A PRELIMINARY ARCHAEOLOGICAL STUDY OF IRON SLAG AND ORE SHRINES AT KORMANTSE, CENTRAL REGION OF GHANA. BY ESI KOAH ARKOH (10490308) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF A MASTER OF PHILOSOPHY (MPHIL) ARCHAEOLOGY DEGREE. OCTOBER, 2022 University of Ghana http://ugspace.ug.edu.gh i DECLARATION I hereby declare that this thesis is a result of an independent research I carried out with the support and supervision of Dr. Gertrude Aba Mansah Eyifa-Dzidzienyo and Dr. David Akwasi Mensah Abrampah and that no part of this work has been submitted to any other institution for the award of a degree. ………… …………… ESI KOAH ARKOH (10490308) STUDENT 06/10/2022 …………………………………………………………… DR. GERTRUDE ABA MANSAH EYIFA-DZ1DIENYO SUPERVISOR DR. DAVID AKWASI MENSAH ABRAMPAH CO-SUPERVISOR 06/10/2022 06/10/2022 University of Ghana http://ugspace.ug.edu.gh ii DEDICATION With love to Mr. Desmond Apedu, my brother-in-law, Mrs. Anita Apedu, my sister and Mrs. Ernestina Arkoh, my mother, for their immense support throughout my MPhil academic period. God bless you all. University of Ghana http://ugspace.ug.edu.gh iii ACKNOWLEDGEMENT I am enormously grateful to the Almighty God without whom I would not have come this far. To Professor Emmanuel Kofi Agorsah for the opportunity given me to work on his site under the auspices of Kormantse Archaeological Research Project (KARP) and for all the comprehensive support I received from him. Thank you to Improving Africa’s Future Using Lessons from the Past (IAFF) project and Canadian SSHRC for funding my MPhil program and research. My profound gratitude to my supervisors, Dr. Gertrude Aba Mansah Eyifa-Dzidzienyo and Dr. David Akwasi Mensah Abrampah, who saw the messy nature of my work at the beginning but have greatly helped shape it into a better work. I am very appreciative of their advices and numerous editing done on my work. My appreciation to Heidrun M.V. Stebergløkken for her supervisory role and contribution to my work during my study at the Norwegian University of Science and Technology Norway (NTNU) and to the other lecturers who taught me at NTNU. My thanks to some faculty members of the Department of Archaeology and Heritage Studies for their advice and encouragement especially to Mr. Daniel Kumah for his leads to Caroline- Robion-Brunner (a researcher at French National Center for Scientific Research) and Prof. James Boachie-Ansah for helping with the identification and analysis of some of my materials. I am grateful to my research team which comprised Mr. Gideon Agyare, Mr. Kelvin Asare, Miss Addison Odarkor and Mr. Gerald Ganyo for their support during the ethnographic and archaeological work at Kormantse. To the land owner, field assistants and the chiefs and people of Kormantse for hosting us and supporting us in their best way possible. I also thank Miss Beatrice Darko Yeboah for her assistance during the post field work, to Mr. Raymond Agyenkwa for his support during the field work and to Mr. Eric Zatey for assisting University of Ghana http://ugspace.ug.edu.gh iv with some of the post field activities. Special thanks go to Miss Dora Fianyo of the Department of Earth Science for her accommodating nature and for assisting with the chemical analysis. I also owe a debt of gratitude to Prof. Daniel K. Attuquayefio of the Department of Animal Biology and Conservation Science for working on the faunal remains and his generosity. Also, I am grateful to Mr Emmanuel Klubi for identifying all the shells that were obtained from the archaeological excavations. Lastly, a heartfelt gratitude to Mr. and Mrs. Apedu, Mrs. Ernestina Arkoh, Mr. John Papa Arkoh, my siblings, Miss Erica Odi Oduro and some very good friends for the care, encouragement, prayers, financial support and advice throughout my MPhil journey. Thank you all. University of Ghana http://ugspace.ug.edu.gh v TABLE OF CONTENTS DECLARATION ..................................................................................................................... i DEDICATION ........................................................................................................................ ii ACKNOWLEDGEMENT .................................................................................................... iii TABLE OF CONTENTS ....................................................................................................... v LIST OF TABLES ................................................................................................................ ix LIST OF MAPS ...................................................................................................................... x LIST OF FIGURES .............................................................................................................. xi ABSTRACT ......................................................................................................................... xiv CHAPTER ONE ..................................................................................................................... 1 BACKGROUND OF THE STUDY AREA .......................................................................... 1 1.0 Introduction ......................................................................................................... 1 1.1 The Study Area .................................................................................................... 2 1.2 Kormantse in the Context of Fante History ......................................................... 7 1.3 Research Problem .............................................................................................. 11 1.4 Aim and Objectives ........................................................................................... 11 1.5 Research Questions ........................................................................................... 12 1.6 Material Culture Studies .................................................................................... 12 1.7 Significance of the Study .................................................................................. 15 1.8 Challenges ......................................................................................................... 15 1.9 Organization of Chapters ................................................................................... 16 CHAPTER TWO .................................................................................................................. 17 THE CENTRAL REGION REVIEWED FROM HISTORY AND ARCHAEOLOGY 17 2.1 Introduction ....................................................................................................... 17 2.2 Central Region Prior to the Trans-Atlantic Trade ............................................. 17 2.3 Central Region During the Trans-Atlantic Trade .............................................. 21 2.4 Previous Archaeological Researches Conducted at Kormantse ......................... 26 2.5 Shrines, their Composition and Significance .................................................... 27 2.6 Iron Slag in Religious Context .......................................................................... 30 2.7 Summary ........................................................................................................... 32 CHAPTER THREE ............................................................................................................. 34 RESEARCH METHODS .................................................................................................... 34 3.0 Introduction ....................................................................................................... 34 3.1 Forms of Data (Secondary and Primary Data) ................................................... 34 University of Ghana http://ugspace.ug.edu.gh vi 3.2 Ethnographic Field Work .................................................................................. 35 3.3 Population and Sampling ................................................................................... 36 3.4 Interviews .......................................................................................................... 38 3.5 Observation ....................................................................................................... 40 3.6 Photography, Survey and Mapping ................................................................... 40 3.7 Archaeological Field Work ............................................................................... 40 3.8 Field Reconnaissance Survey ............................................................................ 41 3.9 Archaeological Excavations .............................................................................. 42 3.9.1 Locus 1 Trench 1 ............................................................................................ 42 3.9.2 Locus 2 Trench 1 ............................................................................................ 49 3.10 Artifacts Processing ......................................................................................... 54 CHAPTER FOUR ................................................................................................................ 56 SHRINES AND POSUBANS IN KORMANTSE .............................................................. 56 4.0 Introduction ....................................................................................................... 56 4.1 Shrines and Deities in Kormantse...................................................................... 56 4.2 Community Shrines ........................................................................................... 57 4.2.1 Etuadze or Otuadze Shrine ............................................................................. 57 4.2.2 Etsi Baka......................................................................................................... 58 4.2.3 Egya Dadeɛ..................................................................................................... 59 4.2.4 Bosom Akyin ................................................................................................... 60 4.2.5 Nana Kyerekwabena ....................................................................................... 62 4.2.6 Koko Takyiwaa ............................................................................................... 63 4.2.7 Nana Amissah Shrine ...................................................................................... 64 4.3 Clan Shrines ...................................................................................................... 64 4.3.1 Akona Shrine .................................................................................................. 67 4.4 Household Shrine .............................................................................................. 69 4.4.1 Isiw ................................................................................................................. 69 4.4.2 Da Abo Ano .................................................................................................... 71 4.4.3 Akor ................................................................................................................ 72 4.5 The Asafo Company .......................................................................................... 75 4.5.1 The Posuban ................................................................................................... 78 CHAPTER FIVE .................................................................................................................. 82 ANALYSIS OF MATERIAL REMAINS........................................................................... 82 5.0 Introduction ....................................................................................................... 