Filling a Gap in the Ife-Benin Interaction Field (Thirteenth-Sixteenth Centuries AD): Excavations in Iloyi Settlement, Ijesaland Author(s): Akinwumi O. Ogundiran Source: The African Archaeological Review, Vol. 19, No. 1 (Mar., 2002), pp. 27-60 Published by: Springer Stable URL: http://www.jstor.org/stable/25130734 . Accessed: 15/02/2011 15:51 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=springer. . Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Springer is collaborating with JSTOR to digitize, preserve and extend access to The African Archaeological Review. http://www.jstor.org African Archaeological Review, Vol. 19, No. 1,March 2002 (? 2002) Filling a Gap in the Ife-Benin Interaction Field Centuries AD): Excavations (Thirteenth-Sixteenth in Iloyi Settlement, Ijesaland Akinwumi O. Ogundiran1 Previous archaeological studies have indicated that the Yoruba polity oflle-Ife and the Edo polity of Benin, both in southwest Nigeria, belonged to the same sphere of sociocultural interactions before the nineteenth century AD. The spatial and temporal dimensions of this interaction sphere have not, however, been under stood, because the archaeological sequences of the areas between the two polities are largely unknown. One of these intervening areas is Ijesaland. The excavations conducted in Iloyi settlement, northern Ijesaland, provide a new set of data that not only fills a gap in the Ife-Benin interaction sphere but also offers new perspec tives on the process of material culture homogenization in the Yoruba-Edo region during thefirst half of the second millennium. Calibrated radiocarbon dates show that Iloyi was occupied during the thirteenth-sixteenth centuries AD. Using the stylistic and iconographie characteristics of ceramics and the patterns of burial and sacrificial rituals as evidence, it is demonstrated that Iloyi was a sociopolitical and cultural frontier oflle-Ife, and that Ijesaland was part of the Ife-Benin cul tural corridor. The paper strengthens the earlier suggestions that the development of a kingship institution at Ile-Ife helped to widen the interaction networks in the region, an historical process that culminated in the trend toward regional cultural homogenization between the thirteenth and sixteenth centuries. Des ?tudes arch?ologiques ant?rieures ont indiqu? que l'?tat Yoruba de Ile-If e et V?tat Edo du Benin, les deux dans le sud ouest du Nigeria, appartenaient ? la avant le dix-neuvi?me si?cle apr?s J.-C. Pourtant, m?me sph?re socio-culturelle les dimensions spacio-temporelles de cette interaction n 'ontpas encore ?t? enti?re ment comprises, car les s?quences arch?ologiques des r?gions entre les deux ?tats restent ? d?couvrir. Uune de ces r?gions est Ijesaland. Les fouilles entreprises ? Iloyi, situ? au nord de Ijesaland, ont divulgu? l'information nouvelle sur la sph?re 1 Florida International Florida Park, Miami, of History, University, Department 33199; e-mail: ogundira@fiu.edu. College of Arts and Sciences, University 27 0263-0338/02/0300-0027/0 ? 2002 PlenumPublishingCorporation 28 Ogundiran d*interaction entre Ife et Benin ainsi que r?v?l? des nouvelles interpr?tations du de la culture mat?rielle dans la r?gion de d?veloppement de Vhomog?n?isation Yoruba-Edo durant la premi?re partie du deuxi?me mill?naire. Sur la base de tests de carbone, on sait que Iloyifut occup? de treizi?me au seizi?me si?cle AD. Les caract?ristiques stylistiques et iconographiques de la c?ramique ainsi que et les rites de sacrifices laissent ? penser que Iloyi les proc?d?s d'enterrement et culturelle de Ile-Ife et que Ijesaland ?tait situ? ? la fronti?re socio-politique de de culturelle la zone faisait partie Ife-Benin. Ceci renforce Vhypoth?se que le institution d'une royaliste ? Ile-Ife ? ?tendu les r?seaux d'inter d?veloppement action de cette r?gion?un processus historique qui culmina avec la tendance ? r homog?n?isation r?gionale culturelle du treizi?me au seizi?me si?cles. KEY WORDS: regional interactions; Ijesa; Yoruba; Edo; ceramics; burial rituals. INTRODUCTION in the This paper revisits the question of cultural-historical relationships Yoruba-Edo southwest between the sixteenth of thirteenth and Nigeria region centuries (Fig. 1). Despite the region's ethnolinguistic differences, archaeological evidence has shown that Ile-Ife and Benin, two of themajor pre-nineteenth century sociopolitical centers in the region, shared a variety of stylistic and iconographie in material culture (Connah, 1975; Eyo, 1974a; Garlake, characteristics 1977; Omokhodion, 1988; Willett, 1967a). There are also oral historical narratives of and economic interactions between Ile-Ife and Benin during the sociopolitical period covered by this study (Babayemi, 1981 ;Egharevba, 1968; Obayemi, 1985). These lines of evidence prompted Eyo (1974a, p. 410) to suggest, over 25 years it is less harmful to assume that both ago, that "on the Ife-Benin relationship,... centers belong to a cultural continuum, a continuum which may be much wider than we can at present imagine." At the core of this statement is the quest to understand the spatio-temporal dimensions of the Ife-Benin "cultural continuum," an understanding that has eluded us for so long because the archaeological sequence and cultural history of most areas in the region are unknown. One of these areas is Ijesaland, a Yoruba subgroup adjacent to Ile-Ife (Fig. 1). Excavations were carried out at Iloyi settlement in 1997 to assess the cultural historical relationships between Ijesaland on one hand and Ife, Benin, and Owo on the other. This paper presents the results of the excavations in Iloyi settlement, an early political center in northern Ijesaland (Fig. 2). The primary goal of this paper is to identify the diagnostic archaeological parameters of cultural-historical relationships in Ife, Benin, Ijesa, and Owo areas. The task of mapping the topology of cultural-historical relationships in these areas involves identifying the shared of material culture and the archaeological vestiges of cultural prac stylistic aspects tices. To this end, emphasis is placed on the comparison of ceramic characteristics and burial/sacrifice patterns in the region between the thirteenth and sixteenth centuries. The paper also examines, in the light of recent historical analysis and The Ife-Benin Interaction in Ijesaland Field 29 3?E 7 9?N Old Oyo OYO t/ \ t i N \ EGBA Ondo? AWORI # Ijobu-Ode ONDO 0 Lagos 6?NH ATLANTIC 0 40 Fig. 80 120 KM 1. Yoruba-Edo archaeological the diagnostic OCEAN region, showing major towns (Ile-lfe) the articulations of regional indices that characterize archaeological evidence, and subgroups (IFE). interactions that produced the Ife-Benin region. ILOYI SETTLEMENT:ARCHAEOLOGICAL FIELDWORK Background Information Iloyi, otherwise known as Early Hare, is an abandoned settlement in Ijesaland about 50 km north of Ife. The local historical traditions identify Iloyi as the center 30 Ogundiran 4?48'E 4?45'E 2 KM 0 Elevation |-1 Fig. 2. Topography 350-580m Range: of Hare district, showing above sea level the study area. of Eka Osun sociopolitical complex, one of the earliest sociopolitical formations in the Ijesa region (Abiola et al., 1932; Ogundiran, 2001). The excavations at Iloyi were part of the pioneering historical-archaeological in northern investigations were at the and aimed sequence of material Ijesaland defining chronostratigraphic culture. Iloyi is located in the upper reaches of the rainforest belt, and its geology is characterized by metamorphosed quartz, gneiss, and schist rocks. The natural soils of the area are clayey in texture, ranging in color between reddish brown and brownish red, and they contain small amounts of white micaceous minerals. the soils are fairly heavy in clay, the percentage of gravel and quartz Although increases with depth, especially from 25 cm below the ground surface (Ekanade, 1984, p. 26). The abandoned settlement of Iloyi and its vicinity are now intensively used for tree and food crop cultivation. The major crops are cocoa (Theobroma cocoa, L.), kolanut acuminata (Cola (Citrus spp. L.), oil palm cassava (Manihot esculenta, and C. (Elaeis guineensis, Crantz), Excavations and maize nitida, Schott, Jacq.), yam (Zea mays, and Endl.), (Diascorea, oranges spp. L), L.). and Chronology reconnaissance was conducted to determine the spatial extent Archaeological of the settlement. To this end, 20-m grids were laid over the settlement in order to identify and map its archaeological features with a fair accuracy. Linear mounds of The Ife-Benin Interaction Field in Ijesaiand 120cm Fig. 3. Outline of Iloyi settlement and excavation sites. collapsed and decaying house walls, refuse mounds, earth enclosures, intrasettle ment pathways and alleys, and scatters of ceramics are the surface archaeological markers that were identified in the settlement. The settlement's perimeter walls survived as earth embankments that vary between 0.7 and 6.1 m in height and and 2.5 m in width. The extent of the archaeological markers indicates that settlement was about L08 km in width before its abandonment. The outlines collapsed house structures are still visible on the surface, with the exception of 2.2 the of the settlement's southern edge, where farming activities have obliterated the structural remains, leaving only scatters of potsherds (Fig. 3). The archaeological excavations were exploratory, and the goal was to gain a basic understanding of the chronostratigraphic sequence in Iloyi. Two mounds were therefore selected for excavation in the expectation that they would yield deep stratified deposits for understanding the occupation sequence, material culture, and chronology of the settlement. Two test Units, 1-A and 1-B, were excavated in Site 1 (3 x 1m and 2 x 2 m respectively), and a 2 x 1m unit was excavated at the summit of Site 2 (see Fig. 3). Four cultural strata (I-IV) were identified in each of the two units in Site 1 (Fig. 4; Table I). The humus sediment of Stratum I inUnits 1-A and 1-B was succeeded by a matrix of reddish brown, semicompact, gravely sand consisting of hard, reddish clay lumps, and quartz pebbles (Stratum II). The Stratum II deposits inUnit 1-B contained ceramic disks similar to those identified in the archaeological deposits of Ile-Ife as tiles for decorating walls and columns (Garlake, 1977, pp. 71-72). The occurrence of ceramic disks in association with reddish clay lumps and quartz pebbles in Stratum II indicate that the level was 1977, composed of the remains of decayed daub earth structures (see Mclntosh, pp. 191-192). An important feature in Site I was the identification of a human 32 Ogundiran Fig. 4. Stratigraphie profiles and excavation units. in Stratum III of Unit 1-A. Site 2 is a refuse mound, and the test Unit revealed 2.2 m deep deposits consisting of seven stratified cultural levels. (2-A) burial Three charcoal samples, one sample from each excavated unit, were submitted for 1-A, a radiocarbon date of 730 ? 