105. Ilesha's population, at its maximum before the sack of 1870, is likely to have been between 25,000 and 40,000. Detailed argumentation of this point will be presented in a later work.

106. These were often associated, however: thus Chief Esira was responsible for rituals connected with the crowns as well as with organizing their manufacture, while Osunmu's woodcarving was connected with his knowledge of the spiritual properties of trees. The Owena festival, which precedes the main parts of Ogun, involves the cleaning and medication of the Owa's regalia, especially his crowns, by the people of ODO ESIRA. Both worshipped Odua, the ancestor of the Yoruba at Ife, as an aspect of these functions. Interviews with Chief Tewogbade at ODO ESIRA and Osunmu and Chiefs, 8 and 27 August 1974.

107. Thus after the royal lineage at 13% of the total population come the lineages of the three most senior-ranking hereditary titles: Ogboni at 7%, Odole at 7.2%, Loro at 5.9%, followed by 1.8% for Risawe, 3.8% for Saloro, 3.4% for Arapate (the next Are in seniority), 0.5% on average for other titles in the sample. Data from 1974 Sample Survey.

108. Law, , Oyo Empire, 62.Google Scholar Law's error, very simply, arises from his explicit assumption “that the better-documented conditions of the nineteenth century can be projected backwards into earlier times.” Not merely is this an implausible assumption, granted the enormous upheavels of the nineteen- the century, but it is inconsistent with one of the factors Law invokes to explain the collapse of the old system, the “progressive erosion of political norms” (pp. 254, 298) that occurred from the late eighteenth century at Oyo.

109. Miller, Joseph C., Kings and Kinsmen: Early Mbundu States in Angola (Oxford, 1976), 52–53.Google Scholar Compare the interesting detail noted by Biobaku, S.O., “Historical Sketch of Egba Traditional Authorities,” Africa 22(1952), 47CrossRefGoogle Scholar, that two nineteenth-century Egba leaders “purchased” the prestigious Oyo titles Basorun and Are Ona Kakamfo from Oluyole of Ibadan and Kurunmi of Ijaiye respectively.

110. Cf. Lloyd, P.C., “Political and Social Structure” in Biobaku, , Sources of Yoruba History, 205–06.Google Scholar Yet see above note 72 and note 116 below on the basic iwarefa titles, which are shared with, but probably not borrowed from, Benin.

111. Thus Efon Alaye has Elefosan, Oisinkin, Olodifi, Olokinran, Olosare, Itaji has Olijoka, Elejofi, Olorinkiran, Oye has Oisinkin, Olijoka, Akure has Elejoka, Olisikin, Elekiran, Elejofi, Aramoko has Oisinkin, Elejofi and so on. We also repeatly encounter titles like Obanla, Salotun, Odole, Sajowa, Saloro, which occur in Ilesha. But the former group establish the point that the titles spread from Ilesha to Ekiti since, while their names express the Ilesha quarters from which they originated, they are not so linked to quarter-names in the Ekiti towns. It is noteworthy that the senior military titleholders were so important as bearers of Ijesha prestige in this direction. Details from “Intelligence Reports” for the above towns, compiled between 1933 and 1936 by administrative officers, in C.S.O. 26. nos. 30169, 29800, 31318, 29834 and 10995 respectively (NAI).

112. Though the Elegbaji are linked with the Elegbe is being drawn from the same quarters, the name has no etymological link. Elegbaji means those who pay 400 cowries (egbaji) to the Owa as tribute on the installation of their chiefs. Elegbe means those who lead associations or bands (egbe) of citizens in the quarters.

113. For a persuasive and extended statement of the view that the institutions of the larger Yoruba kingdoms should be viewed as transformations of those found in the “mini-states” of the culture area, see Obayemi, A., “The Yoruba and Edo-Speaking Peoples and Their Neighbours Before 1600” in Ajayi, J.F.A. and Crowder, M., ed., History of West Africa (2d ed., London, 1977), Vol. 1, 201–09.Google Scholar The main limitation of this view, of course, is that the transition from “mini-state” to “mega-state,” involving the institution of kingship, is something of an unexplained quantum-jump, dependent on external influences.

114. Ibid, 207n33.

115. The norm of six for the iwarefa of the community possibly is related to the threefold military organization of the archaic northeastern Yoruba (Ona center, Otun right, Ohi or Osi left), on which see ibid, and Lloyd, , “Political and Social Structure,” 208–10Google Scholar, both referring to E. Krapf-Askari's work. We find them linked in the organization of Ilesha's newer quarters, whose iwarefa begin with the quarter-head, the Lotun and Losi, the lower three being in some sense their alternates (and often styled Risa, Risa Lotun, Risa Losi). Threeness is clearly established as a feature of the Ogboni (see Thompson, Black Gods and Kings, Ch. 6/2), who are clearly connected with the iwarefa of the archaic Yoruba community. See below note 118.