82 University of Ghana http://ugspace.ug.edu.gh vii 5.1 Artefact Inventory .............................................................................................. 82 5.2 Local Ceramics or Pottery ................................................................................. 84 5.2.1 Pottery Fabric Characteristics (Temper and Texture) ..................................... 85 5.2.2 Pottery Surface Finish Characteristics (Surface Treatment) ........................... 86 5.2.3 Surface Decorations ........................................................................................ 87 5.2.4 Vessel Forms .................................................................................................. 90 5.2.4.1 Bowl Form 1 ................................................................................................ 91 5.2.4.2 Bowl Form 2 ................................................................................................ 92 5.2.4.3 Bowl Form 3 ................................................................................................ 93 5.2.4.4 Bowl Form 4 ................................................................................................ 93 5.2.4.5 Jar Form 1 .................................................................................................... 94 5.2.4.6 Jar Form 2 .................................................................................................... 95 5.2.4.7 Jar Form 3 .................................................................................................... 96 5.3 Faunal Remains ................................................................................................. 98 5.4 Botanical Remains ........................................................................................... 105 5.5 Local Smoking Pipe ........................................................................................ 105 5.6 Querns ............................................................................................................. 110 5.7 Metals .............................................................................................................. 111 5.8 Imported Smoking Pipes ................................................................................. 112 5.9 Imported Ceramics .......................................................................................... 117 5.9.1 Creamware.................................................................................................... 117 5.9.1.1 Mocha Decoration ..................................................................................... 117 5.9.1.2 Floral Motif ............................................................................................... 118 5.9.2 Pearlwares .................................................................................................... 119 5.9.2.1 Willow Pattern ........................................................................................... 120 5.9.2.2 Flow Blue .................................................................................................. 120 5.9.2.3 Bouquet Decoration ................................................................................... 121 5.9.2.4 Floral Motif ............................................................................................... 122 5.9.3 Stonewares.................................................................................................... 123 5.9.4 True Porcelain .............................................................................................. 124 5.10 Beads ............................................................................................................. 125 5.10.1 Bauxite Beads ............................................................................................. 125 5.10.2 Plastic Bead ................................................................................................ 126 5.10.3 Mould-pressed Glass Beads ........................................................................ 126 University of Ghana http://ugspace.ug.edu.gh viii 5.10.4 Drawn Beads .............................................................................................. 127 5.11 Glass .............................................................................................................. 128 5.12 Tuyeres .......................................................................................................... 132 5.12.1 Microscopic Analysis of Tuyeres ............................................................... 134 5.13 Slag ................................................................................................................ 140 5.13.1 Microscopic Analysis of Slags ................................................................... 143 CHAPTER SIX ................................................................................................................... 149 INTERPRETATIONS AND CONCLUSION .................................................................. 149 6.1 Introduction ..................................................................................................... 149 6.2 Interpretations of Artefactual Remains from Kormantse ................................ 149 6.3 Evidence of Iron Working at Kormantse ......................................................... 150 6.4 Evidence of Indigenous Technological Activities ........................................... 155 6.5 Food Ways and Subsistence Strategies ........................................................... 156 6.6 Indigenous Medicine ....................................................................................... 159 6.7 Aspect of Indigenous Religion ........................................................................ 159 6.8 Culture Contact ................................................................................................ 160 6.9 Rationale for the Use of Slag and Iron Ore in Shrines at Kormantse .............. 163 6.10 Conclusion ..................................................................................................... 167 Bibliography ....................................................................................................................... 169 APPENDICES .................................................................................................................... 181 APPENDIX 1: Questionnaire 1 ......................................................................................... 181 APPENDIX 2: Questionnaire 2 ......................................................................................... 184 APPENDIX 3: Archaeological Bone Identification ......................................................... 186 APPENDIX 4: Diagnostic bowl and stem fragments retrieved from the Loci ............. 190 APPENDIX 5: Summary of ceramic types excavated from Locus 1 and 2 ................... 193 APPENDIX 6: Table 6.37: Summary description of beads recovered from the excavations .......................................................................................................................... 194 APPENDIX 7: Tuyeres ...................................................................................................... 199 APPENDIX 8: Slags ........................................................................................................... 203 University of Ghana http://ugspace.ug.edu.gh ix LIST OF TABLES Table 3.1: Materials excavated from Locus 1 Trench 1 ............................................. 47 Table 3.2: Materials recovered from Locus 2 Trench 1 ............................................. 52 Table 4.1: A summary of clans at Kormantse and their totems ................................. 66 Table 5.1: Overall artefact inventory from the excavations. ...................................... 83 Table 5.2: Summary of pottery fabric characteristics ................................................. 86 Table 5.3: Summary of surface treatment on potsherds excavated ............................ 87 Table 5.4: Decorations found on the excavated potsherds ......................................... 88 Table 5.5: Identified animal bones from Locus 1 and Locus 2 .................................. 99 Table 5.6: Species of molluscs recovered in Locus 1 .............................................. 101 Table 5.7: Species of molluscs recovered in Locus 2 .............................................. 101 Table 5.8: Habitats of molluscs species recovered from Locus 1 and Locus 2 ........ 104 Table 5.9: Distribution of the attributes of the tuyeres excavate .............................. 133 Table 5.10: Average of the elements obtained from the two regional analyses of the tuyeres from Locus 1 Level 4. (Source: Author, derived from lab results) .............. 137 Table 5.11: Average of the elements obtained from the two regional analyses of the tuyeres from Locus 1 Level 5. (Source: Author, derived from lab results) .............. 137 Table 5.12: Average of the elements obtained from the two regional analyses of the tuyeres from Locus 1 Level 6 (Source: Author, derived from lab results) ............... 138 Table 5.13: An average of the elements obtained from the two regional analyses of the tuyeres from Locus 2 Level 11. (Source: Author, derived from lab results) ...... 138 Table 5.14: Proportion of elements present in all the tuyeres analyzed. (Source: Author) ..................................................................................................................... 139 Table 5.15: Average of the elements obtained from the two regional analyses of the slag. (Source: Author, derived from lab results) ...................................................... 144 Table 5.16: Average of the elements obtained from the two regional analyses of the slag. (Source: Author, derived from lab results) ...................................................... 144 Table 5.17: Average of the elements obtained from the two regional analyses of the slag. (Source: Author, derived from lab results) ...................................................... 144 Table 5.18: Average of the elements obtained from the two regional analysis of the slag. (Source: Author, derived from lab results) ...................................................... 145 Table 5.19: Average of the elements obtained from the two regional analyses of the slag. (Source: Author, derived from lab results) ...................................................... 145 Table 5.20: The percentages of the elements present in all the slags analyzed. (Source: Author, derived from lab results) ............................................................... 146 Table 5.21: X-ray fluorescence analysis of ore samples collected from Kormantse. (Source: Lab results) ................................................................................................. 148 University of Ghana http://ugspace.ug.edu.gh x LIST OF MAPS Map 1: Map of Ghana indicating the location of Kormantse …………..18 Map 2: Distribution of some shrines in Upper Kormantse……………...73 Map 3: Sketch map of the site showing excavated areas in Kormantse…69 University of Ghana http://ugspace.ug.edu.gh xi LIST OF FIGURES Fig. 3.1: An interview with the chief fisherman of Kormantse Number 2 (Field Photo, 2019). .......................................................................................................................... 37 Fig. 3.2: An interview with Asafo Supi (leader) cum Asafo flag maker in Kormantse (Field Photo, 2019) ...................................................................................................... 37 Fig. 3.3: Interview session with JHS three pupils of Hasowodzi Primary and JHS (Field Photo, 2019) ..................................................................................................... 39 Fig. 3.4: Interview with three elderly women at Lower Kormantse Number 1 (Field Photo,2019) ................................................................................................................ 