155 bp was obtained from Stratum II, 715 ? 145 bp for Stratum IV in Unit 1-B; at Unit 2-A a charcoal sample in Stratum V at 140 cm below the surface yielded a 485 ? 65 bp date. The calibrated ages of the radiocarbon dates in Hare district indicate that Iloyi settlement could have been occupied from the late tenth to the early seventeenth centuries (Table II). However, the intercepts of the calibrated dates, shown in parentheses in Table II, and the oral historical framework of settlement trajectories indicate that the archaeological deposits were most likely formed between the thirteenth and radiocarbon dates. InUnit sixteenth centuries AD. Artifacts and Archaeological Features The artifacts from the three excavated units are summarized in Table III. The categories of artifacts represented in the archaeological deposits are pottery, ceramic wall tiles, lithics (grinding stones, querns, and a stone-ax fragment), and artifacts (10 iron knives, 3 nails, 7 fragments of slag, and 1 of each of the following: copper linked chain, copper ring, and iron bangle). The other artifacts metal are shell and glass beads. Decapitated human remains were also excavated in The Ife-Benin Table I. Description Unit Strata Field Interaction of the Stratigraphie Strata brown (5YR 3/4) 1-A, 1-B, and 2-A 1-B Description Unit Strata brown (5YR 3/4) (5YR 5/3) Very compact clay with quartzite stones, reddish yellow (5YR6/6) Very compact clay with quartzite and cobble stones, brown (10YR4/4) (5YR 5/3) IIA Loose ashy silt, brown III Very loose silty sand with very small quartzite stones, brown IV Very loose mixture of ash and sand, dark brown (7.5YR 5/4) (10YR5/3) IIB III Very loose mixture of ash and sand, dark brown (7.5YR4/2) (7.5YR4/2) Very loose sand mixed with ash, reddish brown Very loose silt with quartzite chips and soft chalky stones, pink (5YR 4/4) reddish clay (2.5YR4/6) 2-A Description yellowish (7.5YR 5/4) IV (see Fig. 4) Humus sediment, dark reddish brown (5YR 3/4) Loose sand, dark Semi-compact gravely sand, reddish brown Semi-compact gravely sand, reddish brown Very compact sandy clay interspersed with small gravel stones, brown for Units Humus sediment, dark reddish Humus sediment, dark reddish III Profiles Unit 1-A Description 33 in Ijesaland IV (5YR 7/4) Compact clay with gravel, red VI (2.5YR 4/6) Very compact clay with cobbles and gravel, red (2.5 YR 4/6) VII Very loose silty sand, black (5Y 2.5/2), intermixed with very fine white ash sand with Compact quartzite and cobbles, light reddish-brown (5YR 6/4) VIII Very compact sandy clay with quartzite stones and cobbles, red (2.5YR 4/6) Site I, in a context that suggests burial and/or sacrifice rituals. The focus of the following discussions will be on the ceramics and the archaeological vestiges of rituals because of their diagnostic attributes in understanding cultural-historical region. relationships in the Yoruba-Edo Ceramics Iloyi was one of four settlements where excavations were conducted in the area in 1997, along with Hare, Okun, and Iloja (Fig. 2). A formal classification Ijesa 34 Ogundiran Table H. Location of charcoal Calibrated sample Site 1, Unit A Stratum II (50-cm depth) Site 1, Unit B Stratum IV (90-cm depth) Site 2, Unit A Stratum V(140-cm Age(s) Lab. no. depth) of Radiocarbon C14 age (bp) Dates0 Two-sigma calibration GX-23998 730 ? 155 cal.AD 998 (1280) 1448 GX-24000 715 ? 145 cal.AD 1021 (1290) 1447 GX-23999 485 ?65 cal.AD 1326 (1430) 1620 in Cambridge, MA. All the radiocarbon flThe charcoal samples were dated at Geochron Laboratories dates reported in this study, including the referenced ones, are calibrated with the CALIB software standard error limits rather than one-sigma. The (Stuiver and Reimer, 1993) using the two-sigma chance of missing is reduced to 1 in 20 whereas with one-sigma, the true date with two-sigma the chances are 1 in 3 that a given measurement will not straddle the true age value of the samples. The dates in parentheses are the intercepts of the calibrated dates and represent the closest dates of the samples (Mclntosh and Mclntosh, 1986, p. 416). system, based on decoration attributes and vessel forms, was developed for the ceramic assemblages from all four settlements (see Ogundiran, 2000, pp. 319 416, for details). The terms for describing the decoration attributes of pottery and terracotta sculptures in Yoruba-Edo archaeology adopted in this paper are those used (1983), Eyo (1976), Connah (1975), Garlake previously by Agbaje-Williams (1977), andWillett (1967a,b). However, the decoration patterns of the ceramics are Table Unit/strata Unit 1-A I II III IV Total Unit 1-B I II III IV Total Unit 2-A I II III IV V VI VII Total Pottery fragments 551 908 901 146 2506 710 914 1145 328 3097 201 386 832 156 341 80 79 2075 III. Ceramic Iloyi: Categories tiles Clay lumps of Finds Lithic artifacts Metal objects Beads 15 15 34 10 3 34 13 The Ife-Benin Interaction Field in Ijesaland 35 defined with reference to both themethod/tool used and the specific type of motifs. A combination of both tools and motifs allows us to differentiate, for example, between incised geometric patterns and stamped geometric motifs, or between stamped dot punctates and stamped circular motifs. This approach provides fine scale units of analysis that aremore effective for ceramic stylistic comparison than one that simply tells us what tools or methods were used in ceramic decoration (e.g., roulettes, stamps, incision), since the latter procedures for ceramic decoration tend to be universal. The typological characteristics of the vessel forms in Hare district are defined following the approaches developed by Connah (1975) for Benin and by Garlake (1977) for Ile-Ife. Despite the difference of typological terminology used in these three studies, the goal was to describe ceramic forms in all their morphological and functional manifestations; using body shape and rim, lip, and shoulder profiles as well as decorative attributes. Eighteen vessel forms are identified in Iloyi, all belonging to four ceramic classes: jars, dish-bowls, colanders (perforated vessels), and lids. The vessel forms in each class are coded using a combination of a letter and a number. The codes start with J for jars, B for bowls, C for colanders, and L for lids. Thus, for example, Jl through J5. the five vessel forms in the jar category are numbered Within the three excavated units, 7678 ceramic sherds were collected. The surface colors of the ceramics range between yellowish red and dull brownish yellow, and the inclusions are mostly quartz grains and mica. Only 70.5% of the total sherds are decorated while the remaining sherds (29.5%) are plain, and only 685 sherds have diagnostic rim profile, shoulder morphology, and body contour fea tures that are amenable to identifying vessel form characteristics. Fifteen decorative attributes and 17 vessel forms are identified. Burnishing, cord roulettes, carved and 91.7% of the total roulettes, grooves, and incised lines account for 90.1,93.2, decoration attributes inUnits 1-A, 1-B, and 2-B respectively (Table IV). Of these, cord roulettes account for 61-66% of the attributes. The occurrence of the other motifs and attributes is less than 10% in the ceramic assemblage of Iloyi. These low distribution attributes are however the "trace elements" that are susceptible to the assessment of regional cultural-historical relationships. Similar decorative motifs are found in each of the three excavated units, although there are stratigraphie variations in the proportions of ceramic attributes. Although similar vessel forms are also found in each unit, there are more notable differences in their distribution frequency (Table V). An attempt is made here to assess whether the stratigraphie variations of the ceramic characteristics are tantamount to time-related trends. To this end, the distribution of decorative attributes and vessel forms from the thirteenth century deposits in Site 1 (all the strata) are graphically and statistically compared with the attributes from the post-thirteenth century deposits in Site 2 (Strata I-V). The chi square statistics and Cramer's V2 index are used to clarify, with some precision, <g- .2 t? <+H C? *3 t? IV 2Se 8?s 1 6.3 52.4 9.8 1.9 69.6 5.6 0.6 0.4 10.8 1.9 0.8 4.3 0.4 0.2 0.8 519 II 3.9 0.6 64.6 3.3 0.6 0.2 6.1 15.3 1.4 0.5 0.8 1.9 639 2.8I2.7 0.7 14.0 7.7 1.4 0.7 58.2 4.3 5.0 2.8 17.0 1.4 5.71.4 141 6.8 0.6 62.3 6.1 2.8 0.5 0.2 11.9 6.2 1.1 0.3 1.3 636 11.9 III 5.2 I4.3 0.8 0.3 3.5 0.2 II 2.4 1.0 63.1 7.6 3.1 4.5 0.3 0.7 15.2 1.0 0.7 0.3 290 0.9 ffl 6.4 2.0 67.0 ,.?? =s -s? ?s?2a811 IS ?.?8> SS 2??O >>? III I?3II 5I1gI-S I?S 3.8 0.10.6 0.3 Total 4.8 1.4 65.6 4.1 0.5 0.2 0.6 13.1 5.6 1.1 0.4 1.6 2266 6.3 Total 2.7 2.3 64.7 8.0 0.4 0.2 0.5 0.8 11.3 4.0 1.5 0.5 0.7 2.4 19044.8 61.6 ffl 2.7 2.8 64.8 7.2 0.5 0.3 10.2 4.9 2.0 0.2 0.8 2.5 597 I2.1 1.1 68.6 12.5 0.2 0.5 0.9 1.8 8.9 1.0 0.9 0.2 1.1 439 0.5 6.0 9.3 2.8 0.9 4.6 54.0 217 1.4 1.9 13th 17.1 IV 0.2 201 3.0 5.0 9.0 4.5 2.0 6.0 20.8 1.0 40.7 IV 0.4 3.3 667 0.6 0.3 3.0 69.1 0.1 0.9 1.3 6.7 11.1 2.4 13th IITotal 0.3 Percentages by Stratum0 Attributes: of VII 4.3 62.3 5.8 1.4 14.5 2.9 5.8 1.4 69 Decorative aThe table refers to the occurrence of 0.7 0.1 0.2 13.1 251 8.0 0.4 3.2 2.0 4.8 67.3 0.8 6.0 1.6 4.4 1.2 15th V decorative VI 1.1 2.2 20.9 54.9 6.6 2.2 9.9 2.2 91 number not of sherd attributes, ?The 1-A Unit dates are 1-B Unit based on 2-B Unit the intercepts radiocarbon the of c 2.8 11.1 dates (lidsand Total Unidentified I5.0 45.0 5.0 15.0 10.0 20.0 20 III 17.8 6.5 3.7 15.9 11.2 5.6 12.1 0.9 7.5 3.7 7.4 6.5 107 I23.0 2.6 20.5 20.5 5.2 17.9 7.7 2.6 17.9 39 1.1 1.3 0.6 1.1 4.2 2.2 3.7 1.7 1.2 18.4 7.2 23.2 III 16.0 6.41.1 16.0 13.8 5.3 14.9 12.8 6.47.5 23.2 3.7 94 35.4 25.0 1.2 12.5 25.0 3.6 16.8 24.1 5.3 0.3 3.2 12.1 9.9 14.5 6.7 2.5 0.3 11.0 2.5 4.6 2.8 283 5.3 9.9 14.7 Total 6.0I26.4 9.4 1.9 7.5 3.8 24.5 5.7 11.3 1.9 5.7 1.9 53 7.3 6.1 11.1 4.2 8 11.1 III 12.5 3.8 2.5 8.8 28.8 3.8 5.0 1.3 7.5 17.6 80 V 15th 20.8 6.3 37.5 10.4 12.5 2.18.3 48 13.6 4.5 4.5 9.1 13.6 23.0 4.5 9.1 22 11.1 II 16.8 3.2 0.6 30.2 9.5 33.3 7.8 11.1 0.44.3 6.3 9 1.2 5.0 Stratigraphie Distribution Table V. of Vessel Forms: Stratum by Percentages