CONFLICT, WAR, DISPLACEMENT AND ARCHAEOLOGY IN PARTS OF OSUN STATE, SOUTHWESTERN NIGERIA By Y R A R BENJAMIN ADISA OGUNFOLAKAN IB (Matric No. 68943) L B.A. Archaeology (Moscow, RussiaN) M.A. Archaeology (Moscow, Russia) M.Sc. Archaeology (Ibadan, NigAeria) M.Phil. Archaeology (Ibadan, Nigeria) BA D I OF AY the sis in the Department of ARCITHAEOLOGY/ANTHROPOLOGY, Submitted to the Faculty oSf Science in partial fulfilment of the requirements for the Degree of R DOCTOR OF PHILOSOPHY IV E Of the N U UNIVERSITY OF IBADAN i 1 NAME: Benjamin Adisa OGUNFOLAKAN 2 MATRIC. No.: 68943 3 TITLE: CONFLICT, WAR, DISPLACEMENT AND 4 ARCHAEOLOGY IN PARTS OF OSUN 5 STATE, SOUTHWESTERN NIGERIA 6 ABSTRACT 7 Most archaeological works in southwestern Nigeria are concentrated in Ile-Ife, 8 Esie, Old Oyo and Owo. In these areas, the focus of archaeological studies had been on 9 different works of art in bronze, terracotta, wood and stone. Studies on cultural themes 10 related to the issues of conflict, war and displacement which have implications foYr 11 landscape archaeology of the area are often relegated to the background. The mainR goal 12 of this research was to highlight how conflict, war and displacement impacteAd on the 13 settlement history of parts of Osun State, southwestern Nigeria. The study also appraised 14 human interactions with the environment and the concomitant effecBts Ron emergent 15 settlement configurations. 16 Oral and written data were collected from Ile-Ife, Ikire, IpetumIodu, Ila-Orangun, 17 and Ajaba to generate anthropological data. Investigations aim eLd at identifying and 18 collecting surface artifacts involved reconnaissance and detailed surveys of the studied 19 sites. Excavations were carried out on potsherd pavements aNt Ajaba and Asi and on a 20 refuse mound at Ajaba. Artifacts from surface collections aAnd excavations were classified 21 according to types, decoration and functional attributeDs. Analysis of Mo, Cu, Pb, and Ni 22 of sherd samples was done using inductively coupleAd plasma mass spectrometry. Ten thin 23 sections were made from selected sherd samplBes for determination of pottery fabric and 24 inclusions. Palynological analysis of soil samples collected from different depths of the 25 excavated mound was carried out using aF mi crIoscope with an attached camera. 26 Decorative motifs such as single twisted cord impression were common to all 27 sites. With exception of sherds from O Ila-Orangun, those from other areas were related in 28 terms of types, fabric and functional attributes. Some of the sherds bore striking 29 resemblance to those documentYed for Old Oyo and Ile-Ife with regard to type, decoration 30 and function. Stylistically, IthTe potsherd pavements at Asi and Ajaba were similar to those 31 documented for Ile-IfeS. A C-14 date of AD 1263 was obtained from charcoal at a depth 32 of 80cm from the Ajaba mound excavation. Maize cob decoration was absent which th33 indicated that Ajaba site was occupied prior to 16 century when maize was introduced 34 into West AfricEa. PRollen of forest species and ornamental plants of Asian origin, such as 35 LagerstroemVia indica, Casuarina equisetifolia and Delonix regia was identified from the 36 excavated mound. These were abundant at the lower levels of the excavated mound. 37 HoweNver, Ipollen of ornamental plants disappeared completely at the upper levels while 38 secondary forest species and artifacts increased in abundance which was indicative of 39 inUcrease in human population and subsequent impact on vegetation. There was evidence 40 of increased peopling of the area from around AD 1263. Oral and written records 41 suggested that conflicts and war caused displacement and re-occupation of most of the 42 settlements. 43 Conflict and war resulted in the abandonment and reoccupation of all the sites, 44 resulting in the delineation of several historical phases of occupation. Human impact on th 45 the environment was noted from the 13 century. 46 Keywords: Conflict, War, Displacement, Archaeological data 47 Word Count: 500 ii ACKNOWLEDGEMENTS Many people in one way or the other assisted and encouraged me to finish this work. I sincerely owe the completion of this thesis to the patience, thoroughness, intellectual guidance and fairness of my supervisor, Dr. Jonathan Olu Aleru. My heartfelt gratitude goes to him for encouraging, guiding and putting me through this work. His patience in going through the draft meticulously is worthy of mentioning. To the Head of Department Prof. Bayo Lawuyi, I say a very big thank you for words of advice anYd constant questions on the state of the completion of the thesis, words of the moutAh arRe not enough to express my sincere appreciation. In addition, I thank Professor DRavid Aremu who, not only insisted on the completion of this project by constantly phBoning to remind me of the need to complete the work but also stood by me in praye rsL. SpIecial thanks go to Prof. (Mrs.) M.A. Sowunmi who with the assistance of Mr. A. E. Orijemie carried out the palynological analysis and interpretation of soil samples from NAjaba. Mama, I know you are always there for us. I pray that God Almighty wDill Agrant you long life and good health. A My director, Professor Akin Ige of N aItuBral History Museum, Obafemi Awolowo University, Ile-Ife is a wonderful personF, a man of God (in fact, he is an Archdeacon of the Anglican Communion) not only Osupported me but actually `stood on my neck‘ to see that I complete the project. Profess or Akin Ige not only encouraged me to be steadfast in the project but he also assisTtedY me in the chemical and material analysis of the ceramic samples from the studyS arIea and other parts of Yorubaland for the understanding of the source of the raw Rmaterial. Mr. Moshood Olayiwola and Babatunde Babalola took the soil samples for the palynological analysis. Special appreciation goes to Mr. Babalola who took IacVtive E part in the excavation. Mr. Olaleye Otunla and Peter Dada provided me with aNll support, most especially, in drawing most of the maps in this work while Ojoniyi TiUmothy worked with me on the field and took most of the photographs. I will also like to acknowledge the support of other members of staff of the Natural History Museum - senior, intermediate and junior – who were all so anxious and concerned about the quick completion of this doctoral thesis. My sincere gratitude goes to Owaloja of Ajaba, Oba Adedotun Adetoyese and his chiefs, the Edigbon of Edemosi, the Orangun of Oke-Ila and all other royal fathers who iii gave me remarkable support during this exercise. My sister in-law, Mrs Modupe Popoola who introduced me to her father in-law, Chief Joshua Popoola, the Tewogbade of Iresi who in turn showed me the potsherd pavement in Ajaba and other informants from this area are worthy of special recognition. The cooperation and support of these people made this work a success. Chief Popoola actually became an advocate of cultural preservation by going round the houses of prominent people in Ajaba to explain my mission in carrying out excavations at Ajaba. He stood by us throughout the period of thYe excavations. R Lastly, I will like to thank my dear wife Mrs. Dorcas Olusola OgunRfolaAkan (Iya Adisa) and the children for their patience and understanding during myB numerous field trips and when at times I had to sleep in the office. Their to tLal Isupport made the completion of this Ph.D work a great success. I appreciate youN all, may God continue to bless you, and solidify our love for each other. Finally, to all of you too numerous to mention, I say bigD thaAnks to you. IB A F Y O T RS I E NI V U iv CERTIFICATION I certify that this dissertation was carried out by Mr. Benjamin Adisa Ogunfolakan under my supervision. Signature……………………… Date:…………………………… Supervisor Y Dr. Jonathan Olu Aleru R Reader A Department of Archaeology/Anthropology, R University of Ibadan, Ibadan. IB L N A BA D F I O ITY ER S V UN I v DEDICATION This work is dedicated to Almighty God and my loving late mother (Alhaja) Chief (Mrs.) Juweratu Abegbe Ogunfolakan who saw the beginning of this project but did not see the end. ‗Iya Muri‘, I miss your motherly care. You inspired, influenced and encouraged me to become what I am today. Forever, you will remain in my heart. May your soul, rest in perfect peace. AR Y R LI B N AD A BF I Y O IT RS E IV UN vi LIST OF PLATES Plate 1.1 A typical market day 5 Plate 1.2 Erosion washes at Iresi (Drainage) 12 Plate 1.3 Deforestation of the sacred grove 17 Plate 3.1 Potsherd pavement (Ila Orangun) 48 Plate 3.2 Oke-Ipole Shrine at Iree 50 Plate 3.3 Igbajo Old place building 51 Y Plate 3.4 Igbajo New palace building 52R Plate 3.5 Okuta Latosa (Kiriji war site) RA55 Plate 3.6 Kiriji War Peace Treaty Site B 56 Plate 3.7 Ijoko: rock chair at the entrance of the route to the sheLlteIr 58 Plate 3.8 Inside the rockshelter at Iresi N 59 Plate 3.9 Oju‘du shrine (Rockshelter at Iragbiji) A 60 Plate 3.10 Gbekan-rukan D 61 Plate 3.11 Okan yi lule o ku okan rocks A 62 Plate 3.12 A building over the Iragbiji pa vIeBment 64 Plate 3.13 Iragbiji potsherd pavement 65 Plate 3.14 Motin shrine at Igbajo F 74 Plate 3.15 The king making sac rOifice to the ancestor during Ori-Oke festival at IragbijYi 79 Plate 4.1 Hoe bla TSde aIt the mouth of the ajere pot (perforated pot retrieved from the exposed Ajaba pavement 92 Plate 4.2 PEetrRol Station under construction on the Ila-Orangun IVpavement site 96 Plate N 4.3 1 by 2 metres pegged for excavation 98 PlUate 4.4: Spit level one 100 Plate 4.5: Excavation in Progress 101 Plate 4.6: Roots and rootlets occurred right from spit level one to level 10 102 Plate 4.7: Sterile Layer of the Excavation 103 Plate 4.8 Measuring and Drawing of the Stratigraphic Units 104 Plate 5.1 Jaw of an animal 110 vii Plate 5.2 Iron slag 110 Plate 5.3 Iron Object 110 Plate 5.4 Iron Object 110 Plate 5.5: Comparing the Iron Objects 110 Plate 5.6: Human tooth 110 Plate 5.7: Baked clay/Fragment of Furnace? 111 Plate 5.8: Plain 118 Y Plate 5.9: Carved roulette 11R8 Plate. 5.10 Plain rim/Twisted cord RA119 Plate 5. 11 Roulette IB 119 Plate 5. 12 Groove/Wavy Groove L 120 Plate 5. 13 Twisted cord N 120 Plate 5. 14 Incision/roulette 121 Plate 5. 15 Plain DA 121 Plate 5. 16 Bossing/Twisted cord A 122 Plate 5. 17 Wavy/Groove B 122 Plate 5. 18 Composite F I 123 Plate 5. 19 Burnished 123 Plate 5. 20 Burnished/GroovYed O 124 Plate 5. 21 Incision 124 Plate 5. 22 PerforatSed PIo Tttery 125 Plate 5. 23 Painted Pottery 125 Plate 5. 24 SEherRds for geochemical analysis from Iloran (Ile-Ife) ancient community 129 Plate N5. I25 V A building under construction at Ita-Yemoo, Ile-Ife 130 PlUate 5. 26 Cutting through potsherd pavement during excavation of a building foundation at Ita-Yemoo, Ile-Ife 131 Plate 5.27 Sherds from Ajaba Pavement excavation for Geochemical analysis 132 Plate 5. 28 Sherds from Ajaba Mound Excavation for Geochemical analysis 133 viii Plate 5.29 Sherds from Iragbiji for Geochemical analysis 134 Plate 5.30 Sherds from Iresi for Geochemical analysis 135 Plate 5.31 Collection of Soil Samples from each layer for Pollen analysis 147 Plate 5.32 Photomicrographs of palynomorphs (all magnification x1000) 150 Plate 5.33 Photomicrographs of palynomorphs (all magnification x 1000) 151 Plate 5.34 Section of Ajaba Wall 162 Plate 5.35 Buried Pot at a Section of Ajaba Wall 163 Y Plate 6.1a Femur of an animal (a large mammal) 16R9 Plate 6.1b Jaw of an animal BR A169 LI N AD A B OF I Y SI T R E NI V U ix LIST OF FIGURES Fig. 1.1 Map of Nigeria showing Osun State 4 Fig. 1.2 Map of Osun State showing the study area 7 Fig. 1.3 Map showing Osun Northeast showing the study area 8 Fig. 1.4 Generalised geological map showing the study area. 10 Fig. 1.5 Soil Association of study area 14 Fig. 3.1 Kiriji War Site map 54 Y Fig. 3.2 Geomagnetic profiling of Ajaba Potsherd Pavement 68R Fig. 4.1 Site plan of Aganju Tapa Potsherd Pavement RA (Aganju Tapa) Excavation B 87 Fig. 4.2 Sketch map of Ajaba town showing the location of thLe exIcavated mound 89 Fig. 4.3 Site Plan of Ajaba Potsherd pavement excavatioNn 90 Fig. 4.4 Stratigraphy of the western flank of the sDite A 93 Fig. 4.5 Plan showing the exposed excavatedA pavement 94 Fig. 4.6 Stratigraphy of Ajaba Mound eIxcBavation 104 Fig. 5.1 Pottery typology 115 Fig. 5.2 REE patterns of sherds noFrmalized to chondrites and compared patterns of regoliYth f ro Om granite 143 Fig. 5.3 Principal comTponent analysis and cluster 144 Fig. 5.4 ComparSisonI of abundance of forest trees (forest resources) with PottRery abundance (index of human population) from level 1E00-10cm 157 IV UN x LIST OF TABLES Table 3.1: Table of X and Y coordinates of most promising anomalies 71 Table 5.1: General Artefact Inventory Ajaba Refuse Mound I. 109 Table 5.2: Distribution of Major Pottery Decoration Types (Body) 116 Table 5.3: Distribution of Major Pottery Decorations (Rims) 117 Table 5.4: Major element composition of pavement samples and raw materials 13Y7 Table 5.5: Trace element composition of potsherds from south-western A R 138 Table 5.6: Rare earth element composition of potsherds from south-wesRtern Nigeria B 139 Table 5.7: Chondrite normalized rare earth composition of pots hLerdIs and granite regoliths from southwestern Nigeria 140 Table 5.8: Phytoecological groupings: N 152 Table 5.9: Figures of Tree Resources and Pottery DA 157 IB A OF TY SI VE R I UN xi TABLE OF CONTENTS Title page i Abstract ii Acknowledgements iii-iv Certification v Dedication vi List of Plates` vii-ix Y List of Figures x R List of Tables xiR A Table of contents IBxii-xvi CHAPTER ONE: INTRODUCTION L 1.1 Introduction N 1-3 1.2 Geographical Background A 3 1.2.1 Location D 3- 6 1.2.2 Climate BA 6 - 9 1.2.3. Geology and Geomorphology I 9-11 1.2.4 Relief F 11 1.2.5 Drainage system O 11-13 1.2.6 Soil Y 13 1.2.7 Vegetation T 13-16 1.3 Historical BackSgrouInd 16-18 1.4 Previous ArRchaeological Investigations in the area 18-20 1.5 Scope oEf the Study 20 1.6 ReIseVarch Aims and Objectives 21 1.7 NMethods of Research 21-23 U CHAPTER TWO: THEORETICAL FRAMEWORK AND LITERATURE REVIEW 2.1: Theoretical Framework 24 2.2: Historical Archaeology 24-27 2.3: Ethno-archaeology 28-29 xii 2.4: Oral Tradition Account 29-33 2.5: Settlement and Settlement Archaeology 33-34 2.6. Archaeology of Conflict and Migration 34-38 2.7: Literature Review 38 -39 CHPTER THREE: ARCHAEOLOGICAL SURVEYS 3.1: Archaeological surveys 40 Y 3.1.1: Archaeological surveys of Oyan, Asi, Ila-Orangun, Iresi, Ire, Ajaba 40R-43 3.2: Account of Archaeological surveys of Oyan, Asi, Ila-Orangun, RA Iresi, Ajaba B 43 3.2.1: Oyan LI 43 (i) Potsherd pavement 43 (ii) Town Wall N 43-44 (iii) Igbo-Ogun DA 44 3.2.2: Asi A 44 (i) Igbo ‗Riro/Potsherd Pavement B 44 (ii) Igbo-Igbale F I 44 3.2.3: Ila-Orangun 45 (i) Ila Kodomu OY 45 (ii) Ila-Yara 45 (iii) DefenseS TreIn Tch and Embankment 45-46 (iv) MouRnd (Oladotun Akanbi‘s Farm) 46 (v) IEbu Ogun (Ogun‘s Water fall) 46 (viI) VIla-Orangun (Ile Ila) 46-47 N(via) Potsherd pavement 47 U (vib) Igbo Atokun 47 (vic) Town wall 47 3.2.4: Iree 47 3.3: Archaeological surveys (Igbajo, Iragbiji Oke-Ila, etc) 49 3.3.1: Igbajo 49-53 3.3.2: Kiriji War Site 53 xiii 3.3.3: Iresi 53 3.3.4: Iragbiji 57 3.3.4.1: Ojudu 57 3.3.4.2: Gbekan-ru-kan 57 3.3.4.3: Okan yi lu‘le o ku okan 57-63 3.3.4.4: Potsherd pavement 63 3.4: Geo-physical methods of Archaeological prospecting 63-66 Y 3.4.1: Data Acquisition 66-67 R 3.4.2: Data Processing 67R-69A 3.4.3: Data Interpretation B69-70 3.5: Summary of Collected Oral Tradition LI 70 3.5.1: Oral History of Igbajo 70-77 3.5.2: Oral History of Iragbiji N 77 3.5.3: Oral History of Edemosi/Ajaba DA 78-80 3.5.4: Oral History of Asi A 80-81 3.5.5: Oral History of Ila-Orangun B 81-83 3.5.6: Conclusion F I 83-84 CHAPTER FOUR: EXCAVATI OONS AT AJABA AND ENVIRONS 4.1: Excavations at AjabaT anYd Environs 85 4.1.1 Excavations of SpotsIherds pavement at Aganju Tapa near Asi 85-88 4.1.2 Excavation at Ajaba Pavement site (KJAJ) 88-91 4.1.3 ExcavatEion Rat Ipetumodu (IPIS) 91-95 4.1.4 ExIcVavation of Potsherd pavement at Ila Orangun 95 4.1.5 NAjaba Mound Excavation (KJAJ RM1) 95-99 4.U1.6 Stratigraphy Description 106-107 CHAPTER FIVE: ANALYSES OF FINDS 5.1. Analyses of Finds 108 5.1.1 General Artifact Inventory 108 5.1.2 Pottery Analysis 112 xiv 5.1.3 Pottery analysis KJAJ RM1 112-113 5.1.4 Pottery Decoration Classification 113 5.1.5 Pottery Typological Classification 113-114 5.2. Geo-Chemical Mineralogical Analysis 114-127 5.2.1 Materials and Method (Pottery Description for Geochemical Analysis) 127 5.2.1.1 Ile-Ife 127 Y 5.2.1.2 Ajaba 127-128 R 5.2.1.3 Iresi 12R8 A 5.2.1.4 Ila Orangun B128 5.2.1.5 Ipetumodu LI 128 5.3 Geo-Chemical Analytical Methods N 128-136 5.3.1 Results and discussion 136 5.3.2 Mineralogical Analysis DA 136-141 5.3.3 Geo-Chemistry Description A 141-142 5.4 Palynological Analysis IB 145-146 5.4.1 Material and Method of pFalynological analysis 146 5.4.2 Palynological AnalysOis Methods 148 5.4.3 Palynological AnYaly sis Result 148-149 5.4.4 Palynological Analysis Discussion 153-156 5.4.5 Orname TSntalIs in Palaeo-ecological studies 156-160 5.5 OtheRr Archaeological Feature 160-161 CHAPTER SIEX: DISCUSSION AND CONCLUSION 6.1. VNDiIscussion 164-171 6.U2. Conclusion 172-174 6.3. Future challenges 174-175 References 176-186 Appendix 1 Glossary of Yoruba Words 187 Appendix 11 List of Informants 188-189 Appendix 111 Geo-Magnetic data readings 190-205 xv Appendix 1V Geo-magnetic data graph 206-255 Appendix V Data Interpretation Tables 256-272 Appendix VI Calibrated Date result 273-274 RY BR A N LI DAA F I B Y O T RS I IV E UN xvi CHAPTER ONE INTRODUCTION 1.1: Introduction This study entails archaeological and anthropological investigations in parts of northeast Osun. It also touches on aspects of previous archaeological works carried out in the study area. The work examines the concepts of conflict, war and displacement in the light of archaeological and anthropological evidences fromR thYe area. Before now, most archaeological works done in southwestern NigeAria were concentrated on Ile-Ife, Benin, Old Oyo and Owo focussing mostly Bon Rart works in bronze, terracotta, and wood. Other themes especially conflict, wLar Iand displacement vital to the areas‘ history and archaeology have been neglected. To understand this neglect, and other related issuAes iNn parts of Osun Northeast of Osun State, Nigeria, series of archaeological workDs were carried out since 1992 by the author. The first phase of the research work wAhich took the author to Ila-Orangun, Ila-Yara, Oyan, Asi, Asaba, Iresi and AjIaBba (Kajola) was meant to expose the archaeological potentials of this paFrt of Osun northeast for future intensive archaeological investigation (O gOunfolakan 2006). The second phase was a continuation of the previous exercise with visits to other notable towns and villages in Northeast Osun such as OTke-YIla, Ila-Magbon, Para-Oke, Otan-Ayegbaju, Ire, Ikirun and revisit to Iresi whicIh had earlier on been visited. This second phase was also aimed at exposingR theS archaeological potentials of these towns and villages. Also a potsherd paveEment site at Ajaba earlier identified during the first phase was excavatedI.V This excavation was meant to identify the material content, orientation, patterNn and design of the pavement, so as to relate it to other known pavements in oUther part of Yorubaland (Ile-Ife, Ila-Oranguun, Iragbiji, Otan-Ayegbaju, etc (Ogunfolakan 2007). It was also meant to assess the relationship between the pavement and a sacred grove ‗igbo‘aiko‘. The sacred grove was later revealed as part of the abandoned settlement. Its investigation forms one of the bases for the study. In realising these goals, both archaeological excavations and anthropological investigations were carried out in Ajaba an abandoned settlement earlier identified during the previous works in the study area. A refuse mound earlier identified within 17 Ajaba abandoned settlement was excavated. The mound was identified during archaeological survey of the area. It is expected that the study would throw more light on the culture history of Northeast Osun in particular and Yorubaland in general. It is also expected that this investigation would contribute to the emerging body of archaeological and anthropological data with regard to the peopling of this part of Yorubaland. Archaeological investigation in the study area started only in the 1990s when the author and other colleagues reconnoitered the area (Ogunfolakan 1994, 2007Y, Akpobasa 1994). R Initial investigations into the archaeology of northeast Osun weRre gAenerally targeted at exposing the archaeological potentials of the area as earlier stated. The first two phases of the project was able to reveal quite a large array oLf cIuBltural entities in the area. Chapter One introduces the subject matter of this studNy and examines the issue of conflict, war displacement and archaeology of NorthAeast Osun State, Nigeria. It further went in-depth into the geographical bacAkgrDound, location climate, geology, relief, drainage system, soil, vegetation,I Bhistory of the region, and previous archaeological works carried out in the s tudy area. Previous archaeological works around the study area were also higOhlighFted. Chapter Two provides the theoretical framework for the thesis. Migrational factors such as conflict, waTr anYd displacement, trade, religion, personal adventure and other parameters of peopIle‘s movement were discussed. The theoretical prop for the study, historical RarchSaeology, ethno-archaeology, geo-archaeology and settlement archaeology arEe the discussed. Chapter Three examines the archaeological surveys carried out in the study area. NTheIse V include the 1994 pioneering as well as current archaeological works in the arUea by the author. This chapter thus discusses the preliminary reconnaissance surveys carried out in towns and villages (Oyan, Ila-Orangun, Ila-Magbon, Ila-Yara, Para-Oke and Oke-Ila), Iresi, Ikirun, Iragbiji and Ajaba) were also investigated. There were potsherd pavements in almost all of these towns and villages most especially, in all of those towns that claimed affinity with Ile-Ife (Ila-Orangun, Iragbiji, Oyan, Ire, Iresi, Ajaba and Oke-Ila). Excavations of some of these pavements are discussed in chapter four. 18 Chapter Four focuses on archaeological excavations at Ajaba and environs. The first was the excavation of a potsherd pavement site within the study area at Asi (Ogunfolakan 1994). The second was at Ajaba (Ogunfolakan 2007). The first and second excavations were meant to identify the orientation of the pavements, the materials and technology for comparison with the known pavements in other parts of Yorubaland. In continuation of the excavation exercise, another pit was opened on one of the mounds earlier identified in the course of archaeological survey of Ajaba. This chapter, therefore, discusses in detail, the excavation of this refuse mound. Y Chapter Five discusses the analyses of finds. The analyses include theA geRneral inventory of both the organic and non-organic materials. The chapter also discusses palynological analyses of soil samples taken from different layers oBf thRe excavated refuse pit. In this chapter, geo-chemical analysis of pottery shIerds from Osun Northeast and some other parts of Yorubaland was carried ou t Lto provide data for comparative analysis. N Chapter Six attempts an integration of the analyAsed data. It also provides an interpretation of the integrated data and offer sugAgesDtion for future work around this area. IB 1.2: Geographical Background: F Ajaba is located in Northe aOst of Osun State, Southwestern Nigeria (Fig. 1.1). The abandoned site is withinY the present day Ajaba town at a junction linking the northern and southern parIt Tof Osun State, on one hand and Igbomina and Ekiti land on the other. Ajaba Ris aSlong Iresi-Ila-Orangun road. It also links Oke-Ila and other satellite villagEes within Ila-Orangun region while it links the south via Edemosi to Esa-Oke IinV Ijesa land. It is a market town that enjoys patronage from different part of the coNuntry as traders from both the northern and eastern part of the country patronize thUis market on weekly basis for farm products (Plate 1.1). The present site is to the western side of the present town of Ajaba (Fig. 1.3). In discussing the geographical background of Osun northeast it is necessary to note the location of the study area, its climate, geology and geomorphology, relief, drainage system, soil and vegetation. 1.2.1: Location: Osun State where northeast Osun is situated is located towards the northeast of o Southwestern Nigeria. The northeast area of Osun State lies between latitude 7 50‘ 19 Y AR LIB R DA N IB A Figure 1.1: Nigeria showing Osun FSta te (Ogunfolakan, 1994) O SI TY ER IV UN 20 Y AR BR N LI DA Plate1.1: A typical market day in Ajaba (KAajola). F I B Y O T SI VE R UN I 21 o o o and 8 05‘ N and longitude 4 40‘ and 4 50 E of the equator. The area covers six Local Government Areas namely Odo-Otin, Ila-Orangun, Boripe, Ifelodun, Boluwaduro, Ifelodun and Ifedayo (Fig. 1.2). Ajaba is a small town at the boundary of two Local Government areas in the study area namely Boluwaduro and Ila-Orangun Local Governments Areas. It is along Iresi/Ila-Orangun road. It also links Oke-Ila and other satellite villages within Ila-Orangun environ while it links the south via Edemosi to Esa-Oke in Ijesaland. Interestingly, Ajaba is a ‗resurfaced‘ ancient town as it grew and is still growing on an ancient settlement abandoned some centuries agoY. According to oral information, it first developed as a market place in the middle oRf the last century (1950s) where all the neighbouring villages came on weekly RbasAis to sell their goods. Today, Ajaba has returned to her ancient status of a beaded crown king‘s town with a market in front of the Oba‘s palace. The present stuLdyI sBite is located to the western side of Ajaba town (Fig. 1.3). N 1.2.2: Climate: A The study area has the climate characteristics of theD savanna and tropical rain forest region of Nigeria. To the south of the studyI aBrea Ais a tropical rain forest region while the northern part is predominantly of thFe sa vanna type. For that reason arable farming is practiced more in the northern paOrt of the study area than in the southern part of it. oThe annual temperature averages between 26 C and 27°C, while the average rainfall ranges between 114.3 mmT tYo 139.7 mm annually (Adeoye, 1988). It is humid throughout the year. HigIhest humidity is witnessed around late February and early March. Osun nortRheaSst experiences two seasonal climatic variations similar to those of other WestE African rainforest regions. The two seasons are determined by two airwaves;I oVne is the rain bearing moist Atlantic ‗monsoon‘ wind that travels from the southNwest and the second one is the dry Northeast Trade wind that comes from AUrabian Desert and sweeps through the Sahara. The raining season begins in March and last till early November. In August, the rain becomes weak with light drizzles which the local people refer to as oda ‗August break.‘ This is the period that allows one to clear and prepare the farmlands for dry season planting. ‗August break‘ consists of about two weeks of no-rains and allows residents to clear the farm and burn the grass, as enough sunshine to dry the grass is available during this short break. This period is known as late planting season, whilst the March-July period is the early planting season. These two seasons provide for two periods of maize 22 Y RA R B N LI A BA D F I Fig. 1.2: Osun State showing Os uOn Northeast (Ogunfolakan 1994) TY SI VE R NI U 23 Y AR LIB R AN D Fig. 1.3: Osun Northeast showing areas of stuAdy (Ogunfolakan 1994) IB F O SI TY R VE UN I 24 planting which is referred to as ‗agbado eerun and agbado ojo’ (dry season maize and rain season maize). The latter is punctuated in December and early part of January by dry and cold weather known as harmattan. The temperature is always high and almost uniform throughout the year. 1.2.3: Geology and Geomorphology: Generally, crystalline and sedimentary rocks dominate the geology of Nigeria (Fadipe, 1991). Crystalline rocks are distributed in three major areas of the countryY: (i) a roughly circular area in the North-Central part of Nigeria; (ii) a triangular arRea in Southwestern part of the country and; (iii) an area with three main units in theA eastern and north-eastern parts of the country. Sedimentary rocks bound theseR crystalline rocks area (Mabogunje 1985). The predominant rocks consist of LbanIdBed gneisses and pegmatite whilst the schists are composed mainly of mica sc hists and sillimanite quartzites. A prominent feature of this geology is the oAccuNrrence of basic and ultra-basic rocks which are commonly found as amphDibolites, metapyroxenites, meta-ultramafites and talc-rich schists. The main minAerals of the amphibole schist parent rocks are green hornblendes, plagioclases, wiBth secondary chlorites, epidote and little quartz. The bedrocks of Ile-Ife areaF co nIsist of coloured gneiss of tonalitic to grandioritic composition, which ocOcur as prominent hills and inselberg around Ile-Ife and surrounding towns. Several t extural varieties of this rock have been found. The most common type is a mTediuYm-grained rock with regular and persistent banding of milimetric to centimetricI thickness. A variety of these rocks is coarse-grained, pink coloured gneiss wRith Slarge lensoid alkali-feldspar porphyroblasts 2-3 cm long. The bedrock of AjaEba, Iragbiji, Ipetumodu and Ikire area are broadly similar to that of Ife except foIr Vvariation in outcrop patterns of mica schists and mafic-ultramafic rocks. In Ila-ONrangun, the bedrocks consist mostly of quartz mica-schist with sub-ordinate oUccurrence of granitic rock. (Ige et al: 2009). From Ikirun to Ire and Iragbiji, granite gneiss dominate. From Iree to Iresi via Otan-Ayegbaju, rock types are mainly porphyrite granite. The Ila-Orangun and Oyan area is dominated by schist and epidiorite complex (Geological Survey map of Nigeria, Iwo Sheet 1: 25000). The study area in essences belongs to the undifferentiated basement complex and undifferentiated meta-sediments (Fig. 1.4) Principally, most of the towns here are located within and on rocky hills. Because of this situation, the towns are today 25 RY BR A LI AN Fig. 1.4: Generalised geological map showing AtheD study area (Adapted from Ige et al 2009). B OF I ITY RS VE UN I 26 developing along the stretch of the roads that cut through these ridges (emergence of linear settlement in the area). 1.2.4: Relief The relief features of the south-western Nigeria, where Osun State is situated may be roughly divided into three belts. The first belt is formed by peninsulas, islands and sand-bank, together with the southern edge of the mainland, including the intervening lagoon and swamps. It is about twelve miles wide, raised only slightlyY above the sea-level and consists of sea-sand and river sand, alluvium, and decaRying vegetable matter, the whole being more or less covered with aquatic plaRnts Aor forest vegetation (Fadipe, 1991:22). As the movements go inland, the secondB belt is noticed with the ground rising gently yet steadily northward to an altitude oIf about 198.12m. The belt consists of mainly unstratified red clayey-sand, kno wLn as Benin sands. Northward of the second belt, is the third belt which mAakNes Yoruba country hilly, being mainly a dissected northward-rising plain withD isolated hills or groups of hills, usually of granite, rising out of the plain to hAeights varying between 15.4m and 243.84m. Osun northeast under study falls wiBthin this northward belt. The rising hills played a prominent role during the cFonf liIcts and wars that led to migration and displacement that took place in thOis region. Looking at the relief of the main study area, Ajaba, the relief within the vicinity of Ajaba and Ila-Orangun to Ede and Ila- Yara, they fall within valleTy ofY the Yoruba ridges that encapsulate the study area. I 1.2.5: Drainage RsystSem: Osun nEortheast is well drained by numerous but small seasonal streams and rivers thaIt Vhave their sources from the hills on which most of the towns in the region are siNtuated. Most of the streams are seasonal. Some of these streams bisect the toUwns; the Afo stream bisects Ikirun, while Isoko stream virtually splits Iresi into two. These streams later merge as they flow to form a large river known as Oyi River in the Ila region to the north of the study area and Osun River in the Ikirun axis to the western part of the study area. Archaeological survey in Osun northeast shows that the gentle slopes of the valleys were favoured for settlement location in the past. Most of the identified abandoned settlements are located on the valley slopes, less than 200 metres from the closest stream. But because the substantial part of the study area is within rocky and hilly region, erosion has continued to wash the cultural materials in 27 RY RAB N LI A AD Plate 1.2: Erosion washes at Iresi (D raIiBnage) F Y O T SI R VE UN I 28 the area. In some cases, it has virtually washed the foundation of buildings thereby exposing the stratigraphy of the soil on which the building was built (Plate 1.2). This scenario is well pronounced in Igbajo, Iree, Otan-Ayegbaju, Ila-Orangun, Oke-Ila, Iresi, and Iragbiji where gully erosion is vastly eroding the area. 1.2.6 Soil: Osun northeast is characterized by gentle to moderate slopes, and the soils of this area belong to both the Iwo, Egbeda and Okemesi soils associationYs (Reconnaissance Soil Survey, Western Nigeria, 1961). This soil type as clRearly identified occurs on the foot-slopes of quartz ridges that dominate this arAea. The texture of the soil ranges between very coarse, coarse and fine textureB (FRig. 1.5). The soil colour also ranges between greyish brown and brown, redIdish brown and brownish red and pale greyish brown to brown. To the northern p aLrt of the study area (Ila-Orangun, Asi, Asaba and Oyan) the soil texture is AheaNvily close to the surface, and the percentage of gravel and quartz increases wDith the depth. It is of sandy to fairly clayey soil, overlying weathered rock material, fresh rock at depth of 2.4 to 2.75 m shallower on steep slopes with fairly IfBreq Auent rock outcrops (RSS, Western Nigeria, 1961). To the south and speciFfica lly in Ajaba area, the soil there belongs to the Iwo soil association. Since the Ostudy area falls within this soil association i.e. Iwo association and that the characte ristics of the soil is of sandy to fairly clayey soil, permanent crop farming likeY cocoa, oranges, kolanut and other arable plants are adequately encouraged. TIhTis then clearly shows why these crops thrive better in the area. This could haSve been the reason why in the first instance the first settler of Ajaba settled aErouRnd this place. 1.2.7 NVIeg Vetation: U Nigeria is covered by three types of vegetation: forests (where there is significant tree cover), savannas (insignificant tree cover, with grasses and flowers located between trees), and montane land (Fadipe 1991). The latter is the least common, and is mainly found in the mountains near the Cameroonian border. Both the forest zone and the savannah zone are divided into three parts. Some of the forest zone's most southerly portion, especially around the Niger River and Cross River deltas, is mangrove swamp. North of this is fresh water swamp, containing different 29 Y AR LIB R AN Fig. 1.5: Soil Association of study area (AdAapDted from Reconnaissance Soil Survey, Western Nigeria, 1961) B OF I SI TY VE R I UN 30 vegetation from the salt water mangrove swamps, and north of that is rain forest. The savanna zone's three categories are divided into Guinean forest-savanna mosaic, made up of -plains of tall grass which are interrupted by trees, the most common across the country; Sudan savannah, similar but with shorter grasses and shorter trees; and Sahel savannah patches of grass and sand, found in the northeast. In Yorubaland, there are four principal vegetation belts. The first is within the coastal region (Fadipe 1991). The vegetation of the northern part of the study area is similar to that of other savanna areas. To the south is tropical rain forest where timbeYr and kolanut (Cola acuminata or C. nitida) are exploited while cocoa Rtrees (Theobroma cacao) are also cultivated in abundance (Cocoa actually form thAe major permanent and exportable crop in this area). Palm trees (Elaeis guinBeenRsis) are also abundant in the forest region (Ogunfolakan 2007). I The vegetation of Osun northeast is generally derived s aLvanna with relics of trees typical of the drier type of rainforest vegetation. ThNis suggests that Ajaba is within the northernmost extension of the rainforest zone. A In the immediate vicinity of the present sAtudyD area and most especially within the excavated mound, about three to fiBve metres (3-5m) away, are Khaya grandifoliola, Chrysophyllum albidum,F Na uIclea diderrichii, Ceiba pentandra, Elaeis guineensis, Raphia vinifera, Millicia excelsa (Syn.chlorophora excelsa) Tetracarpidium conophorum (A fOrican walnut tree), Alstonia boonei, Dracaena africana, Vernonia amygdTalinYa, Cordia millenii, Chromolaena odorata, Cassia sp and members of AmaraInthaceae. There are also cultivated plants such as cocoa (Theobroma cacaRo), cSassava (Manihot utillissima), yam (Discorea bulbifera), pepper (Capsicum sp)E, banana (Musa paradisiaca), plantain (M. sapientum), and cocoyam (ColocasiIaV sp.). Others are pawpaw (Carica papaya), orange (Citrus sinensis), coconNut (Cocos nucifera) and pineapple (Ananas comosus) (Orijemie et al. 2010). U To the northern fringe of the study area (Oyan, Asi, Asaba and Ila-Orangun) the vegetation is derived savanna, trees are more scattered here. In some areas, most especially along the valley or along the course of the rivers and streams that cut across Osun northeast, porch of forest is noticeable. To the southern fringe, secondary forest can still be noticed and economic plants like cocoa, Kolanut trees and in some cases, cashew are still planted. The following plants were found within a radial distance of four hundred to eight hundred (400-800m) metres from the excavated site: Blighia sapida, Cocos nucifera, Newbouldia laevis, Tectonia grandis, Carica papaya, 31 Thaumatococcus danielli, Samanea saman, Duranta repens (yellow bush), Thevetia peruviana and Napleona sp. (Ogunfolakan 2007). It is sad to note that the sacred grove (‗Igbo riro‘) the traditional reserved forest which should have given us the idea of the primary/secondary forest within the abandoned site has of recent been badly deforested. Most of the trees have been cut down without replacement (Plate 1.3). Out of about five acres of land left as sacred grove, about four acres have been desiccated. But the remnant clearly shows that the area was once a forest or better that the area originally was a thick forest before theY deforestation. R A 1.3: Historical Background: R Osun northeast as an entity does not have a collectively writteInB history, though the area remains that part of the Yoruba ethnic group that gaine dL prominence during the Yoruba wars of the 19th century. Specifically, IgbajNo the camp site for the Latosa/Ibadan group during the Kiriji War of the 19thA century remains a centre of political/military history for Yorubaland. UnfortunateDly, apart from the records of the Yoruba War written by Samuel Johnson iInB his A famous book ‗The History of the Yoruba…‘ (Akintoye 2010), most of thFe b ooks relating to Osun northeast, the towns and specifically, Igbajo, deal with Othe Kiriji War and not with the early history and archaeology of the area. Historica l records for each town and city in this region still remain in the archives of tThe oYral historians. Though, more write-ups are coming out now for towns such as IIla-Orangun (Adetoyi 1974, Pemberton and Olaniyan 1996, Adebayo 1996 amRonSg others) Igbajo (Tugbiyile 1989, Fashiku 1995) and Otan-Ayegbaju). MEost of their histories are only recorded on the pages of pamphlets for chieftaincIyV installation, social programs or long essays submitted by graduating studeNnts of history in some of our universities. U The people of Osun Northeast are united by overlapping socio-cultural institutions, historical traditions, religion and world view. The intermingling of the people, most especially before, during and after the Yoruba Wars of the 18th and 19th centuries led to demergence of three sub-groups of the Yoruba entity (Ogunfolakan 2007). These groups include the Oyo Yoruba group of Ikirun, Iragbiji, Oyan, and Asi; Ijesa Yoruba group of Igbajo, Iresi, Otan-Ayegbaju and Igbomina Yoruba group of Ila-Orangun, Ajaba and Oke-Ila. This dialectical division is reflected in the Yoruba language dialects each one of these groups speaks. 