39 Fig. 3.5: A bottle of schnapps and an undisclosed amount of money being presented to the chief through the linguist (Okyeame) of Kormantse Number 1(Field Photo, 2020). .......................................................................................................................... 42 Fig. 3.6: Slags and potsherd seen during the surface survey (Field Photo, 2020) ...... 43 Fig. 3.7: Surface finds of imported ceramic and bottle (Field Photo, 2020) .............. 43 Fig. 3.8: Setting of the North-South baseline for gridding Locus 1 (Field Photo, 2020) .................................................................................................................................... 44 Fig. 3.9: Laid out test pit of Locus 1 (Field Photo, 2020) .......................................... 45 Fig. 3.10: The sterile level of Locus 1 Trench 1 (Field Photo, 2020). ....................... 46 Fig. 3.11a: Stratigraphic profiles of the South wall of Locus 1 Trench 1 .................. 48 Fig. 3.11b: Stratigraphic profile of the East wall of Locus 1 Trench 1 ...................... 49 Fig. 3.12: Excavation and recording of Locus 2 (Field Photo, 2020)......................... 52 Fig. 3.13a: Stratigraphic profiles of the South wall of Locus 2. ................................ 53 Fig. 3.13b: Stratigraphic profiles of the East wall of Locus 2. ................................... 54 Fig. 4.1: Etuadze or Otuadze shrine at Upper Kormantse Number 2 (Field Photo, 2020) ........................................................................................................................... 58 Fig.4.2: Etsi baka shrine at Upper Kormantse Number 2 (Field Photo, 2020) .......... 59 Fig. 4.3: Egya dadeɛ at Upper Kormantse Number 2 (Field Photo, 2020) ................. 60 Fig. 4.4: Slags and other stones in the circular walled shrine (Field Photo, 2020) .... 60 Fig. 4.5: Bosom Akyin at Upper Kormantse Number 2 .............................................. 61 Fig. 4.6: Shrine at Asafo meeting ground (Field Photo, 2020) ................................... 62 Fig. 4.7: Nana Kyerekwabena with a pot embedded in the door step (Da abo ano) (Field Photo, 2020) .................................................................................................... 63 Fig. 4.8: Koko Takyiwaa shrine (Field Photo, 2020) .................................................. 64 Fig. 4.9: Depiction of Nsona clan totem on the wall of the family house (Field Photo, 2020). .......................................................................................................................... 66 Fig. 4.10: Rectangular wall around Akona shrine (Field Photo, 2020) ...................... 68 Fig. 4.11: Iron slags and ores deposited in the Akona shrine (Field Photo, 2020) ..... 68 Fig. 4.12: Akona shrine with schnapps and an embossed pot (Field Photo, 2020) .... 69 Fig. 4.13: Isiw with three crossed incisions (Field Photo, 2020) ................................ 70 Fig. 4.14: Two isiw with plain surface (Field Photo, 2020) ....................................... 70 Fig. 4.15: Isiw with three crossed incisions and an orifice on the left (Field Photo, 2020) ........................................................................................................................... 71 Fig. 4.16: Slag collection called Da abo ano (Field Photo, 2020) ............................. 72 University of Ghana http://ugspace.ug.edu.gh xii Fig. 4.17: Slag and cement-built shrine with pot embedded in the middle called Akor, around it are different drink bottles (Field Photo, 2020) ............................................ 73 Fig. 4.18: Bosom bukyia (ritual hearth) (Field Photo, 2020) ...................................... 73 Fig. 4.19: Asafo drummers and dancers at a funeral (Field Photo, 2020) .................. 77 Fig. 4.20: A hand sewn Asafo flag (Field Photo, 2020). ........................................... 78 Fig. 4.21: Asafo posuban at Upper Kormantse Number 2 (Field Photo, 2020) ......... 79 Fig. 4.22: Asafo posuban at Upper Kormantse Number 2 depicting Nana Adaaban (Field Photo, 2020). .................................................................................................... 80 Fig. 4.23: Asafo posuban at Kormantse Number 1 (Field Photo, 2020) .................... 81 Fig 5.1: A bar graph representation of the vessel parts recovered .............................. 85 Fig. 5.2: Sample of decorative motifs on pot sherds recovered .................................. 89 Fig. 5.3: Halfmoon crescent and incision decoration ................................................. 90 Fig. 5.4: Sherds with multiple decorations ................................................................. 90 Fig. 5.5: Bowl Form 1 ................................................................................................ 91 Fig. 5.6: Variety of Bowl Form 2 ............................................................................... 92 Fig. 5.7: Bowl Form ................................................................................................... 93 Fig. 5.8: Variety of Bowl Form 4 ............................................................................... 94 Fig. 5.9: Jar Form 1 .................................................................................................... 95 Fig. 5.10: Variety of Jar Form 2 ................................................................................. 95 Fig. 5.11: Variety of Jar Form 3 ................................................................................. 96 Fig. 5.12: Sample of molluscs recovered from the excavations ............................... 103 Fig. 5.13: Cypraea sp from Locus 2 ......................................................................... 103 Fig. 5.14: Palm kernels recovered from Locus 1 ...................................................... 105 Fig. 5.15: Local smoking pipe bowl with triangular stamps (Left) .......................... 107 Fig. 5.16: Local smoking pipe bowl with a pointed end ........................................... 107 Fig. 5.17: Local smoking pipes................................................................................. 108 Fig. 5.18: Local smoking pipe with square base and round projections at the corner .................................................................................................................................. 109 Fig. 5.19: A brown burnished local smoking pipe .................................................... 109 Fig. 5.20: Querns obtained from the excavations ..................................................... 111 Fig. 5.21: Metals recovered from the excavations ................................................... 112 Fig. 5.22: Smoking pipe stem fragments recovered from the excavations ............... 113 Fig. 5.23: A smoking pipe with the inscription DA .................................................. 114 Fig. 5.24: Gryphon relief on a smoking pipe ............................................................ 114 Fig. 5.25: The heel of a smoking pipe with a GOUDA shield (left) and a crowned L (right). ....................................................................................................................... 116 Fig. 5.26: A bowl of a smoking pipe with floral motif ............................................. 116 Fig. 5.27: Creamware with black and light brown annular bands with mocha design .................................................................................................................................. 118 Fig. 5.28: Creamwares with floral motifs ................................................................. 119 Fig. 5.29: Blue and white willow printed pearlware ................................................ 120 Fig. 5.30: Flow blue designs ..................................................................................... 121 Fig. 5.31: Pearlware with bouquet decoration .......................................................... 121 University of Ghana http://ugspace.ug.edu.gh xiii Fig. 5.32: Pearlware with hand painted floral motif ................................................. 122 Fig. 5.33: Pearl ware with polychrome hand painted decorations. ........................... 122 Fig. 5.34: German stone wares ................................................................................. 124 Fig. 5.35: Floral decorated porcelain ........................................................................ 125 Fig.5.36: Varieties of beads retrieved from the excavation. ..................................... 128 Fig. 5.37: Bottomed bottle from Locus 1 ................................................................. 129 Fig. 5.38: A medicine bottle from Locus 2 ............................................................... 131 Fig. 5.39: A medicine bottle from Locus 2 ............................................................... 131 Fig. 5.40: Ointment container ................................................................................... 132 Fig. 5.41: Samples of unvitrified tuyeres .................................................................. 133 Fig. 5.42: Tuyere with slag in its interior. ................................................................ 134 Fig. 5.43: Some of the vitrified tuyeres recovered. .................................................. 134 Fig. 5.44: Tuyeres cut in cross section and mounted on stub ................................... 136 Fig. 5.45: Some slags recovered from the excavations ............................................ 142 Fig. 5.46: Some slags recovered from the excavations ............................................ 148 University of Ghana http://ugspace.ug.edu.gh xiv ABSTRACT Memories are preserved for future generations and most often oral history and traditions are its conduit. In this study, it has been observed that African Traditional Religion shrines apart from serving as dwelling places of the spirits and deities, serve as a memory bank for past iron working activity in Kormantse, a coastal settlement in the Central Region of Ghana. The persistent use of slags and iron ore in the religious context in Kormantse is analogous to some religious practices in some past iron working societies in Africa. This suggests that not only was the iron working activity enmeshed in rituals but the by-product from the iron working activity. Furthermore, the discovery of Kormantse as a past iron working society has added to the number of known iron working sites in the Central Region of Ghana and makes this work a precursor to further studies of the extent of the industry in Kormantse and the Central Region as a whole. The study employed ethnography, archaeology and earth science approaches to derive the data presented in this thesis. University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE BACKGROUND OF THE STUDY AREA 1.0 Introduction This research seeks to investigate the significance of iron slag and ore shrines to the people of Kormantse as well as to substantiate or refute the claim of a past iron-working industry in the settlement as suggested by Agorsah and Butler (2008). The name Kormantse which comes in various spellings such as Cromanti, Kromantin, Cormantine, Coromantee (Agorsah, 2014) became a household name during the trans-Atlantic trade. The general history of the settlement is documented in the light of its trans-Atlantic encounter with the various European nations beginning from the 17th Century. Kormantse appears to have existed prior to the European arrival in the area (Agorsah & Butler, 2008), yet little is documented about the pre-European period of Kormantse. Archaeological work conducted at the site by Agorsah & Butler (2008) offered some clues to pre-Atlantic Kormantse. The work revealed the prevalence of piles