Unit/strata (century ADf JlJ2 J3 J4 J5 J6 28 3.6 14.3 39.3 13th IV B2 B3 Bl B4 B5 B8 B9BUB12colanders) rims number aThe dates are based on the intercepts C14 Others Approximate 1-A Unit 1-B Unit 2-A Unit radiocarbon the of calibration 38 Ogundiran whether there is a statistically significant difference in the distribution of ceramic attributes between the thirteenth and the fourteenth-sixteenth century deposits in on The statistical is based the settlement. analysis Iloyi following hypotheses: Hq: There is no relationship in the distribution of decorative motifs and vessel forms between the thirteenth century and the post-thirteenth century contexts (null hypothesis). H\ :There is a relationship in the distribution of decorative motifs and vessel forms between the thirteenth century and post-thirteenth century contexts. The probability significance (a) level for accepting or rejecting either of the hypotheses is set at 0.05. That is, in statistical parlance, we are prepared to reject or accept Ho or H\ if the asymptotic significance of the chi-square value is lower than the probability significance (SPSS, 1998, pp. 70-71). This means that the null hypothesis will be accepted if the distribution of the ceramic attributes is so unusual that they would equally occur 5 times out of 100 or less in the archaeolog the hypothesis of no difference (H\) ical deposits of the two periods. Otherwise, will be accepted (Sherman, 1988, pp. 78-79). The analysis demonstrates that the asymptotic significance (two-sided) of the chi-square value is below the a level (Figs. 5 and 6). Hence, the H\ is accepted, indicating that there is a statistical probability of difference in the distribution of decorative motifs and vessel forms between the thirteenth century (Site 1) and the Cramer's V2 values, centuries (Site 2). However, the fourteenth-sixteenth the statistical indices for measuring the strength of statistical significance (usually ranging between 0.0 and 1.0) for the distribution of decorative attributes and vessel forms are 0.12 and 0.39 respectively (Figs. 5 and 6). These low values indicate that the strength of the relationship in the distribution of ceramic attributes between the two periods (and sites) is weak. The chi-square analysis therefore demonstrates that there was no temporal trend in the ceramic sequence between the thirteenth and the sixteenth centuries in Iloyi. In other words, the choice and use of decorative motifs and vessel forms were broadly the same throughout the 300-400 years of Iloyi's occupation. THE DIAGNOSTIC ATTRIBUTES OF CERAMICS: THE ILOYI ASSEMBLAGES IN THE IFE-BENIN REGIONAL CONTEXT ceramic styles at a regional level is often The occurrence of homogenous indicative of the time-specific intersocietal contacts, especially when the styles are ascertained to be contemporaneous. The use of ceramic stylistic attributes in examining intersocietal relationships in this study is predicated on the premise that the degree of stylistic homogeneity between social groups varies directly with the degree of social and historical distance between them (MacEachern, 1994, I I I S o'E 5* 80.685 0.12 (less than probability 0.05) of value Significance Asymptotic 0.00 14th-16th |century Chi-Square Value 113th century index V Cramer's Fig. 5. Distribution of decorative attributes, thirteenth fourteenth-sixteenth and centuries. c8 40 Ogundiran 92.777 Chi-square Value 0.00 Asymptotic Significance (less than probability value of 0.05) Cramer's V index 0.39 13th century I ^ Fig. 6. Distribution of vessel forms, 14th-16th century *& *> *> ^ <&^ $r <& <%> ^> <%> thirteenth and fourteenth-sixteenth centuries. 1977). In this section, a comparison of the p. 220; Stark, 1998, p. 8; Wobst, ceramic attributes in Iloyi with those previously defined for archeological sites in Ile-Ife, Benin, and Owo is undertaken as a basis for assessing the cultural historical relationships in the region. It should be noted at the outset that this comparative endeavor is hampered by the lack of a standardized and uniform re gional typological scheme for reporting and describing ceramic attributes. This the level of precision to which two or methodological lapse no doubt weakens more assemblages can be compared in the region. Likewise, the quantitative or sequences to represent the dering of ceramic attributes into chronostratigraphic rise and fall of the popularity of ceramic attributes has been largely ignored by most previous excavators (for the use of battleship curves for comparing stylistic indices, see Ford, 1962; Marquardt, 1978). Moreover, a high level chronological resolution is not attainable by any dating technique, including radiocarbon dating, for the short period of occupation of Iloyi, and this makes fine-scale chronostrati 1986, p. 416). Given graphic sequencing very difficult (Mclntosh and Mclntosh, these problems, the ceramic assemblages in Ife, Benin, Owo, and Iloyi have been compared on the basis of the delineation of the diagnostic attributes in the eleventh sixteenth century deposits. We will first examine the nature of the archaeological sequence in Ife, Benin, and Owo, and then examine the relationships between the ceramic assemblages in these areas and those of Iloyi. The Ife-Benin Interaction Field in Ijesaland 41 ARCHAEOLOGICAL SEQUENCE IN ILE-IFE, BENIN, AND OWO The archaeological major cultural-historical sequence of Ile-Ife has been broadly delineated into three 1967a). These are: periods (Eyo, 1974a, p. 409; Willett, "Classic" (twelfth-sixteenth "pre-Classic" (pre-twelfth century), century), and Historical (sixteenth-nineteenth "post-Classic" century) periods. investigations indicate that the period began about the fifth century AD with the fusion of scattered independent villages into multivillage polities, each characterized by a central agency of coordination but without powerful royal dynasties, centralized governments, or urban centers (Obayemi, 1985, p. 261). A formal kingship in stitution and an urban center were forged from these loose sociopolitical unions between the tenth and eleventh centuries to herald what has been described as the Oduduwa or Classical period (Adediran, 1992; Olomola, 1992, pp. 51-61 Willett, ; 1967a). The Classical period has been the focus of most archaeological investigations in Ile-Ife. The period was characterized by the florescence of the production of naturalistic terracotta brass/bronze sculptures, extensive construction of potsherd pavements, and the construction of walls that delineated the urban space of Ile-Ife among others (Eyo, 1974a; Garlake, 1974, 1977; Ozanne, 1969; Willett, 1967a). The post-Classic period refers to the period after the sixteenth century when the production of naturalistic terracotta and bronze sculptures of the elite and royal patrons, and the construction of potsherd pavements, subsided. There are indications that the ceramic attributes of the pre-Classic contexts are different from those of the Classic and post-Classic periods in Ile-Ife. The study at Orun Oba Ado site, for example, revealed preliminary archaeological levels dated to the sixth-ninth centuries, and the ceram pre-Classic occupation ics from these levels are markedly different from the ones that were evident in post-ninth century sites in Ile-Ife (Willett, 1973, p. 130). The morphological and decorative attributes of ceramics that were established during the tenth and eleventh centuries at Ita Yemoo continued during the twelfth-fifteenth centuries inWoye Asiri and Obalara, and lasted until the sixteenth-eighteenth centuries in sites (Ogundiran, 2000, p. 362; Willett, 1973, pp. 126 Lafogido and Odo-Ogbe 127). In other words, there was continuity in the ceramic forms and decoration attributes at Ife from the early stage of the Classic to the post-Classic period. As was the case at Iloyi, cord and carved roulettes constitute about three-quarters of the total decorations atWoye Asiri, Obalara, Lafogido, and Odo-Ogbe sites (Eyo, 1974a; Garlake, 1977, p. 87). Paint, circular stylus impressions, and relief mo tifs, especially keloids, cowryforms, bosses, and animal representations, have low frequency in the archaeological deposits of Ife, but they are primarily associated with ritual vessels, whereas cord roulettes, straight-line grooves and incisions, and brushmarks are common to domestic vessels. Garlake (1977, p. 89) concludes, with reference to the ceramic assemblages at Obalara andWoye Asiri, that "there 42 Ogundiran is a clear impression of continuity and lip forms, the... placement of in the varieties of vessels, decoration, and their shapes,... the association of... types rim of between the twelfth and sixteenth centuries. vessel and decoration..." Connah (1975), on the basis of his excavations at theMuseum, Clerks' Quar ters, Ogba Road, Usama, and Reservation Road sites, has classified the archae ological sequence in Benin into three periods: Early period (thirteenth-sixteenth centuries), and the Late pe centuries), theMiddle period (seventeenth-eighteenth is based on the chronostratigraphic riod (nineteenth century). This classification trajectories of pre-European contact marked by lack of European imports, fol lowed by their initial introduction ca. the sixteenth-seventeenth century, and then an increase in European imports and long-distance ceramic products in the nine teenth century. Both Connah (1975) andWesler (1999) have demonstrated that the local ceramic sequence in Benin changed across the three phases, but the fact that the ceramic attributes of the thirteenth-sixteenth-century stratigraphie levels are similar is also well illustrated by Connah (1975). Unlike Ile-Ife and Benin, the archaeological sequence of Owo is poorly de fined. Our knowledge is currently limited to the indications that the kingship insti tution at Owo was established around the thirteenth century (Eyo, 1976, p. 37), and that Owo was aligned with the Ife cultural sphere from the thirteenth to the four teenth centuries. Owo came under the Benin influence during the early fifteenth century as a result of the latter's hegemonic expansion into the eastern Yoruba area and, between the fifteenth and eighteenth centuries, the archaeological sequence in Owo shows a fusion of the Ife and Benin cultural influences. Our knowledge of the sequence inOwo is limited to only the materials from Igbo'Laja, a archaeological site that seems to have been initially established as a religious center dedicated to royal rituals during the fifteenth century. Igbo'Laja was a single-phase site and its stratigraphy shows that caches