116. The Benin variants of these titles are Oliha, Edohen, Ezomo, Ero, Eholo, borne by the Uzama, the ancient title-holders who welcomed the existing dynasty's putative founder, Oranmiyan, from Ife (Bradbury, R.E., “Patrimonialism and Gerontocracy in Benin Political Culture,” in his Benin Studies, 129–146).Google Scholar Since two of the seven Uzama - Oloton and Edaiken - were of later foundation, it is evident that during the so-called Ogiso period before the constitution of the full sacred kingship, Benin too must have had the typical council of six, the ruler included. In Ijesha these titles occur in the oldest stratum of the title-system and in communities established well before close links were established with Benin in the time of Atakunmosa - along with one or two titles more peculiar to Ijesha, such as Saba and Sajuku - so it is unlikely that they were borrowed from Benin. Most likely Bradbury was correct in his suggestion that they are evidence for a “pre-Oranmiyan phase of political development affecting both the Yoruba and the Edo” (“Historial Uses of Comparative Ethnography With Special Reference to Benin and the Yoruba,” in ibid. 11-12). Other titles found in Ijeshaland or the southeastern Yoruba seem more likely to have been borrowed from Benin: Sasere, Logbosere, Osodi, and Sawe.

117. Biobaku, S.O., “Ogboni: the Egba Senate,” Third International West African Conference 1956Google Scholar; idem, “Historical Sketch;” Morton-Williams, P., “The Yoruba Ogboni Cult in Oyo,” Africa 30(1960), 362–74CrossRefGoogle Scholar; Atanda, J.A., “Yoruba Ogboni Cult: Did It Exist in Old Oyo?” JHSN 6(1973), 365–73.Google Scholar

118. There was a tradition in nineteenth-century Abeokuta that “Ogboni chiefs are, according to the original constitution of the country, the rulers.” Iwe Irohin (1865)Google Scholar, quoted by Abraham, R.G., Dictionary of Modern Yoruba, (London, 1958), 453Google Scholar; see too Biobaku, “Historical Sketch.” Does this mean that what had been merely the chiefly order of the archaic community, the iwarefa, assumed the form of a cult group in the disorder of the nineteenth century; or, conversely, that a titled cult group was the controlling body of the old towns of the Egba forest? Biobaku implies the latter, but does not pose the problem of whether the features of the nineteenth-century Ogboni at Abeokuta can be retrojected to the old Egba towns. But a shift in the character of the Egba Ogboni is strongly suggested by the fact that its titles are of two quite distinct kinds: old communal iwarefa titles like Lisa (cf. Ijesha Risa, Edo Oliha) or Aro, and those that denote the specific form of the cult group, such as Oluwo (“master of mysteries”) or Apena (“convenor”), which do not exist outside the area of the modern Ogboni cult. I am grateful to Dr. Agneta Pallinder for clarifying some of these points as they regard Abeokuta.

119. A new Owa has to visit Ijebu-jesha before his installation at Ilesha, and sit in the lap of its Ogboni. Arojojoye II, Itan Kukuru fun Isedale Ilu Ijebu-jesa ([Ilesha?] n.d.).

120. ChiefMalomo, J.O., Agbayewa, interview, 23 July 1974.Google Scholar

121. A number of titles are said to be “under” the Ogboni, which means not that they are tied to his lineage (though Sajowa and Batisin are) but that they approach the Owa through him and that he plays a role appointing their holders. It may derive from their originally having been clients of the Ogboni, though in the case of the Salotun (who came from Erin) there is no specific tradition of this.

122. Ode has the general meaning in Yoruba of “outside” and so its meanings of “town” or a kingdom's capital (as in Ijebu-Ode or Ode Ondo) seem to follow a primary reference to the outside of the king's Afin. The derivation of Iwude, the central rite of the Ogun cycle, would seem to be “entering the outside” (iwo ode), and reminds us that for the rest of the year the king's place was in his Afin.

123. Inquiries among Alapokurudu chiefs brought no consensus as to why they are called this. It was suggested that they might have carried “short bags” (apo kurudu) to the Afin to take away their portion of whatever was to be distributed, but no such satchels appear to be part of their turn-out today.