of iron slag and ore all over the site some of which have been used by the inhabitants as shrines. Shrines are indicators of African Traditional Religion across African societies, however the particular shape or form it takes is site specific. Shrines in Africa are cultural sign post, place of worship and symbolic vessels of the spirits of ancestors and deities who must be regularly placated, and petitioned for blessings (Allman & Parker, 2005; Apoh & Gavua, 2010; Field, 1939). Aside from their spiritual significance, shrines may be considered as a physical mnemonic that aid in recounting the past (Mather, 1999). The material make-up of shrines are varied and these materials hold both physical and spiritual connotations. At Kormantse, iron slag and iron ore are common materials used in majority of the shrines. Ores are the raw material employed in iron working from which the end product is a bloom and the by-product derived from it is a slag. Slags are regarded as waste while the bloom is further worked on to be furnished into tools and University of Ghana http://ugspace.ug.edu.gh 2 implements. These materials are observed in shrines belonging to the entire community and individual households. The dominance of these materials in majority of shrines in Kormantse is peculiar and worth studying. This observation is the first of its kind in Ghana. An archaeological enquiry bolstered by ethnographic studies and scientific analysis was therefore undertaken to comprehend the phenomenon observed in Kormantse. Considering the variant spellings of the site name in historical literature, this study uses the name Kormantse in accordance with the way it is spelt by the Mfantseman Municipal Assembly when providing primary information except where referenced sources are cited. 1.1 The Study Area Kormantse is a hill top settlement found in the Central Region of Ghana. It lies on longitude 05°12ˈ256ˈˈW and latitude 01°04ˈ85ˈˈN and is one kilometer long and half kilometer wide. It is located at an elevation of 80 meters above sea level (web site 1) (Map 1). This settlement together with other communities such as Mankessim, Essarkyir, Dominase, Anomabu, Abandze, Otuam, Narkwa and Yamoransa are part of the communities that make up the Mfantseman Municipal Assembly (web site 2). Saltpond is the administrative seat of the Municipal Assembly and is bounded on the west by the district of Abura-Asebu- Kwamankese, in the northeast by the district of Ajumako-Enyan-Essiam, in the east by the district of Ekumfi and in the south by the Gulf of Guinea (the Atlantic). The district extends from Eguase to Mankessim, which is the Municipal Assembly's most western and eastern area respectively. University of Ghana http://ugspace.ug.edu.gh 3 Map 1: Map of Ghana indicating the location of Kormantse. Source: Centre for Remote Sensing and Geographic Information Services (CERSGIS) University of Ghana, Legon, 2020 Kormantse is segmented into Upper (Old Kormantse) and Lower Kormantse (New Kormantse). Old Kormantse which is on a hill is gradually being abandoned for the New Kormantse site located along the coastline waterfront. The two main groups in the community are the Bentsir and Nkum, each of which claims ownership and control of one University of Ghana http://ugspace.ug.edu.gh 4 of the two areas of the settlement (Agorsah & Butler, 2008). The groups in the settlement reside in a clearly demarcated area of the settlement of both Upper and Lower Kormantse. Thus, the Bentsir are found in Upper and Lower Kormantse No.1, and Nkum in Upper and Lower Kormantse No. 2. The people of Kormantse are Fante speakers although they claim the original inhabitants or first settlers were the Etsi people. Shumway (2011) intimated that the Etsi were one of the first settlers on the Fante Coast. Fante is the principal Akan language spoken in Central Region of Ghana which is bordered on the west by Ahanta speakers and in the East by the Ga and Guan (DeCorse, 2001) and North by Kasem, Gonja and Dagbani speakers. The Fante are part of the Akan or Twi speaking group which belongs to the Kwa sub family of the Niger Congo family of language as classified by Joseph Greenberg (1949). The settlement has a population of about 8,501 people according to the 2010 Ghana Statistical Service census. In recent years, the Kormantse landscape has been disturbed by activities of local gold miners who pillage the site for antique materials and gold, which disturbs the archaeological record. Upper Kormantse is surrounded on all sides by a deep depression except to the north- west. The depressions surrounding the hill area are characterized by compact clay soil some of which contain mica inclusions. The hill area has soil which is loamy whereas the settlement at the Atlantic waterfront has loose quaternary sandy soil (Ewusi & Kuma, 2011). The land is rich in iron ores. Soils in this region are classified under the Savanna ochrosol soil group which is greyish brown, sandy loam (ibid). The Central Region within which Kormantse is situated is underlain by the Cape Coast Granitoid Complex (CCGC) which is post-Tarkwaian. The granitoids consist of well-foliated, sometimes magmatic, University of Ghana http://ugspace.ug.edu.gh 5 potash-rich rocks and come in medium-grained varieties of muscovite–biotite granite, granodi-orites, porphyroblastic biotite gneiss, aplites, and pegmatites (Kesse, 1985). The Municipal Area has an average temperature of 24o C with relative humidity of about 70 percent. Kormantse is located in the Coastal Savanna ecozone which has double annual rainfall; May-June with peaks in July and September to October with dry seasons usually from December to February and from July to September (web site 2). The vegetative cover of the region within which the study area is found is moist semi-deciduous forest (Ewusi & Kuma 2011). The study area consists of dense scrub tangle and grass and pockets of forest which are remnants of the past forested area. The area especially in the depression has tall trees and some grasses. Drainage features in the study area include Otsi lagoon and the coastal water front (web site 2). The settlement history of Kormantse was narrated differently by the two groups in the settlement, namely the Nkum and the Bentsir. Nonetheless a line of similarity exists in both accounts. Each group claims to have emigrated from or passed through Techiman (the ancestral home of the Fante people). Both groups also claim to have been led by a male leader(s) with similar name(s). According to both the Nkum and Bentsir people, the name Kormantse is derived from two words Kormir or Korme and Antsi. Kormir or Korme was the name of the hunter who led both groups from Techiman to discover the hilltop settlement. The Bentsir group claim they began their journey from Sudan and passed through Ngedum before reaching Techiman and later proceeded to their present location. Oral account has it that Kormir or Korme in his search for a place of abode for his people chose an area that could provide a panoramic view of low-lying areas so as to spot University of Ghana http://ugspace.ug.edu.gh 6 possible intruders from beneath. This compelled him to choose the present settlement which is on a hill. On arrival, Korme found that the area was heavily forested and was habitat for various kinds of wild animals. He then killed an elephant and set it afire at a spot which is presently a dried pond referred to as Komnsu. It was also revealed in other narrations that he discovered the settlement with another group of people, but the latter deserted the place due to the dense and impenetrable nature of the vegetation. The news of the discovery of the hilltop settlement was announced to the rest of the population who were then camped at different locations within that enclave. However, the other group that discovered the settlement with Korme or Kormir were not elated with Kormir’s decision to settle there. The group therefore retorted Kormir antsi….which translates as ‘Kormir didn’t heed our advice that the place is inhabited by wild animals’. The settlement subsequently assumed the name Kormir-antsi which later become Kormantse. The Nkum group asserts that their ancestors were led by two leaders named Kormir Panyin and Kormir Kakraba. During their search for a place of abode on the hill, Kormir Panyin lost sight of his partner (Kormir Kakraba) and shouted his name several times without response until he discovered him under a tree and exclaimed to him Kormir didn’t you hear me calling you? (‘kormir antsi?’). Other accounts recount that both Kormir Panyin and Kormir Kakraba settled on the landscape for a while, after which Kormir Panyin decided to leave the settlement with his brother. However, Kormir Kakraba refused to accompany him. Soon the Asante invaded the settlement and Kormir Kakraba was arrested. So Kormir Panyin was aggrieved when he heard of that incident and exclaimed “if Kormir Kakraba had listened” (kormir antsi), he would not have been captured and it was abridged as Kormantse. Some sources also claim that the name was University of Ghana http://ugspace.ug.edu.gh 7 derived from the Asante warriors who invaded the settlement. They exclaimed ‘Me kɔe manti’ when they returned from the war and were asked what ensued during the war literally meaning, I partook in the war but I did not hear or I do not know the outcome of the war. This was said to conceal their defeat (Asante’s defeat) from their people. Both groups, the Nkum and Bentsir claim to be the first settlers and were present prior to the arrival of the Borbor Fante who were led by Oburumankuma, Odapagyan and Osͻn (the three ancestors that led the Borbor Fante). However, Shumway (2011) intimated that the Borbor or Boribori Fante migrated southward from the Akan heartland (Techiman) around the 14th Century CE and are the ancestors of the modern-day Fante speakers. They spoke a language that was the progenitor of the modern Fante language. The Fanteland prior to 1700 was home to people speaking Guan, Etsi and Akan languages (Shumway, 2011, p.13). Both Nkum and Bentsir are Fante speakers, but both claim to be Etsis as well. Etsi were presumably one of the early languages on the Fante Coast hence the ‘fight’ among the two groups in Kormantse for prominence as the early settlers. However, the Fantes are an amalgamation of cultural traditions of the Akan, Etsi, Asebu, Fetu, Eguafo and Agona (Shumway, 2011, p.15). The Etsi language is now defunct because the Etsi were subdued as a result of series of war with the Asebu, Afutu and Assin in the1660s (Ajayi, Espie & Dike, 1965, p.162) or most probably assimilated into other Akan speaking groups. 