of ritual artifacts, including pottery and sculptures, were deposited over a short period of time between the fifteenth and sixteenth centuries (Eyo, 1976, pp. 38-39). The short period of occupation of Igbo'Laja trends in does not therefore offer the opportunity to examine chronostratigraphic the material culture of Owo. Previous studies have demonstrated to the same "ceramic sphere" between that Ile-Ife, Owo, and Benin belonged the thirteenth and eighteenth centuries 1988; Willett, 1967a). (Connah, 1975; Eyo, 1974a; Garlake, 1977; Omokhodion, There are, however, variations in the proportions of the similar decorative attributes that have been found in these areas. Cord and carved roulettes, for example, account and Usama sites for 40, 49, and 53% respectively at Clerks' Quarters, Museum, (Wesler, 1999, p. 249). In contrast, these two roulettes account for 68-73% of the decoration attributes at Iloyi, and 73-78% atWoye-Asiri, Obalara, Lafogido, sites in Ile-Ife (Agbaje-Williams, and Odo-Ogbe 1983, p. 290; Garlake, 1977, p. 87). Additionally, whereas circular stylus, rustication, and applied/relief motifs have accounted for more than 20% of the decorative attributes in the three Benin sites during the thirteenth-sixteenth centuries, these attributes were below 10% in The Ife-Benin Interaction Field in Ijesaland 43 1,2 applied cordon and stamped geometric motifs; 3. cordon and herring Fig. 7. Decorative motifs. bone carved roulette; 4. cordon; 5. stamped circular stylus; 6. applied geometric motifs; 7. cowry-form in incisions and cowry-form bosses; 8. hyphenated bosses; 9. ribbed (grooved) rim, with diagonal 11. incised leaf motif; 12. snake cisions and rustication; incisions with rustication; 10. criss-cross carved roulette; 14. rosette; 15. guilloche motif. relief; 13. basket-work Ile-Ife and Iloyi sites. Moreover, unlike at Ile-Ife and Iloyi, where ceramic painting occurred in the thirteenth-sixteenth-century stratigraphie levels, this feature did not appear in Benin until the seventeenth century (Connah, 1975, p. 132). The diagnostic decoration indices shared by Iloyi with Ile-Ife, Benin, and Owo are reliefs of bosses, cordons, keloids, and anthropomorphic representations, stamped impressions, circu hyphenated cross-hatched incisions, geometric-shape lar stylus stamps, rustication, and the selective use of red paint on ceramic rims in the vessel forms so far identified at (Fig. 7). The similarities and differences are illustrated in Table VI. The vessel forms that and Owo Ile-Ife, Benin, Iloyi, mostly characterize the ceramic assemblages of the the four areas during the pe riod under review are flared and ribbed jar rims (Jl form) and bowls of types B2, B3, B4, B5, and Bll (see Fig. 8). These attributes can be considered diag nostic of the Ife-Benin ceramic sphere of interaction between the thirteenth and sixteenth centuries because they sharply contrast with those that characterize the other parts of Yorubaland and the areas adjacent to the Edo culture area, especially Old Oyo, Igbomina, Igalla, and Igbo areas (Agbaje-Williams, 1983; Oghuagha and Okpoko, 1984; Omokhodion, 1978; Usman, 1998; Willett, 1961). Although 44 Ogundiran Table VI. Correlation of the Vessel Forms in Iloyi With (Garlake's Jl typology) (Connah's J3 J4 J5 Bl B2 B3 B4 Benin, and Owo Owo typology) 1, 1A, IB Shape 1, Type A; Shape 1, Type A3 Shape 2, Type Al; Shape 3, Type A2 Shape 13, Type H Shape 12, Type G Shape 14, Type I J2 in Ile-Ife, Benin Ile-Ife Iloyi Scheme Forms (Eyo's typology) 1,2A,3A,4A 19 3 2,20? Shape Shape Shape Shape Shape Shape B5 B8 B9 Bll B12 Cl LI L3 L5 18,25 7,14,17,21,22 8,11,12,13 9 8, Type D 8, Type D 10, Type F 5, Type B 10, Type F 7, Type C 6D 6D 6D 6D 4B 6A 6E 4,5 10 15,16 23 ? Present 7 Shape 16, Type Present ? 7 ? 7 9 J the decorative attributes listed above have a low range of distribution, their diag nostic value as indices of cultural-historical relationship is illustrated by the fact that they are associated with both ceramics and the "Classical" terracotta sculp tures of Ife between the twelfth and sixteenth centuries. For example, red paint is found primarily on the lips and rims of B5 and Bll bowl types, on the ribbed rims and necks of Jl jars in Ife and Iloyi, and on the facial parts of the terracotta human figures in Ile-Ife (Garlake, 1974, p. 131). In fact, the prolific occurrence of red paint on the Ife terracotta figures has ledWillett (1967a, p. 70) to suggest that "probably all the terracotta sculptures were originally painted." The occur rence of rustication, and relief motifs of cowryform, bosses, keloids, and cordons are pervasive symbols in the iconographie expressions of the Ife cultural sphere. These motifs are found not only on ceramic vessels but also on terracotta figures. The rectangular cordons on the shoulders of B2, B4, and Bll bowls in Iloyi, Ife, Owo, and Benin are represented as beads on the crowns, ankles, and wrists of the terracotta figures in Ife and Owo between the twelfth and sixteenth centuries (see in 1967a, p. 87). Likewise, herringbone and hyphenated cross-hatched Willett, cised motifs were used at Ife, Owo, and Benin during the same period to decorate ceramic vessels, and terracotta sculptures, including human and animal figures and stools (see Willett, 1967a, pp. 83, 86, 144; Eyo, 1976, pp. 43-45). Some of the herringbone motifs at Iloyi are pictorial representations of leaves, one of the The Ife-Benin Interaction Field 45 in Ijesaland Fig. 8. Vessel forms. iconographie vocabularies that are found mainly in the ritual and religious con texts of Ife, Benin, and Owo (Eyo, 1976, p. 43). Among the animals represented on the vessels excavated in Site 1 at Iloyi are stylized frog, fish, and snake. These are iconographie symbols common to the Yoruba-Edo region. There was a par ticular snake relief on the fragments of a Jl vessel in Site 1 (Iloyi). The snake bears an ovisac, and it is depicted in a posture indicating that itwas biting its own tail (Fig. 7). Similar postures of snakes and mudfish are depicted on the ceramics from religious sites in Ile-Ife and Owo during the eleventh-sixteenth centuries. It has been noted that the representations of animals biting their tails or feeding on themselves is a Yoruba metaphor of eternity conceived as an unending repetition of self-destruction and regeneration (Barnes, 1997, p. 18; Lawal, 1996, pp. 246 247). Two pictorial decorative motifs that are present in the ceramics and terracotta sculptures of Benin and Owo, but absent in Ile-Ife and Iloyi, during the thirteenth sixteenth century are guilloche and rosette designs (see Fig. 7). These patterns are particularly prevalent in Benin, where they have been found in the artistic media of both clay and metal (Eyo, 1976, p. 47). 46 Ogundiran BURIAL CONTEXT: UNIT 1-A The burial context from Stratum III of Unit 1-A provides another archae relationships in the Yoruba-Edo ological index for exploring cultural-historical was at The human skeleton region. Iloyi decapitated and in a fragmented condi tion. Only the head, the torso (consisting of seven pairs of ribs attached to the sternum), and the right arm (including scapula, hum?rus, and ulna bones) of the individual were intact (Fig. 9). The bones of the right hand (carp?is, metacarpals, ly/:^^ -V^V^'c^ Fig. 9. Burial floor, Unit 1-A. The He-Benin Interaction Field in Ijesaland 47 and phalanges) were placed on the right ulna/radius bones, and the left hand and ulna/radius bones were placed above the skull. The locomotive and the left hum?rus bones of the individual were not included in the burial. Fragments of a land snail shell (Achatina sp.) were found on top of the left-hand bones, and the skull of a sheep/goat was found 40 cm from the human skull. The other finds in association with the human remains are two quartz slabs, a quartzite cobble, two grinding stones, and fragments of ceramic vessels. Jars and bowls were represented in the grave, and the interior of some of the vessels contained dry and burnt residues, possibly of food remains. The stratigraphie position of the burial indicates that the remains were interred ca. the thirteenth century, the most likely period for the formation of all of the archaeological deposits in Site 1. The individual was identified, from the maxilla sutures of the skull, as an adult. The archaeological contexts in other parts of Yorubaland (see Discussions), and information from ethnographic sources, suggest that postmortem decapita tion was part of the burial and even sacrificial rituals in Yorubaland from ca. the thirteenth to nineteenth century (Abiodun, 1976, pp. 5-6; Garlake, 1974, pp. 121 123; Poynor, 1987, p. 81). Three different ways of postmortem decapitation have been identified. First, decapitation could be carried out after the desiccation of the remains. The desiccation procedure, usually lasting for several weeks, al lowed for the removal of the bones, and different parts of the body would then be accorded special treatment. Second, the fragmentation of remains could re sult from a "second burial ceremony," a procedure that involved the exhumation of the body after flesh decay, the sorting of the bones, and the reburial of the 1966). The third fragmented skeleton (Abiodun, 1976; Poynor, 1987; Willett, was associated often with human and it involved sacrifice, procedure draining the blood of the victim followed by the amputation of certain parts of his/her body. These parts would then be put together for immediate burial or for cura tion in the shrine or temple (Idowu, 1994, p. 119). Preliminary observation of the skeletal remains at Iloyi showed no evidence of cut marks on the bones, in dicating that the postmortem decapitation of the remains could have been carried out after an elaborate desiccation or as a result of second burial procedures. It also appears that the remains of the individual were buried in more than one location. The decapitation of the deceased and the burial of the remains in associa tion with a goat/sheep skull and artifacts including grinding stones, querns, and ceramic vessels at Iloyi manifest not only different layers of symbolic meanings but also different schedules of burial and sacrificial ritual activities. An enriched understanding of the ritual context and social actions associated with the human burial at Iloyi can be gleaned from similar types of archaeological deposits in Ife-Benin region during the twelfth-seventeenth centuries. We will now turn our attention to these other cases for contextual comparison. 