32 RY RA LI B AN AD Plate 1.3: Deforestation of the sacredI gBrove F Y O SI T R VE UN I 33 Oral history of most of the people in the study area points to the fact that they migrated either from Oyo or Ile-Ife. For instance the people of Ila-Orangun claimed that their forebearer came from Ile-Ife before finally settling at their present site after more than four different places of settlements (Ogunfolakan 2006). In Asaba near Oyan, the oral history recorded that they came from Oyo and here Egungun festival is prominent (Ogunfolakan 1994). In Oyan it is claimed that their ancestor, Epe, came from Ile-Ife and settled first at Igbo-Ogun (Ogun grove) and finally at their present site on the instruction of Ifa oracle after wandering through thick forest (Ogunfolakan 1994). RY In some towns, there are mixtures of people migrating from differeAnt areas and coming together to form a town. In such cases, each group of peopleR maintains a separate identity and their area known either as a quarter or familyI Bcompound. For instance, the people of Ede settled at Ila-Orangun in the area kno wLn today as Oke-Ede this group of people are also found in Igbajo, Imesi-Ile anNd Otan-Ayegbaju. Igbajo town of today also displays an interesting relationshDip oAf people coming together to unite under the same town polity but maintaininAg their self-identities. For instance, Igbajo is known as Igbajo Iloro i.e. Igbajo Bof Iloro. Today chieftaincy title of the paramount chiefs is duplicated. For inFsta nIce we have the Ejemu Iloro and Ejemu Igbajo, Odoffin Iloro, Odofin IgbajOo etc, but all of these chiefs are subject to Owa the paramount ruler of Igbajo. How had the conflict aYnd war leading to dispersal of people from one town or the other influenced the IgTrowth of others in this area? How did this influence the socio-cultural affiRnityS of the people? These are pertinent questions that form the focus of this present Ework. 1.4: PrIeVN vious Archaeological Investigations in the area U Archaeological investigations in this area started only in early 1990s when the author and others carried out an archaeological survey of the area around Ila-Orangun, Oyan, Ajaba, Asi and Asaba (Ogunfolakan, 1994; Akpobasa, 1994; Usman, 1995). These early works were generally targeted at documenting the archaeological, cultural and historical sites in this area so as to expose the area archaeologically. The author was thus interested in documenting the occurrence of potsherd pavements in this area so as compare and relate the pavements to known one in Ile-Ife and Old-Oyo. After exploring the area, an exposed pavement in one of the sites (Asi) was opened so as to 34 identify its orientation, design and material content. The excavation then did not go below the pavement level. In all, the first phase of the work by the author was aimed primarily at mapping archaeological, historical and natural features for future detailed archaeological investigations and also to locate potsherd pavements in the area. Akpobasa (1994) also carried out archaeological investigations of the abandoned site of Ila-Yara in his pursuit of the archaeology of Igbomina land. In his work, he identified series of occupational features including a defence trench and embankment, refuse mounds, some natural formations such as a small water fall. HYe archaeologically surveyed these features for further work. R In 1995, in his pursuit of survey of fortifications in the northern RperiAphery of Old Oyo, Usman et al, (1995) carried out a survey description of Ila-yara trench and embankment earlier identified by Akpobasa, (Akpobasa, 1994). AccIorBding to Usman, the aim of the work was twofold. ‗The first is to provide a surv eLy description of the walls, and the second is to explore the functional interpretatiNons of the walls‘ (Usman et al, 1999:45). Usman et al (1995: 45) is of the opinion tAhat ‗The study of fortification in this part of Yorubaland will provide an understaDnding of the nature of Old Oyo frontier and the prehistoric social relations IinB the A area, especially between Igbomina, and the Nupe‘. In 1997, the author carried ouFt further archaeological explorations of other parts of Osun northeast with vis itOs to other areas that were not earlier visited and revisited Iresi and Ajaba. TAlsYo at Ajaba, a test geological probing of archaeological site for archaeological feIatures and artifacts was carried out with geo-magnetic and electro-resistivity RsurvSey of the pavement site area. The idea was to test the usefulness of this methodE in location of archaeological sites. Thereafter, the author carried out excavatioInVs of a potsherd pavement earlier identified within the vicinity of Ajaba (OgunNfolakan, 2007). This was carried out in other to (i) confirm the geo-magnetic anUd electro-restivity data generated from the survey carried out; (ii) know the extent of the pavement; (iii) examine the orientation and design; (iv) check the materials used and; (v) identify if possible, other associated materials and finally determine the relationship of the pavement to the whole Ajaba abandoned site. The result of the investigation shows that (i) evidence of archaeological materials; (ii) the potsherd pavement bears resemblance with known Ife pavements with pots embedded in the middle of the pavement and; (iii) the result of the excavation shows that different people were working on the pavement at almost the same time because of different 35 orientations the pavement faced. From these surveys it was possible to identify the extent and cultural compositions of the Ajaba abandoned site i.e. refuse mounds, a semi-circular defence wall and other features. Usman et al (2005) carried out excavations of a mound earlier identified by Akpobasa in 1994. This work according to Usman et al (2005) was to examine the ceramic of Ila-Yara and serrate it with other Igbomina ceramics and also relate it to the Old-Oyo ceramic. The essence is to examine the relationship of Igbomina with Old-Oyo and probably establish the origin of the Igbomina people. Y In continuation of the archaeological research work in the area, in 2005/2R006, the author carried out excavations of a cultural mound within the previouslyA studied abandoned site of Ajaba where a pavement was earlier excavated. The daRta from this excavation with other previous data form the core data for this prIeBsent work. The present work is thus a continuation of archaeological investigatio nLs in northeast Osun. 1.5: Scope of the Study AN The present work is focused on a better underDstanding of the effect of conflict, war and displacement in parts of YorubIalBand A, with specific reference to Osun northeast of Osun State, Nigeria, fromF a rchaeology, oral history and ethnographic perspectives. This is done so as Oto examine how this has affected the settlement pattern of this area within the con text of Yoruba settlement. It is also to decipher the inter-regional relationship oTf YYoruba towns and villages. What is or are the Irelationships among the people that moved and intermingled over periods of timRe wSithin this region and other Yoruba towns, most especially Ile-Ife and Old-OEyo? Within the context of understanding Igbominaland and culture, what linkIeVd these groups of settlements together in the beginning, at certain periods and oNver time? U The present research work which includes excavation of a cultural mound within the abandoned settlement site at Ajaba is aimed at relating and inferring the relationship between the cultural materials from this area and the materials from previous excavations and surface materials from other sites earlier identified by the author and other researchers in northeast Osun on the one hand, and Ile-Ife and Old- Oyo cultural materials on the other hand. 36 1.6: Research Aims and Objectives: The goal of this research is to understand how conflict, war and displacement have affected the settlement history of the study area. The objectives are therefore to: (i) examine the relationship amongst Ajaba and other settlements within the study area on the one hand and other Yoruba groups in southwest Nigeria on the other, so as to have a better understanding of the effect of conflict, war and displacement on the archaeological landscape of the area; (ii) ascertain human interactions with the environment in this region using the cultural materials from both archaeologicaYl excavations and ethnographic surveys and; (iii) examine the effect or otherwisRe of war, conflict and displacement on the people of the study area. The RconcAomitant effects of these interpretations on the emergent settlement configuBration are also inveestigated. I In order to realize the goal and set objectives;(i) a criti caLl assessment of the local oral historical model of settlement history, migrationsN and displacement in the study area was carried out; (ii) archaeological data DincAluding dates from the region were assessed and compared to establish the Anature of cultural and genealogical relationships among the settlements within thIeB region. 1.7: Methods of Research F Direct historical approach iOs seen as the appropriate methodology in carrying out this work. The historTicalY approach is integrated with the archaeological data, ethnographic and historiIcal information (most especially, oral traditional historical accounts). Since RoralS tradition encompasses a series of entities – folklore, praise-poems (oriki)E, myths, festival, poetry and host of others, its application in the archaeoloIgVical investigation of northeast Osun is appropriate and expedient in the invesNtigation and interpretation of archaeological data for a better understanding of conflict, war and displacement of the people of Osun Northeast. This will help to uUnderstand what is responsible for the ‗push and pull‘ factors in migration and displacement in the study area. Also, taking into account the sacred sites and places that are still valued by the people of northeast Osun, the: …. critical use of direct historical approach in African archaeology has been found very relevant for situations in which there is a cultural and historical continuum between the archaeological past and the ethnographical present (Ogundiran 2002: 8). 37 The archaeologist has to look for material culture to supplement, understand, and decipher the meanings of the minds and hands of the ancient residents and their conditions. One can also use oral and written historical materials to document and interpret archaeological material or site. The following research methods were employed to establish the migration and displacement pattern, settlement strategies, ceramic use and factors of change in our area of study: i. Re-evaluation of oral history earlier collected among the people in the areaY, paying attention to key informants especially custodians of tradition and cuAltuRre; ii. Archaeological reconnaissance and detailed surveys; iii. Excavation of a mound earlier identified and which is within the aRrea where a potsherd pavement was earlier excavated; IB iv. Laboratory analyses and classification of cultural materi alLs. Classification of cultural materials entailed a general artefact inAvenNtory, pottery decoration analysis, and pottery typology. The laboratory anDalyses include both geo-chemical and mineralogical characterization of potterAy sherds from the excavation and pottery sherds from other sites within nBortheast Osun State and Ile-Ife, most especially from the potsherd pavemFent. Iv. In order to ascertain the aspect Oof the environment of Ajaba and the interaction of the people with this environm ent, a palynological study of some of the excavated soil samples was carried ouYt. In re-evaluating tIhTe historical account of the study area, and specifically, traditional heads oRf eaSch town was always the first target as they are the ‗owners of the land‘ and custodian of the tradition. The traditional rulers are the ones that know who to cIonVtac Et or invite for information. There are times when the traditional ruler himself is not vast in the history of his town, in such a case, elders in the town are uUsualNly invited to the palace of such a ruler for interview. In carrying out the interview, direct interview with the custodian of history and culture was employed instead of using questionnaire. Questionnaire is viewed as limiting the respondent to say what the interviewee wants and not what the respondent know or want to say. Questionnaire a times do not give the required result as majority of the custodians of history and tradition could not read or write, though they are literate in the African sense of it (they have good understanding and memory of the oral history and tradition of their community (Ogunfolakan 2007). Those interviewed 38 were chosen based on the recommendation of the traditional ruler, based on age and observation of their interest in tradition and culture. Also, in some instances, a meeting of the chiefs and traditional title holders were convened to interact with them on the oral history and understanding of their environment (Plate 3.1). Serious attention was not paid to gender equity because from observation, the women always give the men the right to give historical account. Often the women believed that they are wives and may not know more than their husbands who were born and grew in the same community. Y After collecting and synthesising the oral historical account, archaeoloRgical survey of the area was carried out with visits to cultural, natural and historicaAl sites as mentioned in the oral history. The informants often piloted visits to Rhistorical or natural sites. It is now left for the archaeologist to identified archaeIoBlogical, cultural and historical features. This was done in the identification of sev eLral sites and features identified in the study area. Also, with the picture of Npotsherd pavement and rockshelter shown to the informants, pavement at ADjabAa, Ire, Otan and Ila-Orangun and the rockshelters in Iresi and Iragbiji were able to be identified. In the archaeological reconnaissanc AIeB, simple traversing round the site was adopted. Each feature was marked andF su rface material collected for analysis. The road where vehicle was parked alwaOys serves as the reference point. After the archaeological re connaissance, identified pavement at Asi and Ajaba were excavated while a refuseY mound at Ajaba was also excavated. The result of the mound excavation forms IthTe core of this present work. In carryinRg oSut the laboratory analysis, both the surface and excavated materials retrieEved were sorted out (this is discussed in chapter 5) and analysed based on their mIVaterial (organic and inorganic) then ceramics were sorted out according to their Nfeatures rim or body then decoration motives. The analysis went further by the uUse of geo-chemical, geo-physical and mineralogical methods for the identification of material composition, fabrics and mineral contents so as to relate the sherds from Ajaba with the selected sites in the study area and Ile-Ife, Ikire and Ipetumodu. Finally, palynological analysis of soil collected from the mound excavation at Ajaba was carried out (details are discussed in chapter 5). The Palynological analysis was carried out so as to understand the past vegetation of the area and have insight into the man-environment interaction in the past. 39 CHAPTER TWO THEORETICAL FRAMEWORK AND LITERATURE REVIEW 2.1: Theoretical Framework: The archaeologist is committed, among other things to the business of explaining the past. This has been encouraging the development of numerous concepts and methods (Ogundele 2004). Therefore, theories in archaeology are to solve problems and not to further confuse people. Theories are to guide archaeologistYs in their operations so that the archaeological research being carried out can be mRore valued. In carrying out this work, some theories and models employed Ainclude historical archaeology, ethno-archaeology, oral tradition and settlement aRrchaeology, archaeology of conflict and migrations so as to understand anIdB have a good understanding of the effect of conflict, war and displacement in t hLe study area. N 2.2 Historical archaeology DA Historical archaeology developed as a diAstinct branch of archaeology first in the United States and Canada. Here ar cIhaBeologists draw a distinction between prehistoric archaeology, which studies Fthe pasts of indigenous peoples, and historical archaeology, which studies EuropOean sites and indigenous sites with evidence of European contact. Scattered refer ences exist for excavations in European sites from the eighteenth and nineteeTnthY centuries, but it is only in the 1920s and 1930s that historical archaeologyS deIveloped as a coherent field of study. This was a period of great interest in hRistorical reconstruction. Numerous private foundations and the U.S. and Canadian Egovernments sought to study and reconstruct important historical sites. These effIoVrts led to the establishment of many of the major living history museums. HistoNrical archaeology began in places such as Jamestown, Port Royal, Plymouth PUlantation, Saint Marie One, and Colonial Williamsburg (McGuire and Paynter 1991). Archaeologists contributed to these efforts by supplying information on the correct placement and reconstruction of buildings, forts, roads, wells, and other features. They also provided details, such as what artifacts should be placed in the reconstructions. These reconstructions emphasized great men, first settlements, and great events. 40 Historical archaeology has been propounded and defined primarily by European and American archaeologists, and members of post-colonial societies, who were to be more concerned with their own society than with visualizing more universal concepts of archaeology. Their definition may reflect the spread of European culture (Deetz, 1965) or stress the availability of written account of the society under study with tacit limitation to European documents. Posnansky and Decorse (1986) follow Deetz‘s definition as a practical matter to limit the scope of a brief survey, but note the inadequacy of this and related definitions for Africa. WesleYr 1998:13 defined historical archaeology as `the archaeology of peoples who canR also be studied or investigated through historical document‘. Wesler‘s explanatioAn of his definition of historical archaeology is that it draws from two sourcRes of data, archaeological and historical. But his submission that historical archIaBeology includes the archaeology of literate societies, observed by literate societi esL manifests certainly his euro-centrictism. N In Africa, there are orally literate societies whoseA, members could neither read nor write, but they recount, recollect and reciAte Devents as they happened. They perform experiments and administered resuIltBs as deemed fit. Literacy here is more than one who can read and write. Literate s here are those who are conscious of their environments and transmit and transFlate this consciousness into practical forms suitable for their daily living c onOditions and for both their mental and physical development (Ogunfolakan 20Y07). Therefore, historiIcaTl archaeology could be defined as the archaeology of a people, and whose arSchaeological investigations could be carried out through their means of historicaRl documentation. This in essence means that the archaeology of the people cIouVld Ebe studied through their festival, folklore, written and unwritten historNical facts, songs and others. For instance, Yoruba people are known for their `oUriki‘ (praise poem/cognomen). Through this `oriki‘ their history is often recited (Aleru and Alabi, 2010). In a situation where we use the presence of European, foreigners or colonial structures as a sign of historical archaeology, more euro centrism than universal phenomenon are explained. Historical archaeology continues to be divided between humanists and scientists, but much of the fervor of the debate has been lost. This reflects the fact that, since the 1970s, anthropology has become more historical, and history more anthropological. It also indicates a growing realization that good work can be done 41 through both approaches. The core methodological question remains how to integrate artifacts and documents in the study of historic sites. Historical archaeologists promote two complementary positions that stress the value of the study of material culture. The first position grows out of historical archaeology‘s original role of filling in missing information from the documentary record. It focuses on the incompleteness of the documentary record and how material culture can be used to complete historical knowledge. Historical archaeologists recognize that not all time periods and events are equally well represented in the documentary record. In the EuropeanY history of North America, the number of documents that were produced, andR that survive, increases geometrically over time. This makes historical archAaeology especially useful in early periods when few, and often only fragmentaryR, documents are available. Certain events may also not leave a rich documenItBary record. For example, at the Battle of the Little Big Horn (Fox, 1993), none o f Lthe Seventh Cavalry who rode with George Armstrong Custer (1839–1876) surNvived the battle, and the Native American victors left no written accountsA of the battle. Historical archaeologists, such as Richard A. Fox, have been aDble to reconstruct what happened to Custer‘s command on that day and subIstBant Aially confirm Native American oral accounts of the action. In Africa, wFher e few written records exist prior to the European encounter with the regioOn, the use of oral and material records to augment the written evidences from the beginning of the European knowledge of African continent becomes imperatTivvYe. Historical archaeoIlogists note that the documentary record tends to focus on great people and gRreatS events. Even when common documents survive, such as diaries and letters, theEy tend to focus on the personal and emotional concerns of individuals. People doI Vnot generally write about the mundane and day-to-day activities that make up mNost of their lives. In contrast, the archaeological record is primarily made up of thUe remains of these day-to-day activities. Many of the great events of history (e.g., the dispersal of the prince and princesses from Ile-Ife at a point known as ‗Ita-A-jo- ro’ (Ita’Jero meaning a point of decision making) leave little in the way of material remains, but the day-to-day activities of life do. More so, the early Europeans that visited Africa did not believe that Africans could be so creative to have done most of what they saw. No wonder when Leo Frobenius visited Ile-Ife between 1910 and 1911 and saw Ife arts, that he likened Ile-Ife to the lost Atlantis! (Frobenious, 1921). 42 Historical archaeologists stress the consistent biases to the documentary record. It is primarily the literate and the powerful that produce documents and see to their survival. In North American history, wealthy European males tend to dominate the documentary record. The vast majority of people who have lived in North America are poorly represented in the documentary record or not represented at all. This is particularly true of groups that were disenfranchised and the oppressed, such as enslaved Africans, Native Americans, workers, and women. These people did leave physical traces of their lives that historical archaeologists can study. Through thesYe traces, historical archaeologists can learn about the day-to-day experienAce,R the nutrition, and the health of everybody in history (Fox, 1993). Finally, much of the information in the documentary record is Rprescriptive (i.e., it tells people how they should have acted). People did not alwIBays act the way that they were told to. The archaeological record provides a Lcheck between the prescriptive literature and what people actually did. It alsAo prNovides a record of a wide range of illicit and illegal activities ranging fromD opium smoking to smuggling (McGuire and Paynter 1991). The second line of argument begins wBithA the position that archaeology is not simply a poor substitute for the writtenF wo rId. It is, instead, a unique way of studying the world that allows us to learn thOings we could not learn otherwise. People create material culture, and it bears the mark of that creation. From these marks, archaeologists can reconsTtrucYt technologies, costs, trade, and even aspects of the organization of work thatI created these objects. In these ways, material culture gives archaeologists a mReaSns to reconstruct past behaviours. Material culture is also an active agent iEn people‘s lives. Material culture structures what people do, and it carries mIeaVning. Thus, non-material culture may embody emotions, aesthetics, social relatioNns, and beliefs, and these things affect how people will think and act. In this wUay, material culture becomes an active instrument that affects what people will do. Documents are valuable from this perspective because they allow the historical archaeologist greater access to meaning and belief. In this way they supplement the study of material culture, rather than the material culture just filling in what the documents lack. 43 2.3 Ethno-archaeology The aim of ethno-archaeology is focused at discovering general principles that link human behaviour and material culture variability, thus it streses the importance of linking ethno-archaeological studies to archaeological questions. The possibilities for archaeological inference by applying ethnographic insights to archaeological case studies are there. Ethno-archaeologists need to be generalists, with good grounding in both ethnographic and archaeological methods, with a clear sense of how their theoretical leanings shape and potentially limit the models that result from their workY, and with a more clearly defined sense of how their ethnographic research canA fuRrther archaeological interpretation. The preoccupation of ethno-archaeology is the study of the ‗the roRle of human behaviour in forming archaeological accumulations‘ (Gifford 198L0), IanBd its main goal is to aid the interpretation of the archaeological data and d etermine as much as possible the associated behaviour (Atheron 1983). AArcNhaeology is sometimes, according to Kramer, (1979) refers to as fossilized hDuman behaviour that is, a living record of past life preserved through time and aArtifacts or the lack of knowledge of these artifacts. This is when the archaeologisBt starts to employ a type of archaeology called Ethonoarchaeology. Okpoko in Fhis Iinvestigation of Anambra valley defines ethnoarchaeology as the study of aspects of the history (oral tradition), material culture and linguistic data of t hOe present inhabitant of the valley to elicit the information useful for (i) the Ylocation of archaeological site; (ii) the interpretation of such sites, features and IaTrtifacts and (iii) the reconstruction of material and non- material aspects of theS ways of life of former inhabitants of the valley (Okpoko 2006) Folorunso (1993: R98) argued that ‗it would not be correct to treat the use of oral traditionsI aVnd Elinguistic evidence in archaeology as falling under ethnoarchaeology. TheseN fields of study are, on their own, useful for archaeological interpretation and caUnnot be subsumed under ethnography‘. His submission is that the use of oral history in archaeology should be considered within the realm of historical archaeology. Aliyu (2006) defines ethnoarchaeology to mean the use of present cultural information about a group in order to understand the material cultures and ways of life of the ancestors of the same or related people or otherwise, especially in the areas of methods of production, use and discard. Therefore ethnoarchaeology is the theoretical use of analogies derived from present observations to aid in interpretation of past events and processes. In using this 44 method the archaeologist uses the application of deductive arguments to archaeological evidence in order to discover how human beings in different prehistoric societies may have behaved in the past. The idea of ethnoarchaeology can be a very helpful and intriguing tool to the archaeologist when there is a lack of archaeological evidence to support an idea or theory. The best way of explaining past events and practices of peoples is through solid archaeological research and facts that come from the artifacts recovered from a site. But, when there is a question as to what a tool was used for or why certainY buildings are positioned in such a way, ethno-archaeology can be a great aid. It wRorks with the known facts and helps create ideas and hypothesis about the RproAblem or question being asked. Ethno-archaeological research can be most useful when used in an area where cultural continuity is great, for example in most plaLcesI Binn sub Saharan Africa, where traditions and customs survive for generation s. When considering modern day rituals and customs of a culture, it is important Nto remember that they are only modern day rituals and customs and should only beA regarded as a hypothesis on how an ancient custom was performed and must bAe teDsted before being accepted as an explanation for the custom in question. ThisB is important because it doesn‘t matter where an interpretive hypotheses comeFs fr oIm, but rather how it stands up to, and fits the archaeological record. Every archaeologist know s Othat there are limitations in the recovery of historic data and artifacts posed Tby Ythe decay of archaeological material and deposits, sampling errors, seriationI errors, and plain human procedure error. All these problems can affect behavRiourSal interpretations of past events. Archaeologists attempt to understand paEst human behaviour and to constantly find better ways to do this. The methods IoVf ethnoarchaeology step up to the challenge as a legitimate aid in the sciencNe of archaeology as archaeologists develop different ways to overcome liUmitations posed by traditional means. 