1.2 Kormantse in the Context of Fante History The Fante people are known to inhabit both the forest and coastlands of the Central Region of Ghana. Prior to European contact, there were settlements scattered along the coast and throughout the adjacent hinterland. This pattern of settlement probably University of Ghana http://ugspace.ug.edu.gh 8 extended at least 1000 years or much earlier as gleaned from archaeological data (DeCorse, 2001, p.18). Pre-Atlantic Fante coast comprised hilltop and lowland sites some of which had entrenchments surrounding the settlements for defense purposes (Chouin & Decorse, 2010, p. 125). The Fantes are identified as Akans inhabiting greater part of the southern half of Ghana (Christensen,1958; Ellis,1887; Law, 2010; Priestly,1969). Prior to the arrival of Europeans, the general Akan ethnic group in the forest areas lived in widely scattered small chiefdoms or villages based on kinship (Daaku, 1964, p.4; DeCorse & Spiers, 2009, p.30) and were characterized by similar social organization. According to Hornsey (1979, p.98), the Fantes have never been organized under a single political entity but were composed of a number of independent chiefdoms. The family was the linchpin of the pre- Atlantic Fante society, each of which was headed by the family head who was subservient to the chief of the society (Daaku, 1964, p.4). Although there were state boundaries, movement of people were unrestricted on account that family clan membership transcended tribal and political boundaries (Daaku, 1964, p.5). The Akans (Fantes) in the forest and the coast of Gold Coast were involved in long distance trade in gold and kola nuts to Western Sudan (Daaku, 1964, p.3-4) and intra trading activities long before the advent of the trans-Atlantic Trade. The Fante coast was also dotted with small fishing, palm wine tapping and salt producing villages such as Anomabo, Kormantse and Cape Coast, as well as agriculturalists who relied on shifting hoe cultivation and fallowing (Arhin, 1995; Daaku, 1970, p.2; DeCorse, 2001, p.18-19; Fynn, 1987, p.108). According to Chouin and DeCorse (2010, p.123), data from the Central Region University of Ghana http://ugspace.ug.edu.gh 9 Archaeological Project and a number of archaeological researches in southern Ghana point to the existence of substantial settlement of agricultural population before the European trade. Archaeological excavations carried out in the Central Region of Ghana have yielded fragments of slags and pieces of iron objects in areas such as Akrokrowa (Chouin & DeCorse, 2010), Elmina (Decorse,2001), Coconut Grove (DeCorse, 2005), Beifikrom near Mankessim (Nunoo, 1970), and the border between Apewosika and Amamoma (Penfold, 1971), which is indicative of iron working activities on the Fante coast. Radiocarbon date from Akrokrowa earthwork site for instance suggests that the site was occupied around the 2nd half of the 1st millennium CE. This is most likely during the 8th Century which is long before the first European landed on the Elmina Coast (Chouin & DeCorse, 2010, p.138). The other sites mentioned above have been speculated to be pre-Atlantic iron working sites based on their stratigraphy and the absence of European materials. Additionally, in terms of religion, ancestor worship has been a prominent feature in the daily life (Welma, 1930, p.18) of the Akans in the southern coast of Ghana. The Fantes before the spread of Christianity believed and still believe in a Supreme Being, lesser deities and ancestral spirits (Abosom) who normally inhabit natural objects such as the sea, rocks, rivers and in trees (Welma, 193018). Shumway (2011, p.133), intimated that the abodes of the spirits or shrines specifically that of the Nananom Mpow (sacred grove) which predates European arrival played central roles in the ordinary lives, households and communities and in the development of the Fante culture. Kormantse existed before the coming of Europeans on account that it served as a rallying point and outlet of both the trans-Saharan and trans-Atlantic trades (Agorsah & Butler, 2008, p.1). However, University of Ghana http://ugspace.ug.edu.gh 10 historical literature is silent on the socio-political organization of Kormantse prior to the arrival of Europeans on the Gold Coast now Ghana. The available information on the pre- Atlantic nature of the Fante coastland offers scanty details on the study area. The Portuguese were the first Europeans to arrive in the Gold Coast in the 15th Century for trading purposes. Other European nations such as France, the Netherlands, Britain, Denmark, Sweden, and Brandenburg joined the trade at different times (Dickson, 1969; Fage, 1969). This trade which was material in nature at its inception was later replaced with the inhumane trade in enslaved people to feed the plantations in the New World. Two of these nations that had interactions with Kormantse are Britain and the Netherlands. Britain was one of the European nations that traded with the Gold Coast in the seventeenth century. The British built the first English trading post in Little Kormantin which served as the headquarters of the British trade in 1631 and occupied it till 1665. The fort was named Fort Kormantin after an existing settlement which was on a hill north-east of the edifice (Fage, 1969). The Dutch, in February 1665 (thus during the second Anglo-Dutch war), dispossessed the British of the fort owing to the lucrative nature of the trading activities there. The new occupants of the fort renamed it Fort Amsterdam. However, in 1867 when the slave trade had been abolished and profits were on decline, the English regained possession of the fort. The area within which the fort was built according to Green (1745) was called Little Kormantin or Cormantyn (Bosman, 1721) to distinguish it from Great Kormantin; a settlement on a high hill which is a ‘cannon shot below’ Fort Amsterdam (Green, 1745). Bosman (1721) described little Cormantyn as so small and wretchedly poor that it did not deserve their attention, but Great Kormantin was so large and populous that it deserved its University of Ghana http://ugspace.ug.edu.gh 11 name. This settlement (Great Kormantin) together with towns such as Axim, Shama, Komenda, Elmina, Cape Coast, Anomabu and Accra was described as larger and more prominent of the coastal settlements (Dickson, 1969). Great Kormantin was categorized as a major trading center during the trans-Atlantic slave trade (Dickson, 1969). Currently, Little Kormantin is known as Abandze or Kormantse number 3 whereas Great Kormantin still remains the same, but spelt as Kormantse, and is the focus of this research. 1.3 Research Problem Kormantse is replete with different forms of shrine, especially in Upper Kormantse. The area has held strongly on to the African Traditional Religion for a long time. Shrines according to Parker (2011) are of varied array of materials. However, in Kormantse, iron slag and ore are the prevalent materials in majority of shrines in the old Kormantse hilltop settlement (Upper Kormantse). This observation was brought to light by Agorsah and Butler (2008) in the work “Archaeological Investigation of Historic Kormantse: Cultural Identities.” Surprisingly, this observation has not been documented in the literature even though Kormantse has received substantial attention in European literature. No researcher has also found interest in probing this phenomenon and its implication for unearthing the past of the settlement. This research therefore, used ethnographic and archaeological data to help understand the cultural transformation in the utilization of iron slag and ore in shrine context in Kormantse. 1.4 Aim and Objectives The aim of this research is to study and understand the use of slag and iron ore in shrine context at Kormantse. The objectives of the research are to:  Establish the presence of a past iron working industry and the uses of iron University of Ghana http://ugspace.ug.edu.gh 12 working remains in Kormantse.  Document the types, function and configuration of the slag and iron ore shrines at Kormantse.  Examine the rationale for the use of iron slag or ore in creating shrines. 1.5 Research Questions Key among the questions that guided the study are:  What is or are the evidence of iron working and its remains in the past and at present?  What types of shrines are found in Kormantse?  What role(s) does ore and iron working remains play (s) in religion and shrine formations at Kormantse? 1.6 Material Culture Studies This study benefits from material culture studies to interpret the findings of the research. Objects created (historical or pre-historical) are relics that continue to exist in the present and serve as a window to understanding and interpreting the past in many ways. The material forms or objects which represent the daily life of past people, in this case the people of Kormantse, find their way into the archaeological record as artefacts or material culture. These material forms have intimate relationships with the society within which they are created (Deetz, 1977; Oestigard 2004; Prown 1982). The relationship can be gleaned from the objects regardless of temporality (past or present), space and scale (local or global) through material culture studies (Miller and Tilley, 1996, p.5). Material culture studies evolved from the fields of anthropology, sociology and archaeology. It is defined as the investigation of material culture through which the beliefs, values, ideas, attitudes and assumptions of a particular community or society at a given time can be ascertained (Prown, 1982, p.1). According to Yonan, (2011, p.232), material culture studies are the University of Ghana http://ugspace.ug.edu.gh 13 most reliable way of reconstruction of long lost or inaccessible ways of life or importantly, information written records have been silenced on. It is upon this premise that material culture studies is employed in this study to bring to light an aspect of the past of Kormantse that literature has not touched on. Material culture as a facet of social morphology can be subjected to observation and description which reveals the form and understanding of people within the society from which they were recovered (Miller, 1994; Patnaik, 1995). Specific material culture mostly man-made recovered from a society indicates the existence of human intelligence operating at the time of fabrication (Prown, 1982, p.1). Material culture is thus, the tangible aspects of cultural interactions with the environment and other available resources (Patnaik 1995, pp. 59-64). Artefacts may embody different kinds of inherent and attached values, namely spiritual, economical, and technological, based on the context. The archaeological context gives direct sensory experiences of surviving pre- historical and historical events (Parrott, 2001) and in cases of persistence in use of the exact artefacts in the ethnographic context, may offer a clear picture of its value(s) especially those that are difficult to glean from the archaeological record for instance spiritual values. This offers a unique source of knowledge about the contemporary settlement of Kormantse and its ties to the past (Rathje, 1979, p.29). Furthermore, material culture studies reflect the physical evidence of cultures that are made by people and which are attributed to the present or the past of that particular society. Renfrew (2004, p. 23) aptly posits that materials have histories and that the memory of that particular history is evoked when the material surfaces, through archaeological excavations. The concept of material culture underscores the importance University of Ghana http://ugspace.ug.edu.gh 14 of archaeology in the reconstruction of history especially, long lost ones and may serve as a supplement to a known history. A common theme connecting interpretations of material records of the past is how people presently engage with the cultural expressions in their daily interactions (Beaudry, Cook & Mrozowski, 1991). The persistent use of the same material forms amongst the contemporary society advertently or inadvertently serves as a memory bank. These materials serve as window into the past and an avenue to recover more from the archaeological record. These artifacts in the contemporary society are more important in those societies that have no other means of transmitting information and preserving memory apart from oral communication (Kus and Raharijaona, 1990). These ethnographic observations can be explicated as perpetuation of meaningful ancestral ways of doing things. Material culture studies has been employed by several archaeologists in Ghana. Apoh and Gavua (2010) applied material culture studies to interpret archaeological assemblages retrieved at the Katamansu site in Ghana. The interpretive framework helped to shed light on past events and occurrences which transpired at the site. In the mid-west area of the Volta Region, Ghana, Gavua (2015, p. 137) combined ethnographic research and material culture studies to better understand the traditional