48 Ogundiran The Burial at Iloyi in Comparative Regional Contexts Decapitated human remains, similar to those found in Iloyi, have been exca vated atObalara site in Ile-Ife, Igbo'Laja inOwo, and at theMuseum site in Benin, all dated to between the twelfth and sixteenth centuries (Eyo, 1976; Garlake, 1974; Connah, 1975). Similar decapitated human remains dating to the seventeenth nineteenth century have also been excavated at Ilesa (Willett, 1973, pp. 135-136). At Obalara, in Ile-Ife, six potsherd pavements were exposed in a context dated to the early fourteenth century. At the eastern section of the pavement site were ten concentrations of fragmented human remains (see Garlake, 1974, p. 121, fig. 7). One of the concentrations consisted of 8 complete crania, 31 calvar?a, and 2 long bones. About 2 m north of this was a mass of human long bones, a single human skull, human teeth, and sheep/goat mandibles. All the human and animal bones at the site were decapitated and the anatomical parts of ribs, scapulae, or verte brae were absent in the piles of skulls and long bones. The artifacts associated with the human remains are iron nails, rods, and a staff, as well as glass beads, a polished stone ax, grinding stones, querns, terracotta human figures and both complete and broken vessels, a number of which bore pictorial anthropomorphic features. Garlake (1974, p. 122) indicates that the terracotta figures images on the ceramics represent three classes of people: and the relief 1. Elite individuals, depicted in idealized, serene, and unemotional postures 2. Diseased individuals suffering from elephantiasis and umbilical hernia 3. Human sacrificial victims in gagged, decapitated, and bound postures, some with expressions of malevolence and horror. Each of the three classes of human terracotta figures is deposited in different ways. The terracotta heads with rope gags were found within the mass of human skulls, and were not adorned with any jewelry. In contrast, the terracotta figures with idealized postures are elaborately decorated and are arranged separate from the other terracotta figures and human remains. Garlake (1974, p. 143) notes that the fragmented nature of the pottery and the terracotta sculptures at Obalara indicate that they had been deliberately smashed, perhaps in imitation of the postmortem decapitation, before they were buried. At Igbo'Laja site in Owo (fifteenth-seventeenth century), a fragment of hu man femur was found in an excavated ritual site in association with the teeth of about five different cattle (Eyo, 1976, p. 49). The dominant finds at the site were ter racotta objects and smashed ceramic vessels. Two terracotta heads from Igbo'Laja also depict rope gagging, and the terracotta sculpture of a basket containing decap itated human heads was also excavated at the site. One of the decapitated heads has the representation of slashed blood vessels, and two have deep cuts on the cheek and on the head (Eyo, 1976, p. 47). These representations of gagged and mutilated The Ife-Benin Interaction Field in Ijesaland 49 figures in terracotta sculptures indicate human immolation as part of ritual sacrifice activities. The site also contained terracotta sculptures of 20 other human heads and faces, 36 postcranial human features (especially torso and limbs), 12 animal figures (tortoise, horse, mudfish, cock, parrot, and leopard). Fruits and leaves were also represented in the terracotta sculpture, and some vessels were adorned with reliefs of human and animal figures. Eyo (1976, p. 43) interprets the assemblage of terracotta figures as material substitutes of sacrificial goods and human victims. At the Clerks' Quarters site in Benin City, Connah (1975) excavated a mass of disjointed human remains consisting of about 46 human skulls and 62 long bones. Most of the individuals were identified as women between the ages of 19 and 24 years. Although most of the remains were in fragmentary condition, many skulls were still intact, with mandibles. The materials associated with the human remains include fragments of raffia and cotton fabric, 49 bronze bracelets, 5 heavy bronze objects, 3 bronze rings, a fragment of bronze smelting crucible, glass and to the excavator, the agate beads, pottery, and pieces of iroko wood. According shaft containing the human remains was originally dug and used as awater cistern. Following its disuse, human sacrificial victims associated with royal rituals were deposited in the cistern. The human remains were found in the lower part of the cistern, in a context dated to the thirteenth century, the period thatmarked the early phase of the centralized political institution in Benin. At Ilesa, the evidence of decapitated human remains comes from the grave of a high-status male about 40 years old, identified as a royal personage byWillett (1973, p. 135). The individual's head was removed, possibly after death, and placed on his chest. Among the artifacts buried with him are necklaces of blue glass and red stone beads, 12 bronze bracelets, 2 iron daggers with bronze fittings, an ivory trumpet, cowry shells, and ceramic vessels. The nine individuals thatwere buried with the royal personage included two adult males, two adult females, two adolescents aged between 13 and 17, and three children between 6 and 12 years old. Their remains were decapitated and parts of their bones were scattered about the grave. The precise date of the burial is not clear, but both oral traditions and the presence of cowry shells in the grave indicate that the deposits dated to some time between the seventeenth and nineteenth century. The foregoing demonstrates the evidence of human decapitation, either as part of burial rituals or as sacrificial rites in the Yoruba and Edo regions between the eleventh and nineteenth century. The practice of decapitation is also seen in animal remains found in these burial and sacrificial contexts. The skulls and terracotta heads of goat/sheep, ram, and dog are the ones most commonly found in religious and burial contexts. The sculptures of these animals have been found on platters at Igbo'Laja, in Owo (Eyo, 1976), and at Lafodigo, Osangangan Obamakin, Abiri, and Ayetoro sites in Ile-Ife (Eyo, 1974a,b; Willett, 1967a, pp. 96-97). Likewise, the skeletal remains, especially skulls, of sheep/goats have been found in the burial and sacrificial contexts at Iloyi, Obalara, and Igbo'Laja. 50 Ogundiran The archaeological contexts of fragmented human burials in Iloyi, Ile-Ife, Owo, Ilesa, and Benin are similar to the ethnographic examples of postmortem decapitation of corpses as part of burial rituals in the Yoruba-Edo region (Poynor, 1987;Willett, 1966). Until the nineteenth century inOwo, for example, part of the burial rites of the king involved the desiccation and decapitation of the remains of the deceased, which were then buried following ritual procedures (Poynor, 1987, the historical narratives in Benin and Ile-Ife indicate that the p. 81). Likewise, heads of the deceased kings of Benin were removed and brought to Ile-Ife for burial at a place called Orun Oba Ado (etymologically meaning the spiritual realm of the King of Benin). We do not know when this practice began, but historical evidence shows that it lasted until 1888 (Egharevba, 1968, p. 9). The historical confirmed traditions associated with Orun Oba Ado are not yet archaeologically at the site The excavation account, 1973, (Willett, 129). p. despite preliminary however, illustrates the practice of elaborate postmortem mutilation of human remains in the burial and sacrificial rituals of Benin, a practice that survived until the twentieth century. The evidence at Ile-Ife, Benin, Owo, and Ilesa attests to the fact that post mortem decapitation rituals were carried out on people at different levels of social hierarchy. Apart from chiefly status individuals, whose remains were mutilated before burial, people who died of unusual diseases and those who were victims of ritual sacrifice were also decapitated. Moreover, as the archaeological deposits at in Ile-Ife, and at Site I in Iloyi, vividly show, the ritual action of decap itating corpses is similar to the practice of smashing sacrificial objects, including terracotta sculptures and vessels bearing food offerings at temples, shrines, and graves. We are not sure when the practice of postmortem mutilation began in the Ife-Benin area, but it was prevalent in the areas between the Gulf of Guinea and the River Niger in the rainforest and coastal belts of West Africa during the seventeenth-nineteenth centuries (see Law, 1989; Poynor, 1987). The ritual actions associated with postmortem decapitation will be a useful in similarities in the Ife-Benin region only when dex for inferring cultural-historical we have access to a broad range of ritual deposits from several sites. The identifica tion and comparison of ritual ensembles in the Yoruba-Edo region, especially the Obalara, patterns of decapitation, the layout of burial, and the associated materials, will bring us closer to understanding the archaeological signatures of burial and sacrificial rituals that are common to the Ife-Benin cultural corridor during the thirteenth sixteenth centuries. Likewise, we need to know more about the symbolic meanings of iron implements, granite querns and grinding stones, snails, and the skulls and terracotta figures of sheep/goat, ram, and dog that are often associated with these sacrificial and burial contexts. We currently have a poor understanding of the pre eleventh century burial practices in the Yoruba and Edo areas. However, there is ample evidence that postmortem human decapitation, multiple burial ceremonies for the elite, human sacrifice, the sequencing of offering ceremonies at specific The Ife-Benin Interaction Field in Ijesaland 51 altars and shrines, and the use of terracotta and copper-alloy sculptures to service burial ceremonies, were elaborate practices in Ile-Ife during the eleventh-twelfth centuries. Aspects of these practices were present in Benin, Iloyi, and Owo by the thirteenth century, and these ritual actions continued until the seventeenth nineteenth centuries in Ilesa, where kingship was established during the sixteenth century (Peel, 1979, p. 141). Although it is likely that these practices were carried over from the predynastic period in the region, the ritual actions were certainly elaborated in Ile-Ife, Benin, Owo, and Iloyi with the advent of kingship institutions (Connah, 1975; Eyo, 1976; Ogundiran, 2000). Comparative studies of burial and sacrificial ritual contexts would be useful to understanding whether postmortem decapitation