2.4 Oral Tradition Account Any researcher working on the archaeology of Yoruba in particular and Africa in general must rely seriously on the account of oral history as recorded by the tradition. Oral tradition can be defined as the ‗recollection of the more distant past that has been transmitted for several generations and has become more or less the common property of a society‘ (Henige 1982: 106). Vansina (1960: 56) defines oral 45 tradition as `testimonies of the past which are deliberately transmitted from mouth to mouth‘. They concern past events and are distinct from rumors which always bear the character of sensational `news‘ which are not deliberately transmitted from generation to generation the same way‖ (Andah & Okpoko 1979). Such traditions form the main available source for a reconstruction of the past; ‗in those parts of the world without writing…. and even among peoples who have writing, many historical sources including the most ancient ones are based on oral tradition‘ (Vansina 1960). Oral testimonies should be regarded as those versions of history that are not widely knownY and according to Henige (1982), `if they relate to recent events, they belong toR the realm of oral history. Oral information could be seen as coming from twoR maAjor ways or forms. First is oral history and the second is oral tradition. Either of the two infers a means of transmission of history through the word of mouth from an IinBterviewee to an interviewer (Bakinde 2009). The former refers to the history o f La people; society or group over period of time while the latter is contemporary Nnarrative of an event that might be on going at the time of the interviewD. OAral history then is personal reminiscence solicited by the researcher in an intAerview format, on specific events of interest to the researcher, or on the subjecIt‘Bs perhaps idiosyncratic memories of a family, neighborhood, community or mFove ment (Cooper 2005). Oral tradition on the other hOand, relates to a group of society‘s history. It tries as much as possible to capture the history of a society that has not been documented. Secondly, it concentrates oTn tYhe role or roles played by individuals such as warlords or kings in shaping the Ihistory of a group or society. Oral traditions are generally stories about the pRast Sthat local populations produce and reproduce orally, as a means of preserving their worn history and consolidating or contesting a sense of belonging and identIitVy (C Eooper 2005). NOral history has been and continues to be a `guide‘ in the archaeological inUvestigation of Yoruba country. Oral tradition as history in its own right and as a source of history and archaeology has always been indispensable in the study of the historic and proto-historic periods, for archaeologists often depend on the information acquired from it to locate various archaeological features, sites and abandoned settlements (Ogunfolakan 1994). Oral tradition plays an important role in the location of archaeological sites. As Artherton (1983) and Ki-Zerbo (1981) rightly pointed out, for a proper archeological research in Africa, one needs to start with the collection of 46 both oral tradition and ethnographic data. By so doing, archaeologist will be able to identify sites that would otherwise not have been known. The problem with oral tradition in archeological investigation lies with individual archaeologist in their approach to its use and adoption. Archaeologist making use of oral tradition must allow informants lead in the investigation rather than researcher guiding informants with questionnaires. Guiding informant with questionnaire will lead the informant to say what one wants him/her to say rather than what he knows or wants to say. Also, objectivity may be a problem in the selection oYf those to be interviewed in the collection of oral history, but I tried to be near objeRctive by selecting those to be interviewed across all levels of the society. In thReseA days of political insincerity and the fight for supremacy, one must be very careful in the use of oral history in archaeological interpretation. IB In Oral societies, personal recollections are usually los t Lin the ever-moving current of oral traditions. By surviving into the ApreNsent, oral tradition has demonstrated a capacity to appeal to the interest Dof enough people to ensure its success. In this it differs markedly from personalA recollections, which seldom survive the lifetime of the individual or, at least, the liBves of any children (Henige 1982: 107). Awe (1974; Adeeko 2001) sugges teId that historians of non-literate societies should view oral traditions as valuabFle sources for chronicling the evolution of African social and cultural cons ciOousness, because in societies without permanent archives, creative oratures Tdo Ycapture the mentalities of their milieu of production and consumption. For the archIaeologist who is using material evidence to authenticate the historical facts supRplieSd by written or oral history, he/she must count seriously on oral tradition for hiEs investigation (locating sites) and interpretation of data collected. UIntVil recent past, African history, most especially Yoruba history, remains `awigNbo‘ (hearsay) and according to Yoruba sayings, `awigbo‘ (hearsay) is the father oUf history (awigbo ni baba itan). A historian can rely on `awigbo‘ (hearsay) but to an archaeologist, awigbo is not satisfying. Therefore the archaeologists are not contented with hearsay. An archaeologist must have material things, objects, artifacts that he can hold in his hands, count, measure, and photograph and relate to the ‗hearsay‘. Archaeologist do not necessarily rely entirely on `hearsay‘ of the historian or adhere strictly to their written documents. They do realise that not all that is worth knowing is written by literate societies, or recited by oral societies or even photographed and/or video-taped by computerized societies (Ogufolakan 2007). The aspect of oral tradition 47 that archaeologists have not been paying attention to is Oriki (praise names) and names (personal and place names). The use of Oriki (praise name), personal and place names in the archaeological survey of abandoned settlements in Yorubaland and interpretation of archaeological data is advocated by Aleru and Alabi (2010). In Oriki, there are lots of salient information that are of importance to the archaeologist in his investigation most especially, in the identification of abandoned settlements and in the interpretation of archaeological data. Oriki, a source of historical information is a variety of Yoruba poetry whichY, as an aspect of literature, is `one of the universal means of communicatingR the emotional, spiritual or intellectual concerns of mankind‘ (Akinyemi in AdeekAo 2001). Oriki reminds one of the real life incidents details of which mBightR have been forgotten. In Yorubaland, oriki plays an important role in preserLvinIg the history and culture of the people. It is believed that when ‗oriki‘ is chanted f or a person, it inspires the person and it even invokes the spirit of the ancestor Aof sNuch a person. If a child is crying, it is Yoruba custom and tradition for his/heDr mother or the grand-mother to chant the ‗oriki‘ of such a child. It is believed that ‗oriki‘ cools down the nerves of babies. Ideally, oriki is meant to inspire AIthBe subjects addressed to pursue loftier attainments. (Adeeko, 2001). Also, ‗oriFki‘ inspires people not to derail from the good name of the family. In this wayO it is performing a cultural role of making the addressee see the past, assess the present and project into the future. But importantly, salient information are eTmbYedded in this ‗principal means by which a living relationship with the paIst is daily apprehended and reconstituted in the present‘ (Barber 1994:15)R. ThSerefore, in the oral history of the people of Osun northeast, ‗oriki‘ is partlEy examined in the identification and interpretation of archaeological data colleIcVted. NFestivals were also examined in the study area as part of oral tradition since pUast events are recited and rekindled during the festivals. Festival involves religious and ritual activities and practices. These activities and practices should be seen as concrete practices that left interpretable material remains for archaeological studies and interpretation of past social behaviour among Yoruba people (Whitley and Loendorf 1994). There are features, land marks and event that serve as reference points of oral tradition and these are sacred groves and hills. In the study area, abandoned settlement or sacred grove (Igbo ‗Riro) or rock shelter (Ota ebora i.e. the stone of the demons) 48 are not just relics of the past, they represent the home of the ancestors, from where they breathe a breath of life, where they derive their daily subsistence and way of social life, where their ‗fathers sit and oversee their children‘ (Ogunfolakan 2007). Most of the sacred places, usually abandoned settlements, homes of the ancestors are located on the rocky hills. Hence there is the annual ritual of ‗visiting the ancestors‘ (Ebekun festival in Iresi, Oroke festival in both Igbajo and Iragbiji. In the course of the festival, the king renews his authority, replenishes his strength and re-affirms his loyalty to the course of the ancestors and the development of his peopleY. Here he prays for the year ahead and for the success of his ‗children‘ wherever Rthey are (children means all the sons and daughters and everybody within the RcomAmunity) and commune with the ancestors on the problems ahead. The annual and constant ritual practices at the sacred points established and re-establishL tIheB long standing history of the people by keeping the historical narratives alive. N 2.5 Settlement and Settlement Archaeology A Settlement studies are an examination ofA hoDw people organized their use of space, and can pertain to a continuum of pheBnomena ranging from the use of space within structures, to the arrangement ofF str uIctures or other features within the context of a local occupation, to the locatioOn of occupations within the broader landscape. A settlement approach formulates a rchaeological units at a scale beyond that of the discrete artifact; units are eTsseYntially formed that pertain to clusters, or aggregates, of discrete artifacts. ArchaeoIlogical units at this scale are particularly appropriate for the study of subsistenRce, Stechnology, exchange, and the various ways in which human groups organizEe themselves with respect to one another. ThIeV term settlement has been defined in several ways by archaeologist among othersN largely because their experience is pooled from different parts of the world (OUkpoko 1979). Settlement pattern depicts the social relation which forms the framework of a society and is necessary for our understanding of any society‘s economy including the location and use of resources (Okpoko 1979). Two approaches seem to dominate settlement patter studies. The first is primarily ecological and it is based on the assumption that a settlement pattern is a product of the interaction of two variables – the environment and technology. This approach merely studies how settlement pattern reflects the adaptation of the society and its technology to its environment (Okpoko 2006). 49 The other approach is the used of settlement patter data as a basis for making inferences about social, political and religious organization of prehistoric culture. Information can be extracted from individual structure such as building. This information would include economic, social and religious activities of the people associated with such building. According to Okpoko (1979) ‘subsistence regime is one of the factors that determine the structure of individual building of a society‘. In any inquiry about the social past, the first question to address is size or scale. Settlement archaeology therefore includes an array of techniques and theorieYs dedicated precisely to understanding these scalar questions. Archaeologists geneRrally try to address spatial concerns first in the process of decoding past human behAavior. Settlement archaeology is defined as ‗the study of societal relationRships using archaeological data‘ (Trigger, 1967:149). Settlement archaeology iIs Balso defined as the study of the selection criteria and implantation of settleme nLts in the landscape, interrelationships between cities and their rural surroundinNgs, the impact of human occupation on the natural environment and viceD veArsa under past conditions. Settlement archaeology has as its aim the holisAtic reconstruction of the cultures of ancient settlements and urban communiItiBes and their hinterlands. Settlement archaeological research is by definitFion a multidisciplinary enterprise requiring expertise from the disciplines of thOe natural and social sciences, architecture and city planning, as well as specialized te chniques related to the retrieval, recording, analysis and data bank managemeTnt Yof archaeological data (GIS), site conservation and cultural resource managIement. Disciplines and interdisciplinary sub-disciplines required in additioRn tSo archaeology include geology, environmental geomorphology, archaeozoologEy, paleontology, paleobotany, archaeometry, ancient history, anthropolIoVgy, sociology, urban geography, classical architecture and city planning. SettleNment pattern describe the ways in which people distribute themselves across the laUndscape and can be very revealing about that culture‘s structure and relationships with others. The aim is to understand the nature of the activities that took place there, and of the social group that used it. 2.6 Archaeology of Conflict and Migration Conflict archaeology is a new and interdisciplinary study of conflict and its legacies. Conflict archaeology focuses on conflict as a multifaceted phenomenon, whose variety of physical traces possesses multiple meanings that change over time. It 50 is not restricted to battlefields, or to large scale wars between nations, but embraces every kind of conflict and its diversity of social and cultural legacies. Conflict archaeology combines the strengths of many different disciplines: anthropology, heritage and museum studies, cultural geography, military history and art history. Archaeological investigation of conflict primarily relies on four lines of material evidence: settlement data, injuries in human skeletal remains, wear weaponry and iconography (Lambert 1994, LeBlanc 1999). The first of these data, settlement data, is particularly useful for identifying both concern with defense and thYe consequences of failed (or absence of) defensive measures. The time and matRerial resources people deem necessary for protection can help define perceptions ofA thereat. Defensive behavior might include a shift in village location from a vBalleRy floor with ready access to agricultural fields to a steep slope or inacceLssiIble rock shelter requiring greater energy expenditure for day-to-day living. Walls, forts, towers, moats and other defensive struNctures require investment of labour and resources for construction and maintenanceAs, so the appearance of these features should also correlate with a perceived nAeedD for defense sufficient to warrant reallocation of resources to these ends. The bIuBrning of structures, on the other hand, is a common consequence of war. ArchFaeo logical evidence of burned structures and settlements can therefore help docuOment actual attacks, although other possible causes of burning (accidental fires, intent ional clearing) must also be considered. Migration has alsoT beYen defined as "specialized behavior especially evolved for the displacement of tIhe individual in space" (Dingle 1980) in such a way as to "allow exploitatioRn oSf different habitats as life history requirements alter or as environments change seasonally or successionally" (Dingle 1980). Faced with environmIeVntal E change, an organism can either adapt, migrate (also a form of adaptNation), or become extinct. "Movement is, therefore, a fundamental biological reUsponse to adversity, and all populations are spatially fluid in some measure" (Gauthreaux 1980). Population dispersal can occur (1) when the local environment is deteriorating but better conditions occur elsewhere, (2) when environmental parameters change of fering opportunities for a population in a previously uninhabited area, or when stress on resources occurs through population growth and expansion is necessary for survival. Change in the natural and/or social environment is the prime mover, and it is clear that many non-human species have the capacity to react to change by moving to a new location either for shorter (seasonal shifts) or for longer 51 (relatively permanent moves) periods of time. Animals migrate, then, only if they are forced to do so, and if the social and physical environment allows it. They migrate because to do so confer an adaptive advantage (usually defined in terms of inclusive fitness) on the species, population, or individual involved. "Animals move in order to place themselves in optimal conditions for as long as possible" (Sinclair 1983). Unless humans are somehow biologically unique, it follows that they are similarly dependent on their environments. Migration is a difficult concept to define because it includes people who mRovYe for different reasons across different spaces. A migrant can be a person who mAoves to another city or town within a nation; a refugee who crosses an internationRal border to escape religious or political persecution; a jobseeker who moves toI Banother country for better economic opportunities; a slave who is forcibly mLoved; or a person displaced by war or natural disaster. Demographers laNck a single, operational definition for migration because it occurs under differentA conditions. People move for different reasons. These differences affect the oDverall migration process. The conditions under which a migrant enters aB reAceiver population can have broad implications for all parties involved. The ex pIression migration experience refers to the fact that different causes for migrationF will produce different outcomes observable from a sociological perspective. E xOplanatory frameworks premised on the push-pull hypothesis tend to overemphaYsize the role of the individual in the migration process. Critics of this perspectivIeT argue that the decision to migrate is based on group experience, in particuSlar the costs and benefits to the family. Rather than being an isolated calculatioRn, an individual's decision to migrate is conditioned by multiple social, econVomEic and political factors. NMIigration has become a continuous phenomenon in the history of human soUcieties. Migration is synonymous with the history of Africa itself. The migration of individuals and groups over time is associated with the emergence of cultures and of civilizations throughout the world. The centrality of migrations and movements of people in the historical evolution of African peoples and societies lies in war, conflict social and economic factors. Migration was a multifaceted phenomenon, which varied in nature and character, over time and in different places (Aribidesi and Falola 2009) 52 There are a number of reasons why people in Africa migrate or leave their original homes for places sometimes dissimilar to their prior location. Such reasons have been discussed at length by various scholars. They include wars, droughts, and floods; regional inequality of economic development and income; severe population pressure with low agricultural productivity; poverty and hunger in specific regions; the attraction of towns and cities as centers of education, higher incomes, and social amenities; the presence of affiliated ethnic groups and/or kinsmen; the presence of people from the same religious denomination; safety from persecution; and simplyY personal adventurism. It is not easy to determine the impact of any of these faRctors individually or as a whole. In most cases, they also depend on the existiRng pAolitical, economic, social, religious, and personal circumstances. Migration has occurred on a substantial scale many times in the history of the people of the cLontIinBent, and it goes back in time to the origin of human beings. Migration and pop ulation movement in Africa can best be understood within the context of theN political and historical evolution of African societies. A It is generally believed that it was populationD movement that disseminated the Bantu languages over the subcontinent. TwoI Btheo Aries have been suggested to explain the Bantu expansion from the original homelands. First, it is claimed that the abandonment of hunting–gathering subFsistence economy in favor of an agricultural subsistence economy led to popul aOtion growth, which consequently led to migrations of people seeking new lanTd. YAccording to this theory, the migrations of the Bantu peoples from West AfricIa to central Africa are said to have involved agricultural communities, andR theS movement is said to have increased after banana and yam, introduced by Ethe Indonesians, spread to the forest peoples of central Africa (Andah 1979). ThIeV second theory provides a link between the expansion of the Bantu and the beginNnings of the Iron Age. The working of iron encouraged agricultural production bUy providing more effective tools that allowed the Bantu to dominate the peoples in the areas where they settled. It is suggested that the Bantu ―were a dominant minority, specialized to hunting with the spear, constantly attracting new adherents by their fabulous prestige as suppliers of meat, constantly throwing off new bands of migratory adventurers, until the whole southern sub-continent was iron-using and Bantu-speaking (Aribidesi and Falola 2009) Factors, such as famine—leading to the search for better living conditions in the form of good farming and grazing land— 53 epidemics, wars and a sheer spirit of adventure could also have motivated the early Bantu movements. 2.7. Literature Review The history of the Yoruba people is one of the most researched and most written up of any people in Sub-Saharan Africa. However, the archaeological documentation of parts of the region remains inadequate. Efforts had been made, and are still being made, by some educated elites in some of the towns in the study area toY document their history in small monographs which they themselves referrAed tRo as ‗jottings gathered from the chroniclers‘ (Igbajo Development AssociaRtion, 1995). Though, there are few monographs and books written on some of the towns, most especially Ila-Orangun, yet, there is dearth of archaeological wriIteB-up before the initial effort of the author and others in the early 1990s. L The Master‘s Thesis of the author (Ogunfolakan 1A99N4) gave the account of the first phase of the archaeological reconnaissance of tDhe study area which include Ila-Orangun, Oyan, Asi, Iresi and Ajaba. He also puAblished an article on the same work (Ogunfolakan 2006: 73-94) where he enumIeBrated the results of the archaeological survey carried out in the area between 199 2 and 1994. The article revealed that Osun northeast is rich archaeologicallyO anFd called for more detailed archaeological investigations of the area so as t o open the place archaeologically and to reveal its tourism potentials. The woTrk Yitemized the result of the investigations town by town and site by site. In contIinuation of the archaeological exercise in Northeast Osun State, the author eRxteSnded the scope of his study to include archaeological survey of Iragbiji, Iree, EOtan-Ayegbaju, Igbajo, and Oke-Ila with a revisit to Iresi and subsequeIntV excavation of a potsherd pavement earlier identified in the course of the first pNhase of the investigation. Ogunfolakan (2009), examined the issue of peoples‘ dUispersal and how it has influenced the growth of cities and towns in this area. This work also tried to answer the question of the influence of dialect and the socio- cultural affinities of the people of the area. In his own work, Akpobasa (1994 and 2006) also carried out archaeological investigations of one of the abandoned cities of the Ila people (Ila-Yara). According to th Akpobasa (2006) Ila-Iyara was occupied prior to the 17 century when the town was abandoned. In his article, Akpobasa (2006: 98) proposes ‗a case study for socio- historical considerations in settlement studies‘ with the objectives of considering ‗a 54 typical Yoruba town in a developmental perspective; attempting an identification of the traditional facet of the town; and show how this facet can be described as reflecting the social history of the Yoruba people‘. Aribidesi (Aribidesi et al, 2005) in his works on Igbominaland reviewed previous works in the study area and carried out excavations of a cultural mound earlier identified by Akpobasa. He got two dates from the site, 1442-1531AD and 1545-1631AD (These dates corroborate the ones we got from Ajaba). Aribidesi et al (2005) used the earth science methods in the determination of ceramic production andY exchange within the study area by using Instrumental neutron activation anaRlysis (INAA) on ceramic and clay samples from the area. The result of the analysiAs shows that INAA can be effective in differentiating ceramics and raw clayR sources in northern Yorubaland. The analysis offers tremendous potentialIs Bfor examining regional interaction in Yorubaland immediately before and durin gL the Old Oyo period (fifteenth to late eighteenth century), a time of socio-politicNal and settlement change that led up to the collapse of Old Oyo, the Yoruba CiAvil War, and the large-scale settlement abandonment of the nineteenth centuAry. DThe Igbomina ceramics fall into four compositional groups corresponding toB distinct village units and geographic locales (Esisa, Ilere, Ipo-Erese, and Ilor inI), which are tentatively linked to clay sources in these areas. The analysOis alFso indicates that pottery vessels were moved between localities and possibly b etween sites in each area. Apart from these works, there is a dearth of literature oYn the archaeology of the study area. This present work is therefore to contributeI tTo emerging literature on the archaeology of the study area and to arouse the iRnterSest of other scholars in the area. E IV UN 55 CHAPTER THREE ARCHAEOLOGICAL SURVEY 3:1. Archaeological Survey. Archaeological survey entails a variety of procedures such as surface collection of visible materials, systematic transects, combination of systematic and random sampling according to superimposed local or regional grid. However, the value of a survey depends on the level of recording details and nature oYf archaeological record under survey (Ogunfolakan 2007). R In Osun northeast, series of archaeological surveys were carried ouAt in the study area since 1992 when the author pioneered archaeological researchR work in the area. Though, others (Akpobasa 1994, Aribidesi 1999, AribidesLi eItB al, 2005) have continued to carry out archaeological research works in the same area. The first archaeological survey by the author waAs aNimed at documenting the archaeological features in the area with special emDphasis on potsherd pavements (Ogunfolakan 1994). The second phase was meaAnt to extend the frontiers of previous works and to examine the potsherd pIaBvements earlier identified at Ajaba (Ogunfolakan 2007). The third phase whic h forms a major component of the present work was targeted at the AjabaO abFandoned site. It involved documenting the archaeological data within the ab andoned site. These were compared with data from previous works including TtheY potsherd pavements earlier identified in other sites within northeast Osun aInd other Yoruba towns most especially Ife and Old Oyo. Excavation of a RrefuSse mound located within the Ajaba abandoned site was also carried out. E 3.1.1.N AIrc Vhaeological survey of Oyan, Asi, Ila-Orangun, Iresi, Ire and Ajaba U The first phase of archaeological survey took the author to Ila-Orangun, Asi, Oyan and Iresi. For the purpose of the first phase of the survey, and because of the limited financial resources at the disposal of the researcher, and the vastness of the area to be covered, (over 400 square kilometres), a simple traverse survey was used at each site visited. The sites were surveyed by simply walking over on foot, searching for and documenting cultural materials and features. This apparently was guided by information from oral information collected prior to walking round the sites. The use of oral information found useful in the location of the sites. The main roads in each of 56 the towns/sites visited were also found useful in the linear measurement of the sites while the prismatic compass provided the precise measurement of the site. Archaeological reconnaissance in the forest zone is a tedious one, most especially, during the rainy season when everywhere is green and the forest becomes thicker with the trees having their foliages spreading. This is also the period when most of the potsherd pavements are exposed and in many cases washed away. In Ile- Ife, most of the terracotta or bronze heads discovered were discovered after heavy rain downpour (Eyo 1974). As such, art works are washed down from their primarYy context and re-deposited again and covered with soil from erosion. Probably, thRis is why it has become difficult to have these works of art in their primary contextA. Since most of the archaeological sites in this region are in tBhe Rforest zone, major survey took place during dry season, from October to MarLch.I This is the best time to enter the forest when visibility is clear to some extent. The vegetation of the northern part of the study area, that is, the Ila-OrangunA andN Oyan axis is of derived savanna. Therefore, visibility is made clearer due toD the yearly burning of the bush around this area. According to an informant (Adeyemi 1992), the Fulani cattle herders often set the bush (elephant grass) oBn fAire to provide green grass for their cattle during this season. Within one wee kI of this, the ground comes up with fresh green grass for the cattle. The burning Fof the grass and the eating of the fresh green leaves by cattle give clearer view Oof what is on the surface. Though, in many cases damages were done to some Yartifacts by fire and cattle trampling, most especially, organic materials and inoIrgTanic material like ceramics. The archaReolSogical reconnaissance was also made tedious because of religious/cultuEral factors. For instance, one was not allowed to enter some shrines until a paIrtVicular time of the day or year or until some rituals are performed. Access to someN shrines and groves or sacred forest was denied because of the absence of the CUhief Priest. Often, certain categories of females, like unmarried women or those who still nursing children are not allowed to move near these religious landscapes. In most cases, endless appointments were made to visit a grove or shrine without success. And in a situation where you have limited time or resources, it makes things difficult. The experience in Iree needs to be mentioned. The author had several appointments with the late Oba of Iree, Oba Omotoso. The last and successful appointment did not record the optimal success as he was hosting a crucial meeting. By the next appointment, the man had died. 57 Also, most of the people here are not well informed about the role of archaeology in historical reconstruction and construction. For instance, in some cases, land/site owner will require his educated son or daughter to be present before embarking on archaeological survey. To them, they trust their children and whatever such a child or children say is final. In addition, if you are unfortunate to come across recalcitrant ones, they tend to complicate issues, as they will not own up to their ignorance of what one is trying to do but will misrepresent your own idea. For example, when we were surveying the abandoned settlement features within a cocoYa farm in Ajaba, the son of the owner wrongly informed his father that we wereA looRking for gold and not for cultural remains and subsequently the latter requested for monetary gain if we have to continue our work. Not only that, the son hReard that we were talking of charcoal for dating, he again wrongfully informed hiIsB father that coal has been discovered in his farmland and that his cocoa fa rmL was going to be destroyed, this delayed our work for another one week. N As stated earlier on the area covered by the pDrojeAct is large (almost 400 square kilometres), and because most of the archaeologicAal/historical sites are within built up areas, the archaeological survey was limitedB to systematic walking round the sites. This was done by using the main road to l inIk the site from where we walked through the site in a pacing method of fiOve mFetres interval in a north-south direction. In addition, archaeological reconnais sance was carried out in major towns and villages in the study area with emphaTsis Yon Ajaba, Igbajo, Iragbiji Iresi and Asi, A visit was also made to Ire and OtanI-Ayegbaju where the presence of potsherd pavements were reported to us. S With thEe oRral evidences taken, we were able to comb the area for possible site of archaeIoVlogical importance, which will lead us to a better understanding of the historNy of the area vis-à-vis the pattern of space, and how they have been able to exUploit the environment to their own advantage. In continuation of archaeological survey of part of Osun northeast (Ila- Orangun, Oyan, Asaba, Asi, Iresi), reconnaissance survey in major towns and villages around the area was further conducted and in addition the pavement site at Ajaba was also excavated. Before the survey, several visits were made to familiarize ourselves with the people and topography of the area. With the principal informant (Chief Popoola) with us, we had a smooth encounter with the people and the area. 58 Both oral and written evidences were examined thoroughly. Thereafter, we embarked on the actual location, identification, and documentation of archaeological/cultural sites from these areas. There are peculiar problems in carrying out archaeological survey in tropical rain forest environment because of the presence of thick vegetal cover which makes ground visibility difficult. This also often hinders mobility thus impeading access to surface archaeological features. Movement and visibility during the raining season were extremely made difficult by the heavy plant undergrowth and multi-layered tree canopies in the study area most especially in thYe Ajaba, Iresi and Igbajo zone of the study area. But in open area within the cityR and villages, movement and visibility was made easy and archaeological materiAals were able to be identified and collected. However, erosion has greatly affectedR the cultural features here. The topography of the area gives ways for erosioInB to wash away annually, the cultural materials from the soil. Visible feature sL like cultural/refuse mounds, city wall, possible shrine (because of the presAencNe of ‗peregun‘ Dracaena sp.) were only examined and noted for further examiDnation including excavation. It is important to note that in Yorubaland, peregunA ( racaena sp.) is a boundary or shrine/grove marker. B I 3.2 Account of archaeological sOurveFy of Oyan, Asi, Ila-Orangun, Iresi, Ire and Ajaba In the course of the reYconnaissance survey, archaeological features and sites were identified and recorIdTed. This process was carried out site by site i.e. town by town. S 3.2.1. Oyan ER (i) NPoIt Vsherd Pavement – This pavement made of only pottery sherds has been grUeatly damaged due to erosion and human activities. The remnant found in front of a dilapidated building at Ile-Jagun (Jagun‘s compound) in the north eastern part of the ancient town were over-grown by bush but partly exposed by erosion. The pavement is herringbone by design, and bears resemblance to that of Ile-Ife. (ii) Town wall - Around Jagun‘s compound is a remnant of ‘odi amonu’ (inner wall) which appears like a linear yam-heap. The ravages of time and climatic attacks have rendered invisible the mud wall structure of the town wall. The outer wall could 59 not be seen but my informant, Chief Adeyemi, the Jagun of Oyan, merely pointed to a spot almost 50 metres from the inner wall. (iii) Igbo Ogun - according to the Oloyan of Oyan, Igbo-Ogun is an abandoned settlement, where Epe, the ancestor of the Oyan people settled when they left Ile-Ife. It is situated along Oyan-Ila road, about five kilometres to Oyan. Unfortunately, we could not visit this ‗sacred‘ place, as the hunters who were to take us to this site did not keep to appointment on five consecutive scheduled dates. AR Y 3.2.2 Asi (i) Igbo-‘Riro/Potsherd Pavement – Located to the southern part ofR Asi, off the Oyan-Ila Orangun road along Lapiti Omigbooro village road areL ‗IgIbBo Riro‘ (sacred grove) and a potsherd pavement. No ruins of remains of any bu ilding could be found but small mounds and elevations, which could have beeAn faNll-in building walls were found at the northern part of the pavement site. Potsherds scattered all over the grove. ‗Igbo riro‘ (sacred grove) according to one of the iDnformants (Olagunju 1994), is a sacred land where nobody could farm until pIreBsen At day. It is the belief of the people of Asi that anybody that farms there wilFl d ie. According to the informant, any area where they come across objects such as bronze or brass is regarded as sacred land. Though the pavement was recentl yO exposed as a result of road construction activities, the people of Asi have for a loYng time set aside part of this area as a sacred land where nobody is allowed to farImT. But recently, two-thirds of this area has been tilled by farmers, leaving RbehiSnd only less than 100 metres by 80 metres of the supposed sacred land. TEhe potsherd pavement is situated within ‗Igbo Riro‘ (sacred grove). The location IofV the potsherd pavement is locally referred to as ‘kere apaadi‘ (meaning elevaNtion of potsherds). This indicates that the pavement area is on a peak of an uUndulating elevation. According to oral tradition, the present people of Asi found the pavement there and the area is referred to as ‗aganju Oba Tapa’ (Ededele 1994) i.e. the palace courtyard of Nupe king. The king was known as Ajangidi according to Atolagbe (1994). (ii) Igbo-Igbale (Masquerade’s grove) – Located near the palace of Alasi of Asi is Igbo Igbale (Masquerade‘s grove) This is the grove where the annual egungun festival begins and ends. The grove is demarcated with ‗peregun‘ tree (Dracaena fragrans). 60 3.2.3 Ila-Orangun There were three former Ila settlements before the present site of Ila-Orangun, they are Ila Kodomu (Igbo Ajagunla), Ila Yara and Ila Magbon or Para Oke. The sites of these former settlements constitute good archaeological features for the reconstruction of the unwritten history of Ila people. Ila Kodomu is to the north east of the present Ila-Orangun while Ila Yara is to the south east about 12 kilometres to Ila Orangun, Ila Magbon or Para Oke is within the present Ila Orangun and is located on the top of a hill (Oke Wosem as it is called today) to the north of the present site enY route the town‘s General Hospital. AR (i) Ila Kodomu (Igbo Ajagunla) – This was the first settlement site aRnd is now overgrown with forest, as it now forms part of the Ila Forest RIeBserve (Adetoyi, 1994 Per. Comm.). Because of stringent traditional conditions a ttLached to visiting the site, such as ritual performance, presence of priest, visit Aby nNight etc, this site was not investigated but noted for future archaeological workD when those conditions could be relaxed or met. BA (ii) Ila-Yara– Ila-Yara is to the sFout h Ieastern part of the present Ila-Orangun. Features recorded there included defensive trench, (from where the abandoned site derives its name – Ila-yara (Ila Ya raO means Ila with the defence trench) refuse/cultural mound, grooved rock (oroT n Ypon‘ko) and a minor water fall (ibu-Ogun -Akpobasa 1994). I S (iii) DefensEiveR Ditch and Embankment – This defensive ditch and embankment is a circuIlaVr ditch with embankment that was dug round the abandoned settlement. The tNotal width of the ditch and embankment measured about 10.7m; the ditch alone mUeasured about 5.3 m wide while the embankment is almost 2.20 m high. The width and height of the embankment which has continued to be weathered by both human and natural factors of destruction show how deep the trench was then, and how massive the embankment was. From Aba Ejemu (Ejemu village, a modern village within the abandoned settlement) to the point where Ayegunle road cuts across the defensive trench and embankment is almost 450 metres. From observation during the archaeological survey of Ila-Yara, the defensive wall is almost circular. This in essence bears resemblance to other Yoruba town walls as shown by the works of 61 geographers, historians and archaeologist alike. For example we have Owo and Ilesa walls (Afolabi, 1966b); Ife walls (Ozanne, 1969); Oyo-Ile (Agbaje Williams, 1990). Where Ejemu village of today is situated, which was once a ‗city‘ is now partly forest, partly farmland where cocoa (Theobroma cacao) yam (Dioscorea spp.) oil palm (Elaeis guineansis, cassava (Manihot esculenta) etc. are planted. Alanwo stream, which from all indications might have served as the only source of water supply during the hey-days of Ila-Yara, still serves as a source of water for the present inhabitants of Oko Ejemu. RY (iv) Mound (Oladotun Akanbi’s Farm): This mound, according to oral tAradition, is part of the remnants of the past activities of the past inhabitants of IlBa YRara. Surface finds from this mound show that it is a cultural mound. The mouInd is almost 7.5 metres in diameter and abuts 3.6 metres high. To the eastern pa rtL of the ‗ibu ogun’ (a minor water fall and pool) is another mound. AN (v) Ibu Ogun (Ogun’s water fall): To the nortDh east of the mound at Oladotun Akanbi‘s farm is ‗Ibu-Ogun’ minor water IfaBll. AThe minor water fall comes from a small elevation of about 2 metres wherFe a stream runs through north-south direction. It then forms a pool near a rock formation where water from the stream collects. This water pool might have served as t heO source of water for the people of Ila-Yara. To the south of the water fall poiTnt Yare hollows on the rock surface set in the form of the popular Yoruba game caIlled ‘ayo’. The hollows are referred to as oro pon ‘ko (Oro grinded the rock)R. UnSfortunately, no remnants of any standing house wall could be seen. But theEre are some minor mounds which probably might be remnants of collapsed walls. IV (vUi) NIla Orangun (Ile-Ila): The present Ila is located on a high ground with an undulating topography. Here, some archaeological features were identified. These include potsherd pavement, Peregun Awogbemi (Awogbemi‘s grove), and city wall. Also, it was discovered that in each compound, there is a mound-like conical structure called odu. Odu is said to be the guiding ‗father‘ of the people of Ila. According to Chief Adetoyi (1993 Per. Comm.), before settling down, each compound would have to consult Ifa oracle and the sacrificial prescription of the oracle for prosperity, peace and stability in the compound will be prepared and kept in a covered pot and buried in 62 front or in the middle of the compound. A conical mound-like structure is then constructed as a monument for this ‗guidance father‘. Today, the Odu is cemented to prevent destruction by erosion. (vii) Potsherd Pavement – To the south east of the town at a place called Adeniji almost adjacent Peregun Awogbemi (Awogbemi‘s grove) is a potsherd pavement (Plate 3:1). The pavement is like the classical Ife potsherd pavement with a pot embedded in the middle and herringbone in design. A recent revisit to Ila-OrangunY revealed that the pavement might actually be the burial site of Awogbemi. AccorRding to the present Orangun of Ila, a wicked ruler or chief was then usually buRriedA in a big pot in a vertical position. Awogbemi was said to be Ifa priest of OrangBun Agboluaje 1 who reigned between 1789 and 1845 (Adetoyi, 1994). Here at thIe pavement site, Awogbemi was assassinated and buried. Unfortunately, this Lsite has just been destroyed by the construction of a modern petrol station Aat thNe site. Excavation of this site should have confirmed the information giDven by the Orangun of Ila. Unfortunately the whole heritage is lost, except foAr some earlier photographs taken. B (viii) Igbo-Atokun – This is a groveF de dIicated to Orangun Agboluaje 1 who was said to have been assassinated at Othis grove for going against the cult of egungun (masquerade). (ix) Town Wall – OnlyT a rYemnant portion of this town wall was observed around the ancient town of Ila-orIangun. In fact, there is a quarter known as Eyindi (Eyin Odi) that is behind the towSn wall. Even before we enquired about the town wall, we first tried to locate EtheR wall based on the name of the quarter i.e. Eyindi (behind the town wall) soI Vas to show the importance of place-name in the interpretation of archaNeological data from Yorubaland. Surprisingly, after searching round this quarter, a Uwide bank of earth was identified. The identification of archaeological features in mud is one of the most persistent problems for archaeologists as the ravages of time and climatic factors do often render mud wall structures invisible, especially in West African ancient settlements and most importantly, in sandy soil regions. 