religious identity and practices of the people. Material culture studies is the lens through which the physical evidence from the archaeological record in Kormantse is used to better understand the ethnographic observation in the contemporary society. University of Ghana http://ugspace.ug.edu.gh 15 1.7 Significance of the Study This research is significant because no research in Ghana has focused on the use of iron working remains in shrine context. It contributes to archaeological knowledge of past iron working history in the Central Region of Ghana, and specifically the important roles slags and iron ores played and continues to play in the religious context of Kormantse’s past and present. It is also significant because it highlights the important role shrines play in preserving aspects of a society’s past. 1.8 Challenges The disturbance of the site by erosion and illegal gold mining (galamsey) by the local people made it difficult to settle on an area to excavate. The effect of galamsey operations far exceeds that of erosion as a vast stretch of land in Kormantse have been disturbed archaeologically. The activities of galamsey operators on the landscape is evident as a lot of potsherds and other materials are scattered on the surface of the land. Erosion has also exposed some of the buried cultural materials but this is attributable to the sloping nature of the land and as such cultural materials were seen to be out of context. The ethnic rivalries between the two main groups required me to be circumspect in the questions I posed during the interviews since experience from the first phase of the interview showed that some questions were subjects of contention in the settlement. For instance, issues on the settlement history of the people. This limited the extent to which certain insights about the settlement were elicited. Restricted access to some of the shrines and prohibitions on photography posed another challenge. Shrines are sacred places and for that matter having access to them required the performance of certain rites and sometimes were not open to even members of the community. This deprived me of the opportunity to document all the shrines in the University of Ghana http://ugspace.ug.edu.gh 16 settlement. 1.9 Organization of Chapters The thesis is segmented into six chapters. The First Chapter provides an introduction, background to the study area, a description of the study area as well as the aim and objectives of the research that was conducted. It also provides the interpretive framework that guided the interpretation of materials recovered from the excavation that was conducted. The Second Chapter reviews literature on the Central Region of Ghana prior to and during the Atlantic trade as well as previous archaeological works conducted on the study area. Also, literature on the composition and significance of shrines is provided. Lastly, the chapter looks at iron slag in religious context and chemical analysis employed in the study of remains of iron working activity. Chapter Three homes in on the methods that were employed in conducting this research. Chapter Four is a documentation of the contemporary life ways of the people of Kormantse by focusing on the various iron slag and ore shrines in the settlement. While Chapter Five provides the analyses of the materials recovered from the excavations as well as chemical analysis of the remains of iron working activity. Chapter Six provides the interpretation of materials recovered, which is a synthesis of the analyses of materials excavated and the ethnography conducted. Finally, conclusion of the research is presented in this chapter. University of Ghana http://ugspace.ug.edu.gh 17 CHAPTER TWO THE CENTRAL REGION REVIEWED FROM HISTORY AND ARCHAEOLOGY 2.1 Introduction This section reviews literature on the state of the Fante Coast prior to the Trans-Atlantic trade, changes the Trans-Atlantic trade wrought on the Fante Coast and then reviews archaeological works conducted at Kormantse. The issue of shrines and their composition is addressed as well as detailing archaeological evidence of reuse of slags in religious contexts. Additionally, review of literature pertaining to chemical composition analysis of iron remains and its products are presented in this chapter. 2.2 Central Region Prior to the Trans-Atlantic Trade Scholarly literature have thrown light on the nature of 17th century Southern Ghana focusing on state formation of political entities. However, the nature of early or pre-15th century sociopolitical organizations of Southern Ghana remains scanty and elusive (Chouin & DeCorse, 2010, p.124). Archaeology has filled this gap by providing insights of societies prior to the introduction of Western form of writing. Archaeological data obtained points to the evidence of well settled agricultural communities in the Ghanaian forest prior to the advent of the Atlantic trade. In particular, Vivian’s (1996) and Shinnie’s (2005) data from Asante indicate the existence of stable village communities as early as 800 CE. Additionally, data emerging from Central Region suggest a pre-Atlantic settlement density that exceeds the post 1500 CE period. These settlements include hilltop and lowland sites of various sizes as well as entrenchments all with substantial evidence for iron production. Sites with entrenchments are the most visible of these pre-Atlantic settlements. Chouin and DeCorse (2010), as part of the Central Region Archaeological Research Project, University of Ghana http://ugspace.ug.edu.gh 18 conducted an archaeological work at the Akrokrowa earthwork site located near Abrem Berase, with the aim of evaluating the age of the earthwork. Earthworks are monumental trench systems; an enclosure often roughly elliptical composed of a deep ditch and raised earthen banks. Several of such have been found in the Birim valley and Kokobin (Chouin & DeCorse, 2010). Six samples of charred palm kernels, charred wood, and palm kernel recovered from the Akrokrowa earthwork site were subjected to radiocarbon dating to yield dates that were compared with dates from Kiyagaa Mulindwa’s (1979) samples from the Monsa and Batabi sites in the Birim valley. The samples dated place the construction of the Akrokrowa earthworks in the 8th / 9th Century which is consistent with the radiocarbon dates obtained for deposits found in the ditch. The Akrokrowa radiocarbon dates point to the construction of the entrenchment system in the 2nd half of the 1st millennium CE, most likely during the 8th century long before the arrival of the Europeans (Chouin & DeCorse, 2010). These results coincide nicely with samples (S1-2718) collected for dating by Kiyaga -Mulindwa in the ditch of Monsa (Chouin & DeCorse, 2010, p. 138). According to Chouin and DeCorse (2010) the fact that the sites belong to the same historical period radically changed the perception of socio- political and economic life in the forest of Southern Ghana before the commencement of the Atlantic trade. Earthworks are believed to have been part of a settlement system. The amount of work involved in their construction suggests a social organization capable of mobilizing, controlling and coordinating labor, and relatively sedentary settlement over long periods of time (Chouin & DeCorse, 2010, p.138). The excavation at the Akrokrowa site yielded large quantities of fragments of palm-nut and fragments of oil lamps suggesting large scale University of Ghana http://ugspace.ug.edu.gh 19 cultivation of palm trees. The inhabitants of the earthwork communities also made iron tools on account of the discovery of slag as well as pieces of iron objects. Other finds were quartz tools, polished axe and stone beads. Iron and iron technology were crucial factors in the process of digging earthworks, the latter being tangible evidence of the working power of this metal and at the same time a display of prestige for the elite who could demonstrate their ability to accumulate and control kin and followers (Chouin & DeCorse, 2010). The nature of the political systems of which the earthworks were part could not be ascertained from the excavations. However, their construction suggest a type of social organization capable of mobilizing material and human resources to build huge embankments (Chouin and DeCorse, 2010). Evidence of iron smelting and smithing were found in all excavated earthworks in the form of slag and fragments of iron. The slags suggest that iron was smelted locally. The excavations also produced early ceramics and no evidence of reoccupation after its abandonment. The function of earthworks has become a subject of debate amongst scholars. Some scholars such as Davies (1961), Wilks (1957), Kiyaga-Malindwa (1979) have posited that earthworks functioned as fortifications, fosse, banks etc. According to Chouin and DeCorse (2010), earthworks were settlements and trenches that marked the limit of these settlements, evidenced by the absence of archaeological deposits outside the area delimited by the ditch at the Akrokrowa earthwork. Chouin and DeCorse (2010) asserted that the earthworks could have served as a social boundary. That is, to raise a community or elite groups to a higher status in a larger sociopolitical system. This social differentiation could explain the presence, during the same period, of small settlement without earthworks. The chronology of Akrokrowa is in agreement with an abandonment of the site by the mid-14th century. This is University of Ghana http://ugspace.ug.edu.gh 20 in turn amplified by data coming from other earthwork sites in Southern Ghana suggesting a major change in settlement patterns before the commencement of the Atlantic trade characterized by the abandonment of earthworks (Chouin & DeCorse, 2010, p.143) possibly following a deadly plague. People abandoned their settlements and dispersed across the farmlands before reorganizing themselves according to new patterns. The advent of people from the Atlantic world may have coincided with the aftermath of this epidemic (Chouin & DeCorse, 2010). Archaeological works by DeCorse (2005), Penfold (1971) and Nunoo (1970), touched on the technological aspect of the societies in Central Region before the coming of the European. In “Coastal Ghana in the First and Second Millennia AD. Change in Settlement Patterns, Subsistence and Technology”, DeCorse (2005) brought to bear a possible past iron working site at the Elmina area. However, little insights could be gleaned from it since the research was not extensive. The site, Coconut Grove, which is located on a low elevation above the shoreline, four kilometres west of Elmina township revealed iron slags amongst other remains through excavation. A possible furnace base was identified, but it was ruled out as an iron furnace on grounds that iron slags and tuyeres were not found in association with the feature. Hence, it was considered to be a fish smoking oven. A chemical composition analysis of the remains would have revealed the microscopic characteristics of the furnace and could have been helpful in ascertaining its original function. Apart from this Coconut Grove site, slag deposits occur on several Central Region sites, including meter-deep slag mounds north of Komenda and Fomena. Penfold (1971) conducted excavations on a hill three miles west of Cape Coast. The town, which is bordered by Apewosika and Amamoma, was inhabited by a group of people for a University of Ghana http://ugspace.ug.edu.gh 21 short period after which the University of Cape Coast levelled it to make way for the construction of bungalows. Penfold (1971) identified prolific potsherds with varied decorations, numerous slags and three kilns on the settlement. Although the area has been disturbed, the kilns and slags recovered through excavation evinces a past iron working society. Penfold (1971) established that it appears that the hills around the coast were inhabited by Fantes who smelt iron and made pottery prior to the arrival of Europeans. However, the area was uninhabited at the time of the research and Penfold (1971) did not indicate its involvement or occupancy during the trans-Atlantic trade. Also, Nunoo’s work in 1970 at Beifikrom near Mankessim yielded iron working remains such as furnace and tuyeres and large quantities of local pottery. 