evolved with the social structure of kingship institution in the region. DISCUSSIONS: CERAMICS, RITUALS, AND INDICES OF CULTURAL-HISTORICAL RELATIONSHIPS INYORUBA-EDO REGION Two major archaeological indices, ceramics and postmortem decapitation, have been considered in the foregoing as evidence that Iloyi, and indeed the Ijesa area, was part of the Ife-Benin cultural sphere between the thirteenth and sixteenth centuries. The similarities in the stylistic and iconographie records inscribed on the ceramics and the ensembles of burial and sacrificial rituals in Ile-Ife, Benin, Owo, and Iloyi demonstrate a high level of cognitive uniformity among the populations of these areas between the thirteenth and sixteenth centuries. All the characteristics that define the Ife-Benin-Owo-Iloyi ceramic assem blages and burial/sacrificial contexts appeared in the city-state of Ile-Ife around the tenth century. These attributes have so far been identified at Owo, Benin, and Iloyi in thirteenth-century contexts; about three centuries after they first appeared in Ile-Ife. This raises a number of questions. Are these ceramic similarities a product of dispersal of materials, symbols, ideas, knowledge, and populations from the core area of Ile-Ife? Is the symbolic knowledge, reproduced by similar rituals of burial and sacrifice along the Ife-Benin axis between the thirteenth and sixteenth centuries, part of the structural practice of new forms of social order and insti tution that developed in Ile-Ife between the ninth and tenth centuries? Are the homologous cultural elements a product of common cultural origins or of exten sive regional interactions that pre-dated the tenth century in the Yoruba and Edo areas? On the basis of the current archaeological evidence, we know that the polity of Ile-Ife had the longest history of sociopolitical complexity in the region and that the city-state of Ile-Ife developed in a regional system thatwas, until the thirteenth century, devoid of comparable centralized polities. Thus, between the tenth and thirteenth century, Ile-Ife seems to have built its influence in the Yoruba and Edo 52 Ogundiran areas, using its ideology of kingship and the material paraphernalia of sociopoliti cal authority to establish and cement networks of relationships with the surround ing areas. Following the nascent development of a kingship-based sociopolitical institution in Ile-Ife between the ninth and eleventh centuries (Adediran, 1992; 1992), the settlement witnessed amarked elaboration both in sociopo Akinjogbin, litical organization and material culture from the twelfth to the sixteenth century is archaeologically indexed (Willett, 1967a). This material culture efflorescence by the presence of walls that delineated the physical and social geography of Ile Ife as an urban center, and by the local production of ornamental and utilitarian crafts, especially glass beads, iron artifacts, bronze and terracotta sculptures, and exotic pottery. Shaw (1978) andWillett (1977) have demonstrated that Ile-Ife was linked with the Niger bend termini of the trans-Saharan trade routes between the tenth and fifteenth centuries, from where it obtained the copper ingots, tin, zinc, and glass that were crucial to its local craft production. These products primarily functioned within the contexts of sociopolitical institutions and rituals of kingship. The material components of the thirteenth-sixteenth-century levels show that Ile-Ife and Benin belonged to the two ends of the same cultural corridor. Whereas Owo is considered as a meeting point of the Ife and Benin cultural influences (Eyo, 1974a), both the ceramics and the burial contexts show that Iloyi was firmly the ambit of the Ife cultural sphere. Thus, given the geography of these units, it appears that the degree of material culture similarity is sociopolitical directly proportional to the spatial distance among the communities. The regional historical dynamics that most likely generated the shared ritual elements and styles of material culture at Ile-Ife, Iloyi, Owo, and Benin have been interac explained with reference to the frontier process of metropolis-periphery tions (Akinjogbin and Ayandele, 1980; Egharevba, 1968; Eyo, 1974a; Kopytoff, 1987; Smith, 1988). Both the local historical narratives and the conceptual frame work of frontier process provide insights into how themetropolises, with advanced political organization, sophisticated ideologies, token objects of elite identities, and sociocultural infrastructures influenced the cultural-historical trajectories of their peripheries with lower scale sociocultural infrastructure in the region. We have learnt, from these narratives and conceptual frameworks, the historical pro cesses through which new polities, new societies, frontiersmen and political en formations trepreneurs, and new ethnicities grew out of the older sociopolitical 1998; Santley and Alexander, 1992). In a seminal (Kopytoff, 1987; Rowlands, within essay by Kopytoff entitled "The Internal African Frontier," the frontier is conceived as a geographical and an institutional vacuum characterized by open resources and sparse population, with less developed institutional structures of social organiza tion compared to them?tropole. This, according to Kopytoff (1987, p. 11), makes the frontier an ideal area "that can offer little political resistance to intrusion" of frontiersmen from them?tropole. And, as an institutional vacuum, the frontier is the ultimate area where the structures of them?tropole could be replicated through po litical conquest, migration, cultural borrowing, and many forms of socioeconomic The Ife-Benin exchange. Interaction In the Field extreme, 53 in Ijesaland this perspective sees the frontier as socially conserva tive, being primarily a recipient of innovations and populations from them?tropole (Vansina, 1990, p. 262). In reality, however, the frontiersmen's success at social construction on the frontier has always been shaped by the political and cultural relations between the m?tropole and frontier, the relative population densities of the societies involved, and the local political dynamics and initiatives at the frontier (Kopytoff, 1987, p. 26). The conceptual framework of the frontier process would suggest that thema terial culture stylistic similarities in the Ife-Benin cultural corridor were a product of the transplantation of segments of Ife population into the frontier settlements, especially between the eleventh and thirteenth centuries. These population move ments then resulted in the regionwide replication of Ife cultural elements, leading to long-term regional cultural homogenization between the new settlements/social units and their parent homeland. Indeed, oral historical narratives suggest that waves of migrants from the political core of Ile-Ife moved into the outlying ter ritories to develop new societies and institutional structures with social relations and political organizations akin to that of Ile-Ife (e.g., Smith, 1988). The popu lation increase and internal political conflicts that followed the urban growth and in Ile-Ife between the tenth and thirteenth centuries sociopolitical development are said to have intensified the emigration from Ile-Ife into the adjacent territo ries (Akinjogbin, 1992, pp. 110-115). These frontiersmen, as Kopytoff would call them, carried with them the ideology and institutions of political organization they were familiar with in Ile-Ife, and replicated these in their new settlements. The occupation of Iloyi and its position as the political center of Eka Osun (otherwise known as Hare polity) lasted through three centuries of the Classical period in He lfe (Ogundiran, 2001). Because of its proximity, Iloyi would have been one of the receiving areas for the immigrants from Ile-Ife. The rich genre of local historical narratives indicate that the advent of a kingship institution in Iloyi and the devel opment of the settlement as the political center of Hare district were accomplished within the context of very strong sociopolitical ties with Ile-Ife (Obayemi, 1985, pp. 274-276; Ogundiran, 2001). The formal stylistic and iconographie elements of ceramics that are shared by Ile-Ife and Iloyi, and the similarities in the burial rituals found in the two settlements, serve as evidence for the historical connections between the two settlements. These connections seem to have included migrations from Ile-Ife into Ijesaland during the thirteenth-sixteenth centuries. A variant of the frontier process, popular in Nigerian historical scholarship, has indicated that princely elite with a small number of followers migrated to replicate the Ife political institution in the surrounding region, including the Ijesa, Owo, and Benin areas. These few Ife migrants are said to have displaced the gerontocratic-lineage-based political regimes in the Yoruba-Edo region through both military and diplomatic tactics and then established the Ife-style sociopolitical and Ayandele, 1980; Beier, 1956; Egharevba, system (e.g., Akinjogbin 1968). evidence shows that the thirteenth century did indeed mark Chronostratigraphic 54 Ogundiran the beginning of regionwide disruption of the autonomy of small-scale village communities in the Yoruba-Edo region, as political entrepreneurs, operating on structures and ideology, developed new forms of the Ife model of sociopolitical linkages that organized village groups into nucleated towns (for patron-client case studies, see Beier, 1956; Obayemi, 1985; Smith, 1988). The advent of these in and Owo took place during the peak of Benin, Ijesa, sociopolitical trajectories centuries. It is also instructive in the twelfth-fifteenth Ife's cultural efflorescence that regional historical narratives in Iloyi (Hare district), Owo, and Benin recognize Ile-Ife as a sociopolitical center that many harbingers of dynastic institutions in for establishing and validating the region visited and allied with as a mechanism their political power between the twelfth and sixteenth centuries. Therefore, Ife played an important role in validating the credentials of the nascent political elite that adopted its ideology of kingship and the associated rituals in the region. The regional influence of Ile-Ife was strengthened not just by its status as the harbinger of the institutional and ideological elements of kingship, but also as the main supplier of the beaded crowns and glass beads that served as the regional identity tokens for kingship and elite personality. Thus, the political influence of Ife was grounded in both thematerial