63 AR Y R LI B Plate 3.1. Potsherd pavement (Ila-OrangunA) N D BA OF I SI TY R VE UN I 64 3.2.4 Iree At Iree, a potsherd pavement site located towards the western side of the town was identified by a polytechnic student who once went with the researcher on field trip. The pavement is similar to the known Ife pavements with herringbone design. The pavement site was said to be part of a former town known as Ayekale, but it was destroyed during the Fulani war of late 19th century. According to the Baale (Chief) of Ayekale, their great grandfather made the pavement. It was at a later period that they came back to resettle at the same area. They see themselves as distinct from otheYr Iree people. It must have been that the present royal compounds form the grouAp oRr one of the groups that came to subjugate the original inhabitants of the pavement area. But unlike other towns, the people of Iree did not claim direct link with IfBe. ARccording to the late Are of Iree, Oba Moses Omotoso, the people of Iree cameL froIm Ipee in Kwara State of Nigeria in the late 19th century (presumably during t he Fulani war). They first settled at Oke-Ipole (Ipole Hill: Plate, 3.2) and lateNr moved down the hill. Annually, the Iree people celebrate Odun Oke IpoAle (Ipole Hill festival) in remembrance of their early settlement and progenitoDr. The festival precedes Egungun festival in the town. A IB 3.3 Archaeological Survey ofO IgbFajo, Iragbiji, Oke-Ila, Otan-Ayegbaju and Edemosi In the second phaseT ofY the survey, the exercise covered Igbajo, Iragbiji, Oke-Ila, Otan-Ayegbaju and EIdemosi towns (Fig. 1.3). S 3.3.1 Igbajo R AIt VIgb Eajo, major sites of cultural and historical heritage were visited, these include, the palace premises housing the old and the new modern palace buildings (PUlateNs 3.3 and 3.4), Ogun Shrine, Kiriji War Site (Fig. 3.1) and other historical points. According to oral tradition, Ogun (gods of iron) was said to be the pathfinder, the watch-dog over the community, and that is why the shrine is dedicated to him. Ogun shrine is often located at the entrance of the palace, middle of market or in front of a house since Ogun is believed to be the protector of the town or family. Most of the groves and shrines in Igbajo are located at the outskirts of the town. Prominent among them are Igbo-Esile, Igbo Ojo-Gbaruku, Okuta Aruka, Igbo-Ogun, Igbo-Ore 65 AR Y BR N LI AD A Plate 3.2: Oke-Ipole shrine at Ire F I B O SI TY ER NI V U 66 RY RAB LI Plate 3.3: Igbajo Old Palace building N A AD B OF I SI TY R IV E UN 67 Y RA R LIB AN AD F I B Plate 3.4: Igbajo New palace builOding Y T RS I VE NI U 68 (Ore grove is a prominent grove in Ile-Ife, the deity is believed to be one of the early settlers of Ife), Igbo-Oju Isao, Igbo Elegbede, Igbo-Igbale, Igbo Ita and Igbo Apon. Ogun shrine is at the front of the palace while Oduduwa shrine is to the eastern side, located at the back of the old palace building. It is interesting to note that most of the Ogun shrines in Yorubaland are always located at the entrance of the palace. Example is the Ogun shrine in Imesi-Ile or the Oke-Mogun shrine in the ancient city of Ile-Ife which is directly opposite the palace. The shrine of Osun, the goddess of the river is located at the valley known as Koto Osun (Osun‘s depression). Other shrines arYe Ajale and Oju-ori-Ijesa (point of hanging the skull of the Ijesa warrior duringR the Kiriji war that ended in 1886). A R 3.3.2 Kirirji War (1887-1896): At the Kiriji war site IgbLajoI B(Fig. 3.1); the following spots were identified with the assistance of an infor mant, Chief Lemikan (1997): (a) Oke Latoosa which is believed to be the rallyingN point for Latosa and his warlords; (b) Okuta Latoosa (Plate 3.5) is a small rocDk sAhelter where Latosa is said to have lived and died. He was said to have been burAied at the shelter after been killed by the Ekiti Parapo army; (c) Igi faragbota – (bBullet proof tree) although, the tree is no more there by the time we visited the site. ItI was said to be a bullet-proof tree used by the Latosa army (Chief Faloye 199F7). That was why it was given the name `faragbota‘ meaning ‗bullet-proo f Otree‘; (d) Oju-Ogun (war front) (e) Odo Fejeboju (river covered with blood);T (f)Y defence trench and; (g) the peace treaty site (Plate 3.6). I 3.3.3 Iresi In the course Eof tR S he survey, more archaeological sites and features were identified. These incIluVde a rock shelter known as ‗Okuta aladura’ (Prayer rock). The rock shelter has bNecome a place of prayer point for Christians. The rock shelter is a massive one alUmost 20 meters high and extending inwards to about 25 meters (Plate 3.8). There is evidence of constant use of the place by farmers as fire-place and modern pots were seen there. According to the principal informant, Pa Popoola (1994 per. Comm.), the shelter served as a place of abode during the Kiriji war of the 19th century. At the entrance of the route to this rockshelter is a rock boulder carved like an easy-chair (Plate 3.7). According to the informant, the rock-chair was used by the principal guard during one of the Yoruba wars. 69 RY BR A N LI DA IB A OF ITYS VE R UN I Fig. 3.1. Kiriji War Site map (Adisa Ogunfolakan, 1997) 70 Y RA R LI B AN Plate 3.5: Okuta Latoosa, Kiriji war site BA D F I Y O IT S VE R UN I 71 RY RA LIB AN D Plate 3.6 Peace Treaty Site (Point whBere Aagreement was signed by the two parties to the Kiriji war) F I O SI TY ER IV N U 72 3.3.4 Iragbiji Some of the sites here include a rockshelter called Oju’du (Oju Odu (Point of Odu, Odu (Ifa corpus) is an aspect of Ifa divination system), gbekan-rukan (rock overhang), and Okan yi lule o ku okan (Rock boulder), pavement site and Oke- Iragbiji (Iragbiji hill). 3.3.4.1 Oju’du (Point of Odu, where Ifa priests are initiated to Ifa priesthood) is a massive rockshelter near a stream (Plate 3:9). The shelter can conveniently housYe almost 100 people. In fact, during our investigation of the shelter, some goRods, believed to have been stolen at night and kept there were observed. Today thAe shelter serves as a shrine where a new babalawo (Ifa priest) is initiated. AlsoB, itR is the point of annual Ifa festival. I 3.3.4.2 Gbekan-ru-kan (Rock overhang: Plate 3:10) is Lanother feature that shows the work of nature. It derived its name, gbekan-ru-kaNn from the formation i.e. one lying on the other. It is not worshipped like othersA but adored by the people of Iragbiji as tourist attraction centre. A boulder is lyinDg on another one as if a person is carrying a load on his/her head. A minor sIheBlter A is at the base of the rock. It offers shelters for farmers during rainy seasonF. 3.3.4.3 Okan yi lu’le o k uO okan (One fell down remaining one) are two historical rocks (Plate 3:1T1) Ythat were said to be standing on each other before a dispute on the land sepaIrated them. According to oral tradition by Chief Muraina Oyelami (pers. coRmmS. 1997.), there was a dispute between the people of Iragbiji and Obaagun durinEg the reign of one of the obas, Late Oba Oloyede Dada. Many towns and villages surround Iragbiji, one of them being Obaagun which is about four kilomNetreIs Vnorth of Iragbiji. During the early reign of Late Oba Oloyede Dada, there arUose a boundary dispute between the Aragbiji of Iragbiji and the Olobaagun of Obaagun on the disputed piece of land where two huge rock boulders were standing on each other. Both the Aragbiji and the Olobaagun laid claims to the land. The then Resident Officer in charge of Ibadan Division in 1945 according to Chief Muraina Oyelami, the Eesa of Iragbiji (per. comm. 1994) set up a tribunal to decide the ownership of the land. Then, the two warring parties were invited to the disputed area. 73 Y RA R LI B DA N A IB Plate 3.7: Ijoko: The rock chair at Fthe entrance of route to the rockshelter at Iresi O TY RS I E IV UN 74 AR Y BRLI AN D Plate 3.8: Inside of the rockshelter at Iresi BA OF I ITY S VE R UN I 75 RY BR A N LI AD A Plate 3.9: Oju’du shrine (Rockshelter at IIraBgbiji) O F TY I RS IV E UN 76 RY BR A LI AND Plate 3.10: Gbekan-rukan at Iragbiji A B OF I ITY S VE R UN I 77 RY RA LI B AN BA D I Plate 3.11: Okan yi lule o ku okan roFcks at Iragbiji Y O T RS I VE NI U 78 The Aragbiji affirmed that without any controversy, the land belonged to him and if it was so, one rock should fall down within seven days. This was received with joy by the then Olobaagun because he believed that nothing could bring the rock down from the top of the other one. On the seventh day, when the people gathered again on this disputed land, to the amazement of all those who were present, the upper rock had fallen down on its own, according to the pronouncement of the Aragbiji. The District Officer then said in his judgment that truly the Aragbiji of Iragbiji was indeed the owner of the land. Since then, there was no more controversy over the ownership oYf the land and since that time, Iragbiji has been variously referred to in tradition Ras – ‘Iragbiji Oloke meji tako tabo lori aagba, okan yi lule o ku kan’ (meaningA Iragbiji land of two rock boulders, male and female, one fell down leaving the othRer one in the original position). Till today, the rocks are worshipped and a shrIinBe marked with peregun (Dracaenae spp.) is indicated at the site. L N 3.3.4.4 Potsherd Pavement - Near the EeDsa (AChief Muraina Oyelami) of Iragbiji‘s residence, is a building over a pavement that spreads both in front and at the right side of the building (Plate 3.12 andI B3.13 A). The pavement was constructed in the same way as that of Ile-Ife, in herrinFgbo ne design. 3.4 Geo-physical Methods of AOrchaeological Prospecting Archaeological proTspeYcting using geo-physical methods are known in many places, and are about fourI decades old. But in Nigeria, especially in Osun North East of Osun StaRte, tShis is the first time this method is being employed to identify archaeologicalE sites. Historically, it is known that archaeological features are buried under dusIt Vand other materials, so some sort of contrast (magnetic) between the buried materNials and the sourrounding host rock may be present. This contrast can be dUetected by using the magnetic method (Ogunlesi 2000). Magnetics have been used for exploration at numerous archaeological sites around the world to detect such features as buried walls and structures, pottery, bricks, roof-tiles, fired pots, buried pathways, tombs, buried entrances, monuments, inhabited sites and numerous object submerged in water such as ships, ballast stones, iron, cannon, various potsherds etc. Most of these objects were detected and mapped as a result of being more magnetic than the surrounding or covering material. A few features such as certain buried walls, and tombs were not themselves magnetic, but 79 Y AR BR N LI DA Plate 3.12: A building over the IragbijIi pBav Aement F O ITY S ER NI V U 80 AR Y LIB R AN D Plate 3.13: Iragbiji potsherd pavement IB A O F SI TY ER NI V U 81 displaced a uniformly magnetic soil which covers them. Still other sites, both historical and archaeological have iron objects which are easily detectable by the magnetic method. Anomalies exist at archaeological sites as a consequence of the contrast in magnetic properties between the cultural features of interest and the surrounding medium, both of which are usually composed of natural origin such as rock or soil or every empty space. This magnetic contrast is a function of the concentration, thermal and mechanical history of magnetic mineral present in either the cultural features oYr its burying medium. The amount of magnetic mineral determine the magRnetic susceptibility and therefore, the induced magnetization, commonly preAsent in materials which have undergone heating is responsible for the mosRt prominent anomalies arising from cultural features (with the exception of iron)I, Band also, in the uniformly magnetized soils. L The use of geophysical methods in this present Nwork was aimed at (i) establishing its relevance in archaeological study; (ii)A to determine the extent or otherwise of the potsherd pavement earlier exposedD in the area under investigation; (iii) to identify other archaeological featureIs Bin t Ahe area; and (iv) to economise time and fund since magnetometry profiling is a n inexpensive, non-destructive, and a very cost effective method of acquiring Fuseful information about the environment especially in locating ground st ruOctures i.e. small findings which could be great archaeological interest. Y In Nigeria, the usIe Tof this method is new, as it has not been applied for long. This present survRey aSt Ajaba potsherd pavement site was conducted and interpreted with the assistEance of a colleague from Germany, Dr. Fred Fielberg, who was then on a visitingI lVectureship at the Department of Geology, Obafemi Awolowo University, Ile-IfeN with some geophysics students of the same department. 3U.4.2 Data Acquisition: In magnetic survey, field procedure can be anyone or a combination of Ground survey, Aeromagnetic survey and Marine survey. For this particular work, only the ground survey was carried out. The area extent was measured to be 100 metres by 100 metres (Fig. 3.2). However, due to time constraint, only 50 metres by 100 metres were mapped. The field layout was divided into two halves; 50 metres north and 50 metres south, with the base point lying directly on the exposed potsherd pavement. Each halves was however divided into 25 grids thus 82 making 50 grids on the whole, with the area extent of each grid being 10 metres by 10 metres grids. Stakes and ropes were used in establishing each grid, after which traverses were then established in directions across the area of interest with each grid having 11 traverses and each traverse having 11 stations and so the number of stations in each grid totals 121. The station towards the north was tagged positive while that towards the south was tagged negative. The traverse-traverse separations used was 1 metre and the station-station separation was also 1 metre. Such small distances were used so thaYt archaeological materials that are small in size i.e. width and lateral extent caRn be delineated. The total number of stations recorded thus equals 6,050. A For this study two magnetometers were used, which consists of a BcheRst mounted console. Each magnetometer was mounted by a person while anotIher person takes down the readings as it is called by the person carrying th eL magnetometer. An important precaution was, however, taken throughout theN period of the survey as everybody involved in the survey were magneticallyD cleAan i.e. no magnetic material such as chains, earrings, belts were allowed to be worn by the participants. Before starting measurements, on each grid,I thBe p Aerson that mounts the magnetometer and his or her recorder goes to the base p oint where at least 10 readings were taken and the time at which these readingOs weFre started and ended were recorded. After this, measurements were then commen ced on each traverse with the first reading being the station position (i.e. the lonYgitude and latitude) through the help of a global positioning system. TimeI wTas taken before and after readings on each of the traverse. On completing measuSrement on each of the grids, 10 readings were again taken at the base point andE theR time was noted, this was however necessary for time correction. An average oIfV 25 to 35 minutes was spent in completing measurement on each grid. It is very Nimportant to note that for this entire survey we were always facing the south dUirection since the concentration of pavement is towards the south. 3.4.3 Data Processing: Computer was used in processing the observed data (Sofer 8 package). The data for the 50 grids were typed in, in a format as shown in Appendix 1. Measurement of each grid was presented as a file i.e. the 50 grids on the field were represented as 50 files on the computer. Each of these files begins with a header and contains information such as the name of the site, the names of individuals that carried out the survey of 83 RY RA LI B AN BA D F I O ITY RS VE UN I Fig. 3.2 Geo magnetic profiling of Ajaba Potsherd Pavement plan 84 that particular grid, the height of the sensor, the global positioning system readings, the direction being faced for each one of the grids, and since we have the x and y positions, it is important to define that our x direction is north and our y direction is east. The header of each file also contains information such as:  x0 meaning station starting position at the beginning of each grid in x direction  y0 meaning traverse stating position at the beginning of each grid in y direction Y  dx meaning the change in the station position in y direction R  dy meaning the change in the traverse position in y direction A  nx meaning the total number of stations along each traverse in xB diRrection  ny meaning the total number of traverses in each grid in y LdireIction  bos meaning the mean magnetic readings at the base point before each grid  boe meaning the mean magnetic readings at the bAase Npoint after each grid  ts meaning the time at which bos is taken D  te meaning the time at which boe is taken A The recorded time were input int oI tBhe computer as the global positioning system time and for processingF it was converted into seconds by multiplying the observed time in hours O(hh) by 3600s and that of minutes by 60s i.e. time (seconds) = hh*3600 +Y mm*60s. After inputting theI dTata in the computer, a 3D pot data files (uncorrected) were plotted. This was donSe by using gnu plot. Also, after imputing these fields collected data into computeRr, the fields were again inspected to check for typing error. Field data in magneEtics as also in other geophysical methods contain a series of noise and signals. TIhVe process involved in the removal of noise to enhance the reliability of our data iNs what is referred to as data corrections. U 3.4.4 Data Interpretation Interpretation in magnetic, as well as other geophysical methods in investigation such as gravity seismic, electromagnetism and electrical resistivity e.t.c can be qualitative, quantitative or a combination of both. However, only the qualitative interpretation of the 3D plots is employed in this work. 85 Qualitative interpretation of magnetic anomalies involves the inspection of maps and profiles for patterns diagnostic of the target. The most reliable targets for magnetic prospection in archaeological sites are fired hearth, fired rock and pottery. These features produce significant magnetic fields in proportion to their bulk, and at least in the case of fired hearts, fired rock and pottery are found virtually on all archaeological sites. These sources are also important in archaeological interpretation. Qualitative interpretation of magnetic anomalies has the advantage of: a) locating the target; b) identifying the altitude of the target and indicating the extent to whichY identified objects are magnetized, thus it is able to suggest possible souArceR and materials creating the anomaly. The qualitative interpretation (i.e. inspection) of the anomalies oRbserved on each of the 50 grids resulted in the table given below (Table 3. 1). SiIncBe the interest is based mainly on archaeological objects and as such is expected dLepth of investigation is expected to be shallow. However, the depth of investigatiNon is expected to be more than the height of the sensor used (i.e. 1.5 m), and can Abe estimated using equations 4.10 and 4.13. Different sizes of anomaly observed Dfrom the plots ranges from spot- like, to small, big and very big sizes. BA From this exercise, it has beeFn o bIserved that the use of geomagnetic and electro-restivity methods in archaeOological investigation of the study area is relevant and economical in the sense that it is easier to take decision on where to excavate as the data revealed possible aTreaY of artifacts and structures (Appendixes III, IV and V). I 3.5: Summary of CSolected Oral Tradition Oral trEaditRions were collected from five extant settlements. The results of the data colleIcVted during the donduct of the exercise are provided below. 3U.5.1:N Oral History of Igbajo As it is the case in most Yoruba towns, Yoruba oral history, especially settlement history is similar. For instance, it is either that they left Ile-Ife or a former place of abode because of chieftaincy dispute since ―competition for the throne is usually not without bitterness among the Yoruba. Today losers often resort to litigation. But in the past, when the winner was eventually installed, the losers used to leave their communities out of shame or anger or the fear that they could be oppressed 86 Table 3.1: Table of X and Y coordinates of most promising anomalies X Y GRID -46 15 10 -5 24 11 -12 37 17 -14 33 17 12 2 27 18 19 32 Y 6 7 36 AR 1 22 36 R 34 29 39 B 13 49 47 LI 15 41 47 N A D IB A F O Y T I ER S NI V U 87 by the winners‖ (Adepegba 1986). In a bid to find another settlement, they either follow the instruction of Ifa oracle or just move in the direction they felt would suite their habitation. Quite often, the immigrants either met aboriginal settlers or different sub-groups accidentally stumbling on each other after they have settled. According to one of my informants, ‗smoke from fire was usually the revealing signal, indicating the presence of human settlement‘ (Chief Lemikan of Igbajo, personal communication 1994). This was the situation at some towns in the study area, for example, according to the Aragbiji of Iragbiji, at Iragbiji (1994), when some group of hunters went to hunYt and as they were settling down, they observed smoke from a distance, they then trRaced the smoke only to find a group of people with whom they finally settled at theA present place of abode. R In choosing settlement sites Ila-Orangun, Oyan, Asi and OIkBe-Ila, emphasis was laid on the instruction of Ifa oracle. But in the case of I gLbajo they settled at different sites before finally settling at their present place of Nabode without consulting Ifa. According to Tugbiyele (1989) what attracted the sAettler from various places to Igbajo was ―apparently the first and perhaps the mAosDt evident factor, the main stream, ―okun‖ (ocean) which oozes clean and clearI wBater out of the hill from the east (of the town). This stream flows through the tFown and it joined other springs along its way. This stream provided the early settOler with fertile land for farming as the bank of the stream must have been very fertil e as the yearly rain must have deposited the top soil removed from the surrounTdinYg hills in the valley‖. Looking at the topography of Igbajo, the settlers might Ihave taken into cognizance the security provided by the hilly nature of the towRn. ESvidently, this was why Igbajo became a war front during the Kiriji-Ekiti-paErapo war of the late l9th century. ThIeVre are different versions of the origin of Igbajo people. All the versions claimNed that different groups came to settle at Igbajo at different periods and that the IfUe group dominates the royal lineage of the town. According to Chief Lemikan (personal communication 1997) Adegbola was the ancestor of the Igbajo Iloro people who came from Ife following a chieftaincy dispute. On their way he dropped dead when his horse slipped. He was succeeded by Ajiboye whom the people referred to as accidental king (because it was the sudden death of Adegbola that led to his becoming the king). Ajiboye led them to Odo-Komu and later to Moyin Akure without giving concrete reason for their departure and finally moved to the present Igbajo-Iloro site. They brought with them soil from Ile-Ife which they buried at a point marked today as 88 a shrine called Motin (Plate 3.14). Yet another version claimed that prior to the Oduduwa era, Esidale, a prince with some people left Ilare Quarter of Ile-Ife and migrated to the present Igbajo precisely Lemikan compound (Chief Lemikan, personal communication 1997). The Oduduwa era version asserts that Esidale one of the children of Oduduwa migrated to the present site on an adventurous journey and settled at Igbajo. Later his ‗footsteps‘ were traced through Ifa oracle by another group of Oduduwa princes who were formally given beaded `Are‘ crown to found new settlements in Yoruba land. Y This group of Oduduwa princes included Orangun of Ila, Oloyan of Oyan, the ARlara of Aramoko, the Ajero of Ijero, the Ore of Otun-Ekiti, the Olojudo of Ido-EAkiti, the Alaaye of Efon Alaaye, the Owa of Igbajo, and the Onire of Ire-Ekiti aBll oRf whom settled first at Igbajo before moving to their different respective Ilocations. Before their departure, they held a meeting where each sat on a stone sla bL. Today the point of their meeting is known as `Olokuta-mewa‘ (ten stone slabAs). NThough, the stone slabs could not be found, but a point in the town was showDn as the spot of departure by the princes. Another version of Igbajo history as recorded in the book: Igbajo The Citadel of the Brave‘ published by the Igbajo DevelopBmeAI nt Association asserts that, the Omo Ejemu first settled at the area known aFs Il oro while another group with crown came later to join them. But there was a mOisunderstanding between the new settlers and the earlier settlers and this led the lat er known as Omo Ejemu to call their occupied area Iloro, whilst the second groTup Ywith the ‗Are‘ crown opted to call their own area Igbajo since their leader was AreIgbajo. From diffRerenSt versions of the oral history of Igbajo, it is clear that a group of people left a pElace and either settled at different points before the present place or that a group fIirVst settled before other group joined them. Which ever is the true version, it is cleNar that Igbajo is a conglomeration of different people from Yorubaland and even frUom Nupeland. Because there was a mutual understanding among them, they were able to live together happily as is reflected in their cognomen: Igbajo-Iloro, Omo al’agogo m’emu, that is, Igbajo-Iloro, the child of he who announces the call for palmwine drinking with iron-gong. The administration of Igbajo is unique in the sense that their traditional council consists of titled chiefs from the two arms of the town that is Igbajo and Iloro. Whilst there is the Ejemu Igbajo, there is also the Ejemu Iloro, Odofin Iloro, Odofin Igbajo 89 RY BR A LIN Plate 3. 14: Motin shrine at Igbajo A AD F I B Y O IT S VE R NI U 90 with Owa as the head of the council. Oriki, the praise poem of Igbajo clearly demonstrates these two arms of the town: ‗Igbajo-Iloro omo al’agogo m’emu that is Igbajo-Iloro that announces the call for palmwine drinking with iron-gong’ as earlier said. If there was not mutual understanding, they couldn‘t have been calling each other to drink palm wine. Though, typical of Yoruba oral history, the first settlers were said to have come from Ile-Ife with soil brought with them and buried at a spot called Motin, which is located today at the center of the town which is also the location of thYe main market. R Also, it was mentioned that in the founding history of Igbajo-Iloro, EsAile (who is now deified) played an important role as he was to become the king.R His mother was said to have come from Nupeland. On being elected as the IOBba, he left for Nupeland to be well prepared magically. On his return, anot hLer person had been crowned as the king. This led to his anger and he left the toNwn in annoyance. He was said to have left for Iresi where he met Ebekun with wAhom he sojourned. In Iresi today, there is Ile esile (Esile‘s compound) witAh aDn akodi, (family hall) typical of some Yoruba royal architecture. B Looking at the oriki (praise poem )I of different groups in Igbajo, one will readily agree with Tugbiyele submOissFion that different groups came to sojourn in Igbajo. This evidence is attested t o from the oriki of the Agbearaka family of Igbajo, in which they are traced to Ire Y(Ekiti) and that they are blacksmith, hunter and warrior: Agbearaka di mo l’enu I T He who gag a person Eru jeje l’abe ogeRde S A fearful man under the banana stem Ewa dota, meeE ni toja oun lo This is 50, I will not fight Mogun OInVire Mogun of Ire Omo Nabule s’owo He who turns the soil into money OUmo amaha owo re ‘mo He who pacifies the child with calabash of money Eji jeki nnaja oore The rain stopped me from going to Oore market Toto jeki nnaja oore The dew stopped me from going to Oore market Ateji ati toto Both dew and rain Ee jeki nnaja Ire kale Should not allow me from selling at Ire market until the evening 91 Eji gbojo, ara, Ila gbede Rain took the day, the man from blacksmith home Omo awurin tun ‘rin ro The child of whom smelts and smiths Omo agunfe giri bi ojoku n’Ire the son of whom ……as if is going to rain at Ire Ire loo ti mbo, loo p’oko fun mi wa He came from Ire and got me roofing grass Poko fun mi wa ngo r’ile fi ko Get me grass to build a house A kii bimo nIre ka pose owo nina We don‘t give birth to a child at Ire andY be miserly R Aha koto ni won fi nbu owo f’aya wo they use big calabash to givRe thAeir wife money Aya Ire mogun, ko sowon nunu ile wa The wife of IremLoguInB, there is no scarcity in our hom e Owo oko ni nwon fi ntara se They spendA thNeir husband‘s money to adorn thDemselves Ogun lo n’Ire l’esin OgunA is worshiped in Ire Igi lasan lara oko mbo ITBhe villagers worshiped ordinary wood Awa lo logun, igi lasan lara oko mbo F We are the owner of Ogun, ordinary O people worshiped wood Ire mogun omo ayeyo Ire mogun, the happiest son A wule wuwo, a wule wu seTgi eYtc He who digs and dig money, he who dig I and digs beads The oriki RtracSes the place of origin of the Agbaraka family, it portrays their economic and social life (Omo awurin tunrin ro; omo am’aha owo re’mo). From the oriki, it isI Valso E clear that the Agbaraka family are iron smelters and smithers (Omo a wu’riNn tun ‘rin ro, Omo awule wuwo, omo a wule wu segi)) that is the child of he who dUigs iron-stone and smith iron, the child of one who digs soil and digs money, the child of he who digs the soil and digs beads. Even, in the oriki (praise poem) we are reminded of the materials of worship for the gods from each clan or family or city. Thus ogun (god of iron) is generally praised as: M’ogun Onire Ogun of Onire Ire M’ogun Ire of Ogun Ogun alara njaja Ogun of Alara eats dog Ogun Onire a ‘jagbo Ogun of Ire eats ram 92 Ogun elekole The Ogun of Elekole Esuu ‘su gborogbo long roasted yam Lo ma nje is what he eats This oriki according to Yoruba tradition applies to anybody working with iron (ogun) that is, hunters, smithers, carvers, and drivers. 3.5.2: Oral History of Iragbiji Y The town is bounded to the north by Oke Iragbiji (Iragbiji Hill) the hill (wRith a cave) which accommodates the famous `Ori Oke festival shrine (a shrine iRs wiAthin this cave on the hill), to the north-east by Iree town, to the north-west by Obaagun town, to the west by Ikirun, to the east by Aagba and Ororuwo towLns;I tBo the south by Egbeda village; to the south-west by Osogbo and southeast by Ibokun. Iragbiji developed on an ―undulating land with an altitude of 213A.36 Nmetres and 274.32 metres above sea level‖ (Atoyebi, 1985:2). The town is suDrrounded by chains of hills. It is drained by rivers Otara, Ladoo, Egundo, OtapeteA and Moori. The central ridge of the town runs from the north-east to the south-wIeBst, and provides a well drained land for residential development. According to the oral historOy, a Fgroup of hunter left Ile-Ife and settled under a tree called Ira (Bridelia ferrugine a). The tree was said to have provided shade as it was widely spread over theY place. The town thus derived its name from this phenomenon. Iragbiji meIanTs Ira (Bridelia ferruginea) (tree) gba (take) iji (shade) – in short, ira tree proRvideSs shade. Another version asserts that some group of people had settled in the aErea before the hunters came. The hunters were said to have observed a smoke froImV a distance and then traced the smoke only to find a group of people with whomN they finally settled at the present place of abode. Another version of the trUadition provided by the chief priest of Ifa, states that the progenitor of the Iragbiji people was a hunter who lived on the hill that over-looks the town. Their ancestor shot an antelope but it could not be killed instantly as the antelope entered a hole on the hill and the hunter followed it. He traced the antelope to the location of the present palace premises where he saw the dead antelope and eventually settled there. To commemorate this event, an annual festival is held around July when the king will go to the top of the hill for sacrifice and annual prayer to the ancestor in a small cave on top of the hill (Pate 3. 15), where prayers are offered to the ancestor. 93 3.5.3: Oral History of Edemosi/Ajaba The oral history of Ede is likened to that of Ajaba as the abandoned settlement at Ajaba is assumed to be the original home of the Ede people as reflected in the following history of Ede people. Also, the people that occupied the present Ajaba town came from different towns within the study area to settle at the last half of last century (1950s). Ede as it was known before now, is an ancient town whose people are now scattered all over the major towns and villages within Osun northeast. Though, today the town is called Edemosi because of what the traditional ruleYr referred to as ‗confusion of identity‘ (Edigbon of Ede 1997 personal communicatRion). There is another town named Ede located close to Osogbo the state‘s capital wAhich is usually confused with Edemosi. The spellings of the names of the twBo toRwns are the same but they are pronounced differently because of the tone markLs. IEde near Osogbo is with middle tone marks on the vowels whilst the second Ede is with lower tone marks on the vowels. To address this confusion in mailsA sorNting and delivery, Ede of the Ajaba area is now called Edemosi. Unfortunately, no single written documenAt isD available on this ancient Yoruba town. Therefore, the study is based solely oInB the oral information collected from the place. The town of about 5,000 peopleF wa s said to have been a big city enjoying the big trade of the north and south of tOhe Yoruba region being a major market town centrally placed at a junction lin king the north and the south and the east and the westv of the Yorubaland.T AYccording to Owolabi (personal communication 1997) Edigbon is the ancestor oIf the Ede people. Edigbon was said to have come from Ile- Ife, after a chieftaRincyS tussle which he lost. He was a hunter and warrior and powerful medicine manE. Because of the magical power of the Ede people, most especially the king, EdeI,V their city was abandoned as a result of dispute between two powerful groupNs in the town. This dispute according to oral tradition, led to subtle plot against thUe king, such that a person died and there was nobody to bury the person. This led to esi gigun (a magical or medicine marked point) which resulted in people deserting the town and moving to different towns and villages in the area. They now settled at Ila Orangun in Oke-Ede, Edigbon, Aworo-Okun, Laaro, Obatufe, Ala and Odosin compounds; Imesi-Ile – Oosa Oba, Okun Iroko, Ariyo compounds; Igbajo – Ile-Eri, Ile-Salako compounds; Otan-Ayegbaju, Ikirun, Iragbiji, Iresi and others. According to Oba Adebisi Adeniyi (personal communication 1997), they decided to return to their 94 RY RA LI B Plate 3.15: The king making sacrifice to the ancestAor dNuring ori-oke festival at Iragbiji AD IB O F ITY ER S IV N U 95 home in 1908 when Oba Dada Arilewoofi, Oladiipo I, brought them back, but settled at about 4 kilometres to the point of their ancient town which is today suspected to be Ajaba from the description of the king. The claim that Edemosi people scattered around many towns and villages within the locality of Osun north east led to visiting these towns and villages to ascertain the claim of oral tradition. In Igbajo, we visited Ile-Ari and Ile Balogun Onifade and Lofemu Olori Awo compounds, where in actual fact; some of the people from these compounds maintained their relationship with their original home aYt Edemosi. According to Chief Salako (personal communication 2004), he vRisits Edemosi regularly, and has built a big house in the town with the hopRe ofA leaving Igbajo to settle at Edemosi in future. In the record of Fashiku (1995) Ede is associated with Ila-Orangun as he indicated that most of the above mentiLoneIdB compounds in Igbajo came from Ila-Orangun. This may be connected with t he immigration route from Ajaba to Ila and finally to Igbajo in the bid to findA a dNefensive area to settle. In Ila-Orangun, we visited Oke-Ede, where majority of Dthe Ede people settled. 