2.3 Central Region During the Trans-Atlantic Trade DeCorse and Spiers’ (2009) work details change and transformation in the political economy of the Akan coast specifically Elmina and Eguafo during the era of the trans-Atlantic trade. This work demonstrates that the situation on the Ghanaian coast before the 17th century can be dramatically contrasted with developments in the following centuries. The Central Ghanaian coast and hinterland can be characterized by its comparative lack of urbanization immediately prior to the advent of the Atlantic trade. Earthwork site at Abrem Berase may have been an exception, but data shows it was unoccupied in the late 15th century. Increasing urbanization is the most striking change in African societies on the coast during the post- European contact period. Prior to the 15th century, villages in the Central Region were small in nature, engaged in localized trade and concentrated around small lagoons as evidenced by the presence of pre- 15th century occupation around Benya Lagoon at Elmina, Brenu Akyinim and Aborobeano University of Ghana http://ugspace.ug.edu.gh 22 lagoon (DeCorse & Spiers, 2009). However, with the coming of Europeans, large populations moved to larger settlements associated with European trade entrepots and the capital of the site in the immediate hinterland, also expanded in size and importance. Eguafo already existed before the Atlantic trade era, but changed in form and organization during the Atlantic era, while Elmina sprang up after the advent of the trans-Atlantic trade. Eguafo incorporated at various times the areas between the Pra river and the Sweet River within its boundaries. The Eguafo Kingdom was one of several Akan polities that first traded with the Portuguese. It served as an intermediary position between the coast and the hinterland. According to Spiers and DeCorse (2009), the kingdom was instrumental in the coastal trade during the 17th and 18th centuries, and this was a period of dramatic expansion of the slave trade. During this period, the Kingdom experienced transformations in its size due to population growth, spatial arrangement and in its social control and the emergence of more centralized political hierarchies. Similar trends are observed in hinterland towns such as Asebu and Efutu (ibid). Archaeological data on the post 15th century period indicate a similar pattern at Eguafo where there was abandonment of a hilltop site and the expansion of a valley settlement during the 17th and 18th centuries. Archaeological data suggest that from the 17th century, the site of the Eguafo settlement increased and there were variety of artifacts as a result of the long distance trade. The settlement of Eguafo experienced a change in settlement from a small defensive settlement to a much larger town during the 17th century. This period of expansion was followed by a period of decline that extends to the present with the current settlement still occupying less area than it did at its apogee (DeCorse & Spiers, 2009). According to archaeological evidence, Elmina; home to the first European trade post in West University of Ghana http://ugspace.ug.edu.gh 23 Africa, prior to the trans-Atlantic trade was confined to two small loci on the Elmina peninsula (DeCorse & Spiers, 2009). Elmina was not an independent polity until after the advent of the trans-Atlantic trade since Eguafo and Efutu may have both claimed territorial rights to Elmina (ibid). After the founding of Castelo de Sao Jorge da Mina, Elmina with the assistance of the Portuguese became increasingly autonomous. Trading activities with the Europeans led to an influx of people into the settlement, leading to population increase in Elmina between the 15th and 19th centuries. Elmina was originally limited to the settlement area and adjacent farmlands. However, during the 18th and 19th centuries the town expanded west along the Elmina peninsula and into portions of the area north of the Benya Lagoon (DeCorse, 2001, 47-52). Presently, the Edina state with Elmina as the capital includes several towns and smaller settlements which is bounded by the stools of Dutch Komenda in the west, Oguaa (Cape Coast) in the east and Eguafo to the north (DeCorse & Spiers, 2009). Oral accounts suggest that the Edina state has incorporated some villages which were formally part of the Eguafo state until the 19th century. There is, however, no evidence for kingship at Elmina until much later and even then, it is an office lacking in hierarchical authority typified in traditional formulation of the state. The office of the Omanhene (Paramount Chief) did not emerge until the 18th century (DeCorse & Spiers, 2009). Currently the head of the Edina state who is the political, military and religious leader rule through monarchical succession. The polities of coastal Ghana do not follow the same pattern of development nor do they present an undirectional evolution from lesser to greater complexity. They thus afford insight into the varied processes that always characterized the evolution of political system (DeCorse & Spiers, 2009). In DeCorse’s (2001) book “An Archaeology of Elmina: Africans and Europeans on the Gold University of Ghana http://ugspace.ug.edu.gh 24 Coast, 1400-1900”, more emphasis is placed on Elmina’s development over the past 500 years. The settlement encapsulates the years of European contact, trade and colonization than any other site in Ghana evidenced by trade list and archaeological data. One visible indicator of change in the Elmina settlement were the many dwellings of multi-storied flat roofed stone construction. These construction methods illustrate a unique aspect of Elmina, probably originating in the latter half of the 17th century with artisans trained by the Dutch. However, archaeology has showed that some of the buildings conformed to African ideals of the use of space. Despite some transformations noted above, variety of source material collectively suggest a great deal of perpetuity in some aspects of religious beliefs, behavior and material culture of Elmina. Although some individuals at Elmina may have imbibed European mannerism in certain contexts, food preparation and consumption practices probably remained unchanged. The social context in which food was consumed has also likely remained the same (DeCorse, 2001, p.178). There are also indications of continuity in ritual expression and the underlying epistemologies over the past 500 years. According to DeCorse (2001), changing values in Elmina appear more prominently in the 19th century after the advent of colonial rule. This period saw direct role in the political hierarchy of Elmina, increasing frequency of houses conforming to European floor plans and the neoclassical features of the surviving 19th century buildings. Although culture contact has been integral in human interactions, the 15th century interaction was between two dramatic different societies (DeCorse, 2001, p. 192). The Elmina data shows Elmina’s adaptation of some European feature and additions to local material inventory rather than a drastic replacement and modification of African culture. Considering the long history of culture contact between Europeans and societies along the University of Ghana http://ugspace.ug.edu.gh 25 coast of Ghana, an archaeological study of shipwrecks and related sites holds great prospects. Maritime archaeology is a new field of endeavor in Ghana and employing it in the study of shipwrecks and other submerged sites will furnish a wide range of material culture involved in the Afro-European interactions, their temporal contexts and provenience. Elmina was one of the coastal towns that engaged and thrived during the maritime trade with Europeans. The siting of the first European trade post; Elmina Castle in Elmina close to an existing settlement, facilitated the trade for centuries. The Elmina maritime landscape is a wave dominated coast interspersed with rocks and sandy beaches (Cook, Horlings & Pietruszka, 2016). Under the Central Region Archaeology Research Project, the Fante Coast was the focus of underwater investigations based on methods such as side scan, sonar, magnetometer, echo-sounding, sediments coring and excavations. In 2009, a field work led to the discovery of a large anchor with one fluke visible and a chain in water 400m south- east of the Elmina castle. In 2007, two anchors with their shanks standing relatively vertical out of the sediments and their arms and fluke completely buried were also discovered. Also, Rachel Horlings led the team to Cape Coast castle and the seascape of Mouri, where three anchor sites were unearthed (Cook et al, 2016). During the 2003 and 2005 field seasons at Elmina, the wreckage of a vessel believed to be the Groeningen, a Dutch West Indian company ship that sank after one of its guns exploded while saluting the castle on its arrival in 1647 and other objects were discovered. Initial assessment suggested mixed context with dates spanning the entire contact period and a prevalence of 19th Century material (Cook et al, 2016). Further investigations of the Elmina wreck site by Pietruszka and Horlings in 2007 led to the discovery of new features, an anchor, a sixth cannon at the northern end of the site and wood samples believed to be Dutch University of Ghana http://ugspace.ug.edu.gh 26 in origin and dated between 1600 and 1730 (Cook et al, 2016). Another field season in 2007 led to the discovery of a second historic vessel in the Benya River; a lagoon system that flows between Elmina Castle and the modern town. The discovery was made following a call to the Central Region Project team after a construction company chanced on remains of a wooden vessel and small cannon balls in the dredged debris. Inspection of the dredged debris by the team exposed 14 disarticulated timbers representing a wide range of structural components including, planking, floor, futtocks, deadwood and beams. The finds were identified as from an early 18th Century Dutch vessel (Cook et al, 2016). 