basis of social and economic transactions and in worldview that validated the regional legitimacy of political 1992, 1992, p. Ill; Horton, 1992, p. 123; see also Demarest, pp. 149-153; Kopytoff, 1987, p. 72; Tambiah, 1976, p. 123). The frontier migration process had more potential to change the regional ma terial culture and cultural practices, than the princely elite migrations. Both frontier migration and interdite networks should however be seen as part of the broader sociohistorical process of intensive intermixing of populations as people moved "by choice and by necessity for economic, political, and social purposes" (Barnes and Ben-Amos, 1997, p. 49). Although these movements and interactions seem to of the kingship institution or Classical period at Ife (Olomola, the advent pre-date 1992), they seem to have intensified with the advent of political centralization and a the mythico-religious scions (Akinjogbin, formal kingship institution, especially after the tenth century. The development of sociopolitical centralization and dynastic institutions would have facilitated closer regional social ties than were possible during the predynastic period, since social units at the level of independent village communities would have had less of a need for social interdependence than peer-centralized political units. In the latter case, internal factional competitions for political power not only influenced intensive population mobility at the regional level, but the need for access to the valuable materials that legitimized political authority also increased the level of regional exchange networks. Thus, political centralization can be viewed as the springboard for heightened cultural-historical relationships and interactions in the Yoruba-Edo region through a complex exchange of materials, people (e.g., itinerary craftsmen), symbols, and knowledge. Under these conditions, one would expect new ideas and stylistic attributes to have spread along the interaction routes traversed by political The Ife-Benin Interaction Field in Ijesaland 55 entrepreneurs, ritual and craft specialists, traders, and by people simply looking for opportunities better than the ones available in their homeland. These interaction networks no doubt created a regional "symbolic reservoir" in the Ife-Benin region between the thirteenth and sixteenth centuries. Mclntosh (1991) defines this reser voir as the core of symbols, beliefs, values, and ideas shared by the social units and settlements in a region, irrespective of the internal differences among them. The of ceramic styles and ritual elements, evident in the trend toward homogenization of records Ife, Hoyi, Benin, and Owo, shows that these communities archaeological drew from a "symbolic reservoir." Some of the elements of this "reservoir" were at Ile-Ife during the eleventh century and were anchored on the an ideology of kingship institution that became the mainpost of of development the cultural conception of social relations in the Yoruba-Edo region between the indi eleventh and thirteenth centuries. Although the regional chronostratigraphy cates that these developments were first manifested at Ile-Ife, this does not mean first manifested that all the characteristics of "symbolic reservoir" that characterize the Ife-Benin cultural sphere originated from Ile-Ife. Rather, these similarities were products that by the thirteenth century took of regional borrowing and interdependence a of within the framework "peer-polity" network (Demarest, 1992, p. 141; place Renfrew, 1986, pp. 2-3). CONCLUSIONS The same cultural group occupied Iloyi for about three to four centuries, and the above comparative study indicates that the areas of Ijesa, Ife, Owo, and Benin belonged to the same sphere of sociocultural interactions between the thir teenth and sixteenth centuries. This study has focused on the similarities in ce ramic styles and aspects of burial and sacrificial rituals as the basis for discerning cultural-historical relationships in the region. We are able to discern that the ce ramic styles and burial/sacrificial patterns in Iloyi during the thirteenth-sixteenth centuries were a replication of the cultural indices that dated to the tenth-twelfth centuries in Ile-Ife. The historical sources enable us to view these similarities in the archaeological data as a product of regional trajectories of sociopolitical interdependency stimulated by the advent of kingship institutions. Ceramics pro vide the most diagnostic archaeological indices that Ijesaland was part of the Ife-Benin interaction field between the thirteenth and sixteenth centuries. How the heuristic interpretive qualities of ceramics as an ever, in order to maximize effective index of regional interaction and cultural-historical relationships, there is the need to develop a uniform regional ceramic classification scheme that would enable us to compare ceramic assemblages with better precision of terms and diagnostic indices. The archaeologists working in the Yoruba-Edo region also of the diagnostic decorative attributes have to explore the symbolic meanings 56 Ogundiran and mentioned above. Some of these attributes are iconic signs with meanings cultural concepts that can be interpreted with the aid of critically assessed oral in his study historical and ethnographic texts. As Ray (1987) has demonstrated of material metaphor and social interaction in eastern Nigeria, these iconic signs constitute an active social force because they embody identities, social valuation, and power relations. Thus, our ability to identify and interpret the iconic signs that are embedded in what we have so far referred to as decorative or stylistic indices would enrich our understanding of the cultural similarities and variations, and the mechanisms of regional interaction along the Ife-Benin cultural corridor. This, I believe, would open new vistas of understanding on the dynamics of interactions and cultural-historical relationships in the region. Likewise, the ritual activities associated with burial and religious worship are a medium through which we can understand the ideology, worldview, and social structure of a society or a geo-cultural region (David, 1992; Morris, 1992). It has been underscored in this essay that the study of the archaeological vestiges of the of and sacrificial contexts, especially burial postmortem human de practice were constructed in how actions ritual be understanding helpful capitation, may as meaningful statements about the actors' perceptions of their world at a partic ular period. These perceptions are a product of the articulation of structural rules and individual choices aimed at addressing time-contingent conditions and expec tations that "rites of passage" or "social transformations" such as death, burial, and offerings, are intended to fulfill (Hodder, 1982, pp. 141-146; Morris, 1992, p. 6; Parkin, 1992, pp. 22-23). The implication of these ritual actions in terms of social formation and social structure in the Yoruba-Edo region is not the pri mary focus of this paper. Rather, the goal has been to establish the typological characteristics of burial and ritual practices in relation to the cultural-historical relationships in the Ife-Benin area between the thirteenth and sixteenth centuries. this paper has inspired the need for a follow-up study on the con Nevertheless, textual and symbolic meanings of mortuary artifacts and burial patterns along the Ife-Benin "cultural continuum." Such a study would help us unravel the link between burial/sacrificial practices and social actions, and it has potentials for un derstanding the ideological and worldview aspects of regional cultural-historical relationships. The foregoing has emphasized only a 300-400-year period with regard to an is similar to the work In that effort Ijesaland in the Ife-Benin interaction field. and Mclntosh Sterner (1992) in the of David (1992), MacEachern (1991), (1994), of Mali and the Mandara of Middle Niger Delta Highlands Nigeria and Cameroon, two it the identifies this paper accomplishes diagnostic archaeologi goals. First, cal indices that show the areas between Ile-Ife and Benin to be part of the same "symbolic reservoir" between the thirteenth and sixteenth centuries. Second, it that the Ijesa and Ife areas belonged to one minor cultural pool demonstrates within the context of the Ife-Benin cultural/interaction sphere. Although the nature The Ife-Benin Interaction Field in Ijesaland 57 of the archaeological evidence at Iloyi has delimited the scope of the preceding should not be based solely "on discussions, it is recognized that our epistemology a on material... that silent record" (Friedel, 1992, p. 118). generalities depend Thus, by weaving local historical narratives with archaeological data, this paper has sought to provide the historical and social contexts that produced the similar ities inmaterial culture and vestiges of ritual action in the region. Future studies similar to those carried out at Iloyi would enable us to assess the local peculiar ities and the contributions made by the different communities that belonged to the Ife-Benin cultural/interaction the That similarities in the stylistic and sphere. crosscut of material and culture ritual practice iconographie aspects ethnolinguistic differences in the region beckons us to develop research agendas on the long-term of ethnogenesis in the Yoruba-Edo region. These agendas would involve regional comparative studies of the archaeological deposits of the fifth-thirteenth centuries, with a focus on the origins and spread of artifacts and the sociocultural structures that are diagnostic of the Ife-Benin symbolic reservoir. As the scope of archaeological investigations extends into other areas, our expanding database should allow us to have a better understanding of how panregional interactions and cultural historical relationships were articulated in the ever changing sociopolitical fields of the Ife-Benin region. processes ACKNOWLEDGMENTS I am grateful to Julie Hansen, Michael DiBlasi, and Christopher DeCorse for their critical comments on the dissertation chapters that inspired this paper. I appreciate the remarks made by panel participants on an earlier version of this paper, presented at the 1999 conference of African Studies Association (A.S.A.) in Philadelphia. The Department of Archaeology and Anthropology, University of Ibadan, under the chairmanship of Bayo Folorunso, kindly provided me with equip ment and office space during my 1997 research. Raphael Alabi, Alex Asakitikpi, and Dele