3.5.4: Oral History of Asi BA Asi is a sub-town to Oyan, aboFut 3 -I4 kilometers from Oyan, just after Asaba on the Oyan-Ila-Orangun road. No Osingle written document t is available on the early history of Oyan. This work depen ded solely on the oral account given by the Alasi of Asi, Oba Olagunju and somTe oYf his people most especially the chiefs and elders in the town. I According to Soral account given by Alasi and corroborated by one of his chiefs, Chief JosReph Atolagbe (personal communication 1994) Olubode was the ancestor oIfV the E Asi people. He (Olubode) left Old Oyo as a result of barrenness and as th a resuNlt of the fall of Old Oyo in the early part of 19 century. It was asserted that OUlubode left Oyo Ile after consulting Ifa Oracle to leave home and go to another new land. He told his father who was then the reigning Alaafin, Alaafin Olunloyo but the later did not allow him to go. He again told his father who once more refused him. His father refused him the third time and the decided to leave without the consent of his father. When he earlier consulted Ifa, the oracle advised him to settle at a place where he will hear the voice of ‗agbigbo‘ birds (perons). He left and passed through many towns before finally settled at the present site. According to chief Atolagbe (personal 96 communication 1994), Olubode did not settle down without problems, as he was constantly attached and forced to leave the town for Ikirun. It was from this off and on that the people of Asi derived the name of the town. Asi literarily means ‗emigrant‘. Oba Olagunju jokingly said that the Oyan people used to refer to them as Asi ma de Oyan, that is, ‗Asi emigrants that do not migrate to Oyan‘ 3.5.5: Oral History of Ila-Orangun There are different versions of the oral history of the ancient city of IlaY- Orangun. But it is common to these versions that (i) the ancestor of Ila people cRame from Ile-Ife; (ii) that they settled at different places before they finally settleAd at the present Ile-Ila (Ila-Orangun); (iii) that the progenitor of the Ila people waRs one of the seven children of Oduduwa and various information placed himL oInB the second or fourth place among these children of Oduduwa. The oral hi story of origin of Ila centered on the popular migration story of the seven chAildrNen of Oduduwa from Ile-Ife. Though, the reasons behind the migration are noDt clear, the fact still remains that they left Ile-Ife, to found their own domain. SoAme people said it was due to over-population, some said it was due to OduduBwa‘s old age and that he wanted his children to establish their own kingdoFms bIefore his death (Chief Adetoyi personal communication 1994). From the various versio nsO of the historical origin of Ila, the names of Ajagunnla and Igbonibi feTatuYre prominently. Ila oral tradition makes references to four different sites of setItlement before the last and present site, Ila-Orangun. While Ifagbamila AjaguRnnlaS was said to be the first to rule at the first site, Ila Kodomu or Igbo AjagunnlEa, Igbonibi was said to be the first king to reign at the last but present site. ThisI mVay be the reason why the two names enjoyed greater prominence in the traditNion of origin of Ila than any other Oranguns that ever reigned in Ila. U One version of the oral tradition states that Ajagunnla left Ile-Ife with his mother, Princes Adetinrin and some emigrants and later settled at Ila-Kodomu. The other version asserted that Adetinrin was grand-daughter of Oduduwa and daughter of Okanbi that, the beauty of Adetinrin made Oduduwa to impregnate her. Realizing the evil deeds and what could follow, Oduduwa decided to terminate the pregnancy (Adetoyi 1974). Unfortunately, his people got the wind of the act before the abortion could be done and they termed it ‗oro egun’ (an act of abomination). Consequently, Odudwa decided to sent Adetinrin out of Ile-Ife, moreso, it has been revealed, after 97 consultation with Ifa oracle that the child she was expecting would be great if she leave Ile-Ife (Adetoyi 1974). Before her departure, Ilaji (half) of every item of Oduduwa properties was shared to her. In addition, she was given two cutlasses called ada ogbo (ogbo cutlass). The cutlasses were used in leading and directing her followers on the ways to follow. Whenever she was asked which way to follow she used to say ogbo mo ona that is ‗ogbo knows the way‘. That is why today the area and the people are known as Igbomina coined from ogbo mo ona. Y The oral tradition further stated that Adetinrin gave birth to a male childR and the mother sent message to Oduduwa saying that ‗oran ti a ni egun , momRo guAn o’ i.e. the case that we called a curse has become a blessing (Adeoye 1988). WBhen Oduduwa learnt of this, he was very happy and sent bacvk that the child beI named Orangun coined from that words ‗oran mi gun’ that is my plans worked o uLt well Chief Adetoyi 1974). N Another version says that Fagbamila was a diDrectA and second son of Oduduwa, that after the first son, Odudwa did not have anothAer male child for long. He consulted Ifa oracle and appease was made to the oraBcle after which a male child was born again. The child was then named IfagbFam ilIa that is, Ifa saved me and oran mi gun, that is, my problem is solved. BecaOuse of his love for him, a big house was built for him in the premises of the pal ace of which Fagbamila was thus referred to as Orangun-Ile-nla that is OraTngYun of the big house. Thus after migrating from Ile-Ife, the word Ila was corrupteId from Nla (big) and Fagbamila was referred to as Orangun Ile Ila. S FagbamEilaR was said to have settled at a thick forest, later known as Ila-Kodomu IaVnd after Fagbamila the settlement was named Igbo Ajagunnla that is, AjaguNnnla forest. Ajagunnla reigned there for almost 130 years (Adetoyi 1974). After hUis death, Amotagesi, one of his sons became the next leader and king. For one thing or the other Amotagesi moved the people from Igbo Ajagunnla to Ila Yara where he became the king. This place is referred to as Ila-Yara because of the defensive trench and embankment constructed round the town. Yara means trench or ditch/pit in Igbomina Yoruba dialectical language. After the death of Amotagesi, succession disputes among the children arose. This went on for a long time and it later had effect on the population in that many of the inhabitants deserted Ila and scattered to various places like Ilawun and Ilase now 98 in Ijesaland, Ajase-Ipo, Oro and Omu-Aran all in the present day Kwara state and even down to Nupe and Bariba land (Adeoye 1988). Accoording to Chief Adetoyi, in the midst of all these disputes, two of Amotagesi‘s children, Arutu and his junior brother Igbonibi decided to migrate with the remaining followers to another place. They left Ila Yara for a place later called Magbon. Before leaving Ila Yara, Arutu consulted Ifa oracle who advised him to settle at any place where the orere staff is placed on the ground. But according to tradition, when they were going they got to a place where they took a rest and the follower who carried the orere unconsciously puYt the staff down. On realizing this, he did not tell the leader and thus they contiRnued their journey till they got to Ila Magbon or Ila ‗Kolo on the top of a hill oveArlooking the present Ila. At this pace, they were frequently attacked by eartBhwoRrms. Arutu consulted Ifa oracle again and Ifa told him that the orere staff has beIen placed on the ground somewhere before they got to Ila Magbon. Arutu then as kLed his follower who carried the staff and the man confessed and promised to takNe Arutu back to the exact spot where he mistakenly placed the staff on the grounAd (Adetoyi1974). Arutu who could no longer move further due to old age, AhanDded over all traddditional royal apparels like beads, crown and the staff, opIaB orere to Igbonibi his junior brother to continue the journey as he gave up the gho st. Igbonibi took up the leadershipF challenge and continued the journey with the followers to the spot where the st aOff was said to have first touched the ground. Here they stopped and settled. ThiYs place is called Isedo in the present day Ila-Orangun according to Chief AdeItoTyi (personal communication 1994). According to oral tradition, Isedo isR a Spre-Orangun settlement. When Igbonibi got to Isedo, he met Tiimo with hisE family well settled. He also met Sangi and Okangi with their families each livinIgV on its own but very friendly with each other. Tiimo was said to have migraNted from Obaile in Ekiti land While Sangi and Okangi came from Nupe land. U When Igbonibi arrived with his group at Isedo they were well received by Timmo, Sangi and Okangi all of whom later submitted to his lordship (Alhaji Sunmonu Adesina 1994 personal Communication). Sangi and Okangi fled Isedo with their people for an unknown destination. 3.5.6: Conclusion From the account of the oral tradition, most of the people from the study area migrated from one place or the other before their final place of settlement. They also 99 migrated as a result of one factor or the other which include conflict, war, and royal/succession disputes. In their preparation to migrate, they, in most cases consulted Ifa Oracle which gave them direction on where to go or settle. Except for Asi where they claimed to have come from Old-Oyo, most of the people from this region claimed their ancestry to Ile-Ife. They also claimed to have stopped at different locations before the final and present location. RY RA B N LI AD A B OF I Y IT S VE R UN I 100 CHAPTER FOUR EXCAVATIONS AT AJABA AND ENVIRONS 4.1. Excavations at Ajaba and Environs This present excavation exercise is not the first in Ajaba. In 1992 a potsherd pavement site Aganju Tapa at Asi on the Oyan - Ila-Orangun road was ‗traced‘ (the excavation then did not go beyond the pavement level) so as to understand the technology, design and orientation of the pavement (Ogunfolakan 1994). During thiYs first phase, visits to Ajaba (Kajola) along Iresi-Ajaba road revealed also a potsRherd pavement near a sacred grove (Igbo‘riro). The pavement was then earmaArked for further investigation including excavation. By this time, the pavement Rwas already exposed at the middle of the road leading to Iresi from Ajaba as a reIsuBlt of the yearly grading of the road by heavy road machine. L Our visit to the place two years (1996) after revealedN that the whole pavement earlier exposed in the middle of the road had been toAtally removed by the same grading of road. By July 1997, more pavements weDre exposed towards the northern end of the road leading to Iresi. The total remIoBval A of the pavement from the main road now gave way for erosion to create minFi gu llies on the road and also exposing more of the pavement and scattered sherds along the road. It was then decided that the pavement be opened with the m aOin objective of determining its extent, design, orientation, material inclusiYon and subsequently determine its relevance and relationship to the abandIonTed site and also relate it to known pavement sites at Ile- Ife, Ila-Orangun, RIragSbiji, and other known pavement sites within the study area. In 1999, excavatiEon of this exposed pavement was carried out after a geo-magnetic and electro-reIstVivity survey was carried out. NThe Ajaba pavement according to Owaloja of Kajola, Oba Jacob Adedotun Adetoyese is part of Kajola abandoned settlement and that the whole site was initially a Usacred grove but people started encroaching on the grove for their farming and most importantly for the exploitation of timber. By 1997 almost three quarter of the grove had been cultivated. A marginal portion is now left as `Igbo-‘Riro that is uncultivable land or ‗Igbo risa‘ (forest of demons). 4.1.1: Excavation of potsherd pavement at Asi near Oyan (Aganju Tapa –Grove of the Tapa) Excavation at Asi was the first to be carried out within the study area. 101 After oral historical information of Osun North East had been collected and the archaeological reconnaissance survey carried out, it was then decided to carry out a lateral excavation (scraping) of the potsherd pavement at this site (Asi). The exercise itself was meant to determine the extent of potsherd pavement around the exposed ones at this site; the orientation of the pavement; determine within that range, any other design other than the exposed one i.e. herringbone design. The exposed side of the pavement shows that the pavement was constructed in a circular form and not in straight rows as known in Ife potsherd pavement. It was also meant to prevent thYe total loss of information on the potsherd pavement as a result of the yearly gradinRg of the road by heavy machine where the pavement was located and the RerosAion that subsequently washes away the pavement on yearly basis th th The lateral excavation took three days, from 16 to 18 L ofI FBebruary, 1994. The site was mapped out in 10 metres by 5 metres, five metres to the north (numbered 0-5) and ten metres to the east numbered A-F, (Fig. 4.1)A. It Nwas laid in a grid system and divided into units of 2 metres by 1 metre eachD for excavation. Five units were then excavated. Part of the pavement at pit AB and BC have been exposed as a result of road construction. In fact, it was this exposBureAI that prompted the lateral excavation of this site in order to determine theF ex tent of the pavement. As a result of the exposure, it was decided to furtheOr open up pits BC-1 and CD-1 to determine the extent of the pavement to the ea st and pit AB 3-5 to the north west, EF 4-5 to the north east EF 0-2 to the souTth Yeast and CD 2-3 to the centre of the site were opened. Because of the difIficulty in getting necessary instrument for the survey then, a simple survey techniqSue was used to get different elevation of each pit using the pavement leveEl at Rpit AB-1 (the exposed part) as reference point. An Empire line level instrumenItV was used for the measurement. At the start of the excavation, the measuNrement reads that pit AB-4 & 5 (North West) was 9.5 cm lower than the pUavement level. Pit EF-5 was also lower by 9.5 cm, pits CD-1, EF-1 & 2 and CD-3 were also brought below the pavement level by 14.5 cm. Pit CD-1, EF-1 and 2 and CD-3 were opened and scrapped to the pavement level using the Empire line level to take the range of elevation of each pit to the pavement level. These pits did not produce any cultural material except pit CD-1 which produced 12 pottery sherds. Pits AB-4 and 5 and EF-5 were also opened and brought down by 10 cm, and no cultural material was retrieved from the pit. After scrapping down to about 14.4 102 RY BR A LI AN Fig. 4.1: Site plan of Aganju Tapa PotsheArd PDavement (Asi) Excavation IB OF SI TY R IV E UN 103 cm, pit CD-1 revealed potsherd pavement. The ‗scraping‘ also showed that there is an unpaved space measuring about 81 cm between the exposed pavement at pit BC-1 and the pavement at pit CD-1 (Fig. 4.1). The exercise itself was meant to determine the extent of potsherd pavement around the exposed ones at this site. 4.1.2: Excavation at Ajaba Pavement site (KJAJ)): In furtherance to the previous excavation carried out at ‗Aganju Tapa‘ at Asi on Oyan-Ila-Orangun road, and in order to confirm the geo-magnetic and electroY- restivity data generated from Ajaba pavement site geophysical survey, it was decRided to excavate the northern part of the pavement site because remnants of the pAavement stretched towards this part. Certain questions became pertinent is this regaRrd.  What is or are the relationships of this pavement LwiIthB the abandoned settlement?  Is there any correlation with possible AmaNterials from this present exercise and the previous excavation carried out?  Is there any link with ceramic maAteriDals from other sites identified in Osun northeast? These are thIe Bpossible question the present excavation is going to profer solutioFn. The Ajaba pavement site is located to the north western part of Ajaba town (Fig. 4.2). It is about 200 meYtres t Oo the middle of the town‘s market and about the same distance to a village T(Aba Paanu) where the owner of the cocoa farm on which the pavement is locaSted Iresides. From the appearance of the exposed pavement it appeared that theR road had cut through the original pavement thus, exposing the pavement and Esherds of different sizes, shapes and thicknesses. To ascertain the area covered bIyV this pavement, the extent of damage and specifically, to confirm the data generated by the geomagnetic and electro-restivity profiling, an area of about 50 by 5U0 mNeter was cleared. The measurement cut across the road that leads to Iresi from Ajaba. A datum point was established towards the eastern part of the site. A 12 by 12 meters area was then demarcated for excavation. Since the road cuts across this pegged area, a 5 by 5 meter square was designated for proper excavation at the western section of the site with a zero point pegged to the south. The squares were then numbered 1 to 25 (Fig. 4. 3). 104 RY BR A LI AN BA D F I Fig. 4.2: Sketch map of Aj aOba town showing the location of the excavated mound. SI TY VE R NI U 105 AR Y R LI B N Figure 4:3: Site Plan of Ajaba Potsherd pavemeDnt eAxcavation A IB O F TY RS I VE UN I 106 After establishing the grids and datum point, the excavation then started at the spit of 10 cm exposing and going along the pavement line. At about 20 cm (spit level 2) the pavement layer was struck. Pits 1-3 and 8-10 were excavated to the pavement level. Realizing that the pavements spread towards the south, it was then decided to extend pits 1 and 10 towards this direction. These pits were tagged the extension of the previous pits 1 and 10, 11, 20 and 21. Also, at about 20 cm, the pavement layer was also struck, but it was observed that the pavement was slanting. At pit 9 and 10 the excavation went below the pavement level so as to determine other culturaYl materials below the pavement layer. The pit did not go beyond spit 4 (40 cm)A whRen it got to the matrix. Also, in order to further determine the extent of the pavement, a test pit was dug towards the northwest of the pegged area, exactly on pit 17. AR total of 255 potsherds, 8 lithics materials and one iron object were recovereIdB with series of potsherds with different decorative motifs. Mat-impression, tw iLsted cord and plain sherds dominated the motifs. No organic material was ArecoNvered. Also, a pot (ajere, perforated pottery), probably embedded (since the pavement around the pot has been destroyed) in the middle of the pavement was retrievDed. It has been badly crushed by heavy traffic. The pot was in an inverted posBitioAn with a small iron hoe blade (Plate 4.1) at the mouth. The stratigraphy of Fthe Iwestern wall of the site was drawn (Fig. 4.4). It shows that the topmost layer is of top soil followed by dark brown soil with gravel, and reddish brown with fi nOe gravel. Root and rootlets also prevail across the layer. Reddish-brown, more cYompact, and reddish brown are the last two layers. A total of 255 pottery sherdIsT, eight lithic materials and one iron object were recovered. The plan showingR the Slayout of the excavated pavement was then drawn (Fig. 4.5). 4.1.3 ExcVavaEI tions at Ipetumodu Iron Working Site (IPIS) NIpetumodu is about 15 km west of Ile-Ife. The site belongs to an extinct itUinerant Oyo smelting guild. The excavation of a smelting site at Baakun – Ipetumodu was carried out by a colleague from the Ife Museum in collaboration with a geochemist from Natural History Museum of the Obafemi Awolowo University, Ile- Ife (Ige and Adesina: 2008). The excavation (Ige et al. 2008) revealed three distinct layers. Layer I measured between 15 cm and 40 cm in thickness and composed of gritty and dark soil. The finds in this layer consist of potsherds, ores and some iron slags. Layer II was between 40 cm and 75 cm and consists of homogenous loose and 107 Y RA R LI B AND A Plate 4.1 : Hoe blade at the mouth o f IthBe ajere pot (perforated pot) retrieved from the exposed AjabaF pavement O SI TY ER NI V U 108 Y AR LIB R N DA Fig. 4.4: Stratigraphy of the westerBn Aflang of the excavated potsherd pavement at Ajaba I OF ITY ER S NI V U 109 AR Y R LI B DA N A Fig. 4.5: Plan showing the exposed eIxBcavated pavement. F Y O IT RS E V UN I 110 light brown soil containing potsherds, laterite coated pieces of iron slags and furnace lining. Layer III was between 75 cm and 1 m and it contains mostly slag lumps. The soil was compact and deep reddish brown. Finds from this level include broken pieces of tuyeres and pottery intermixed with iron slags. Below this level was the bedrock which consists of amphibolites and quartzites. From the evidence of the pottery recovered from layers I, I, and III, it appears that the industry belong to one cultural unit which is iron working industry. 225 pieces of potsherds were recovered from the excavation. The potsherd finds also include 15 pieces of complete tuyeres foundY encrusted together with slag in sets of two or three. Samples of sherds are browRn to greyish in colour. Samples IPC 1-1, 3-2, 3-7 and 5-4 were selected and collecAted from the researchers for analyses. IB R 4.1.4: Potsherd pavement at Ila-Orangun L No excavation was carried out here. Erosion and huNman factors have greatly exposed this pavement to destruction. Potsherds werDe scAattered all over the site. The sherds are both coarse - and fine-grained in fabArics, are decorated and came mostly from a pot vessel. Like most of the sherds Bfrom this region, they contain mica as inclusion. The pavement in Ila-OrangunF is rIemarkable, as the sherds were arranged to flush with the lip of a pot, similar to the Ife tradition. Seventy-two potsherds were surface collected from the site (O gOunfolakan 1994). Samples KM 1, 3, 4 were taken from this site for analysis.T UnYfortunately the site has been destroyed as a result of a petrol station built on the Isite (Plate 4.2). S 4.1.5: AjabaE MoRund Excavation (KJAJ RM1): OIneV of the earlier identified refuse mounds (KJAJ RM1) within the Ajaba abandNoned settlement was excavated. It is to compliment the earlier excavation at the pUotsherd pavement site within the abandoned settlement complex. A 1 x 2 meters trench was marked for digging (Plate 4.3). The 1 x 2 meters was pegged because of some reasons: (1) time constraint; and (2) the surface were heavily littered with potsherds; hence, concentration of material is likely to be high. The pit was sunk on a refuse/rubbish mound over a gridded area of about 6 x 6 metres, towards the southern o o edge of the mound. The pit was sunk right on the peak of the mound (7 55‘N & 4 111 RY BR A LI AN AD Plate 4.2: Petrol Station undIerB construction on the Ila-Orangun potsherd pavement site OF TY RS I VE NI U 112 53‘E) in order to: (1) have a complete section of the geographical horizon (2) examine the vertical concentration of material and (3) establish a cultural sequence for the site. A 10 cm interval was chosen as our spit level. The trench consists of twelve spit levels, which yielded varieties of materials at each level. Before the commencement of the excavation, having gridded the site and chosen a portion to be excavated, spot height readings of the four corners of the trench were taken with the quick-set (Plate 4.3). The following readings were obtained: corner Southwest 1.10 m, Northwest corner 1.10 m, Northeast corner 1.23 m and Southeast corner 1.27 m with the centraYl 1.17 m. As a result, the excavation proper commenced at the spit of 10 cmR for convinenicy and photograph taken at the end of each spit (Plate 4.4) till it gotA to level 12 (112cm; Plate 4.8). R The excavation yielded varieties of material remains,L bIotBh organic and inorganic. These include pottery sherds, iron objects, slag (few), cowry shells, bivalve shells, animal bones, snail shells, grinding stones, flakes, Ntuyere fragments, baked clay, charcoal, palm kernel and plant seeds among otAhers. These finds were then bagged according to their provenance and mateArialD. Throughout the levels, pottery sherds was the most frequent and most abBundant of the materials. Although the quantity diminishes as the pit was getFting Ideeper. Because of the vegetation of the area and cocoa plantation, roots of cocoa tress as well as other plants within the excavation area disturbed the mat eOrial in-situ as the roots perched through potsherds, breaking some of them intoT smYaller pieces. It made the excavation process slow (Plate 4.6. Also the exercise at Ithis point did not go smoothly, as some of the local people raised objections to oSur digging at the site. This they viewed as an exercise to ‗mine charcoal‘ becaEuseR one of them heard us discussing the possibility of getting charcoal for datingI. N At V level eight (70-80cm), there was pronounced ash intrusion at the flank of thUe western and southern walls, thus, lumps of charcoal specks were collected. This level also produced a human tooth and mica. The sterile layer was reached at 110- 120cm below the surface with compact lateritic sediment (Fig 4. 7). This is the bottommost unit. It is a very moist, fairly compact, fairly hard, silty clay deposit contains some tiny rootlets and it is completely devoid of cultural materials. It is Yellowish brown in colour (5YR 4/6 Munsell colour chart). 113 Y RA R LI B AN BA D I Plate 4.3: 1 by 2 metres pegged trFenc h for excavation Y O IT RS IV E N U 114 At the end of the excavation and after taking soil sample for palynological analysis, the stratigraphy was drawn using the Munsel colour chart to discern the colour of each layer (Fig. 4.8). Finds were documented and sorted according to each layer (Table 5.1). AR Y BR LI DA N IB A O F ITYS VE R I UN 115 Y RA R LI B DA N A IB Plate 4.4: Spit level one F O TY RS I VE UN I 116 RY BR A LIN DA IB A Plate 4.5: Excavation in ProgressF Y O IT ER S IV N U 117 Y AR LIB R DA N BA Plate 4.6: Roots and rootlets occuFrre dI right from spit level one to level 10. O SI TY R IV E UN 118 RY RA LI B AN BA D I O F ITY RS Plate 4.7: VESterile Layer of the Excavation I UN 119 RY BR A N LI AD A Plate 4.8: Measuring and Drawing o f IthBe Stratigraphic Units O F TY I RS IV E UN 120 Y AR BR N LI A D Fig. 4.6: Stratigraphy of Ajaba MoundB EAxcavation F I Y O IT RS IV E UN 121 4.1.6: Stratigraphy Description Layer 1: The top soil is a dry, very compact, hard, powdery silty deposit that breaks in clods. It contains abundant roots and rootless of various sizes, sits conformably on top of the uppermost cultural layer. Contains some compressed (weathered) cultural materials, especially potsherds. Colour: The top soil is dark reddish brown in colour (5YR 2.5/2Y Munsell colour chart). A thin potsherds line directly uAndeRrlies it. R Layer II: The layer is a very hard, very compact, moIisBt, gritty, sandy, silty deposit. It breaks in clods and co ntLains some roots of varying sizes. It also contains abuNndant cultural materials (pottery, snail shells etc). Sits Adiscomformably in the mid section of the north wall andD contains some chunks of burnt bricks. A Colour: It is very da rkI Bgrey in colour (5YR 3/1 Munsell colour chart) OF Layer III: The layYer is a moist, fairly gritty sandy silty deposit that conItTains very abundant burnt brick, some roots/ rootlet, chunks Sof varying sizes, charcoal specks, very abundant potsherds and R broken parts of tuyeres. VE Colour: It is reddish brown in colour (5YR 4/3 Munsell colour I chart). UN Layer IV: The layer is a moist, compact, powdery and silty deposit. It rolls between fingers when wet. It contains very abundant cultural materials (mainly pottery), some burnt brick chunks, some roots and rootlets, and snail shells among others. It is reddish brown in colour (5YR 4/3 Munsell colour chart). 122 Layer V: This is the bottommost unit. It is a very moist, fairly compact, fairly hard, silty clay deposit contains some tiny rootlets and it is completely devoid of cultural materials. Colour: It is Yellowish brown in colour (5YR 4/6 Munsell colour chart). RY BR A N LI A AD F I B Y O T SI R IV E N U 123 CHAPTER FIVE ANALYSES OF FINDS 5.1. Analyses of Finds. Finds from this present excavation of Ajaba refuse mound 1(KJAJ RM1) were classified into organic and inorganic material. Organic materials were sorted out according to their type. Inorganic materials were sorted out according to the individual materials. Lithic, iron and ceramic materials were identified and classifiedY. Apart from classifying ceramic into their decoration and morphological atAtribRutes, geochemical and mineralogical analysis of ceramics from previous and present i.e. Ajaba potsherd pavement and Ajaba refuse mound 1 (KJAJ PP and KRJAJ RM1) excavations and materials from other sites in Osun [namely: IpetumIoBdu (IPC), Ikire (IKRE), Iresi (IRC) and Ile Ife (IFE 1,2,3)] were carried out. TLhis is to enable us compare and relate their mineralogy and material conAtenNts with each other. Data derived from the geochemical and mineralogical analysis is expected to throw light on the ceramic source for Ajaba. The data derived from Dthe analysis has also thrown light on interrelationships between the ceramic tradBitioAns from these four areas (Ipetumodu, Ikire, Ile Ife and Ajaba Kajola). PalynoFlog icIal analysis of soil samples collected from different layers of KJAJ RM1 was carried out to throw light on some aspects of past vegetation of the area. O Y 5.1.1 General ArtefactI ITnventory Finds fromR KSJAJ RM1 excavation consist of both organic and inorganic materials incluEding metals (Table 5.1). As stated earlier, pottery sherds form the bulk of the finds from this excavation (3654 sherds). Other finds include iron objects (11piNecesI; VFigs. 5.3-5.5), slag (5 pieces, Fig. 5.2), cowry shell (3), bivalve shell (13), anUimal bones (72; Fig. 5.1), Human tooth (1; Fig. 5.6), snail shells (175), lithics materials (63), flakes, tuyere fragment (16), baked clay/furnace fragment (96; Fig. 5.7), charcoal, palm kernel and plant seeds among others. During the excavation, finds from daily excavations were packed according to their spit levels. These finds were also separated based on the materials types (i.e. pottery, lithic and organic etc.). Each material type was also subjected to classification based on form. For instance the pottery sherds were classified according to forms and decoration. Further analysis of other finds such as iron, human tooth, snail shells, animal bones were also carried out. 124 Table: 5.1: General Artefact Inventory – Ajaba Refuse Mound I. + 0-10 371 34 - - 2 9 - 20 - 1 - 47 2 8***Y 1 1 x 41 - - - - - 537 1 ++ 10-20 387 42 - 1 1 5 - 8 - 8 1 9 - 7 2 20-30 645 70 66 - - 15 1 15 - 8 - 34 - 2R - 13 x - - - - - - 482 ++5 - 2 x 6 4 - - - 1 892 ++ 30-40 222 43 158 - - 25 - 14 2 4 - 5 A- 5 - - x 3 - 8 7 - - 496 3 40-50 392 40 - - - - - 24 - 4 - - R- - - - x 1 - 1 - - - 462 ++ 50-60 240 15 38 - - 12 - 18 - 34 4 B- - 6 1 - x - - 9 - - - 377 ++ 60-70 167 39 33 - - - - 45 - - - I - 1 5 1 - x - - - - - - 291 70-80 178 10 34 - - 4 - 19 9 N5 L- - 1 - - - x - - - - 19 2 281 +++ 80-90 192 19 66 - - 1 - 9 2 6 6 - - 4 - - x - - - - - - 305 90-100 75 4 35 - - 1 - 2 - A 2 - 1 1 - - - x - - - - - - 121 100-110 26 2 4 - - - - 1 D- - - - - 3 - - x - - - - - - 36 110-120 3 - - - - - - -A - - - - - - - - - - - - - - - 3 Total 2898 318 434 1 3 72 1 175 13 72 11 96 5 63 3 16 51 4 18 7 19 3 4283 + = ‗Opoto‘ (Ficus Capensis); ++ = Grinding stones (x = present; *** = Quartz or Granite flakes; present; 1 = 1 Phalanges, 1 pelvic bone and 6 fragments; 2 = 1 st erInuBm, 1 vertebrae bone and 6 fragments; 3 = 1 Humerus and 3 fragments; +++ = Soap stone; OF TY SI ER V NI U 125 Level (cm) Body Sherds Rim Sherds Undiagnostic Sherds Perforated Sherd Pot Lid Knob Palm Kernel Burnt Kernel Snail Shell Bivalve Bone* Metal Pieces** Hearth Slag Lithics Cowry Tuyères (fragments) Charcoal Plant Seed Furnace fragment Burnt Clay Charred food (fragments) Ash Lumps Mica Schist Total Plate 5.1: Jaw of an animal Plate 5.2: Iron slag RY RAB N LI DA A Plate 5.3: Iron Object IPBlate 5.4 Iron Object OF ITY RS VE Plate 5.5:I Comparing the Iron Objects Plate 5.6: Human tooth UN 126 Y Plate 5.7: Baked clay/Fragment of Furnace? AR R LI B N AD A F I B Y O SI T R VE UN I 127 5.1.2. Pottery Analyses The nature of ethnographic ceramic research differs fundamentally from that performed by archaeologists in that the former examines whole pots in systematic context while the latter most often deal with vessel fragments from archaeological deposits (Bollong, 1994). Most often descriptive analysis is carried out without consideration for geochemical and mineralogical analysis of pottery sherds in order to relate these sherds from different sites to one another. As observed, there is `no standard regional typological scheme developed foYr describing ceramic in Yoruba archaeology‘ (Aleru, 2006: 125). This set bacRk is usually blamed on the inadequacy of ethno-archaeological research on AYoruba pottery‘ (Allworth-Jones 1996; Fatunsin 1992). Secondly most stuBdiesR on Yoruba pottery have tended to focus mainly ‗on procedures for pottery mLanuIfacture (forming the vessels, decoration and firing), types of raw materials, forms , types and functional attributes‘ (Aleru, 2006: 125). Also each researcheAr teNnds to use his/her own procedure depending on what he/she is aiming at. ThDis has culminated into the present state of inadequate understanding of cultural relationships in Yorubaland. For a better understanding of human groupings and relaItiBons A, culture process and movements of people among other things, pottery analFysi s from our area (Yorubaland in general and Osun North east in particular) we mOust proceed beyond technological and typological definitions (Aleru, 2006). Since clay productT isY one of the most indestructible evidences of human existence, we therefore neIed to carry out thorough examination of clay products from this area and otherR siteSs within the area and Ile-Ife. 5.1.3: PoItVtery E analysis KJAJ RM1 NThe bulk of the finds consist of pottery, which forms about 85.3% of the total fiUnds compared to the non-pottery finds (14.7%). A total number of 3654 pottery materials were collected from the excavation. Of these a total of 3220 sherds were diagnostic. The pottery was classified into five major groups. These are body sherds, rim sherds, undiagnostic sherds, perforated sherds and pot lid with frequencies of 2898 (body), 318 (rim), 434 Undiagonistic sherd), 1 (perforated) and 3 (pot lid) respectively. Pottery materials occurred in all the levels, although the number pottery sherd decreases as we approached the sterile layer. Level 3 (20 cm - 30 cm) has the highest number of pottery (781) followed by level 5 (40 cm – 50 cm) with total 128 numbers of 432 pottery materials. Level 1 (0 cm – 10 cm) has 407 pottery while level 12 (110 cm – 120 cm) has the lowest number of 3 (Table 5.2). 5.1.4: Pottery Decoration Classification The decoration types on the pottery materials were determined. Eighteen decoration types were discerned on the body sherds while thirteen were identified on the rim sherds. The morphological analysis was also carried out and form part of this discussion. Y Plain body sherds, occurred in all levels and has the highest percenRtage occurrence with a total number (1088, 38.5%). This is followed by singleA twisted cord with a total number of 829 sherds. This amounts to 31.6% ofB theR total body sherds. It occurs in all the levels apart from levels 10 and 12. DoIuble twisted cord forms the third most frequently occurring decoration type. It is p reLsent in all the levels except levels 4 and 12. A total of 371 sherds were classifieNd in this category. This is 15.1% of the total body sherds. Grooves and burnished sAherds are also present in high numbers (185 and 125 respectively). Other decAoraDtion types recorded are painted, incision, bossing, scallops, circle stylus, hIeBrringbone design, carved roulette and groove/punctuate among others (Table 5 . 2). Similarly, the rim sherds were also classified based on the decoration oOn thFem. Like with the body sherds, plain rims have the highest percentage occurrenc e (135 with 44.1%). However, unlike the body sherds, burnished rim sherdTs hYave the second largest percentage occurrence (74 sherds - 24.2%). Grooves have Ithe third highest percentage occurrence with 25 rim sherds. This amounts to 8R.2%S of the total rim sherds. Other decoration types diagonised on the rims are sinEgle and double twisted cord and incision, among others (Table 5. 3). 5.1.5 VNI Pottery Typological Classification U For cultures that produced pottery, archaeologists invariably spend a great deal of time defining ceramic "types." Each type is a series of attributes which distinguishes one group of pottery (whether whole vessels or potsherds) from all other groups of pottery, such that each type was produced in a single time and place. Ideally, the attributes used to identify types are ones that are identifiable with the naked eye, and are found on small fragments of pottery, so that the sorting of potsherds into types is quick and straightforward. 