commentary on p. 51. Lobban, op. cit., p. 10. Wolfgang, op. cit., Table 24, p. 207.

^{page 68 note 2} Oyo Yoruba includes Ibadan, Oyo and Oshun in the Central and Northwestern part of the state. Ekiti refers to the people of the Northeastern parts. Egba Yoruba are those from the Southwest, near the Dahomey border. Ijebus are from the area above Lagos; Ijesha would be the Northeastern area between Oshun and Ekiti.

^{page 68 note 3} Table 6 must also be read in conjunction with the available census data. The ethnic breakdown in the 1963 census is: Yoruba, 9 million (92%); Urhobo and Edo, 139,000 (1.4%); Ibo, 134,000 (1.4%); Hausa and Fulani, 64,000 (0.7%); with the remainder distributed among the smaller ethnic groups (Tiv, Nupe, Ibibio, non-Africans and other Africans).

^{page 68 note 4} Lobban's data also seem to confirm this finding. In the Sudan inter-ethnic homicide accounted for a far smaller proportion of the total than was expected. Tanner's study finds 90 inter-tribal homicides for 280 intra-tribal homicides, in a country where tribal differences are marked by language differences as well as cultural differences.

^{page 69 note 1} Apart from colonial police records the only internal comparisons I have seen are in Milner, A., The Nigerian penal system, Sweet and Maxwell, London, 1972, cap. 1, “Crime in Nigeria”. Colonial police records seem to show a homicide rate in the North which is much higher than that in the West or in the East.Google Scholar

^{page 70 note 1} It has been conservatively estimated that over 160 homicides occurred in Western State during the Agbekoya rioting. In 1969, 27 accused were charged with homicide and later came to trial in 14 cases. The official police toll of deaths from October 1965—January 1966 was 153. The group of accused studied here included 41 prosecuted in 1966 and 1965; they were tried in 22 cases. The number of cases of homicides charged in 1969 was about one third below the average for 1966, 1967, and 1968. For estimates of deaths during the Agbekoya incident, see Beer, C. E. F., “The farmer and the state in Western Nigeria: the role of farmers’ organizations and co-operatives”, Ph.D. Thesis, Political Science Dept., University of Ibadan, 1971, 393.Google Scholar For the 1965–1966 figures see Robin, Luckham, The Nigerian military, Cambridge University Press, 1970, 219.Google Scholar

^{page 70 note 2} Lobban reports that about 18% of all homicides among men in the North of the Sudan result from situations involving alcohol. Tanner reports that alcohol was a factor in 38% of the killings of men and 29% of the killings of women. Wolfgang reports that alcohol was a factor in 64% of the 588 homicide cases he studied. See Lobban, op. cit., 11; Tanner, op. cit., 42 and Table 33; Wolfgang, op. cit., 134 ff. and Table 14.

^{page 72 note 1} The difference between the analysis regarding victims and the figures presented in Table 7 is that the victim figures are calculated on a base of 114 victims. The percentages of victims are: 83% of all Ijebu Yoruba victims were killed by other Ijebu Yoruba; 75% of Egba Yoruba victims were killed by Egba Yorubas; 67% of Oyo Yoruba victims were killed by Oyo Yoruba; 53% of Ekiti Yoruba victims were killed by Ekiti Yorubas; 50% of Ijesha Yoruba victims were killed by other Ijesha Yoruba; 38% of Ibo victims were killed by other Ibos; No Hausa victims were killed by other Hausa.