124. Interview with Babatope, J.A., then Odole-elect, now Chief Odole, 27 August 1974.Google Scholar

125. ChiefOyewumi, D. the Risinkin (interviewed 22 July, 11 Sept. 1974)Google Scholar claimed that the Obanla title had been hereditary until the time of his own ancestor Obanla Ojege, a famous and powerful chief. Ojege got his fried Uyiarere installed as Owa and subsequently his influence provoked a riot in which he and most of his family were killed, and the title was taken away from his lineage. The son who escaped did later become Obanla (Ogbolu), but the title remained open. Obanla Ojege's lineage is a well-known one in Ilesha (Peter Apara, referred to in note 51, was connected with it) and its members have often taken open titles (Obaodo, Lokiran, and Risinkin).

126. E.g. Saba, Sajuku, Aralopon.

127. Segbua traditions are confused. In Itan Ilesa Abiola says the first Segbua came from Efon and does not call it an open title (oye omo ilu). The current Chief Segbua in 1974, who had then held the title for forty years, said it was an open title and that its first holder came from Ijero, also in Ekiti. Such contradictions are likely in the case of open titles, since any traditions about the origin of the first holder may be confused with traditions about the origins of the lineage of the current holder. The origin of Ejemo is also disputed. Abiola says “Ado” which, unqualified, is more likely to refer to Benin than to Ado-Ekiti. It was the present Ejemo, Chief Kayanfada (interviewed 23 August 1974) who specified Owo. In any case a southeasterly provenance seems clear. No explanation was given for why the title did not become hereditary in the lineage of this first holder. Itaji-Ekiti tends to be assumed (e.g. Obayemi, , “Yoruba and Edo-Speaking Peoples,” 253)Google Scholar, but ChiefAduroja, , the Lejoka in 1974 (interviewed, 20 August)Google Scholar, said that Itaji was a small village, existing no longer, near Ilesha. This fits with the strongly indigenous character of Owari, whom the Lejoka worships (see below). Itaji is also the name of one of the quarters of Ibokun.

128. Thus three or four agnates or other close kin of Risinkin Oyewumi (note 125 above) have held that title during the past hundred years, while the Lejofi title has been dominated by two families, Alaka and Esan (ChiefAlaka, M.O. the Lejofi, interviewed, 1 August 1974).Google Scholar

129. IJAMO people do not carry Ondo marks but their chief, the Sawe, has a title of Ondo origin. At Ondo it is Sagwe, which is likely derived from the Osague title at Benin. The IJAMO people under Sawe related to the center through Chief Loro, the earlier Ondo immigrant, whose origin was also preserved in his cult of Ogun Ijamo, a distinct version of the god of iron: ChiefAkinola, S., the Sawe, interviewed 23 March and 13 April 1974.Google Scholar On Ondo, see also Lloyd, P.C., Yoruba Land Law (London, 1962), chapter 5.Google Scholar These well-retained Ondo memories find an interesting echo in the brief published history produced by the Awosika Family Association at Ondo: Iwe Itan Mode (Ondo, n.d.). This large and locally renowned lineage descends from a daughter of the Osemawe of Ondo whose son, having failed to achieve the Lurowo title, migrated to Ilesha where he was made Loro. His daughter subsequently returned to Ondo, being advised by a diviner that only in this way would she conceive children, and from her the later Ondo lineage descends. Precise details of names and times are impossible to establish - in particular whether the original migration to Ilesha is really to be identified with the arrival of the Loro or with the later, chronologically more likely, arrival of the people who settled IJAMO - but the tradition strongly confirms the retention by the Ondo migrants of a distinct identity.

130. See Lloyd, P.C., Political Development of Yoruba Kingdoms in the Eighteenth and Nineteenth Centuries (London, 1971), chapter 4Google Scholar; Awe, Bolanle, “Ibadan, its Early Beginning” in Lloyd, P.C., Mabogunje, A.L., and Awe, B., eds., The City of Ibadan (Cambridge, 1967), chapter 2Google Scholar; Elgee, C.H., The Evolution of Ibadan (Lagos, 1914)Google Scholar; and Morgan, Kemi, Akinyele's Outline History of Ibadan, rev and enl (2 vol,: Ibadan, n.d.).Google Scholar

131. Thus Adetugbo, A., “The Yoruba Language in Yoruba History” in Biobaku, , Sources, 184.Google Scholar

132. Something of the converse happened in Ilesha between the 1870s and the 1890s, when the prior traditional chieftaincy structure, undifferentiated compared with Ibadan's since even the Elegbe had important civil roles as quarter-heads, could not adapt to the need to maintain a permanently mobilized army. An informal Ibadan-type group of titles, under Ogedengbe as Seriki, began to emerge in the war camps but was not institutionalized at the end of the war, its leaders being re-integrated into the old structure, variously as Obanla, Odole, Lejoka, Lejofi, etc.