129 The use of a traditional ceramic typology allows one to compare different assemblages since the typology consists of a set of shared traits used by analysts. This type system permits the archaeologist to date a site without reliable chronometric data via comparison to known dated ceramic assemblages. This system also provides a basis of identifying locally produced and imported ceramics and their approximate source area. It also helps to identify the source or sources of ceramic as it relates to same group of community (Colton and Hargrave 1937). In Yorubaland, clay transformation (pottery production) has given poRtteryY products different names to distinguish their uses. The most important oneAs being, Ape (cooking pot), Amu (water-jar), isaasun (soup-pot) and oru (kettle)R Ikoko (big cereal collection pot) and they are reflected all in the pottery claIssBification of the Ajaba pottery (Fig. 5.1). Most transient of the pottery typology Lin Ajaba are kolobo (Fig. 5.1a), isaasun (soup pot Fig. 5.1b and c), ape (small Ncooking pot Fig. 5.1 d-f) and ikoko (big cooking or storage pot Fig. 5.1 g-i) and Aamu (water jar Fig. 5.1 j and k). AD 5.2. Geochemical and mineralogical A nIaBlysis In drawing attribute inferencesF on the cultural-historical relationship among the people of Osun Northeast and bOetween them and other parts of Yoruba land, most importantly with Ile-Ife, ceramYics attributes, specifically, decorative motifs were used on one hand and geocheImTical and mineralogical analysis on the other. Also, the stylistic attributes ofS pottery are used to draw inferences on cultural historical relationship amonRg settlements. The aim of any scientific study of ceramics should be consideredV froEm two frames of the production of pottery; (i) the technology; and (2) characterIistics. A principal objective of the analysis is to elucidate the technology of potterNy, the method by which clays were prepared, shaped, finished and fired (Rice 1U999, Hodges 1968, Sayko 1965; Wirska (1967). In this work the aim is to use geochemical methods to answer the following questions: (i) What are the compositional groups present in ceramics from different archaeological sites and how do they differ from each other? (ii) What are the differences in manufacturing techniques? 130 RYA LIB R N A BA D I OF ITY S VE R UN I Figure 5:1: Pottery typology 131 Table: 5.2: Distribution of Major Pottery Decoration Types (Body) Levels (cm) Decoration Types Plain 179 70 313 64 92 86 91 51 100 29 10 3 108 38.5 8 Burnished 6 41 24 12 9 2 1 13 11 - 6 R-Y 125 4.4 Single Twisted 136 112 248 107 96 80 - 54 33 22 4 - 892 31.6 Cord Double Twisted 17 60 78 - 90 46 31 29 12 4R A4 - 371 13.1 Cord Incision/Excisio 4 - - - - - - 3 1 IB- - - 8 0.3 n Groove/Wavy 13 7 13 16 20 21 32 14 L28 19 2 - 185 6.5 Groove Twisted - - 20 23 8 - - A2N - - - - 53 1.9 Cord/Groove Painted - - - - - - AD2 4 2 1 - - 9 0.3 Bossing - - - - - - - 1 - - - - 1 0.04 Carved Wooden - - 3 - 6 3 7 4 5 - - - 28 1.0 Roulette- IB Checked Pattern Wavy 5 - 7 - OF- - - - - - - - 12 0.4 Groove/Twisted Cord Roulette Groove/Twisted 10 31 T- Y - - - - - - - - - 41 1.5 Cord Scallops - -S I 2 - - - 2 3 - - - - 7 0.2 Carved Wooden 1 Roulette- R- 1 - - - - - - - - - 2 0.07 Multiple Design E Boss/TwistedI V - - 1 - - - - - - - - - 1 0.04 Cord Circle StyNlus - - 1 - - 1 - - - - - - 2 0.07 HerrinUg Bone - - - - - 1 - - - - - - 1 0.04 Design Groove/Punctate - - - - - - 1 - - - - - 1 0.04 Total 371 321 711 222 321 240 167 178 192 75 26 3 2827 100 132 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120 Total % Table: 5.3: Distribution of Major Pottery Decorations (Rims) Levels Plain 26 18 32 14 10 1 16 5 8 3 2 135 44.1 Burnished 5 14 13 7 13 6 8 1 7 - - 74 24.2 Single Twisted Cord - - 7 4 - 3 4 - - - - 18 5.9 Double Twisted Cord - 4 - - 7 2 - - - - - 13 4.2 Incision/Excision - - 1 - 6 - - - 1 - - R8Y 2.6 Groove/ Wavy Groove 1 - - 7 - 2 11 1 3 - - 25 8.2 Twisted Cord/Groove 1 - - - 2 1 - - - - - 4 1.3 Groove/Twisted Cord 2 2 - 4 - - - - - -R A- 8 2.6 Boss/Twisted Cord - - 1 - - - - - - 1 - 2 0.7 Wavy Groove/Twisted Cord - - 4 - - - - - - B- - 4 1.3 Excision - 1 10 - - - - - L- I - - 11 3.6 Wavy Excision - 2 - - - - - - - - - 2 0.7 Twisted Cord/Punctate - - 1 - - - - N1 - - - 2 0.7 Total 35 41 68 36 38 15 3A9 8 19 4 2 306 100 AD B I F O SI TY VE R I UN 133 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100-110 Total % Y RA R LI B N Plate 5. 8 Plain A AD F I B O ITY RS IV E UN Plate 5. 9 Carved roulette 134 RY BR A LI AN Plate 5. 10 Plain rim/Twisted cord D A F I B O SI TY ER NI V U Plate 5. 11: Roulette 135 Y AR LIB R AN Plate 5. 12: Groove/Wavy Groove D A F I B O ITY RSE NI V U Plate 5. 13. Twisted cord 136 AR Y LIB R Plate 5. 14: Incision/roulette N A AD IBF O SI TY ER IV UN Plate 5. 15: Plain 137 AR Y BR Plate 5. 16: Bosing/Twisted Cord N LI AD A F I B O ITY RS VE UN I Plate 5. 17: Wavy/Groove 138 Y RA R B I Plate 5. 18: Composite L AN AD F I B Y O ITS VE R UN I Plate 5. 19 Burnished 139 AR Y R LI B Plate 5. 20 Burnished/GroovedN A D BA F I Y O T RS I VE UN I Plate 5.21 Incision 140 RYA R Plate 5.22: Perforated Pottery IB LN DA IB A O F ITY ER S IV Plate N5.23 Painted U 141 (iii)Is conflict, war, displacement and migration a factor in technological evolution of the ceramics? (iv) Are the clay compositions a reflection of the local geology? Sourcing or chemical fingerprinting of archaeological materials is becoming increasingly important in our understanding of prehistory, especially in helping to reconstruct past mobility and exchange systems. Obsidian, andesite and basalts sourcing, has been a mainstay in provenance studies (Jones et al. 1997), but recent attempts to source steatite or soapstone (Allen et al. 1975; Ige and Swanson 2008)Y, and even trees (Durand et al. 1999) have shown that these lines of inquiry can be Rquite informative. Despite success in many areas worldwide with sourcing pottery A(Bishop et al. 1988; Lynott et al. 2000), this avenue of research has been virtuallyR ignored by Nigerian archaeologists. A recent pilot project in the Igbomina areIa Bof the northern part of Yorubaland using instrumental neutron activation analys isL (INAA) (Usman et al. 2005) is a significant contribution to the characterizatioNn of these objects. As a whole, chemical analyses of ceramic in Nigeria lag far Abehind the analysis of other artifact categories. It is hoped that this work wiAll fuDrther add significant data to the technological evaluation of ancient Yoruba poBttery technology. The aim in this present analysFis isI to address this problem by creating a typology based on chemical properOties of prehistoric southwestern Nigerian ceramics using ICP-MS and other analytica l facilities. The goal of the study is similar to that of lithic sourcing, that is, to TdivYide artifacts on the basis of where they are from. It is assumed however, that thIis goal will not be as straightforward as in lithic sourcing (Steponaitis et al.,R 199S6). This is because clay is relatively common and clay sources are generally Elarger than obsidian source zones, thus making ceramic sourcing less accurate IthVan obsidian sourcing in a spatial sense. Moreover clays form under a numbNer of conditions and are often mixed with other source clays thus creating a coUntinuous distribution of chemically varying clays across an area, unlike obsidians. Finally, raw clay is subject to a number of transformations by people before it actually becomes a pot and ultimately a sherd in the archaeological record (Arnold et al., 1991; Blackman, 1992). At the Materials and Archaeometry Unit of the Natural History Museum, Obafemi Awolowo University, Ile-Ife, there is an ongoing long-term effort to characterize the provenance of archaeological artifacts from the Yoruba area of southwestern Nigeria (Ige and Ogunfolakan, 2001; Ige and Rehren, 2003; Lankton et al., 2006; Ige and Swanson, 2008). 142 5.2.1. Materials and Methods (Pottery Description for geochemical analysis) The sherd samples for the analysis are mostly from body part of pottery vessel. Rim portions of pottery vessels were also used (although this is rare). Generally, the sherds for this present analysis are decorated. Some of them are plain and burnished sherds. Fabric wise, some are coarse while some are fine-grained. Majority of them also came from potsherd pavement. Potsherd pavements are common in Ile-Ife and other Yoruba towns and even as far as Togo and Benin Republics. It is a West African cultural phenomenon. In Ile Ife and related areas the potsherd pavements arYe stylistically herringbone in structure. In Old-Oyo, Togo and related areas theyR are structurally laid flat on the ground. In order to have a good representatioAn of the pottery geochemical analysis, sherds were collected from six differentB siteRs with three sites coming from the study area. These sites include Iresi, AjaLba Iand Ila-Orangun, and Ipetumodu, Ikire and Ile-Ife outside the study area. N 5.2.1.1. Ile Ife A In Ile-Ife sherds each were retrieved from twDo different sites. These are the Iloran and Mologun ancient communities, IreBpre Asenting aborigine, pre-Oduduwa (97 Samples IFE series) and Ita Yemoo (F100 IFE FUV series) representing Oduduwa settlements. Here we also retrievOed several fragments of glass making crucibles. These samples were retrieved fro m different layers, ranging from 14cm to 1.4 metre depth. Samples from otherT sitYes from Ile-Ife came from abundant surface scatters at Mologun, Iloran, Ita YemIoo. Most of these sherds are fine- grained in fabrics. They are mostly brownR inS colour although, some exhibit some dark colour. Some are decorated whiEle some are smooth surface. In addition, majority of them have mica as inclusionI. SVherds from Ita-Yemoo came from a disturbed layer of about 35cm when a foundNation of a building was being dug (Plate. 5.25). It came from the line of a pUotsherd pavement within the foundation dug for the building (Plate. 5.26). Sherds from other parts are from pavement scatters. The sherds‘ thickness ranges from 0.5cm to 1.5cm. 5.2.1.2 Ajaba For the geo-chemical analysis, samples were also taken from Ajaba. Samples are taken from the sherds from both pavement and cultural mound excavations of an abandoned settlement dated by radio carbon 14 methods to between the 13th and 17th 143 centuries (Ogunfolakan, 2007). Both pavement and mound are about 140 m apart. The pavement is part of the Kajola abandoned settlement and oral tradition noted that it was initially a sacred grove. Potsherd samples from Ila-Orangun, Ikire and Iresi were retrieved from different disturbed layer ranging from 10 cm to 1.4 metre deep. Sherds from Ajaba came from excavations of both pavement and cultural mound. The pavement site was exposed at the middle of a road along Iresi-Ajaba road when the road was being graded. Both pavement and mound sites are about 240 metres apart. The mound iYs towards the north-eastern part of the pavement. The pavement layer is about 3R0cm from the surface. The mound depth after excavation is about 1.2 metres. ThAe sherds came from different layers. Unlike sherds from Ife, sherds here are mostlyR of coarse in texture. They also have mica inclusions while a few are fine-grainLed IinB texture. 5.2.1.3 Iresi N Sherds from Iresi came from a rockshelter (PlatAe. 3.11). The morphology of some of them can be distinguished. They are maAinlyD from pot body. One is black in colour whilst others are dark-brown. B I 5.2.1.4 Ila-Orangun F Sherds from this site came fOrom pavement scatters. Erosion and human factors have greatly exposed thisT paYvement to destruction. They are of coarse and fine grained in fabrics. They aIre decorated and came mostly from pot body. Like most of the sherds from thRis reSgion, they contain mica as inclusion. They are brown in colour. 5.2.1.5 EIV Iragbiji N The sherds from Iragbiji came from a pavement site. They are of fine grain in faUbric. They are brown and dark-brown in colour. Only one is decorated. 5.2.1.6 Ipetumodu Sherds from Ipetumodu came from excavation of an iron smelting site. 5.3 Geochemical Analytical Methods: Ten thin sections were made from selected samples from each site for petrographic examination. Out of the samples about three samples representing the 144 RY BR A N LI DA Plate 5.24: Sherds for geochemical anaBlysAis from Iloran (Ile-Ife) ancient community I OF ITY ER S NI V U 145 AR Y R LI B DA N Plate 5.25: A building under constructioBn atA Ita-Yemoo, Ile-Ife F I O Y T I ER S NI V U 146 Y RA R LI B N DA A Plate 5. 26: Cutting through potsher dI Bpavement during excavation of a building foundation at Ita-Yemoo, IleF-Ife O ITY RS VE UN I 147 RY RA LI B AN BA D Plate 5.27: Sherds from Ajaba PavFem eInt excavation for Geochemical analysis O ITY S ER IV N U 148 AR Y LIB R N DA BA Plate 5.28: Sherds from Ajaba MoundF E xIcavation for Geochemical analysis O ITY RS E IV N U 149 RY RA LI B AN BA D I Plate 5.29: Sherds from IragbiOji foFr Geochemical analysis Y T I S ER NI V U 150 AR Y LIB R N A AD Plate 5.30: Sherds from Iresi for Geoc hIeBmical analysis O F ITY RS E NI V U 151 characteristic mineralogy were selected for geochemical analyses. In all, 21 samples were selected for major, trace and rare earth element analyses. IFE RAW1 is a sample of decomposed granite gneiss, near a clay deposit from Ile-Ife which was analysed as a possible source rock for the clay used in making Ife pots. A portion of each potsherd was ground to a particle size of 5–10 mm using a cobalt ball mill. 10 major elements, 31 trace and rare elements have been measured. Maximization of the number and range of elements was important because it was not known beforehand what types of elements or combination of elements, if any, might be important in distinguishing RY regional clay and pottery assemblages. For non-refractory trace elements (Mo, Cu, Pb, Zn, Ag, Ni, As, Au,A Cd, Sb, o Bi, Hg, Tl, and Se), 0.25 g of each sample was leached with hot aqua regiaR at 95 C for one hour. For refractory elements (Ba, Be, Co, Cs, Ga, Hf, Nb, Rb, SInB, Sr, Ta, Th, U, V, W, Zr, and Y) and rare earth analysis, 0.5 gm sam pLle was fused with LiBO4/LiB2O7 and leached with nitric acid. GeochemicAal Nanalyses were carried out using Inductively Couple Plasma Mass SpectrometrDy (ICP-MS) at ACME analytical laboratories in Vancouver Canada. Detection for trace and rare earth element analytical precision calculated from replicate AI aBnalysis of one sample (IKC) is better than _1% (table 5.2). Chondrites REE were normalized after abundances of chondrites in Wakita et al, 1971.O To Fassist in the creation of reference groups, a principal components analysis (P CA) was performed on the ICP-MS trace and rare earth element data. PCA isY a convenient way to capture and view complex multidimensional data, IsTuch as compositional data composed of 31 different dimensions (i.e., RelemSents) in a much smaller number of dimensions (Eekens et al. 2002.) We usEed PAST Software developed by Hammer et al 2001, updated 2006. PCA is pIarVticularly effective when the original variables are correlated, as is expected with cNompositional data from discrete chemical sources. 5U.3.1 Results and discussion The description of the mineralogical and geochemistry analysis is given here. The result of the analysis is also discussed. 5.3.2 Mineralogical Analysis Sherds from Ife are grouped into two according to location. Group one consists of samples IFE 1-3. The non-plastic materials consist of broken quartz and 152 Table 5.4 Major element composition of pavement samples and raw materials Sample SiO2 Al2O3 TiO2 Fe2O3 MgO K20 CaO Na2O P2O5 MnO {IRC-1} 67.18 16.73 4.53 0.58 2.42 0.02 1.61 0.02 0.33 0.12 {IRC-2} 68.01 17.23 0.77 5.11 0.47 2.45 0.01 1.31 0.31 0.34 {IRC-3} 66.88 16.98 0.91 4.82 0.59 2.05 0.03 1.23 0.31 0.11 {IPC 1-1} 67.12 15.84 1.34 7.13 0.72 2.38 0.19 0.84 0.38 0.15 {IPC 3-2} 66.89 18.33 1.15 6.43 0.48 2.42 0.05 1.26 0.32 0.06 {IPC 3-7} 67.12 15.85 1.43 7.59 0.58 1.77 0.06 0.95 0.33 0.07 Y {IPC 5-4} 66.95 16.12 0.78 8.12 0.45 1.94 0.04 1.21 0.41 0.12 R {IKRE1} 67.12 17.04 0.69 4.72 0.59 2.33 0.01 1.29 0.35 0.09 A {IKRE3} 67.66 17.09 0.75 5.22 0.55 1.65 0.03 1.12 0.35 0.12 R {KP 3} 70.12 16.79 0.74 3.56 0.32 3.14 0.12 1.31 0.29 0B.14 {KPC 1} 69.98 16.96 0.88 4.65 0.34 3.13 0.01 0.94 0.41L I 0.18 {KPC 2} 70.12 16.12 0.76 3.88 0.31 4.11 0.12 0.96 0.2 7 0.17 {KM 1} 69.56 15.18 0.71 4.12 0.45 4.23 0.02 1.16 N 0.34 0.15 {KM3} 66.87 18.12 0.70 5.07 0.44 3.44 0.03 A1.45 0.32 0.09 {KM 4} 67.12 17.99 0.64 5.88 0.34 3.12 0.0D3 1.94 0.33 0.11 {IFE RAW1} 69.09 17.01 0.45 2.55 0.71 3.54 A 0.92 1.56 0.27 0.07 {IFE 1} 68.89 17.36 1.02 3.99 0.77 I1.B89 0.03 1.12 0.31 0.12 {IFE 2} 68.27 17.01 0.99 4.22 F0.69 2.67 0.06 0.97 0.27 0.14 {IFE 3} 69.03 17.19 0.88 3 .7O7 0.72 2.45 0.05 1.84 0.28 0.18 {IFE FUV1} 68.77 16.12 1.2Y4 5.12 0.68 1.56 0.06 1.43 0.25 0.15 {IFE FUV2} 67.23 16.55I T 1.33 4.55 0.88 2.12 0.03 1.44 0.4 0.17 ER S IV UN 153 Table 5.5 Trace element composition of potsherds from southwestern Nigeria Sample Ba Be Co C Ga Hf Nb Rb Sn Sr Ta Th U V W Zr Y {IRC-1} 317 3 34.4 1.5 25.2 6 17.3 22.8 3 282.3 0.8 2.4 0.7 205 75.1 221.2 19.3 {IRC-2} 486 5 25.7 7.5 24.1 8.5 16.7 72.9 6 65.6 0.9 4.1 6.7 166 70.7 308.5 26 {IRC-3} 386 3 21.3 2.7 24.8 6.4 11.1 28.6 4 457.3 1 8.2 4.1 118 62 238.3 17.1 {IPC 1-1} 713 3 21.4 4.4 19.3 21.7 20.1 110.6 3 121.1 1.6 19.5 4.7 72 122.4 722.3 37.4 {IPC 3-2} 756 2 16.5 4 17.5 19 19.3 104.2 2 119.2 1.4 19 4.1 77 69.7 669.8 30.9 Y {IPC 3-7} 3279 2 17.7 1.8 22.4 26.2 18.5 105.1 2 461.1 0.9 1.7 1.5 88 49.2 1055 16.1R {IPC 5-4} 656 3 32.7 4.5 18 19 18.8 105.1 3 114.8 1.6 14.8 3.7 72 165.4 647.4 A 40.1 {IKRE1} 343 3 37.2 1.2 27.9 4.3 10.6 40.1 6 301.3 0.7 2 1.8 197 85 155 R 21.4 {IKRE3} 1260 5 48.6 1.1 23.6 13.5 22.9 63.2 4 646.3 1.7 13.2 1.5 152 I 2B94.6 514.7 49.2 {KP 3} 270 2 190.5 1.6 8.9 1.9 5.9 23.2 1 13.6 0.6 4.1 1 L 236 56 62.4 10.8 {KPC 1} 143 2 126.5 1.3 8.5 2.6 8.9 20.2 2 14.6 1 3.6N 0.9 119 64.1 91.8 14.7 {KPC 2} 243 2 175 1.5 10.5 27.1 21.9 1 11.3 0.5 3.A5 1 153 56.9 68.7 12.6 {KM 1} 543 4 32.7 7.2 22.7 5.7 16 97.7 2 93.9 1D.2 10.2 1.9 128 86.7 189.5 31.8 {KM3} 472 5 66.5 4.4 16.3 5.4 13.8 87 2 B 34.3A 0.9 10.8 2.3 151 57.7 194.1 34.1 {KM 4} 498 11 56.1 8.6 22.6 7.3 26.9 1 10I.2 9 40.4 2.6 8.2 2 176 78.7 251.4 43.4 {IFE RAW1} 409 2 36. 6 9.9 16.8 4.4 12.3F 133.5 3 159.7 1.1 7.6 1.5 126 169.5 144.5 25.1 {IFE 1} 421 3 23.5 2.8 20.8 17 .7O 21.1 40.9 3 284.7 1.1 3.3 1 185 51.9 665.6 28.6 {IFE 2} 435 2 42.3 3.2 22.Y1 11 18.4 47 3 292.4 1.1 2.7 0.5 208 160.9 401.4 23.8 {IFE 3} 3169 2 17.4 2.I3 T 22 41 25.1 138.2 2 313.3 1.2 3.2 2.3 106 48.2 1674 16.6 {IFE FUV1} 967 2 29.7S 13.9 30.1 3.5 19.4 213.1 13 60 3 7.7 4.9 28 361 109.4 31 {IFE FUV2} 958 E 2 R29.6 13.8 29.9 3.7 20 210.7 12 57.6 2.9 8.2 5 26 364.3 108.8 30.5 V I N U 154 Table 5. 6 Rare earth element composition of potsherds from south-western Nigeria Sample La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu {IRC-1} 21.5 56.5 6.75 27.6 6.04 1.49 5 0.58 4.09 0.61 1.95 0.2 1.88 0.16 {IRC-2} 49.8 97.9 13.2 55.7 10.01 2.23 7.99 0.91 5.63 0.84 2.6 0.27 2.29 0.22 {IRC-3} 26 79.1 6.55 26.4 4.38 0.9 3.43 0.42 3.21 0.51 1.88 0.19 2.07 0.2 {IPC 1-1} 72.1 156.9 16.82 64.3 10.68 1.15 8.96 1.08 6.91 1.13 3.62 0.42 3.52 0.43 Y {IPC 3-2} 59.8 129 14.26 54.5 9.51 1.09 7.73 0.94 6.32 0.93 3.06 0.34 2.99 0.33 R {IPC 3-7} 35.4 68.3 8.72 37.9 5.82 2.19 4.45 0.4 2.88 0.43 1.49 0.13 1.51R 0.15A {IPC 5-4} 52.2 113.6 12.16 47.6 8.12 1.04 7.01 0.91 6.89 1.23 4.03 0.49 3.8B6 0.47 {IKRE1} 30.1 59.8 9.07 41 7.63 1.75 6.22 0.63 4.31 0.63 1.91 0.1L6 1I.58 0.12 {IKRE3} 112.5 154.6 28.13 108.6 18.1 3.88 14.73 1.64 9.73 1.48 4N.3 0 .49 3.06 0.42 {KP 3} 12.8 27.8 3.42 13.8 2.76 0.48 2.36 0.24 2.15 0.31A 1.17 0.1 1.27 0.07 {KPC 1} 12.2 24.9 3.37 14 2.66 0.45 2.47 0.3 A2.67D 0.42 1.53 0.11 1.39 0.1 {KPC 2} 13.3 32.5 3.67 15.7 2.92 0.55 2.58 0I.2B8 2.29 0.35 1.36 0.11 1.35 0.1 {KM 1} 43 82.3 10.56 41.1 7.57 1.44 6.6 0.83 5.73 0.92 3.29 0.34 2.91 0.31 {KM3} 39.3 78.7 10.63 42.3 8.1 1.6 F6.97 0.91 6.59 1.07 3.72 0.42 3.82 0.43 {KM 4} 29.4 67.1 8.54 34.8 7.22 O1.5 7.15 1.02 7.64 1.43 5.04 0.56 4.54 0.55 {IFE RAW1} 24.1 53.2 6.15I T 23Y.7 4.85 1.14 4.33 0.5 4.28 0.67 2.63 0.26 2.74 0.27 {IFE 1} 38.1 71.4 1S0.48 43.4 7.92 1.95 6.97 0.77 5.51 0.88 2.97 0.3 2.68 0.26 {IFE 2} 37.3 74R.3 10.77 45.6 8.51 1.81 6.56 0.7 5.04 0.71 2.61 0.23 2.48 0.19 {IFE 3} 29E.6 68.4 7.47 31.3 5.54 1.76 4.51 0.41 3.34 0.51 1.95 0.17 2.12 0.19 {IFE FUV1I}V 43.2 72.4 10.73 40.8 7.5 1.48 6.45 0.78 5.62 0.82 2.75 0.26 2.46 0.21 {IUFE FNUV2} 44.1 76.4 10.87 41.1 7.59 1.5 6.37 0.76 5.44 0.82 2.79 0.24 2.55 0.2 155 Table 5.7: Chondrite normalized rare earth element composition of potsherds and granite regoliths from southwestern Nigeria Sample La/N Ce/N Pr/N Nd/N Sm/N Eu/N Gd/N Tb/N Dy/N Ho/N Er/N Tm/N Yb/N u/N {IRC-1} 63.2 62.0 56.3 43.0 30.0 30.5 19.2 12.3 13.6 7.8 9.8 6.3 8.5 4.7 {IRC-2} 146.5 107.7 110.0 87.0 50.1 12.3 30.7 19.4 18.8 10.8 13.0 8.4 10.4 6.5 {IRC-3} 76.3 86.9 54.6 41.3 21.9 15.8 13.2 8.9 10.7 6.5 9.4 5.9 9.4 5.9 {IPC 1-1} 212.0 172.0 140.2 100.5 53.4 14.9 34.5 23.0 23.0 14.5 18.1 13.1 16.0 12.6 {IPC 3-2} 175.0 141.0 117.9 85.2 47.6 30.0 29.7 20.0 21.1 11.9 15.3 10.6 13.6 9.7 {IPC 3-7} 104.0 75.0 72.1 59.2 29.1 14.2 17.1 8.5 9.6 5.5 7.5 4.1 6.9 4.4 {IPC 5-4} 153.5 124.8 100.5 74.4 40.6 24.0 27.0 19.4 23.0 15.8 20.2 15.3 1 7.5 13.8 Y {MIFE 1} 88.5 65.7 75.0 64.1 38.2 53.2 23.9 13.4 14.4 8.1 9.6 5.0 7.2 3 .5 R {MIFE 3} 331.0 170.0 232.5 169.7 90.5 6.6 56.7 34.9 32.4 19.0 21.5 15.3 16.4 A 12.4 {KP 3} 37.6 30.5 28.3 21.6 13.8 6.2 9.1 5.1 7.2 4.0 5.9 3.1 5R.8 2.1 {KPC 1} 35.8 27.4 27.9 21.9 13.3 7.5 9.5 6.4 8.9 5.4 7.7 3B.4 6.3 2.9 {KPC 2} 39.2 35.7 30.3 24.5 14.6 19.7 9.9 6.0 7.6 4.5 6.8 I 3.4 6.1 2.9 {KM 1} 126.5 99.5 87.3 64.2 37.9 21.9 25.4 17.7 19.1 11.8 1 6L.5 10.6 13.2 9.1 {KM3} 115.6 86.5 87.9 66.1 40.5 20.5 26.8 19.4 22.0 1N3.7 18.6 13.1 17.4 12.6 {KM 4} 86.5 73.7 70.6 54.4 36.1 32.2 27.5 21.7 25.5A 18.3 25.2 17.5 20.6 16.2 {IC 1} 227.6 134.5 149.9 105.0 51.6 17.4 29.6 16.6 16.4 9.0 11.6 6.3 9.4 4.4 {IC 2} 175.0 119.0 110.5 78.8 47.6 12.1 32.4 23.4 D25.4 15.5 21.4 14.4 17.7 12.4 {IC 3} 156.7 92.3 91.6 60.3 28.5 12.7 16.6 8A.1 9.3 5.3 7.6 3.1 6 .2 2.1 {IC 4} 422.0 383.0 305.3 204.7 116.5 15.6 6I9.B8 46.2 43.4 26.5 33.5 22.8 25.3 18.5 {IFE RAW1} 70.8 54.5 50.8 37.0 24.3 2F6.7 16.7 10.6 14.3 8.6 13.2 8.1 12.5 7.9 {IFE 1} 112.0 78.5 86.6 67.8 39.6O 24.8 26.8 16.4 18.4 11.3 14.9 9.4 12.2 7.6 {IFE 2} 110.0 81.5 89.0 71.3 4 2.6 24.1 25.2 14.9 16.8 9.1 13.1 7.2 11.3 5.6 {IFE 3} 87.0 75.2 61.7 48.9 27.7 20.3 17.3 8.7 11.1 6.5 9.8 5.3 9.6 5.6 {IFE FUV1} 127.0 79.5 88.7 63.8 Y 37.5 20.5 24.8 16.6 18.7 10.5 13.8 8.1 11.2 6.2 {IFE FUV2} 129.0 84.0 S89.8 I 6 T4.2 38.0 0.0 24.5 16.2 18.1 10.5 14.0 7.5 11.6 5.9 ER IV UN 156 mica grains, which are uniformly distributed, with a significant absence of feldspars. There is also the presence of green amphibole, which is identified to be hornblende. The second group consists of IFE FUV1 and 2. Mineralogically, they are characterized by very fine and uniformly distributed quartz crystals, mica as well as inclusions of feldspars. No other minerals could be identified. Potsherds from Ipetumodu (IPC) are characterized by inclusions of quartz and feldspar, the latter are very few although uniformly scattered. Other mineral inclusions are hematite, tourmaline and hornblende. Y Potsherds from Ikire (IKRE) are characterized by inclusions, highly weathRered coarse crystals of quartz, mica and feldspars. The major non-plastic compoAnents of the Iragbiji sherds consist of abundant scattered medium size quartz and Rmica. There is complete absence of feldspar which is a reflection of the qLuarItzB mica bedrock geology. Other minerals include hematite, ilmenite and tourmalin e. The Ila-Oragun samples are highly enriched in fineN grained quartz and large mica crystals. Archaeological potsherds from AjaDba A(KPC) are characterized by abundant medium-grained mica, feldspar and quaArtz. Tourmaline was also identified. In all, the predominant mineralogy oBf the sherds for all six sites is quartz, mica, and varying amounts of feldspar, wIith traces of tourmaline, ilmenite and amphiboles. Thus, there appear to be nFo temporal or pronounced regional changes in the mineralogy of the sherds. O Y 5.3.3 Geochemical DesIcTS ription Major, tracRe and rare earth element data of sherds, together with one sample of regolith from aE granitic protolith are presented in Tables 5.4 – 5.7. Sample IFE RAW is used asI Va reference material. This is a sample from a regolith from granite gneiss bedroNck. U The results of major element analysis (Table 5.5) show only small variation among the samples in spite of geographical separation. This is clearly demonstrated 2 2 3 2 by the diagram SiO –Al O –K O where all samples plot close together. From the composition of trace elements, most samples are broadly similar in composition. The unsystematic variation in trace element is due to the high mobility of these elements. The lines joining the data points of the analyses for each of the sherds and the raw material have patterns showing a striking resemblance to each other. When compared to the sample of granite regolith (IFE RAW1) the sherds largely follow the elemental 157 patterns of the raw material. The elements mostly affected by the alteration processes in ceramics are Ca (Calcium) and the alkali metals Cs (Cesium), Rb (Rubidium), K (Potasium) and Na (Sodium) are also considered to be soluble elements and are depleted easily (Schwedt et al., 2003). The concentrations of Cs and Rb are controlled mainly by the presence of K-feldspar and mica where Cs and Rb substitute for K. The contents of elements Y (Yttrium), Th (Thorium) and Ta (Tantalum) are generally low to moderate. The REE, Th and Sc (Scandium) are considered to be the most confidenYt elements for provenance studies of ancient ceramics because these elementsR are insoluble and the effects of metamorphism, weathering and diagenesis upon tAhem are minor. On the normalized REE diagrams, all the pottery sherds showB typRical granite signature when compared to the analysis of a granite regLolIith. Despite the differentiations due to some variation in element concentrations, most of the analyzed potsherd and raw material present chemical similaritieAs pNointing to the use of the same kind of raw materials for producing the ceraDmics. Moreover REE pattern of most sherds are similar to that of the raw materials. The outliers are from Ipetumodu which came from a smelting site and Ila-OIBrang Aun whose bedrock is composed of quartzites and quartz mica, amphibole sFchi st. The distribution curves of average REE composition from each locality noOrmalized to chondrites are quite similar to each other and display a negative Eu- anomaly and depletion toward HREE. This pattern falls close to granite-derTiveYd saprolite (Fig.5.1). The elemental concentrations obtained from ICP-MS wIere used as variables in the statistical treatment. We used the Pass (Hammer eRt alS., 2001) packet, cluster analysis, and principal components analysis. In the cluster analysis, we used Euclidean distance as a measure in n- dimensioInaVl sp Eace. As a result of the cluster analysis, the samples were divided into two gNroups, the Ife group and the Ipetumodu group, determined from the chemical coUmposition of these groups (Fig. 5.2). The principal component analysis (PCA) was used to identify the significant elements that contributed to each cluster. There is only one major cluster with one sample each from the IP and IF series forming a lesser cluster. Using both trace and REE at 95% confidence level all samples except IP and one IFE fall within the 95% confidence ellipse showing that the potsherd have a common history or source materials and correlate with the geologic setting. 158 RY BR A N LI DA Fig. 5.2: REE patterns of sherds normalizedA to chondrites and compared to patterns of regolith (IFERAWI)B from granite. F O Y T SI VE R I UN 159 Y AR LIB R N Fig. 5.3: Principal Component analysis and CluDsterA BA I OF Y IT RS E IV N U 160 5.4 Palynological Analysis It is now very well known that archaeological investigations are more comprehensive if carried out within an environmental context; consequently, in order to ascertain the botanical environment of Ajaba and the interaction of the people with this environment, a palynological study of some of the excavated soil samples was carried out. Since the pioneering work of Lennart von Post in 1916, there have been great advances in the application of palynology in palaeo-environmental studies and relative dating. Lennart von Post in 1925 was the first to examine fossil pollen froRm Y archaeological sites (Bryant and Holloway, 1996). Palynology subsequently hAas been employed with great success in many archaeological researches throughouRt the world. The palynological analyses of excavated soil samples, humaInB coprolites and fossilized honey, for example, have shed light on prehistoric d ieLts, food preparation techniques and subsistence economies, human impact onN the natural vegetation, beginning of agriculture, burial practices, and historyA of man-made fires, among others (Bryant 1974; Sowunmi 1985; Kirch 1996; JenDsen 2004; Duffin 2008). In Nigeria, palynological studies have beAen carried out, with positive results, at four archaeological off- or on-sites in Ni gIerBia: (1) Palynological evidence fromF an off-site location in the Niger delta suggests that by about 2,800 BP there wa s Ohuman interference with the natural vegetation in the adjacent forest zone and pYerhaps parts of the deltaic plains, through agricultural practices (Sowunmi 1981Ia)T. (2) The palynoSlogical study by Sowunmi and Awosina (1991) of soil samples from both the terraRce and the interior of a Late Stone Age rockshelter at Kariya Wuro, northeasterVn NEigeria, in the Northern Guinea-Sudan savanna zone, excavated by AllsworthI-Jones (1989) indicated that the vegetation at the time of occupation was essenNtially similar to that of the present day. Furthermore, there seems to be coUntinuity from the Late Stone Age up till today with regard to the exploitation of plants in the area for food and medicinal purposes (Sowunmi and Awosina, 1991). (3) Oyelaran (1998) carried out a palynological and sedimentological study of a core from Osaru pond, which is about 1 km southwest of the Itaakpa rock shelter excavated by him in the northern part of southwestern Nigeria. He obtained a very clear evidence of an upsurge in oil palm pollen and a change in vegetation from wet forest to dry savanna, both attributable to human destruction of the forest. 161 (4) Pollen analysis at Ahanve, a village in the Badagry area, southwest Nigeria, gave a strong indication that the felling and burning of forest trees by humans, from some time after ca. 3,109 ±26 BP, contributed to the destruction of forest vegetation and the concomitant phenomenal rise in Elaeis guineensis, resulting in the extension of the coastal park savanna (Sowunmi 2004). Archaeological finds of a ground stone axe, charred palm kernels, and charcoal at a level above the one dated to 2,670 ±90 BP at a site just east of Ahanve (Alabi, 1999) seem to support this indication. Y The vegetation of Ajaba is derived savanna with relict trees typical of the Rdrier type of the rainforest. This suggests that Ajaba is within the northernmost exteAnsion of the rainforest zone. R In the immediate vicinity of the excavated mound, aboLut I3B-5m away, are Khaya grandifolia, Chrysophyllum subnudum, Nauclea diderric hii, Ceiba pentandra, Elaeis guineensis, Raphia vinifera, Milicia excelsa, TAetrNacarpidium sp. (African walnut tree), Alstonia booeni, Draceana africana, Vernonia amygdalina, Cordia millenii, Chromolaena odorata, Cassia sp anAd ADmaranthaceae. There are also cultivated plants such as cocoa (TheobromIaB cacao), cassava (Manihot utilissima), yam (Dioscorea spp), pepper (CapsicuFm spp), banana (Musa paradisiaca), plantain (M. sapientum), and cocoyam (CoOlocasia sp.). Others are pawpaw (Carica papaya), orange (Citrus sinensis), coconut (Cocos nucifera) and pineapple (Ananas comosus). The following plants werTe fYound within a radial distance of 400-800m from the excavated site: Blighia saIpida, Cocos nucifera, Newbouldia laevis, Tectonia grandis, Carica papaya, TRhauSmatococcus danielli, Samanea saman, Duranta repens (yellow bush), ThevetiaE peruviana and Naploena sp. 5.4.1 NMIa Vterials and Methods of palynological analysis U Pollen analysis of six soil samples from levels 98 cm, 76 cm, 51 cm, 34 cm, 20 cm and 11 cm of the western wall of the excavated trench at Ajaba refuse mound was carried out (Plate 5.32). The analysis was carried out in the palynological laboratory of the Department of Archaeology/Anthropology, University of Ibadan by Mr. Oriejeme under the supervision of Prof. (Mrs) M.A. Sowunmi. The method, result and interpretation are given below. 162 Y AR LIB R N Plate 5.31: Collection of Soil Samples from each layAer for Pollen analysis in progress BA D OF I ITY RS E IV N U 163 5.4.2 Palynological Analysis Methods One gram of each sample was subjected to standard pollen analysis procedures (Faegri and Iversen 1989). After acetolysis, each sample was mounted on a pair of slides. Microscopic study of the slides was done with an Olympus CH 30 microscope with an attached camera; identification and counting of palynomorphs were done under both x20 and x40 objectives. Identification of palynomorphs was done by comparing them with the reference slides collection at the Palynology Laboratory, Department of Archaeology and Anthropology, University of Ibadan. This referencYe collection contains about 3600 pollen slides of extant species. PhotomicrographRs in albums at this Laboratory as well as those in relevant literature were alsAo used. Identification was facilitated because the grains were well preserved. PhoRtographs of the most important palynomorphs encountered were taken at a magnIifBication of 1000. All palynomorphs encountered were used in calculating the pol leLn sum. The various taxa were classified into phytoecological groups based Aon Ntheir present-day natural distributions (Hutchinson and Dalziel 1954, 1958, 19D63, 1968, and 1972; Keay 1959). The pollen diagrams were prepared by Dr. Monique Tossou of Abomey Calavy University, Benin Republic. She used the TILBIAA and TILIA software (Grimm 1991) in preparing the pollen diagram. F I 5.4.3 Palynological Analysis R esOult All palynomorphs TencYountered during counting were recorded. Twenty-eight forms were recognized. OIf this number, 25 were pollen grains while the other three were spores. TweRnty Spollen types were identified, 14 of them to species level, one to genus level anEd five to family level. The remaining five could not be identified. Two out of thIeV three spores were identified to species level; the other could not be identiNfied (Plate 5.33). Three pollen zones were delimited (I, II and III) ZUone I (98-34cm): Zone I was sub-divided into zones Ia (98-76cm) and Ib (76- 34cm). (a) At the beginning of this sub-zone the predominant species is the ornamental, Lagerstroemia indica, constituting 71.6% of the pollen sum. There was only a trace occurrence of another ornamental, Casuarina equisetifolia. Elaeis guineensis is 5.0%, Alchornea spp. constitutes 4.2% while Poaceae is 10.0% and the dry forest species, Morus cf. mesozygia is 2.5%. By the end of this 164 sub-zone there was a total absence of L. indica and a minor, relative increase in C. equisetifolia concomitantly with marked increases in Elaeis guineensis (20.0%) Poaceae (35.0%) and weeds (1.6% to 10.0%); ferns and savanna species appeared and increased to 10.0% and 5.0%, respectively. There was a slight increase in primary forest species (5.0%). (b) Forest species and ornamentals begin a gradual increase until each reaches a peak at 14.3% and 18.3% respectively. A slight increase occurred in Alchornea (4.7% to 8.0%). There were marked reductions in savanna specieYs (4.7% to 0.0%) and Elaeis guineensis (19.0% to 4.0%), but less decreasRes in ferns (9.4% to 8.0%) and Poaceae (33.3% to 24.0%). Towards theR enAd of this zone, there is a significant reduction in Poaceae (24.0% to 5.5%), a slight decrease in Alchornea (8.0% to 5.6%), and a completeI Babsence of E. guineensis. Ornamentals (Delonix regia, Lagerstroemia inLdica and Casuarina equisetifolia) register a slight reduction to 16.6%. SNavanna species reappear and increase to 5.9%. Ferns (Pteris mDohasAiensis and an unidentified monolete) decrease (8% to 5.5%), while weeds reappear and rise to 5.5%. Most importantly, forest elements (CeIltBis c Af brownii, Pavetta owariensis and Syzygium guineensis) increase to 38.9%. Zone II (34-20cm): The mOajoFr components of this zone are ferns, Poaceae, and two secondary forest element s --Alchornea sp., Elaeis guineensis. The latter two increase from the previous T5.6Y % for Alchornea sp. and 0% for E. guineensis to 18.2% and 9.1% respectively. PoIaceae register a marked increase (5.6% to 36.4%) while fern spores which are RhithSerto absent reappear and increase to 18.2%. The forest species show a markedE decrease from 38.9% to 9.1%. Ornamentals also experience reduction from 16.6I%V to 9.1%; while savanna species (Grewia mollis) and weeds disappear compNletely from this zone. ZUone III (20-11cm): Ferns (Pteris mohasiensis and Polypodium vulgare), Poaceae and Elaeis dominate this zone. Ferns reach the peak of their occurrence (33.8%), Poaceae is 29.4% and E. guineensis increases to 14.7% from 9.1%. Cyperaceae, which occurred only at the beginning of the column, reappears and registers 2.9%. Primary forest species continue to decrease until they reach 5.9%. Towards the end of this zone Alchornea decrease from 18.2% to 7.4%. Weeds are present again (1.5%). 165 Y AR LIB R AN BA D F I O SI TY R IV E UN 16 17 18 Plate 5.32: Photomicrographs of palynomorphs (all magnification x1000) 1-3: Pavetta owariensis;4: Amaranthaceae/Chenopodiaceae;5-6: Alchornea sp.7: Celtis cf. brownie;8: Unidentified;9: Hymenocardia acida;10: Chromolaena odorata;11-12: Alternanthera repens;13: Casuarina equisetifolia;14-15: Lagerstroemia indica; 16-17: Grewia mollis;18: Poaceae 166 Y RA R LI B N DA IB A OF SI TY R IV E UN Plate 5.33: Photomicrographs of palynomorphs (all magnification x 1000) 1-2: Pterocarpus santalinoides; 3: Cyperaceae; 4: Elaeis guineensis; 5: Delonix regia; 6: Asteraceae; 7: Pteris mohasiensis; 8: Aspillia Africana; 9: Lagerstroemia indica; 10-11: Combretaceae/Melastomataceae 12: Syzygium guineensis; 13:Unidentifed; 14:Monolete spore; 15:Polypodium vulgare; 16:Morus cf mesozygia 167 Table: 5.8: Phytoecological groupings: 1 Savanna Hymenocardia acida, Grewia mollis, 2 Elaeis guineensis Elaeis guineensis 3 Alchornea sp. Alchornea sp. 4 Forest Celtis cf. brownii, Pavetta owariensis, Syzygium guineense, Pterocarpus santalinoides and Morus cf. mesozygia. 