Odunmbaku provided me with field assistance. I am grateful to Lea Koonce Ogundiran for her generous support and editorial comments, to Nicolas Luchsinger for his enthusiastic help with text translations, and to the anonymous reviewers for giving me useful insights into the interpretations of the data. Finally, I acknowledge the financial contribution of the Boston University Humanities Foundation tomy research inYorubaland. REFERENCES CITED of the function of Ako, second burial effigy in Owo. Africa Abiodun, R. (1976). A reconsideration 46: 4-20. J. D. E., Babafemi, J. A., and Ataiyero, S. O. S. (1932). ?t?n Il?s? (History of Ilesa), Ilesa. Abiola, 58 Ogundiran I. A. (ed.), The Cradle of a of the Ife state. In Akinjogbin, Adediran, B. (1992). The early beginnings to 1980, Sunray, Lagos, pp. 77-95. Race: Ife From the Beginning to the Archaeology B. (1983). A Contribution of Old Oyo, PhD Dissertation, Depart Agbaje-Williams, of Ibadan, Ibadan, Nigeria. ment of Archaeology, University to 1800. In Akinjogbin, I. A. (ed.), The I. A. (1992). The growth of Ife from Oduduwa Akinjogbin, to 1980, Sunray, Lagos, pp. 96?121. Cradle of a Race: Ife From the Beginning before 1800. In Ikime, O. (ed.), Groundwork I., and Ayandele, E. (1980). Yorubaland of Akinjogbin, Ibadan, pp. 121-143. Nigerian History, Heinemann, S. O. (1981). The Oyo, Ife, and Benin relationship reconsidered. African Notes 8(2): 15-26. Babayemi, Barnes, S. (1997). The many faces of Ogun: Introduction to the first edition. In Barnes, S. (ed.), Africa 's Press, Bloomington, pp. 1-26. Ogun: Old World and New, Indiana University P. G. (1997). Ogun, the empire builder. In Barnes, S. (ed.), Africa's Ogun: Barnes, S., and Ben-Amos, Old World and New, Indiana University Press, Bloomington, pp. 39-64. Beier, H. U. (1956). Before Oduduwa. Odu: Journal ofYoruba and Related Studies 3: 25-32. in and Around and Other Researches Connah, G. (1975). The Archaeology of Benin: Excavations Benin, Nigeria, Clarendon, Oxford. of ideology: Mortuary David, N. (1992). The archaeology practices in the central Mandara highland, In David, N., and Sterner, J. (eds.), An African Commitment: Papers inHonor northern Cameroon. of Calgary Press, Calgary, pp. 181-210. of Peter Lewis Shinnie, University Demarest, A. (1992). Ideology in ancient Maya cultural evolution: The dynamics of galactic polities. In Demarest, A. A., and Conrad, G. W. (eds.), Ideology, School of American Research Press, Santa Fe (New Mexico), pp. 135-157. J. (1968). A Short History of Benin, 4th edn., Ibadan University Press, Ibadan. Egharevba, 148: Ekanade, O. (1984). A socio-geographical analysis of the present Ijesaland. Nigeria Magazine 25-38. at Ife and Owo, and Their Implications for Ife, Owo, and Benin Eyo, E. (1974a). Recent Excavations of Ibadan, Nigeria. of Archaeology, Studies, PhD Dissertation, University Department at Odo Ogbe Street and Lafogido, Ife, Nigeria. West African Journal of Eyo, E. (1974b). Excavations 4: 99-109. Archaeology 6: 37-58. Eyo, E. (1976). Igbo'Laja, Owo. West African Journal of Archaeology Brief No. 9, Cultural Chronology, Museum Ford, J. A. (1962). A Quantitative Method for Deriving Columbia. of Anthropology, of Missouri, Museum University In lowland Maya civilization. Friedel, D. (1992). The trees of life: Ahau as idea and artifact in Classic Demarest, A. A., and Conrad, G. W. (eds.), Ideology, School of American Research Press, Santa Fe (New Mexico), pp. 115-133. at Obalara's land, Ife, Nigeria. West African Journal of Archaeology Garlake, P. (1974). Excavations 4: 111-148. on the Wove Asiri family land in Ife, Western Nigeria. West African Garlake, P. (1977). Excavations 7: 57-95. Journal of Archaeology Studies of Material in Action: Ethnoarchaeological I. (1982). Symbols Culture, Cambridge Hodder, Press, Cambridge. University I. A. (ed.), The Horton, R. (1992). The economy of Ife from c. AD 900-c. AD 1700. In Akinjogbin, to 1980, Sunray, Lagos, pp. 122-147. Cradle of a Race: Ife From the Beginning New York (USA edn.). Idowu, I. (1994). Olodumare: God in Yoruba Belief, Wazobia, I. (1987). The internal African frontier: The making of African political culture. In Kopytoff, Kopytoff, Indiana University I. (ed.), The African Frontier, Press, Bloomington, pp. 3-84. Law, R. (1989). "My head belongs to the King": On the political and ritual significance of decapitation in pre-colonial Dahomey. Journal of African History 30: 399-415. in an African Culture, Art, Gender, and Social Harmony Lawal, B. (1996). The G?l?d? Spectacle: of Washington Press, Seattle. University and inter-group relations: S. (1994). "Symbolic reservoirs" MacEachern, Review 12: 205-224. African Archaeological seriation. In Schiffer, in archaeological Marquardt, W. H. (1978). Advances Method and Theory 1, Academic Press, New York, pp. Archaeological of mud structures: An experiment R. (1977). The excavation Mclntosh, Archaeology 9(2): 185-199. West African examples. in M. B. (ed.), Advances 257-314. from West Africa. World The Ife-Benin Interaction Field in Ijesaland 59 R. (1991). Early urban clusters in China and Africa: The arbitration of social ambiguity. Mclntosh, Journal of Field Archaeology 18(2): 199-212. R. J. (1986). Recent research and dates from West S. K., and Mclntosh, Mclntosh, archaeological Africa. Journal of African History 27: 413-442. I. (1992). Death-Ritual in Classical and Social Structure Morris, Antiquity, Cambridge University Press, Cambridge. and M. Yoruba and A. their before 1600. In (1985). The Obayemi, neighbours Edo-speaking peoples Press, Ajayi, J. F. A., and Crowther, M. (eds.), History of West Africa, 3rd edn., Oxford University Oxford, pp. 255-322. in Eastern Nigeria: A. I. (1984). History and Ethnoarchaeology P. A., and Okpoko, Oghuagha, to the Anambra A Study of Igbo-Igala Relations With Special Reference Valley, Cambridge in African Archaeology 7, British Archaeological Reports International Series 195, Monographs Cambridge. in Central Yorubaland, AD 1200 Interactions Ogundiran, A. (2000). Settlement Cycling and Regional PhD Dissertation, in Hare District, Nigeria, Boston University, 1900: Archaeology and History Boston. and settlement cycling in sociopolitical Ogundiran, A. (2001). Factional competitions, development, in a Yoruba community. Oral traditions of historical experience Hare District (ca. 1200-1900): inAfrica 28: 20-40. History I. (1992). Ife before Oduduwa. In Akinjogbin, I. A. (ed.), The Cradle of a Race: Ife From the Olomola, to 1980, Sunray, Lagos, pp. 51-61. Beginning I. (1978). Oyo He Pottery Analysis, MSc Dissertation, of Archaeology, Omokhodion, Department of Ibadan, Ibadan. University I. (1988). Northwest Benin Sites and Socially Determined Omokhodion, Artifact Distribution, PhD Dissertation, State University, Ann Arbor. Michigan Ozanne, P. (1969). A new archaeological survey of Ife. Odu (NS) 1: 28-45. Parkin., D. (1992). Ritual as spatial direction and bodily division. In de Coppet, D. (ed.), Understanding London, pp. 11-25. Rituals, Routledge, Peel, J.D. Y. (1979). Kings, titles, and quarters: A conjectural history of Ilesa I:The traditions reviewed. inAfrica 6: 109-153. History Poynor, R. (1987). Ako figures of Owo and second burials in southern Nigeria. African Arts 21(1): 62-63,81-82,96-87. social interaction and historical reconstructions: Ray, K. (1987). Material metaphor, Exploring patterns of association and symbolism in the Igbo-Ukwu corpus. In Hodder, I. (ed.), The Archaeology of Contextual Meanings, Press, Cambridge, pp. 66-77. Cambridge University Renfrew, C. (1986). Introduction: Peer polity interaction and socio-political change. In Renfrew, C, and Cherry, J. (eds.), Peer Polity Interaction and Socio-Political Change, Cambridge University Press, Cambridge, pp. 1-18. Rowlands, M. (1998). Center and periphery: A review of a concept. In Kristiansen, K., and Rowlands, M. (eds.), Social Transformation in Archaeology: Global and Local Perspectives, Routledge, London, pp. 219-242. systems. In Schortman, Santley, R., and Alexander, R. (1992). The political economy of core-periphery Power, and Interregional E., and Urban, P. (eds.), Resources, Interaction, Plenum, New York, pp. 23-49. Its Archaeology and Early History, Thames and Hudson, London. Shaw, T. (1978). Nigeria: Shennan, S. (1988). Quantifying Archaeology, Press, Edinburgh. Edinburgh University of Wisconsin Smith, R. S. (1988). Kingdoms Press, Madison, WI. of the Yoruba, University Guide Base 8.0, SPSS Inc., Chicago. SPSS (1998). Applications in material choices and social boundaries culture patterning: An intro Stark, M. (1998).Technical duction. In Stark, M. T. (ed.), The Archaeology Smithsonian Institution, of Social Boundaries, DC, pp. 1-11. Washington, in the Mandara Highlands reservoir" of Northern Sterner, J. (1992). Sacred pots and "symbolic In David, N., and Sterner, J. (eds.), An African Commitment: Cameroon. in Honor of Papers of Calgary Press, Calgary, pp. 171-179. Peter Lewis Shinnie, University Stuiver, M., and Reimer, Calibration P. J. (1993). Radiocarbon of Washington. University ternary Isotope Laboratory, Program Rev. 3.0, Washington: Qua 60 Ogundiran Tambiah, S. (1976). World Conqueror and World Renouncer, Cambridge University Press, Cambridge. Relations and Sociopolitical in Igbominaland, North Usman, A. ( 1998). State-Periphery Development central Yoruba, Nigeria, PhD Dissertation, of Anthropology, Arizona State University, Department Tempe. of Wisconsin Vansina, J. (1990). Paths in the Rainforest, University Press, Madison. in Nigerian ceramics. African Archaeological Review Wesler, K. (1999). Chronological sequences 16(4): 239-258. at Old Oyo 1956-57: An interim report. Journal of Historical Willett, F. (1961). Investigation Society 2(1): 59-77. of Nigeria F. (1966). On the funeral effigies of Owo and Benin, and the interpretation of the life-size Willett, bronze heads from Ife. Man n.s. 1: 34-45 F. (1967a). Ife in the History of West African Sculpture, Thames and Hudson, London. Willett, F. (1967b). Pottery classification in African West African Archaeological Willett, archaeology. 7: 44-55. Newsletter F. (1973). Archaeology. In Biobaku, S. O. (ed.), Sources of Yoruba History, Clarendon, Oxford, pp. 111-139. F. (1977). Baubles, Bangles, and Beads: Trade Contacts ofMediaeval Willett, Ife, Thirteenth Melville J. Herskovits Memorial Press, Edinburgh. Lecture, Edinburgh University In Cleland, C. E. (ed.), Essays Wobst, H. M. (1977). Stylistic behavior and information exchange. in Honor Ann Arbor, of Anthropology, of James B. Griffin, Anthropology Paper 61, Museum pp. 317-341. Willett,