133. The Bajimo has responsibility for relations between the Owa and the members of the royal lineage who are Loja or village-heads. Apart from his headship of ITISIN quarter, Batisin's functions in relation to the lineage of Chief Ogboni are not clear to me, though he does not preside over lineage meetings in the way that Yegbata does for those of Loro. The title was vacant during my fieldwork.

134. On his way from Ijebu-jesha to the Afin the new Owa stays a night in Lemodu's house. Lemodu shaves his head, takes him to worship Obokun at the shrine in his house, and presents him with two calabashes, one with gold in it and one with stones, to divine the nature of his reign. The Lemodu takes all the Owa's old clothes and destroys them, and the Owa must never again see Lemodu's house, where he was an ordinary man for the last time, for henceforth he is greeted as an oba. Interview with ChiefAjayi-Obe, J.M., Lemodu, 21 April 1974.Google Scholar

135. Wheatley, P., “Significance of Traditional Yoruba Urbanism,” Comparative Studies in Society and History, 12(1970), 393–423CrossRefGoogle Scholar; though Law, , Oyo Empire, 30Google Scholar, expresses reserved dissent from Wheatley's view, he does not advance much counterevidence, which, indeed, for the early growth and character of Old Oyo must be scanty indeed.

136. As Itan Ilesa puts it (chapter 7): “Owaluse was a very hard ruler (je oba ti o le pupo), since he would flog anyone he had a mind to. The whole town grew angry at that sort of cruelty, so they conspired (di ote mo) against him.”

137. This is not emphasized in Abiola's Itan Ilesa, though it is implicit in Ekundare's account of the founding of IDASA. (See note 144). Egharevba, , Short History, 33Google Scholar, says “Atakumarha” was banished to Benin, on account of his tyranny and that the Oba persuaded Ijesha elders to take him back.

138. Abiola et al., Itan Ilesa Chapter 12.

139. Thus in the itan of Atakunmosa, Abiola et al., begin their account of what took him to Benin with his desire to increase Ilesha's population (“his town was not as full as he wanted,” ilu on ko kun to bi on ti nfe), and his search for a powerful medicine, made from a horse which was buried on the boundary with Benin. He eventually returns from Benin with a further powerful charm and ritual specialists. There follows an account of the general regional hegemony Atakunmosa established for Ilesha especially in Ekiti, and of the new quarters he founded in Ilesha.

140. Hopkins, A.G., ed., “A Report on the Yoruba, 1910,” JHSN 5(1969), esp. 76.Google Scholar This was written by six educated Africans, including the Ijesha C.A. Sapara-Williams. They say that slaves taken in war were delivered to the oba who “causes such redistribution as he considers desirable.”

141. As one rueful proverb puts it: Ijesa omo alale igbajoji: “Ijesha, more welcoming to strangers than to her own sons.”

142. These are represented in tradition as small hamlets and are variously named (for one version, fuller than Abiola et al on this point, see Oni, J.O., History of Ijeshaland, Ile-Ife, n.d., 27).Google Scholar Some of their names (e.g. Ibosirin, Itakogun) survive as localities in Ilesha, but their chiefs seem to have died out, never being incorporated into the developing title system, and one presumes their populations were absorbed into the founded quarters.

143. The Lejofi title is said to have existed before the Loro came, and a number of very ancient rural communities in the close vicinity of Ilesha (e.g. Oke Omo and Oke Awo) had him as their onile.

144. Ekundare, , Iwe Itan, 16–17 and interview, 28 August 1974.Google Scholar

145. Except for these Lotun, nearly all titles in these quarters were open. Of the nineteen quarters, sixteen had Lotun drawn from Loro's lineage and three from Odole's which also secured the iwole presence.

146. Data from 1974 Sample Survey. For practical reasons, the sample was stratified by quarter and the households randomly selected from a randomly chosen bloc within each quarter. This meant that the sample came to include several clusters of houses which had been carved out of a formerly existing large compound of a particular chief (e.g. the cluster of houses of descendants of Saloro Omirin at IWERE). Some of the settlement is modern, especially where members of a lineage have had land granted in the vicinity of the chieftaincy house (e.g. the three not very closely related members of Odole, who have this century acquired house-land in ARAGAN), probably on ground once occupied by slaves and other dependents of the current titleholder. Much of the “older” (i.e. pre-twentieth century) settlement is in the “newer” (Omodeowa-led) quarters, and many of the major recognized segments of the larger titled lineages are identified by the quarter from which they had come.