5 Poaceae Poaceae 6 Weeds Aspillia africana, Chromoleana odorata, Althernanthera repens, Asteraceae, 7 Amaranth/Chenopod Amaranthaceae/Chenopodiaceae Y 8 Ornamentals Lagerstroemia indica, Delonix regia, R Casuarina equisetifolia A 9 Ferns Pteris mohasiensis, Polypodium vulgare, Monolete fern R 10 Fresh water swamp Cyperaceae IB 11 Combret/Melast. Combretum racemosum type. N L AD A F I B Y O T RS I VE NI U 168 5.4.4 Palynological Analysis Discussion The pollen analysis from zone Ia (98cm-76cm) shows that by a short period after cal. AD 1263, the landscape was dominated by the ornamental Lagerstroemia indica with a trace occurrence of another ornamental, Casuarina equisetifolia. Along with them only the pollen of one other arboreal species, characteristic of dry forests, Morus cf mesozygia, was recovered. There are comparatively low occurrences of Elaeis guineensis, Alchornea sp., grasses and weeds. The abundance of L. indica (71.6%) of the entire pollen sum at 98cm at this time shows that the trees were veryY close to the excavated site. Their pollen is most probably insect-dispersed, judRging from the large, showy, purple flowers and the presence of wart-like pArocesses (verrucae) on the outermost wall of the pollen grains. According to HutRchinson and Dalziel (1968), L. indica, is said to be a native of Asia, (probably LIndIiBa as the specific name suggests) and an ornamental commonly cultivated thro ughout West Africa today. Was the beauty of this tree when in flower so muAch Nappreciated by the people of Ajaba that they preserved it around their habitatDion? It is interesting to note that Tutuncu et al. (2007) in their study of exotic plants in Edirne, Turkey, observed that L. indica was an exotic plant usually found IinB pa Arks and gardens there. This confirms the ornamental value of L. indica. The fFoss il record of Lagerstroemia goes back to the Tertiary. O Fossil leaves of LageYrst roemia have been described from India and the Oligocene of Japan (PigIgT and DeVore, 2005). Fossil fruits recovered from the Pleistocene of Japan aSnd found to be similar to extant L. indica of central China, have been assigned to Rthat Chinese species (Pigg and DeVore, 2005). Fossil pollen of LagerstroeVmiaE was reported from Pleistocene deposits of Japan, but it was not stated if it was Ithat of L. indica or L. speciosa (Miyoshi et al., 1999; Fujiki et al., 2001 in Pigg Nand DeVore, 2005). The pollen of L. indica is medium in size (P: 33.5; E: 3U5.00µm), spheroidal and tricolporate; ora conspicuous and clear in polar view; exine is thick, and pattern is distinctly verrucate. (Figure 5.52, 14-15). It is opined that the fossil pollen is L. indica. Modern pollen of L. speciosa, though similar in size (P: 33.75; E: 35µm), have a thinner and comparatively less distinctly verrucate exine pattern. The modern pollen of L. tomentosa (P: 35; E: 3 7.5µm) are somewhat prolate- like in equatorial view. Since fossil records of Lagerstroemia have come from China and India, it seems logical to assume that the species originated from the Indo- 169 Chinese area. If, as stated above, L. indica is of Indo-Chinese origin, how then did it reach Nigeria or West Africa, particularly Ajaba? This and related issues will be discussed later. One pollen grain of another ornamental, Delonix regia, (commonly called The Flamboyant tree or Flame of the forest on account of its bright, orange-red flowers) was recovered at 76cm below the surface. Delonix regia, though widely planted in Nigeria today ―is a native of Madagascar‖ (Hutchinson and Dalziel, 1958, 481). The flowers are insect-pollinated and the pollen grains are fairly large (P: 58.3; E: 39µmY) and very distinctive. Finding this one pollen grain is noteworthy. It probably stucRk to the pod of the tree, which was brought home by the inhabitants of Ajaba. TAhe dried pods of Delonix regia are used in making fire; furthermore these dried podRs are called sheke-sheke in Yoruba. Sheke-sheke is a name given to objects whiIchB, when shaken, make rattling sounds, thus they are used as a play object by child reLn. There is a dramatic change in the vegetation bAy tNhe end of the sub-zone. Lagerstroemia indica is no longer represented, wDhile there are marked relative increases in Elaeis guineensis (5.0% to 20.0%), grasses (10.0% to 35.0%), weeds (1.6% to 10.0%); ferns and savanna specieIsB ap Apeared and increased to 10.0% and 5.0%, respectively. There was a slight r elative increase in primary forest species (2.5% to 5.0%).The sub-zone eOndedF with E. guineensis being the dominant component. This development ind icates that E. guineensis in effect replaced L. indica but not the other forest spYecies. The presence of an increase in snail shells, presumably those of the IgTiant snail (Archachatina marginata) (increase from one in 100-110cm to 19 in 7S0-80cm) suggests the presence of a closed forest. The pottery recovered at thEis Rtime shows a significant increase, from 114 at 90-100cm to 277 at 80-90cm IaVnd 222 at 70-80cm. If increase in pottery is regarded as an index of increaNsed population, then it can be inferred that there was a gradual growth in the pUopulation of the area during this period. The reason(s) for the total absence of L. indica is (are) not evident. In Zone 1b (76 cm-34 cm), the overall development signaled a reinforcement of the trend at the end of the preceding phase with regard to expansion of primary forest vegetation and a reversal in the development of secondary forest. Of particular note are increases in ornamentals and forest species; the latter reached the peak of their occurrence in the entire column by the end of this sub-zone. Alongside these 170 increases was a significant reduction in grasses while Elaeis guineensis decreased markedly and finally disappeared from the record. The impact of humans on the vegetation shows up very clearly in Zones II (34-20 cm) and III (20-11cm). The primary forest species decreased significantly while Elaeis guineensis and Alchornea spp. as well as Poaceae increased. Towards the end of these zones, ornamentals continued to decrease and subsequently disappeared completely while two other forest species, Morus cf. mesozygia and Pavetta owariensis, also disappeared. These changes are indicative of the existence of a morYe open forest. It is striking to note that fern spores become quite abundant reacRhing 18.2%. Increase in fern spores is known to be indicative of a humid enviAronment (Morley, 1995). Such an environment would have been very conducivRe for forest growth but the contrast was recorded. The appearance of PterocarpuIs Bsantalinoides, a riverine forest species, and the re-appearance of Cyperaceae, foLund in fresh water swamps further indicate that the climate was not dry. ConseqNuently, the destruction of forest vegetation most probably was due to human actionA. The palynological indications are reinfAorcDed by archaeological evidence. Archaeological materials recovered were Bmost abundant in these two zones, especially levels 20-30cm (Table 5.1). TFhe reI was also an increase in palm kernel from the previous 12 in 50-60cm to 40O in 20-40cm. It is significant to note that though charcoal occurred in all the levels , it was during this phase that the only burnt kernel was recovered. There was TalsoY an increased presence of hearths - an indication of the use of fire for cooking or Iother domestic activities. The occurrence of burnt clay is yet another evidence Rof thSe use of fire. These developments are regarded as an indication of increased hEuman activity, including the use of fire. The presence of a furnace fragment IfrVom level 20-30cm is another indication of the use of fire and possibly iron-workiNng. If this were the case, trees would have been good sources of fuel. The felling oUf different types of trees for fuel probably accounts for the reduction in the pollen of forest trees. From the graph in which the abundance of forest trees is compared with that of pottery (Fig. 5.9), it can be seen that beginning from 100-90cm to 80-70cm, an increase in pottery (index of population) coincides with a reduction in forest trees. Subsequently, increases in forest trees and pottery run parallel, suggestive of equilibrium between the people and forest vegetation. However, at level 30-20cm there was a significant reduction of forest trees coincident with a phenomenal rise in 171 pottery. This reduction in forest trees is thought to be the resultant effect of an increased human population having a greater impact on the forest vegetation. This is based on the assumption that a high pottery value is an index for high human population. Thus both palynological and archaeological evidence combine in indicating a significant impact of humans on their environment and the intensified use of fire during this period. It can also be inferred that there probably was a change in diet, with more use made of the oil palm tree than hitherto. As said earlier, the people oYf this area are well noted for relishing palm wine tapped from the oil palm tree! R Palynological research in the Department of Archaeology and AnthroApology, University of Ibadan, has shown that in southern Nigeria, humans creBateRd conditions which enhanced the proliferation of the oil palm, an importanLt cIomponent of the subsistence economy during the second phase of the Late Ston e Age in this region. (Sowunmi 1981b, 1985; Oyelaran 1998). The human acAtioNn entailed the creation of gaps in the forest for the cultivation of crops other tDhan the oil palm itself, using the slash-and-burn farming technique. Though the oil palm itself was not likely to have been cultivated, it most probably was selecBtiveAI ly protected. Sowunmi (1999: 205) went further to postulate that the `fossilF po llen of Elaeis guineensis constitutes a very good indicator of anthropogenicO forest disturbance or ‗an index of "palaeo-cultivation" in west and west cent ral Africa…‘ In addition, Zeven (T19Y64: 123) noted that in modern sediments the occurrence of oil palm pollen, whichI can be up to 30% of the total, "reflects the cultivation of the oil palm as well asR meSthods of land use favourable to its spread." Indeed in the surface sample from the Niger delta the oil palm pollen was 38.7% of the pollen sum (SowunmIi V198 E1b, 1985). 5U.4.5 NOrnamentals in palaeo-ecological studies Ornamentals are plants considered to be of aesthetic value to humans. As a result, humans deliberately cultivate, nurture and protect them in areas close to the home or the immediate surroundings of a home, in order to enjoy their beauty. Thus they are closely linked with human presence and sedentism. Thus, the correct identification of their pollen and spores in fossil samples or in an archaeological context such as the present study will further strengthen the establishment of a direct link between humans and the study site. Though some ornamentals are exotic to Nigeria, many are 172 Table: 5.9: Figures of Tree Resources and Pottery Depth(cm) Tree resources Pottery 100-90 25 114 80-70 47 222 . 60-50 143 293 40-30 387 423 30-20 91 781 20-10 59 429 Y R BR A LI AN BA D F I O ITY S ER IV N U 173 Fig. 5.4: A comparison of abundance of forest trees (forest resources) with pottery abundance (index of human population) from level 100-10cm 900 800 700 600 500 Series1 400 Series2 300 AR Y 200 100 BR0 100-90cm 80-70cm 60-50cm 40-30cm 30-20cm 2 0L-10cIm Percentage values of forest trees (range of 2.5%-38.7%) have been multiplied by 10 for ease of comparison with pottery figures which are in the Nrange of 114-781. Series 1= % forest trees, Series 2=Pottery. Y axis= pottery aDbunAdance and % forest trees; X axis= depths. A F I B O SI TY VE R NI U 174 indigenous. The recovery of micro- and macro-fossils of exotic plants in certain areas has proved very useful in interpreting and resolving some palaeoenvironmental difficulties. For example, the recovery and identification of the pollen grains of Zea mays have helped in establishing the beginning of food production (agriculture) in certain areas of South America (Piperno et al. 1991). The correct identification of Casuarina pollen provided useful information for understanding, from a palynological point of view, the introduction of exotic species and the antiquity of humans in South East Asia (Haberle, 1994). Y In the present study, three woody, exotic, ornamental specAies, Ri.e. Lagerstroemia indica (Lythraceae) [native of Asia], Casuarina equisetifolia (Casuarinaceae) [native of Australia], and Delonix regia, (CaesalpiniaBceaRe) [native of Madagascar], were recovered from different levels of the excavIated trench. The recovery of the pollen of these plants becomes the earliest an dL first palynological evidence of ornamentals in Nigeria. Two of these plants N– C. equisetifolia and D. regia -- are trees while Lagerstroemia indica is a shrub. AOf significant note is the fact that these plants are all found in the forest regAion Dof Southern Nigeria today. The pollen diagram (Plate 5:44) shows that L. inIdBica and C. equisetifolia were present at the deepest level of the pit (98 cm) andF th ey both constituted a very high percentage (72.4%) of the pollen sum at that lOevel. L. indica, in particular (71.6%) was not only present but abundant within the c atchment area of the excavated site. In view of its aesthetic attraction, this abuTndYance is suggestive of some human intervention. This suggestion is further supIported by the occurrence of Elaeis guineensis and weeds (Chromolaena odoRratSa, Aspilia africana and Asteraceae) in appreciable quantities. These weeds aEre known to be associated with human habitation (Sowunmi, 1981c). QualitativIeV vegetation analysis carried out by (Oriejieme, per. Comm, 2011) in the forestN areas of southwestern Nigeria confirms Sowunmi‘s (1981b) observation. The foUllowing weeds are some of the commonest found around homes and in disturbed areas in this part of Nigeria: Chromolaena odorata, Aspilia africana, Sida acuta, Ageratum conyzoides, and Tridax procumbens. Several questions arise from the presence and the pattern of occurrence of the three exotic ornamentals. If they were cultivated, by who were they cultivated – the local people or foreigners (probably colonialists)? The inferred date of the levels where Lagerstroemia indica was most abundant is sometime soon after AD 1263. This is very much earlier than the known arrival of Europeans in West Africa, i.e. 175 th th between the 15 and 16 centuries. This seems to rule out European cultivators. The specific name of Lagerstroemia, i.e. indica, indicates that it is Indian in origin. Furthermore some other botanical names often reflect the view that cultivated plants found in the African continent were transferred from India to Africa, for example, Tamarindus indica and Sesamum indicum (Blench, 2003), as well as Mangifera indica (the mango tree). However, as far as we know there are no records of the th arrival of Asians in Nigeria prior to the 20 century. Is there any other possible way by which L. indica could have reached this part of Nigeria at such an early dateY? Could they have been introduced through the influence of ―phantom voyagers‖ Rfrom Indonesia, whose coming to sub-Saharan Africa (including Nigeria) and MRadAagascar, according to Dick-Read (2005), much predated that of Europeans? The scholarly and comprehensive research work by Dick-Read (2005) only came toI Bthe attention of Professor Mrs. Sowunmi in a private discussion she had with DLr. Patrick Darling, recently at a conference in Britain. If the identification ofA L. Nindica pollen is correct, it would seem that the Ajaba people recognized its ornamental value and therefore planted and protected it for quite some time. TheA saDme explanation can be given for D. regia which also has very attractive floweIrBs and which could have been brought by the ―phantom voyagers‖. Another possFible source through which could have reached tis part of Africa ight be through migratory birds. Birds such as ‗pathtom voyagers‘ are known to have migrated from Ocentral Europ in early roosting in several parts of Nigeria including SoutweTsterYn Nigeria before movng on in other directions. The occurrence of C. equisetIifolia pollen in the fossil record at Ajaba is at the moment difficult to explain ThoughE itR . S may not be considered likely, it does not ruled out entirely the possibilitIy Vthat these three pollen types might belong to indigenous, forest species whichN is yet to be encountered in this palynological study. Their pattern of occurrence isU similar to that of the indigenous forest species. As exemplified in the Combretaceae and Melatomaceae on the one hand and Chenopodiaceae and Amaranthaceae on the other, it is well known that certain species which belong to different families do have similar, virtually indistinguishable pollen grains. 5.5 Other Archaeological Features An Earthwork, in the form of a wall was observed at Ajaba. This wall stretches from the back of the Oba‘s house (Plate 5.34) to the southwest of the 176 excavated mound. The height, length and width of this wall were determined (About 120 metres in length, 1.8m high and about 2.9m wide). The wall in terms of size and complexity is not comparable to defensive walls found in other parts of Yorubaland, such as Oyo Ile, Orile Owu, Orile Ijaye and Ila-Yara to mention but a few. The wall is small in size (i.e. height, width and length) and not enclosed but concentric in nature. It is relatively linear in shape. Unfortunately, the extent of the wall was to be mapped with the aid of a Global Positioning System (GPS) receiver purposefully to represent it on the map of the area as one of the archaeological features at the site but becausYe the wall is under a thick foliage cover, the GPS receiver did not locate sateRllites therefore, it could not be mapped with GPS. The wall was not associated withA ditches as is common in other places in Yorubaland where such defensive wallRs have been studied. Alternatively, it is possible that the construction of the wallI mBight have been truncated as a result of an impending external aggression, hence thLe absence of visible ditches. N During the reconneiting of this wall, a complete hAalf buried pot was found at a section on the wall (Plate 5.35). According to the ObDa, the pot is a ritual pot that was recently used to offer sacrifice to the godsI oBn b Aehalf of a barren woman who later gave birth to a child. The morphology aFnd decoration motif of this pot was used as an ethnographic model for identifying Oritual pots among the excavated potsherds. SI TY VE R I UN 177 AR Y R LI B N Plate 5.34: Section of Ajaba ditch and embarkmeDnt oAf Ajaba Wall IB A F O ITY RS IV E UN 178 RY BR A N LI A Plate 5.36: Buried Pot at a Section of Ajaba Wall D BA OF I Y IT RS IV E UN 179 CHAPTER SIX DISCUSSION AND CONCLUSION 6.1. Discussion The geographical area known today as Yorubaland has serially been involved in series of conflicts, wars and displacements. This had at several times led to series of movements, realignments and dislocations. In this area, as in other parts of the world, as human beings became more sophisticated in their efforts to adapt to thYe environment they found themselves, they would have made use of resources provRided by nature around them to manipulate the environment to meet their variouAs needs. This would have led to socio-cultural developments and increaseB inR population movements. Invariably, these developments would have also bLroIught about some political and economic changes and sometimes instability. Som e groups would start dominating others. In some cases, new immigrants wouldN dominate the aboriginal peoples and where such aborigines could not cope wDith Athe new elements they would have migrated, accepted or coped with the subjugAation. These actions would have led to displacements and changes in the settlemenBt pattern. As clearly stated by Ogundiran,F (20 0I2: 12) ―soil, water, elevation and climate are potential and real natural resourOces that affect but do not necessarily predetermine, the patterns of settlement, the pur suit of subsistence, and making of human history‖. The early points of settlemTentY for the people in the area are found in the plains or at the gentle, wide slopes oIf the valleys and very close to streams that support their development. ButR at Sa later stage, when the threat of war and the actual war was predominant iEn the region, most of the towns and villages like Igbajo, Iree, Otan-AyegbajuI Vand Ila-Orangun were relocated to the hilly areas. Here they were safer from Nexternal invasion. With the incursion of the Fulani Jihadist in the early decades thoUf the 19 century Yoruba wars, relocation to hilly terrains became very rampant in the area. Accounts of oral tradition from most of the towns in the study area indicate that people of the area migrated either from Ile-Ife or Old-Oyo and that they settled at different places before their final place of settlement. At different point of settlement, they have to relocate either as a result of conflict, chieftancy dispute, or show of magical prowess as revealed by the oral history of thee Ede people or in a bid to find a new area to settle. This migration has the ‗pull and push‘ factors. For instance, the 180 people of Ede were dislodged from their settlement and were accepted in other towns within the vicinity of their former settlemment. Today, they are found in such towns as Igbajo, Ila-Orangun, Otan-Ayegbaju, Imesi-Ile, Iresi among others. It is in the light of the above observations that archaeological and anthropological investigations carried out in areas of Northeast Osun is viewed as veritable insights into the complex consequences arising from conflicts, wars and displacements. In the context of settlement dynamics, the consequences have far reaching implications. They are visibly manifested in settlement demography andY settlement pattern of the area. They are also manifested in the interrelatedness oRf the cultural entities of both the ‗aboriginal‘ peoples and the immigrants. They Aare also temporally and spatially manifest in the abandonment and resettlementB of Rmany towns and villages in the area. I Archaeological excavations at Ajaba and its environs in OLsun northeast have revealed some salient aspects of the culture and general lAife-Nways of the people. These excavations yielded potsherds that have illuminDated aspects of socio-political formation and intra-and inter-group relations beAtween the ancient Ajaba inhabitants and their neighbours, through pottery seriatioBn. Over the years, Old Oyo pottery has been used as a yardstick for measurin g Iother Yoruba groups‘ and subgroups‘ interactions. Although there is debate oFver the use of Old Oyo pottery as model for pottery classification in Yorubala nOd, yet Old Oyo and Ife still hold a strong place in the socio-political formation oYf many Yoruba settlements. The chronological attribute of Old Oyo pottery was vIeTry useful in seriating Igbomina ceramics of which Ajaba is a component partR (UsSman 2003). Although Ajaba oral tradition claims affinity with Oyo Yoruba, iEts material culture constituents is more similar to those of the Igbomina than the IOVyo. It should, however, be noted that in certain regards the Igbomina materNial repertoire bears some semblance with those of Oyo. In fact oral traditions of mUost of the Igbomina communities claimed migrations from either Oyo or Ife (Aleru 2006). The association of Ajaba with Oyo might therefore not be totally misplaced. The interaction and migrations between some of the Yoruba groups can be determined through pottery studies. Old Oyo pottery decoration motifs which have been extensively studied have been used as yardstick for studying pottery from other Yoruba groups and sub-groups. In view of this, the results of the pottery seriation of the Igbomina sites seem to follow a sequence of analyses suggested by scholars such as Agbaje-Williams 1983, 1991; Soper 1975, 1983 and Willet 1960, 1962). Old Oyo 181 pottery tradition is classified into two phases. These are the ‗Diogun‘ and the ‗Mejiro‘ periods (Willet 1961). Soper‘s excavations at Old Oyo from 1973 to 1979 further divided ‗Diogun‘ and ‗Mejiro‘ wares into early and late ceramic types based on the decorations (Soper 1975; Usman 2003). The ‗Diogun‘ pottery which is the earlier (dating to AD 1100 + 110) is characterized by brush or broom making incisions, rocked-comb impressions, impressed arcs (scallops), knotted roulette and frond roulette (Agbaje-Williams 1983). On the other hand, ‗Mejiro‘ pottery which characterized the later phases (dating to AD 1300 + 80), is characterized by carvedY roulette, snail shell markings, and maize cob roulette (Usman 2003). R Following the pottery seriation in Yorubaland which is greatly attriAbuted to Old Oyo and Ife, one will expect that any settlement or regions that claimeRd migration from Ife and Old Oyo direction should, to a considerable extent, reIflBect Ife/Old Oyo pottery patterns. From the result of the excavated pottery from LAjaba, the supposed Old Oyo tradition is not predomiinant as expected. Thus, iNncision, carved wood and scallops have percentage of 1.57% of the total percentagAe of pottery decoration types on body sherds. This assertion however, does noAt mDean that the people did not at any time interact with the people from Old OIByo. They might have interacted at a particular time, through diverse means F(e. g. migration, war, marriage or trade) but it does not adequately reflect in theirO pottery patterns or decorations. The site has been th thradio carbon dated to between 13 and 17 century (1263 A.D. and 1609 A.D.). Ajaba pottery has vTariYeties of decoration motifs, which could make inference from pottery decorationsI in terms of relatively dating the site a difficult task. The pottery has combiRnatioSn of Old Oyo, Ife and Nupe attributes. Thus, clear affinity with a particular YEoruba group or subgroup cannot, for now, be determined. Maize cob roulette iIs Vsaid to be the most commonly used rouletting tool among Yoruba potters (FatuNnsin 1992), yet it is totally absent at Ajaba. The Old-Oyo/Ajaba relationship is hUighly contestable not only from the archaeological point of view but also from the record of oral traditions. According to information gathered from oral tradition, Ajaba and its environs were under the political hegemony of Ilesa, during the reign of Obokun as the ruling monarch in Ilesa. The Obokun for a very long time did not have contact or relationship with Old Oyo. Hence, this could have accounted for the absence or low representation of Old Oyo pottery tradition at Ajaba. There was an attempt to ascertain the relationships between the Ajaba and other Yoruba groups using pottery affinity as a criterion. This is because interactions 182 and migrations among the Yoruba groups may be ascertained through a study of their respective pottery. The Old Oyo pottery is used as a reference in considering pottery from other Yoruba groups or sub-groups, although this usage is still controversial. Willet (1961) classified the Old Oyo pottery into two phases, namely the ‗Diogun‘ and ‗Mejiro‘ phases. Based on his work at Old Oyo, Soper (1975) considered ‗Diogun‘ and ‗Mejiro‘ as representing the early and late phases of ceramic types, respectively, based on their decorations. While ‗Diogun‘ pottery (dating to AD 1100 ± 110yrs) is characterised by brush or broom-made incisions, impressed arcs (scallops)Y, knotted and frond roulette, ‗Mejiro‘ (dating to AD 1300 ± 80yrs) is characteriseRd by carved roulette, snail shell markings and maize cob roulette (Usman, 200A3). Any settlement which had close interactions with Old Oyo is expectBed Rto have an appreciable number of these decorative motifs in its pottery. ALlthoIugh most of the Igbomina communities claim to have migrated either from Old Oyo or Ile-Ife (Aleru, 2006), the oral tradition about Ajaba makes a very diffeArenNt claim. According to the latter tradition, Ajaba wish is assumed to be the fomDer settlement of the Ede people and its environs were under the political hegemony of Ilesa; and the king of Ilesa, Owa Obokun, did not have contact with OlBd OAyo for a very long time. This oral tradition seems not to be supported by evi dIence from material culture since some of the pottery sherds retrieved from this exFcavation bore resemblance to that of Old Oyo, Nupe and Ile-Ife. O From the morphologicYal analyses on the rim of the pottery sherds, there is an indication that some of ItThe pots or pottery materials represented in the pottery assemblage vary RfromS small bowls, small and large cooking bowls (Isaasun), Amu (water pot), IkEoko (storage pot) and Ajere (perforated pot) among others. The ―Ajere‘ could havIeV been used for roasting or smoking meat or other food items (the meat or other Nfood items like yam or plantain is placed inside the ‗Ajere’ pot and placed on nUon-flaming fire) or for ritual purposes. Also pottery with bossing or applied decoration type may suggest trade mark or is suggestive of ritual/sacrifices container, although, the number of pottery with this decoration type is very few. The low distribution of pottery with bossing decoration type could be because of its probable usage for ritual purposes. This is because such pottery is not commonly used except for special ceremonies or to offer sacrifices, which is not an everyday event, so they are most likely to be made on request. In other words, it could be that bossing or applied decorations are meant for trade mark identification of individual potter. For 183 instance, Plate 5.9 shows two cross marks on the potsherds while Plate 5.16 shows an applied decoration. The pottery assemblage varied from small bowls through medium-sized and large pots to perforated pots. These pots might have been used for cooking, storage of water and food items, or for rituals. The iron slag, hearths, tuyere, some fragment of tuyeres and furnance and furnace fragments are suggestive of iron working and the use of fire. The people probably used iron implements presumably for felling trees and for farming. However, there are no traditional or an ethnographic account to support an ironY- working tradition in the area though, not far from this area is a village knowRn as ‘agbede’ (iron smith). The village is a recent one. A more extensive arcRhaeAological survey is required in order to ascertain more definitely whether there indeed was iron- working in the area. IB Additionally, the iron debris is very fragmentary. The na tuLre, scope and extent of iron working in the area are therefore still shrouded iAn obNscurity. Or is it possible that the objects were imported into the area? It is poDssible that the inhabitants might have engaged in iron-working at a distant past and place, probably somewhere close to the source of raw materials. BA Organic materials such as animFal b oInes, prominent among which are the jaw and femur (Plate 6.1a & b) indicOate presence of large and small animals. Other organic materials such as snail sh ells, bivalve, plant seed, palm kernel (charred and non-charred) could give aTn inYsight into subsistence means of the inhabitants. They could also provide informIation for vegetation/environment reconstruction of the study area. S The wEall eRarlier mentioned is an important archaeological feature at Ajaba. ConsideriInVg the size of the wall, the purpose or function is still difficult to ascertain. It couldN have functioned as farm or territorial demarcation. Defensive walls in most caUses are enclosed or concentric in nature with deep ditch and considerable high bank. Ditches are regarded as a mechanism to compliment the defence nature of banks. Consequently, because of the low level or total absence of ditches around Ajaba, it is not certain that the wall was used for defensive purposes. Though, it could be that the construction of the wall was abandoned in the wake of conflict or war that ravaged the th area in the 17 century. Moving away from the traditional methods of pottery analysis, this work took a step further by carrying out geo-chemical and mineralogical analysis of pottery 184 AR Y IB R Plate 6.1a: Femur of an animal (a large mammal) L DA N IB A O F ITYS ER NI V U Plate 6.1b: Jaw of an animal 185 sherds from six sites, including sherds from Ajaba. The mineralogical and chemical composition of the sherds recovered from the six archaeological sites, Ife, Ipetumodu, Ajaba, Ila-Orangun and Iragbiji, Iresi appear to reflect the geological setting of the area. The mineralogy of the ceramics from all sites is grossly similar both temporally and spatially and contains quartz, micas, and trace amounts of feldspars, amphiboles and iron oxide. The mineralogical composition indicates low temperature firing of the clay. Clays, which were used to make the sherds, are typically referred to as hydrous o silicates (Deer et al., 1992). When heated above 500 C, the hydroxyl component iYs driven off. This results in mica-like structures. If the clays are heateRd to o approximately 900 C or higher, the clays become vitreous and form porcelain.A Thus o o the sherds in this study were probably fired above 500 C but below 900 CR. In the areas studied, extensive clay deposits have develoIpBed by tropical weathering of rocks of predominantly granitic composition. OLther rock types are minor quartz rich schists and amphibolitic rocks. Clay deposNits are abundant and there are more than 12 well-known deposits which range inA size from a few meters in length to extensive deposits of a kilometre or twAo loDng. Weathered and unweathered granitic and granodioritic rocks occur quiteI nBear almost all the archaeological sites. The raw materials for making the ceraFmic s came certainly from neighbouring areas close to sites and include clay mateOrial from a saprolite or mottled zone derived from rhyolite/granite. It is assumed that ancient people of Ajaba collected the raw materials around the area, close to thTe sYites. The firing of the potteries in open atmosphere did onot exceed 600 C, as demIonstrated by partial dehydroxylation of clay material seen in the occurrence of RmicSa inclusions. In general, the contents of trace elements found in the ceramic frEagments of Ife, Ipetumodu, Ikire, Igbajo, Iresi and Ajaba, can be well correlatedI Vto saprolites derived either from granites/rhyolites or to granoNdiorites/dacites. The contents of the rare earth elements (REE) fall in the domain oUf the granitic rocks and its weathered products. The trace element data suggest that the main raw materials used for the fabrication of the ceramic artifacts have a composition close to saprolite derived from acid rocks (granite/rhyolite). The rare earth element data are consistent with derivation from a granitic protolith. The chemical compositions of the raw materials probably used for pottery production and the composition of the ancient ceramics can be successfully fingerprinted or classified by basic, geochemical techniques. It should be noted that the similarity of geology could make provenance very problematic. Also it should be 186 noted that some other ethnic groups especially the Nupe people could have been involved in the construction of potsherd pavement in the study area. For instance, the pavement at Oyan is referred to as ‗Aganju Tapa‘ [Tapa‘s (Nupe people are referred to as ‗Tapa‘in Yorubaland) courtyard]. Though, as by stated by Akintoye (1971) ‗among the Igbomina (of which part of the studyarea belongs) there are scattered pocket of Nupe and Fulani settlers who have now been absorbed into local society‘ But further work is needed to ascertain this hypothesis, as there is no identification of such pavements in Nupe land. Y The result of the pollen analysis shows that not long after about AD 1263R the exotic ornamentals, Lagerstreomia indica, dominated the landscape of AjaAba. The abundance of L.indica along with traces of two other exotics Casuarina Requisetifolia and Delonix regia, of which there was only a trace occurrence, suggeIsBts some form of horticulture. The dates of the levels at which they were found aLre earlier than those known for the arrival of foreigners, specifically Europeans, iNn Nigeria, but there might have been a hitherto unknown foreign influence in Dthe Aoccurrence of these species. This might be the first botanical evidence of ornAamentals from an archaeological site in Nigeria. But the possibility of the pollen Bbelonging to indigenous forest species must not be ruled out entirely, especiallFy a s Ithe pattern of occurrence is similar to that of L.indica in particular. FollowinOg its initial abundant occurrence, L. indica soon decreased drastically, then incre ased to lower levels and disappeared completely sometime after AD 1491, fTor Yreasons which are as yet unclear. Similarly, indigenous forest species decreased tIoo then increased, reaching a peak level. Subsequently they decreased appreciRablyS, being mostly restricted to fringes of rivers. This later primary forest decreaseE, concomitantly with notable increases in early colonisers of disturbed forests, eIspVecially Elaeis guineensis, along with weeds and grasses is attributable to increaNsed human activity, including the probable use of fires, presumably preparatory toU slash and burn farming. This period coincided with the climax of human occupation, as evidenced by increased abundance of artifacts. The use of fire is indicated by the occurrence of burnt palm kernel, burnt clay, furnace fragments and tuyeres. Thus both palynological and archaeological evidence combine in indicating a significant impact of the Ajaba people on their environment and the intensified use of fire from about 600 years ago. 187 6.2 Conclusion Oral history, written records and archaeological survey and result of data from excavations indicate some historical phases of occupation. The results of these data show that there was movement of people at different periods for different reasons. This led to settlement and resettlement of people, abandoning one site for another, changing the demographical setting of the area. In many cases, some settlements were totally deserted and abandoned and some settlements have their population increased thereby changing the socio-political and economic status of such settlementsY. Conflict, war and displacement led to movements of people from one place Ato oRther. These movements had effect on the demography, settlement pattern and environment which they abandoned and/or where they resettled. R The archaeological excavations at Ajaba and its environsL anIdB other places in Osun Northeast, Southwest Nigeria, provided information in regards to historical phases of occupation of the study area. As a resAult Nof conflicts, wars and displacements generally in Yorubaland, the period between the 13th and 15th centuries saw increase in population, presence of exDotic ornamental plants. Decrease in primary forest as a result of human popuIlBatio An was noted which resulted in more cultivation of their land resources. AFlso , destruction of forest, (as indicated by palynological analysis) probably as a result of forest slash and burning by man in his bid to prepare land for cultivation Owas also noted . The second phase was the period th th between 15 and 17 ceTntuYries, this period saw technological development as evidenced by potsherd paIvement and the abandonment of most of sites such as Ila- Yara, Ajaba abanRdonSed site and Oyan probably as a result of war as evidenced by defence wall aEnd ditches. It is not unlikely that the burnt materials indicated on the southern IpVortion of the excavated unit at Ajaba was as a result of house burning durinNg conflict (house then was built of thatched roof which is prone to fire disaster). th thTUhe third phase is the period between the 18 and 19 centuries when notable wars broke out in parts of Yorubaland and most especially in the study area. Among these were Jalumi War, Ikirun War, and Kiriji War in Igbajo/Imesi-Ile area (Ogunfolakan 2007). Notable war relics from this period, Defence Trench and Peace Treaty site (Plates 3. 5 and 6) are noted at Igbajo among others. Finally, from oral history, archaeological survey and excavations, the present location of most of the towns around the study area indicate relocation at different periods as a result of conflict brought about probably by slavery, royal conflict (most of the time when there is a 188 dispute or contest of royal stool, those that loose out usually leave the town and establish their own town) which usually lead to population movement and demographic changes. It is envisaged that further archaeological investigations in other areas around Ajaba would complement the data from this study. This is in order to have a regional overview of activities and inter/intra-regional relationship among the people that formerly occupied this place and their neighbours within and around Igbomina, and also to generate hypotheses about the socio-political formation of the people, particularly, in relation to other prominent Yoruba groups/subgroups. In OsunY northeast of Osun State, Nigeria, (Fig. 1.1) conflict, war and displacements impaRcted on the settlement pattern of the inhabitant. A Decorating the body add aesthetic value which represents coBsmoRlogical and religious concepts, and similar patterns of decoration on differentL poIt types expresses coherent underlying perceptions, accounting for continuities in an art form (David, Sterner and Gauva 1998). These underlying ‗perceptionsA‘ isN what today has been the major information archaeologists are scouting for tDo understand. While in Ife, clay was used to produce famous images and pots for ritual purposes, in other areas and most importantly, the study area, pottery beBcamAI e an important aspect of utilitarian materials in their homes. Because it haFs re mained one of the indestructible materials, archaeologists have been using pot Oto probe into those periods that were not known to the written world. Pottery is a very TgoYod chronological indicator; its quantification proves crucial, not only in compIaring different archaeological context but also as an aid in answering certainR arSchaeological and historical questions. It is also good in the investigation Eof relationships between context and fabrics. An important part of archaeological study of pottery is the comparison of ceramic assemblages in terms of their comIpVN osition. U Archaeological analysis of prehistoric and historic materials have shown a remarkable potential for tracking intra and inter-site movement of materials, most especially ceramics, therefore, interrelatedness of pottery decoration and symbolic structures justify widespread use of decoration as the prime index of ethnicity preserved in the archaeological record. Since characterization involves examination of ceramic properties, with a view to isolating material of different origin, and ultimately to establish their source hence, the need for the examination of the ceramic property. In line with this, geochemical and mineralogical analysis is imperative and this was 189 carried out. Arising from this present work and others carried out by the author and other researchers from the region, (Akpobasa 1994, Usman 1995, Usman et al 2005, Ogundiran 2002) historical, anthropological and series of archaeological evidences point to the fact that there were series of settlements, abandonment and resettlement in the study area arising from war, conflicts and displacement. These exercises left series of cultural materials such as pavements, wall and ditches and potsherds as evidence of destruction which could be linked to war, conflicts and displacement. Y Decorative motifs such as single twisted cord impression were common tRo all sites. With the exception of sherds from Ila-Orangun, those from other areAas were related in terms of types, fabric and functional attributes. Some of the Rsherds bore striking resemblance to those documented for Old Oyo and Ile-IfeL wiIthB regard to type, decoration and function. Stylistically, the potsherd pavements ex cavated were similar to those documented for Ile-Ife. A C-14 date of AD 1263 waNs obtained from charcoal at a depth of 80cm of the excavated unit. Maize cob deAcoration was absent from the th potsherds indicating that Ajaba site was probably occDupied prior to 16 century when maize was introduced into West Africa. BA th Human impact on the environmFent wIas noted from the 13 Century onwards. Pollen of forest species and ornameOntal plants of Asian origin, such as Lagerstroemia indica, Casuarina equisetifolia and Delonix regia were recovered from the excavations. These were abTunYdant at the lower levels of the excavated pit. However, pollen of ornamental pIlants disappeared completely at the upper levels while secondary forest sRpeciSes and artifacts increased in abundance which was indicative of increase in hEuman population and subsequent impact on vegetation. There was evidence IoVf increased peopling of the area from around AD 1263. Oral and written recordNs suggested that conflicts and war caused displacement and re-occupation of soUme the settlements. 6.2 Future challenges In line with the challenges of this research, there is a need to carry out characterisation of organic residues in pottery vessels from the study area. This will include characterisation of the lipid fraction contained in the pottery. According to Rotlander, 1999 (cited by Giorgi et al 2010), organic residues contains biomarkers which are produced by biochemical transformation of their 190 original content. Analysis of organic residues in pottery can provide significant information about past cultural and technological activities. They can also provide information on pottery uses, customs and dietary habit of the people. More geo-magnetic and electro-restivity of more sites around the study area still need to be carried out, most especially in Oyan, Kiriji War site, and also at Ajaba abandoned site. Search for iron smelting sites need to be carried out to ascertain the source of the metallic objects found at the excavations. Y AR BR LI AN D A F IB O Y IT S R E IV N U 191 REFERENCES Abimbola, W. (ed.) 1976: Yoruba Oral Traditions. Department of African Languages and Literature, University of Ife. Adebayo, I. (1996): The Orangun Dynasty. Catford Copy Centre, Number 3, Bellingham Road, London. Adeoye O., (1988): The Origin and development of Ila. B.A. Dissertation, Department of History, Obafemi Awolowo University, Ile-Ife. Adeeko, A. (2001): Oral Poetry and Hegemony: Yoruba Oriki. DialecticaYl Anthropology 26: 181-192. R Adepegba, C. O. (1996): The Descent from Oduduwa: Claims of RSuAperiority Among B Some Yoruba Traditional Rulers and the arLts oIf ancient Ife. The International Journal of African Historical Studies. Vol. 19 No. 1. Pp 72-92. AN Adetoyi, A. A. (1974): A Short History of Ila-OranDgun. Iwaniyi Pres, Ila-Orangun, Nigeria A Adewumi, I. (2001): Indigenous 16th Cen tuIrBy Ceramic Wearing Course for Low Traffic Walkways In Ile-Ife: Contemporary Engineering Application . OPape Fr presented at the weekly seminar series of the InstituteY of Cultural Studies, Obafemi Awolowo University, Ile-ITfe, Nigeria. Adeyemi J. (1994): PerIsonal Communication (Oyan, July 18, 1994) Afolabi K. (1994)R: SPersonal Communication (Ire, July 16, 1994) Agbaje-WilliaEms, B. (1987): Ife, Old Oyo and Benin: A Chronological Consideration IV in the Light of Recent Archaeological Work at Old N Oyo‘, Nigeria Magazine 55: 23-31. AUgbaje-Williams, B. (1995): Archaeological Investigation of Itagunmodi Potsherd Pavement Site, Ijesaland, Osun State, Nigeria, 1991/92. IFRA Publication, Institute of African Studies, University of Ibadan Agbaje-Williams, B. (2001): Potsherd Pavements and Early Urban Sites in Nigeria, Yorubaland: A Consideration. Nigerian Field Vol. 65 192 Akintoye, S. A. (1971) Revolution and Power Politics in Yorubaland: Ibadan Expansion and the Rise of Ekitiparapo, 1840-1893, London, Longman. Akintoye, S. A. (2010) A History of the Yoruba People, Amalion publisher, Dakar, Senegal Akpobasa, J. (1994): Settlement Studies of Ila-Yara, Osun State: An Ecological Approach. Master‘s Dissertation, University of Ibadan, Ibadan. Y Akpobasa, J. (2006): Settlement Studies: A Developmental ConsiderAatioRn in the Evolution of a Modern Yoruba town. WRest African Journal of Archaeology, 33: (2) 65-72 Alabi, R. A. (1999). Human-Environment RelationshipL: IAB synthesis of ethno-archaeological evidence of human impact on the environment of the BadaNgry coastal area of southwestern Nigeria. DJouArnal of Science Research 5 (1), 25-31. Aleru, J. O. (2006). Old Oyo andI tBhe AHinterland: History and Culture in Northern Yor ubaland. Ibadan, Textflow. Aleru, J. O. and Alabi, R. A. (2010)O: ToFwards a Reconstruction of Yoruba Culture History: A Toponymic Perspective. African TY Study Monographs Vol. 34 (1): 149-162. Aliyu M. K. (2006): I Definition of concepts: Historical Archaeology and RS Ethnoarchaeology. An African Perspective Ed. Okpoko E A.I and Aliyu M.K. Zaria Archaeological Papers, A.B.U., Zaria. AllisoNn, PI. V A. (1963): A terracotta Head in the Ife Style from Ikirun, Western U Nigeria. MAN Nos. 1 93/94: 156-157 Allsworth-Jones, P. (1989): Kariya Wuro rock shelter, Bauchi State: recent results. Paper presented at a conference in honour of Thurstan Shaw. Ibadan, November, 1989. Allsworth-Jones, P. (1996): Continuity and Change in Yoruba Pottery. Bulleting of the School of Oriental and African Studies, University of London 312-322. 193 Andah W .B. (1979): Perspective on West Africa‘s Past. Special Book Issue of West African Journal of Archaeology, Vol. 9 Anthony, D. W. (1990): Migration in archaeology: The baby and the bathwater. American Anthropologist, 92, 895-814. Aribidesi, U. (2003). Ceramic Seriation, Sites Chronology, and Old Oyo Factor in Northcentral Yorubaland, Nigeria. African Archaeolological Review 20 (3), 149-169. Aribidesi, U. Aleru J.O. and Alabi, R. (2005): Sociopolitical Formation on thYe Yoruba Northern Frontier: A Report of Recent Work at Ila-Yara, NRorth Central Nigeria. Journal of African Archaeology Vol. 3 (1) 139-154. A Aribidesi, U. (2009): Pre-colonial regional migration and settlement abanRdonment in Yorubaland, Nigeria. In Falola Toyin and Usman AribiIdBesi: Migration, Boarders and Identities in Africa. University of Rocheste r LPress, USA. Arkell, A. J. (1944): Archaeological Research inA WNest Africa. Antiquity Vol. XVIII: 147-150. Arnold, D. E., Neff, H., Bishop, R. L., (1991): ACDompositional analysis and ‗‗sources‘‘ of pIoBttery: An ethnoarchaeological approach.F Am erican Anthropologist 93, 70–90. Atherton, J. H. (1983): EthnoOarchaeologyin Africa. African Archaeological Rev iew, 1, 75-104. Atoyebi, J. (1985): TTYhe Historical and Political Development of Iragbiji in I Pre-Colonial times. An essay submitted to History RS Department, Obafemi Awolowo University, Ile-Ife. Awe, B. (1974E): ―Praise Poems as Historical Data: the Example of the IV Yoruba Oriki” Africa 44 (4) 331-349. BakinNde, O. (2006): Oral tradition, linguistics and ethnoarchaeology. U Ethnoarchaeology: An African Perspective Ed. Okpoko A.I and Aliyu M.K. Zaria Archaeological Papers, A.B.U., Zaria. Barber, K. (1994): Polyvocality and the individual talent: Three Women Oriki Singers in Okuku. The Yoruba Artist: New perspective on African arts. (R. Abiodun et el) Blackman, R., (1992): The effect of natural and human size sorting on mineralogy and chemistry of ceramic clays. In: Neff, H. 194 (Ed.), Chemical Characterization of Ceramic Pastes in Archaeology. Prehistory Press, Madison, WI, pp. 113– 124. Blench, R. (2003): The movement of cultivated plants between Africa and India in prehistory. In Neumann, K., Butler, A. and Kahlheber, S. (eds), Food, Fuel and Fields, 273-292. Druck, Hanskock, GMBH, Bielefeld, Koln, Germany. Bryant, V. M. (1974) Prehistoric diet in southwest Texas: The coprolitYe evidence. American Antiquity. 39 (3), 407-420. R Bryant, B.M. and Holloway, R. G. (1996). Archaeological Palynology.A In Jansonius, J. and McGregor, D. C. (edBs), RPalynology: principles and applications. AmericaIn Association of Stratigraphic Palynologists Found aLtion 3, 913-917. Sequence Stratigraphy in SE ANsia, 49-71. Butzer, K. W. (1982): Archaeology as human ecoAlogy: Method and theory for a contextual approaAch.D Cambridge University Press. Cameron, C. M. (1995): Migration andI tBhe movement of southwestern peoples. Journal of an thropological archaeology, 14, 104-124. Chami, F. A. (2007): DiffuOsionF in the studies of the African past: Reflections from new archaeological findings. African AYrchaeological Review, 24: 1-14. Curtin, P. D. (1997): ‗AfIrTica and Global Patterns of migration‘ in Gungwu (ed) Global HisRtoryS and Migrations (Boulder CO: Westview). Deer, W. A., HEowie, R. A., Zussman, J. (1992): An Introduction to the Rock- IV forming Minerals, second ed. Longman Scientific and N Technical, New York, 696 pp. DUurand, S. R., Shelley, R., Antweieler, R. C., Taylor, H. E., (1999): Trees, chemistry, and prehistory in the American southwest. Journal of Archaeological Science 26, 185–203. De-Corse, C (1990): Documents, oral histories and the material record: Historical Archaeology in West Africa. World Archaeological Bulletin 7: 40-50. Deetz, J. (1965): The Dynamics of Stylistics Change in Arikara Ceramics. University of Illinois Press, Urbana. 195 Dick-Read, R. (2005). The Phantom Voyagers: Evidence of Indonesian Settlement in Africa in Ancient Times. Winchester, England, Thurlton Publishing. Duffin, K. I. (2008). The representation of rainfall and fire intensity in fossil pollen and charcoal records from a South African savannah. Review of Palaeobotany and palynology 2008, doi:10.1016/j.revpalbo.2008.02.004 Dupont, L. M., Jahns, S., Marret F., and Ning, S. (2000). Vegetation change in equatorial West Africa: time slices for the last 150ka. PalaeogeographyY, Palaeoclimatology, Palaeoecology 155, 95-122. R Ededele L. (1994): Personal Communication (July 17, 1994) A Eekens, J. W., Neff, H., and Glascock, M. D., (2002): Ceramic prodRuction among small-scale and mobile hunters and gatherers: a case sItuBdy from the southwestern Great Basin. Journal of Anthropological Ar cLhaeology 21, 200– 229. N Ekanade, and Aloba, (1986): The 19th century Yoruba WAarfare: A geographer’s view point. Paper presented at the National cAonfDerence on the centenary of the Kiriji/Ekitiparapo Peace Treaty. UnivIeBrsity of Ife, Ile-Ife, Nigeria. Eyo, E. (1974): Odo Ogbe StreeFt a nd Lafogido: Contrasting Archaeological Sites in Ile-Ife, Western NigOeria. West African Journal of Archaeology (4) 99-109 Eyo, E. (1976): IgboT‘LaYja, Owo. West African Journal of Archaeology (6) 37-59)I Faegri, K. and IveRrsenS, J. 1989. Textbook of Pollen Analysis. Denmark, John Wiley Eand Sons Ltd. Fadipe, AI. V(1991): Sociology of the Yoruba. Ibadan: Ibadan University Press. FasikNu, A. (1995): Igbajo and its People. Writers‘ Fraternity Limited, Lagos, U Nigeria. pp. 11 Fatunsin, A. K. (1992). Yoruba Pottery. Lagos: National Commission for Museum and Monuments. Folorunso, C.A. (1993): Ethnoarchaeology – Its methods and practice: The Tiv example. West African Journal of Archaeology, Vol. 23: 93-103 Fox, R. A. (1993): Archaeology, History, and Custer‘s Last Battle: The Little Big Horn Re-examined. Norman: University of Oklahoma Press. 196 Frobenius L. (1921): The voice of Africa. Benjamin Bloom, Inc., New York and London. Grimm, (1991). TILIA and TILIA GRAPH. Illinois State Museum, Research and Collection Centre, Springfield, Illinois, USA. Garlake, P. (1974): Excavation at Obalara‘s Land, Ife: an Interim Report. West African Journal of Archaeology (4) 111-148 Garlake, P. (1977): Excavations on the Woye Asiri Family Land in Ife, Western Nigeria. West African Journal of Archaeology (7) 57-95 Y Geological Survey map of Nigeria, Iwo Sheet 1: 25000. Federal Ministry of LandR and Survey, Lagos. A Gilman, A. (1991): Trajectories toward social complexity in the later pRrehistory of the Mediterranean. In Earle T. (ed), Chiefdoms: PowLerI, BEconomy and Ideology. Cambridge University Press, Cambridge pp 14 6-68. Gould, R. (1980). Living Archaeology. Cambridge UnivNersity Press, Gould, R. (1990). Recovering the Past. University of AMexico Press:Albuquerque . Graham, C. (2001): African Civilization: Pre-cDolonial cities and state in tropical Africa: An Archaeological pIerBspe Active. Cambridge University Press. Graham, C. (2004): Forgotten Africa: An Introduction to its Archaeology Cambridge University Press. F Haberle, C. (1994). Anthropoge nOic indicators in pollen diagrams: Problems and prospects for Late QTuaYternary palynology in New Guinea. In Hather J. G. (ed.) Tropical ArchaeoIbotany. Applications and new developments, 172-201, London, RoutlSedge. Hammer, O., HEarpRer, D. A. T., Ryan, P. D., 2001. PAST: Paleontological Statistics soIftVware package for education and data analysis. Palaeontologia Electronica N4 (1), 9. HUenige, D. P. (1982): Oral Historiography. Longman Group Limited, United Kingdom. Hutchinson, and Dalziel, (1954 – 1968). Flora of West Tropical Africa Vol. I, Parts 1 and 2 (Second edition, revised by R.W.J. Keay); Vols. II and III, Parts 1 and 2 (Second edition, edited by F.N. Hepper), London, Crown Agents for Overseas Governments and Administrations. 197 Ige, O. A., Ogunfolakan, B. A. (2001): Archaeological prospection using hosting rock characteristics: an example from archaeological sites in Igbajo SW Nigeria. ODU: A Journal of West African Studies 41, 23–31. Ige, O. A., Rehren, Th, (2003): Black sand and ironstone: iron smelting in Modakeke, Ife, Southwestern Nigeria. Institute of Archaeo-Metallurgical Studies 23, 15–20. Ige O.A., Adesina M. (2008): Mineralogical characterization of ancient ceramics: An example from Ipetumoddu relics. Archaeology, Heritage Management and Y th Tourism Development in Nigeria. Proceedings of the 18 Annual ConfereRnce of the Archaeological Association of Nigeria, November 7-19, 2007. A Ige, O. A., Swanson, S, (2008) Provenance of Esie sculptural soapsRtone from southwestern Nigeria. Journal of Archaeological Science 35,I 1B553–1565. Ige O. A., Ogunfolakan B. A, Ajayi E. O. B. (2009): Chemical cLharacterization of some potsherd pavements from parts of YorubalAandN in southwestern Nigeria. Journal of Archaeological Science 36, 90-99. Jensen, C. (2004). The vegetation history of a Dcoastal stone-age and Iron Age settlement at 70°N, Norway. VegetatiBon HAistory and Archaeobotany 13, 269- 284. I Jones, G. T., Bailey, D. G., Beck, C., (1F997): Source provenance of andesite artefacts using non-destruc tiOve analysis. Journal of Archaeological Science 24, 929–943. Y Keay, R. W. (1959). AnI OTutline of Nigerian vegetation. Lagos, Nigeria, Federal Government oSf Nigeria. Kirch, V. P. (1E996R). Late Holocene Human-induced Modifications to a Central PoIlVynesian Island Ecosystem. Proceedings National Academy of Science NUSA 93, 5296-5300. KUi-Zerbo (1981): General History of Africa, Vol. 1, Heinemann. Kramer, C. (1979): Ethnoarchaeology, Implications of Ethnography in Archaeology. Columbia University Press/New York. Lanlehin, L. and Aina, S. (1995): OTAN-AYEGBAJU: The Task of Rebuilding. Otan Ayegbaju Club 74, Otan Ayegbaju Lisa, F. (Ed) (1991): Historical Archaeology in Global Perspective. Smithsonian Press pp. 19. Lynott, M. J., Neff, H., Price, J. E., Cogswell, J. W., Glascock, M. D., (2000): 198 Inferences about prehistoric ceramics and people in southeast Missouri: results of ceramic compositional analysis. American Antiquity 65 (1), 103– 126. Mabogunje A.L. (1962): Yoruba Towns. Ibadan University Press, Ibadan. Mabogunje A. and P Richards (1985): ‗Land and people: Models of spatial and rd ecological processes in West African history‘ History of West Africa 3 Edition: J.F. A. Ajayi and M. Crowder (Eds.), pp 4-47, Oxford, Oxford Press. McGuire, R. and Paynter, R., (1991) The Archaeology of Inequality. Oxford: Y Blackwell. R Morley, R. J. (1995). Biostratigraphic Characterization of Systems Tracts in ATertiary Sedimentary Basins. Proceedings of the International Symposium oRn Nicholas, D. Sterner, M., and Gavua, D. (1998): Why Pots are IdBecorated (with comments). Current Anthropology 29: 365-389. L Nzewunwa, N. (1989): Prehistoric Pavement. WesNt African Journal of Archaeology 19: 93-116. A Ojo, G. J. A. (1996b): Yoruba Culture – A geogArapDhical analysis. Ife and London, University of Ife and University of LonBdon Press. Ojo, E. O. (1992): A brief Analysis Fof thIe importance of the Kiriji War on Igbajo. An essay submitted to History Department, Obafemi Awolowo University, Ile-Ife. O Ogundele, S. O. (2004): TRYethinking West African Archaeology. John Archers (Publishers) LimitIed, Ibadan. Ogundiran, A. (20R02):S Archaeology and History in Ilare District (Central YorubaEland, Nigeria) 1200-1900 A.D. Cambridge Monographs in African AIrcVhaeology. OgunNfolakan, B.A.(1990): Luwo Potsherd Pavement in Il-Ife: 30 years of Nigeria U Independence. Institute of African Studies, Academy of Science Publication, Moscow, USSR Ogunfolakan, B. A. (1994): Archaeological survey of Osun North East of Osun state, Nigeria. A M.Sc. Dissertation submitted to the Department of Archaeology/Anthropology, University of Ibadan, Ibadan, Nigeria. Ogunfolakan, A. (2001): Luwo Potsherd Pavement: A Unique aspect of Ife Culture. Journal of Arts and Ideas, Vol. 3: 88-97. 199 Ogunfolakan, A. (2000): Archaeological Investigation of Ishan-Ekiti Landslide. Nigerian Field 64: 148-154. Ogunfolakan, B. A. (2006). Archaeological Survey of North-East Osun State, Nigeria: West African Journal of Archaeology. Vol. 33 (2) 73-94 Ogunfolakan, B. A. (2007). Historical Archaeology of Northeast Osun. Unpublished M.Phil. Dissertation, Department of Archaeology and Anthropology, University of Ibadan. Okpoko, A. I. (1979): Settlement Archaeology in the Anambra River Valley: A shorYt note. West African Journal of Archaeology, Vol. 9, pp 167-177 R Okpoko, A. I. (2006): Settlement pattrns and burials: Ethnoarcheology: AnA African Perspective. Zaria Archaeoogy Papers, ABU, Zaria R Orijemie, A. E., Ogunfolakan A., Aleru, J.A., Sowunmi, M.A. (2010)I: B The archaeology and palynology of Ajaba, a late iron-age sett leLment in north-east Yoruba land, Nigeria: some preliminary results. WAestN African Archaeology: New developments, new perspectives BAR IntDernational Series, Achaeopress, Publisher of Brithishh rchaeological Report, London. Oyelaran, P. A. (1998). Early settlemeBnt Aand Archaeological Sequence of Northeast Yorubaland. African Arch aIeological Review 15 (1), 65-79. Pemberton III, J. and Afolayan F.O S. F(1996): Yoruba Sacred Kingship. Washinton, Smithsonian institution Pre ss. Pigg, K. B. and DeVore, MT. LY. (2005). Shirleya grahamae gen. et sp. nov. (Lythraceae), LagIerstroemia-like fruits from the middle Miocene Yakima canyon floRra, Scentral Washington State, USA. American Journal of Botany 92, 242-251 Piperno, DIV. R. E, Bush, M. B. and Colinvaux, P. A. (1991). Palaeo-environments and NHuman Occupation in Late-Glacial Panama. Quaternary Research 33, 108-U 116. Posnansky, M. and DeCorse, C. (1986): Historical Archaeology in sub-Africa: A review. Historical Archaeology 20 1-14. Rice, P. (1999): On The Origin of Pottery. Journal of Archaeological Methods and Theory, Vol. 6 No. 1, pp1-55. Rodney, W. (1972): How Europe underdeveloped Africa. London Bogle- L’Ouverture 200 Schwedt, A., Mommsen, H., Stephan, H. G., Gaimster, D., (2003): Neutron activation analyses of ‗Falke-Group‘ stoneware. Archaeometry 45 (2), 233–250 [Late Medieval European]. Smith, R. S. (1967): Yoruba Armament. Journal of African History, Vol. VIII (1) Smith, R. S. (1969): Kingdom of the Yorubas 2nd Edition, Methuen, London Soper, R. C. 1975. Archaeological Work at Old-Oyo. Unpublished Report, Department of Archaeology and Anthropology, University of Ibadan. Sowunmi, M. A. (1981a). The late Quaternary environmental changes in NigeriaY. Pollen et Spores 23 (1), 125-148. R Sowunmi, M. A. (1981b). Aspects of late Quaternary vegetational chaAnges in West Africa. Journal of Biogeography 8, 457-474. R Sowunmi, M. A. (1985). The beginnings of agriculture in WestI ABfrica: botanical evidence. Current Anthropology 26 (1), 127 - 129. L Sowunmi, M. A. (1999). The significance of the Aoil Npalm (Elaeis guineensis Jacq.) in the Late Holocene environments ofD west and west central Africa: A Reconsideration. Vegetational History and Archaeobotany 8 (3), 199-210. Sowunmi, M. A. 2004. Aspects of NiIgeBria An coastal vegetation in the Holocene: some recent insights, in R. W. FBat tarbee, F. Gasse, and C. E. Stickley (eds), Past Climate Variability thOrough Europe and Africa, 199 – 218, Dordrecht, the Netherlands, Kluwer A cademic Publishers. Steponaitis, V. P., BlackmaTn, YM. J., Neff, H., (1996): Large scale patterns in the chemical composiItion of Mississippian pottery. American Antiquity 61 (3), 555–572. S Tite, M.S. (197E2):R Methods of Physical examination in Archaeology. Seminar PrIeVss, London TookNer, E. (1982). Ethnography by Archaeologists. The American Ethnological U Society. Trigger B. C. (1967) Settlement Archaeology: Its goals and promise. Antiquity 32. Tutuncu, S., Dane, F. and Tutuncu, E. (2007). Examination of pollen morphology of some exotic trees and shrubs found in the parks and the gardens of Edirne (European Turkey) I. Journal of Applied Biological Sciences 1(2), 45-55. 201 Tugbiyele, E.A. (1989): A Brief History of Igbajo. Unpublished lecture delivered at the annual celebration of the National Union of Igbajo students at OAU, Ile-Ife. Vansina, J. (1960): Recording the Oral History of the Bakuba. Journal of African History 1 (1): 45-53. Wakita, H., Rey, P., Schimitt, R. A., (1971): Abundances of 14 rare earth element and other trace elements in Apollo 12 samples: five igneous and one breccia rocks and four soils. In: Proc. Second Lunar Sci. Conf. 2. The M.I.T. PressY, pp. 1319–1329 R Watson, P.J. (1995) Archaeology, Anthropology and the culture concepRt. AAmerican Anthropologists, 97: 683-694 Wendy, A. and Knapp, B.A. (Eds) (1982): Archaeologies of ILBandscape – Contemporary Perspectives. Oxford: Blackwell Publishe r.L 124-145 Wesler,K.W. (1999): Chronological Sequences in NigeArianN Ceramics. African Archaeological Review, Vol. 16, (4). Willett, F. (1961). Investigations at Old Oyo, D1956-57: An Interim Report. Journal of the Historical Society of NiBgeriAa 2 (1), 59-77 Zeven, A.C. (1964). On the origin ofF the Ioil palm (Elaeis guineensis Jacq.). Grana Palynol. 5, 121-123 O ITY ER S V UN I 202 APPENDIX 1: Glossary of Yoruba Words Yara – Trench Ila-Yara - Ila -of Defence trench Agbado – Maize/Corn Eerun – Dry Season Agbado Eerun – Dry season maize/corn Agbado Ojo – Raining season maize/corn Apaadi – Pottery sherd Y Apaadi Luwo – Luwo potsherd R Omi-Eku – Residual water from palm-oil production A Olokun – Yoruba goddesses of ocean/river R Aro – A chieftaincy title in Ife IB Oriki - Praise poem/cognomen L Awigbo - Hearsay N Awigbo baba itan - Hearsay, the father of histoAry Igbi-‘riro - Sacred grove D Ota Ebora - Stone of the demBonA Oke - Hill I Odun-Oke - Hill festivFal Oba – Kin g O Ile-Esile - EYsile‘s compound/house Aladura – ITChristian/Prayer group Ijoko - S Seat Okan yi lule o Eku oRkan- One fell down remaining one Gbe kan rIuV kan - Carrying one over another Ira - N A type of tree IjUi - Shade Ajere - Perforated pot meant of drying meat Itele - Underlay 203 APPENDIX II: LIST OF INFORMANTS Name Place Gender Age Date of Occupation Interview Chief Adeyemi, Jagun Oyan Male 80+ 1992/1997 Farmer of Oyan Oba (Dr) N.B. Oyan Male 75 1992/97 King Omotoso Mr. S.O. Olaniyi Oyan Male 65 1992/97 Farmer Mr Olagunju Oyan Male 60+ 1992/97 Civil servan Rt Y Oba Peter Olagunju Asi Male 80+ 1992/94 King A Prince A. Olagunju Asi Male 60+ 1992/97 BTeaRcher Lamidi Ededele Asi Male 70+ 1992/97 IFarmer Pa Atolagbe Asi Male 90+ 1992/97 L Farmer Oba O. Fasade Igbajo Male 60+ 199N2/2006 King Chief James Adesina Igbajo Male 70 A1992/1996 Farmer Chief Adegbite Igbajo Male A70+D 1992/94 Rtd. Civil IB Servant Chief Olabode Faloye Igbajo MFale 70+ 1992/94 Farmer Mrs Modupe Adesina Igbajo OFemale 60+ 1992/94 Trader Chief Ezekiel Ayepola IgbaYjo Male 80+ 1992/94 Farmer Pa I.O. Fatanmi ITgbajo Male 90 1992/94 Farmer Joseph Faloni SIIgbajo Male 90 1992/94 Farmer Oba Timothy OyeRbode Iragbiji Male 80+ 1992/97 King Chief MurainaE Oyelami Iragbiji Male 60+ 1992/2006 Traditional chief/Artist Alhaji AfIoVlabi Dayiola Iragbiji Male 80+ 1992/94 Traditional N chief OUpatola O Sangodoyin Iragbiji Male 50+ 1992 Artist Oba Moses Omotoso Ire Male 70+ 1992/97 King Isaaac I. Omotoso Ire Male 80 1992/97 Farmer Joseph Ojo Ire Male 80 1992/97 Farmer Alhaji Kadiri Afolabi Ire Male 60+ 1994/97 Farmer Prince A.A.Adetoyi Ila- Male 80+ 1992/94 Rtd C. Servant Orangun 204 Alhj Sumonu Adesina Ila- Male 80+ 1992/94 Farmer Orangun Aafa Tiamiyu Ila- Male 70+ 1992/94 Farmer Orangun Oba Sikiru Odugbemi Iresi Male 30+ 1992/2006 King Pa Joshua Popoola Iresi Male 80+ 1992/2006 Gold (Baba Gosi) Smith/farmer Adeleke Iyiola Ajaba Male 70 1992/2006 Farmer Y Pa Adebisi Bakare Ajaba Male 90 1992/2006 Farmer R Oba Adedotun Ajaba Male 90+ 1992/2006 King A Adetoyese R Oba Adebisi Adeniran Ede Male 60+ 1992/2006 IBKing Prince A. Ogunwole Ede Male 60 1992/20 0L6 Farmer Prophet G. Jolayemi Ede Male 90+ 199N2/2006 Farmer Chief Salako Ede Male 80+D A2004 Businessman/faA rmer Emebosa Oriejieme Ibadan Male IB 30+ 2006 Lecturer O F ITY RS E NI V U 205 Appendix III Geo-Magnetics reading data File 1-50 RY RA LI B AN BA D OF I SI TY R IV E UN 206 RY RA LI B N AD A IB OF ITY S VE R I UN 207 RY RA LI B N AD A IB OF ITY S VE R I UN 208 RY RA LI B N AD A IB OF ITY S VE R I UN 209 RY RA LI B N AD A IB OF ITY S VE R I UN 210 RY RA LI B N AD A IB OF ITY S VE R I UN 211 RY RA LI B N AD A IB OF ITY S VE R I UN 212 RY RA LI B N AD A IB OF ITY S VE R I UN 213 RY RA LI B N AD A IB OF ITY S VE R I UN 214 RY RA LI B N AD A IB OF ITY S VE R I UN 215 RY RA LI B N AD A IB OF ITY S VE R I UN 216 RY RA LI B N AD A IB OF ITY S VE R I UN 217 RY RA LI B N AD A IB OF ITY S VE R I UN 218 RY RA LI B N AD A IB OF ITY S VE R I UN 219 RY RA LI B N AD A IB OF ITY S VE R I UN 220 AR Y R LIB AN AD IB O F ITY ER S IV UN 221 Appendix IV Graph data Geo-magnetic reading File1- 50 RY BR A LI AN AD F I B O SI TY VE R I UN 222 RY RA LI B N AD A IB OF ITY S VE R I UN 223 RY RA LI B N AD A IB OF ITY S VE R I UN 224 RY RA LI B N AD A IB OF ITY S VE R I UN 225 RY RA LI B N AD A IB OF ITY S VE R I UN 226 RY RA LI B N AD A IB OF ITY S VE R I UN 227 RY RA LI B N AD A IB OF ITY S VE R I UN 228 RY RA LI B N AD A IB OF ITY S VE R I UN 229 RY RA LI B N AD A IB OF ITY S VE R I UN 230 RY RA LI B N AD A IB OF ITY S VE R I UN 231 RY RA LI B N AD A IB OF ITY S VE R I UN 232 RY RA LI B N AD A IB OF ITY S VE R I UN 233 RY RA LI B N AD A IB OF ITY S VE R I UN 234 RY RA LI B N AD A IB OF ITY S VE R I UN 235 RY RA LI B N AD A IB OF ITY S VE R I UN 236 RY RA LI B N AD A IB OF ITY S VE R I UN 237 RY RA LI B N AD A IB OF ITY S VE R I UN 238 RY RA LI B N AD A IB OF ITY S VE R I UN 239 RY RA LI B N AD A IB OF ITY S VE R I UN 240 RY RA LI B N AD A IB OF ITY S VE R I UN 241 RY RA LI B N AD A IB OF ITY S VE R I UN 242 RY RA LI B N AD A IB OF ITY S VE R I UN 243 RY RA LI B N AD A IB OF ITY S VE R I UN 244 RY RA LI B N AD A IB OF ITY S VE R I UN 245 RY RA LI B N AD A IB OF ITY S VE R I UN 246 RY RA LI B N AD A IB OF ITY S VE R I UN 247 RY RA LI B N AD A IB OF ITY S VE R I UN 248 RY RA LI B N AD A IB OF ITY S VE R I UN 249 RY RA LI B N AD A IB OF ITY S VE R I UN 250 RY RA LI B N AD A IB OF ITY S VE R I UN 251 RY RA LI B N AD A IB OF ITY S VE R I UN 252 RY RA LI B N AD A IB OF ITY S VE R I UN 253 RY RA LI B N AD A IB OF ITY S VE R I UN 254 RY RA LI B N AD A IB OF ITY S VE R I UN 255 RY RA LI B N AD A IB OF ITY S VE R I UN 256 RY RA LI B N AD A IB OF ITY S VE R I UN 257 RY RA LI B N AD A IB OF ITY S VE R I UN 258 RY RA LI B N AD A IB OF ITY S VE R I UN 259 RY RA LI B N AD A IB OF ITY S VE R I UN 260 RY RA LI B N AD A IB OF ITY S VE R I UN 261 RY RA LI B N AD A IB OF ITY S VE R I UN 262 RY RA LI B N AD A IB OF ITY S VE R I UN 263 RY RA LI B N AD A IB OF ITY S VE R I UN 264 RY RA LI B N AD A IB OF ITY S VE R I UN 265 RY RA LI B N AD A IB OF ITY S VE R I UN 266 RY RA LI B N AD A IB OF ITY S VE R I UN 267 RY RA LI B N AD A IB OF ITY S VE R I UN 268 RY RA LI B N AD A IB OF ITY S VE R I UN 269 RY RA LI B N AD A IB OF ITY S VE R I UN 270 Y AR LIB R AN BA D F I O SI TY VE R I UN 271 APPENDIX V Interpretation of Geomagnetic readings File 1-50 RYA LIB R AN D IB A OF SI TY R VE UN I 272 RY RA LI B N AD A IB OF ITY S VE R I UN 273 RY RA LI B N AD A IB OF ITY S VE R I UN 274 RY RA LI B N AD A IB OF ITY S VE R I UN 275 RY RA LI B N AD A IB OF ITY S VE R I UN 276 RY RA LI B N AD A IB OF ITY S VE R I UN 277 RY RA LI B N AD A IB OF ITY S VE R I UN 278 RY RA LI B N AD A IB OF ITY S VE R I UN 279 RY RA LI B N AD A IB OF ITY S VE R I UN 280 RY RA LI B N AD A IB OF ITY S VE R I UN 281 RY RA LI B N AD A IB OF ITY S VE R I UN 282 RY RA LI B N AD A IB OF ITY S VE R I UN 283 RY RA LI B N AD A IB OF ITY S VE R I UN 284 RY RA LI B N AD A IB OF ITY S VE R I UN 285 RY RA LI B N AD A IB OF ITY S VE R I UN 286 RY RA LI B N AD A IB OF ITY S VE R I UN 287 Y RA R B LI DA N A F I B O TY RS I E IV UN 288 APPENDIX VI: RADIO-CARBON DATE RESULT UNIVERSITY OF WASHINGTON QUATERNARY ISOTOPE LAB RADIOCARBON CALIBRATION PROGRAM REV 4.1.2 Stuiver, M. and Reimer, P.J., 1993, Radiocarbon, 35, p. 215-230. Calibration file(s): intcal98.14c Listing file: c14fil.lst Export file: c14res.csv GX-32369 Y (KJ)(AJ)(Rm 1) 40-50 cm charcoal R Radiocarbon Age BP 360 +/- 60 RefeArence Calibrated age(s) cal AD 1491, 1603, 1609 (StuivRer et al., 1998a) IB cal AD/BC age ranges obtained from intercepts (LMethod A): one Sigma** cal AD 1448 - 1637 N two Sigma** cal AD 1432 - 1656 A Summary of above: D maximum of cal age ranges (cal ageAs) minimum of cal age ranges: 1 sigma cal AD 1448 (1491, 16B03, 1609) 1637 2 sigma cal AD 1432 (149F1, 1I603, 1609) 1656 cal AD/BC age ranges (cal ages as above) from probability dist rOibution (Method B): % area enclosed cal AD Yage ranges relative area under probability distribution 68.3 (1 sigma) I Tcal AD 1472 - 1525 .461 1560 - 1628 .539 95.4 (2 sigma) cal AD 1443 - 1642 1.000 R S GX-32370 E (KJ)(AIJ)V(Rm 1) 103 cm charNcoal URadiocarbon Age BP 780 +/- 80 Reference (Stuiver et al., 1998a) Calibrated age(s) cal AD 1263 cal AD/BC age ranges obtained from intercepts (Method A): one Sigma** cal AD 1191 - 1200 1208 - 1290 two Sigma** cal AD 1040 - 1102 1115 - 1142 1151 - 1316 1353 - 1388 Summary of above: maximum of cal age ranges (cal ages) minimum of cal age ranges: 289 1 sigma cal AD 1191 (1263) 1290 2 sigma cal AD 1040 (1263) 1388 cal AD/BC age ranges (cal ages as above) from probability distribution (Method B): % area enclosed cal AD age ranges relative area under probability distribution 68.3 (1 sigma) cal AD 1176 - 1295 1.000 95.4 (2 sigma) cal AD 1040 - 1103 .096 1107 - 1142 .048 1146 - 1315 .799 Y 1353 - 1388 .058 A R BR References for datasets used: I Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W .,L Burr, G.S., Hughen, K.A., Kromer, B., McCormNac, F.G., v.d. Plicht, J., and Spurk, M. (1998a). Radiocarbon 40:1041-1083. A Comments: A D * This standard deviation (error) includes a lab error multiplier. ** 1 sigma = square root of (sam pIlBe std. dev.^2 + curve std. dev.^2) ** 2 sigma = 2 x square root Fof (sample std. dev.^2 + curve std. dev.^2) where ^2 = quantity squar eOd. [ ] = calibrated with Yan uncertain region or a linear extension to thTe calibration curve 0* represents a "neIgative" age BP 1955* denotes inSfluence of nuclear testing C-14 NOTE: Cal agRes and ranges are rounded to the nearest year which may Ebe too precise in many instances. Users are advised to round results to the nearest 10 yr for samples with stIaVndard deviation in the radiocarbon age greater than 50yr. UN 290