147. For these details I am indebted to Michael Ekundare for showing me his manuscript history of IFOFIN (of which quarter he is the secretary), “Iwe Itan Adugbo Ifofin.”

148. For example, OMOFE quarter contains the houses of the descendants of Lokiran Oruru (the historian J.D.E. Abiola's grandfather), and of the Asaolu family, descending from one segment of the household of Lejoka Danaija (others sited elsewhere). These were both nineteenth-century Elegbe chiefs, so the lineages probably do not go back before then to the foundation of the quarter in the seventeenth century. Nonetheless it implies continuity of an earlier pattern.

149. Isape can hardly have survived Ilesha's sack in 1870 and is only tenuously recalled now. Little can be said about its organization, except that its head, the Lusape, was linked with Chief Ogboni. Apparently a kind of overflow from the Afin, it housed slaves who might be required for sacrifice or sent for for some other purpose - hence its name, “run and call” (sa … pe) - or so folk etymology has it.

150. Interview, Luroye and chiefs, 5 January 1975. The site of this compound is recalled, but naturally, no present resident of IROYE claims descent from its members.

151. Ilesha thus seems to exemplify well what has been aptly called “a particular kind of consumer society - one in which social groups have a driving urge to take in new persons:” Miers, Suzanne and Kopytoff, Igor in their excellent collection Slavery in Africa: Historical and Anthropological Perspectives (Madison, 1977), 64et passim.Google Scholar

152. Abiola et al., Itan Ilesa, chapter 16; also interview with ChiefOwolabi, I.A. the Obaodo (who, incidentally, represented Wayero as male), 12 January 1974.Google Scholar

153. Abiola et al., Itan Ilesa, chapter 17.

154. Ogboni of Ibokun, interview, 4 July 1975.

155. Law, , Oyo Empire, esp. 308–312Google Scholar; Lloyd, , Political Development, 47–53.Google Scholar

156. For example, see Wilks, Ivor, Asante in the Nineteenth Century (Cambridge, 1975), 127Google Scholar, where he speaks of government being extended “in range, to control areas far distant from the metropolitan region; in scope, to regulate spheres of activity previously untouched by central authority; and in proficiency, beyond the managerial resources of the hereditary aristocracy of the pre-imperial period.”

157. Political Development, 1, 3.

158. Fortes, Meyer and Evans-Pritchard, E.E., eds., African Political Systems (London, 1940), esp. 5–14Google Scholar; Goody, J., Technology, Tradition and the State in Africa (London, 1971), esp. chapter 3Google Scholar; Bradbury, , “Historical Uses of Comparative Ethnography,” 3–16.Google Scholar

159. By this he must here mean regional concentration: Lloyd, , Political Development, 50–51.Google Scholar

160. That trade routes are a necessary, but not a sufficient condition for the formation of African states, we are reminded by Terray, E., “Long Distance Exchange and the Formation of the State,” Economy and Society, 3(1974) 315–45CrossRefGoogle Scholar and the greater part of Igboland, in southeastern Nigeria, is testimony to the same.

161. The unpublished Iwe Itan Oko Apara (cited in note 51 above) says that “sons of great Ilesha chiefs” (omo ijoye nlanla Ilesa) built houses at the market of Omirinmirin on the border with the Oyo at the river Oshun. Johnson used an almost identical phrase, “all the children of influential Oyo chiefs” of the Oyo-Nupe border market on the Niger at Ogodo (cited in Law, , Oyo Empire, 211Google Scholar). In his statement of 1882 (see note 15) Meffre stated that “chiefs” collected tolls at Ilesha's seven gates, being also responsible for the “welfare” of subordinate towns in that direction. This sounds like a highly simplified account of the onile system. The toll-collecting chiefs were almost certainly the Obanla and members of the Elegbe whose houses, as Map 1 shows, were situated well out on the radial routes in the town.

162. See, for example, the letter of Governor Carter, written from Ilesha on 23 August 1894 (in CMS 1/1 14, NAI) in which he stressed the weakness of Yoruba rulers, and contrasted Ilesha, a typical case in his opinion, with Iwo, whose ruler's power he mistakenly attributed simply to Islam. Lloyd, , Political Development, 38–39Google Scholar, makes it clear that Iwo was also liable to oscillations of power.