2.4 Previous Archaeological Researches Conducted at Kormantse The area known as Great Kormantse in most historical documents, has been archaeologically investigated by two Professors, E. K. Agorsah and James Boachie- Ansah. The investigations carried out in different years focused on revealing different aspects of the trans-Atlantic past of the society. The research by Prof. Agorsah which was the first archaeological work undertaken in Kormantse spanned 2007 to 2010. The research aimed at investigating the cultural formation and transformation of the settlement in response to changes occurring through colonial times. The excavations yielded cultural materials such as local and imported beads, local and imported ceramics, tuyeres, smoking pipes, human skeletal remains among others. The myriad of cultural materials recovered were not all analyzed and the report on the ones analyzed is yet to be published. However, a preliminary report on Kormantse excavation has revealed that the site abounds in iron slag some of which have been incorporated into shrines. Agorsah (2008) intimated that this points to a past iron working tradition in the society. Fairly little information can be gleaned from them University of Ghana http://ugspace.ug.edu.gh 27 because no research has focused on them. It is on this account that this research becomes crucial in order to investigate how the slags have played significant roles in the religious and economic context of Kormantse’s past and present. The second archaeological enquiry was undertaken from 16th-27th June 2008 by Professor Boachie-Ansah. This excavation was carried out to compare the materials he recovered from Fort Amsterdam with those recovered in Kormantse to ascertain the extent to which the cultural materials from Kormantse bear resemblance to or differ from that of the nearby Fort Amsterdam, whose residents traded with the inhabitants of Kormantse. The excavation at Kormantse yielded materials such as European, Japanese and locally manufactured pottery, gun flint, glass bottles and beads. European smoking pipes, bricks and roofing tiles, metal objects, animal bones and mollusk shells. He noted similarities in the pottery from both sites which were regarded as indices of Asante influence and presence on the coast. Additionally, the dietary patterns were similar as inferred from the animal bones and mollusk shells obtained from both sites. The imported materials recovered from Kormantse attest to the trade relations that existed between the Kormantse and the Europeans. Boachie- Ansah (2015) added that indications of acculturation and reliance on European imported goods can be discerned in the archaeological record as numerous European goods were recovered from the excavations. This, just like Agorsah’s work is revealing aspects of the trans-Atlantic past of the society and this has been the focus of most researches along the west coast of Ghana. 2.5 Shrines, their Composition and Significance The Akan belief in the concept of a supreme being and sole creator of the world; Onyame/Nyame (God) who is assisted in the task of managing his creation (humanity) by University of Ghana http://ugspace.ug.edu.gh 28 the abosom who are deities/ divinities (Blay,2009; DeCorse, 2001). The abosom either male or female are in essence, the intermediaries and immediate objects of reverence on earth. The abosom embody various manifestations of nature namely, wind, trees, animals, assemblages of rocks, natural features of the landscape such as mountains, ponds, shaped stones, lakes, rivers or other water features (Blay, 2009; Dawson, 2009). Shrines are built to function as a space and means through which humans can communicate with Nyame via an obosom in or near the locality where the abosom is said to reside (ibid). In literal sense, shrines act as ‘containers’ for the spirits of ancestors and deities who must be consistently placated and petitioned for blessings, requests for intercession and divine sanction (Dawson, 2009). Mather (2009) also posits that a shrine can be a physical object at which animal sacrifices and libations are offered in honor of supernatural and other worldly agents including ancestors, spirits of the land, divining spirits and nature spirits. Furthermore, shrines are foci of ritual practice (libation and sacrifice) because they hold the supernatural agents that the people pay homage to (ibid). According to Parker (2011), shrines within a particular religious tradition appear resistant to precise definition on account of their varying forms and configuration from one locality to the other. The material components of a shrine can be complex depending on the type of shrine in question (Kankpeyeng et al, 2013; Insoll 2011a and b). A shrine may consist of multiple objects or a single object (Mather, 2009). African shrines can be mundane material objects such as ceramic pots, tools used to work the earth such as axes, hoes, iron, spears, bronze and iron bracelets, animal remains and materials ultimately derived from the earth such as worked stone pestles and grinders (Insoll, 2006). These were frequently found to form core elements of smaller shrines amongst the Tallensi (ibid)). For University of Ghana http://ugspace.ug.edu.gh 29 instance, Kusaasi shrines are made from six different types of objects namely, trees, stone, pot, calabash, tail and root (Mather, 1999). These manmade or natural objects used in the shrine context could be explained as appropriate use of abundant materials in the locality. Nevertheless, they assume ideological, religious and spiritual functions besides their primary function through practice and context (Insoll, 2004, 2006). In, ‘Material Culture and Indigenous Spiritism: the Katamanso Archaeological “Otutu” Shrine’, Apoh and Gavua (2010), revealed that objects such as pottery, imported ceramics, bones, beads, metals, glassware, cowries, bottles of schnapps, gin amongst others excavated from the archaeological Otutu shrine, were bound up in spiritual expressions and practice owing to the context and practice. Shrines are categorized on the basis of their attribute (Mather, 2009), function as well as the resident deities or spirits. Shrines therefore, can be classed into lineage shrine, personal destiny shrine, ancestor shrine, household shrine, external shrine, chiefly shrine, earth shrine and community shrine (Insoll, 2006). Dawson (2009) asserts that shrines like collectives are not static and their meanings within the society alters as morphology of the collective changes. They can be modified, rebuilt, renovated, relocated, forgotten, re-legitimised and forgotten once again (Dawson, 2009). According to Insoll, Kankpeyeng, & MacLean (2009), shrines especially in the West African contexts appear to serve as symbolic repositories of information as they are vital for recounting history, for remembering important people and events from the past. In Apoh and Gavua’s (2010) work, the material assemblage in the Otutu archaeological shrine provided insights on the state of the society or nation, and indication of the people of Katamanso’s involvement in local and international interactions or trade in the past. University of Ghana http://ugspace.ug.edu.gh 30 2.6 Iron Slag in Religious Context Slags are the by-product of iron working activity. Unlike the bloom, the end product which is fashioned into tools, the slag is left on the site after the process is completed. According to McCosh (1979), slag is considered waste and irrelevant and as such, is referred to as faeces by the Ngoni of Malawi. However, in other African countries, evidence of the use of slags in shrine, rituals and religion have been observed. In “Ritual and Political Aspects of Iron Working; Iron in War and Conflict. Archaeology of Early Northeastern Africa”, Haaland (2006) reveals that a temple dedicated to the war god Apedemak is built on a slag mound at the south eastern part of the ancient city of Meroe. In Haaland’s (2006, p.143) view, it is tempting to see the building of the temple on the slag mound in Meroe as a manifestation of rituals of iron smelting. Therefore, the ritual significance of the temple is manifested in it being localized on slag heaps (Haaland, 2006, p. 147). Non-religious reuse of slag in Ghana and beyond has also been documented. At Adome, Gavua and Apoh (2011), in “Alternative Site Conservation Strategies in Ghana: The Adome Ancient Ironworking Site” undertook a research in a community where the inhabitants did not know the archaeological significance of the slag. The inhabitants used the slag as foundation stones for buildings and as bathroom flooring materials. In southern Finland and central Sweden, archaeologists have frequently observed iron slag in a number of iron age burials. According to Shepherd (1997), there are several strong reasons for believing that the slag may have a ritual overtone. In trying to establish a symbolic meaning of the slag with the burials, Shepherd (1997) asserts that both are by- products. Thus, the corpse is the by- product of a human life transformed by death into a freed and disembodied spirit and the slag is the by-product of the transformation of iron ore into powerful and dangerous iron. The study of materials involved in the ancient iron working process (ore and slag) and University of Ghana http://ugspace.ug.edu.gh 31 furnished iron tools (iron objects) possess the ability to provide information regarding composition, origin and technological inferences (Ashkenazi et al., 2012). Scanning Electron Microscopy, Energy-Dispersive X-Ray Spectroscopy (SEM-EDX) are touted as important analytical tools for such purposes. Okoro (1989) presents the archaeological, ethnohistorical and ethnographical material on the iron technology at Gambaga in Ghana. In the study, iron ore samples from smelting sites, furnace and ore mining locations were subjected to an elemental or chemical composition analysis by the Geological Survey Department of Ghana. This made it possible to relate the iron ore and slag to different furnace forms found in the settlement. Unfortunately, the specific chemical analysis employed was not indicated in the study. Analysis of ore located 800m southwest of Gambaga and ore on which bowl shaft furnace was excavated indicated similarity in chemical composition, particularly an iron weight of 56%. This demonstrates that the ore used at the bowl shaft furnace was mined from the nodular hematite ore deposit from the same location. For the tall naturally induced furnace, ore samples analyzed had 48% iron content and was a lamellar ferruginous laterite. According to Okoro (1989), this implies that the tall furnace used lower iron content ores. Okoro (1989, p.123) gleaned from the analysis that smaller furnaces employed rich or concentrated ores while the large and tall ones seem designed to utilize poor quality ores. The stylistic variations in furnace built may be explained with reference to technological consideration of the ore used. The samples of slag collected from mounds and furnaces had high iron content of 40%. This hints at an inefficient iron working process since the slag that run off the furnace contained not mainly waste but a high amount of iron oxides, which forms a major component of the end product which is bloom. University of Ghana http://ugspace.ug.edu.gh 32 In Gethlin’s (2016) work, a study of archaeological slags was carried out to glean insights into the metallurgical activities and broaden the knowledge of such finds. The study analyzed archaeological slags from Tell Dhiban and Armenian region, all in Israel using Scanning Electron Microscopy, Energy Dispersion Spectroscopy (SEM-EDS) analysis. According to Gethlin (2016), SEM-EDS analysis is touted over other forms of chemical analysis because of its non-destructive nature and ability to provide large information. The analysis commenced with macroscopic assessment of the slags from both sites. This involved weighing, measuring and photographing of samples. The physical assessment of the slags identified the samples as smithing slags. Further chemical analysis showed high iron content in the samples with low silica and low alumina, and this according to Gethlin (2016), are characteristics of smithing slag. The chemical analysis showed that the slag specimens collected from both Tell Dhiban and the Armenian Garden all originated from smithing practices and were very similar in terms of chemical composition and microstructure. This indicates that the same or very similar general methodologies were being practiced at both sites. Furthermore, the chemical composition and microstructure produced by the SEM-EDS analysis of the samples from Tell Dhiban and Armenian Garden are distinct when compared to smithing slags from other regions and time period. This can be attributed to the use of calcareous rather than siliceous fluxing agents during the smithing operation (Gethlin, 2016). 2.7 Summary Along the Fante coast of Ghana evidence of iron working activity has been in the form of slag, tuyeres and furnace fragments recovered from excavations in sites such as Akrokrowa earthwork site, Beifekrom and the area between Apewosika and Amamoma. Others are University of Ghana http://ugspace.ug.edu.gh 33 surface evidence in the form of slag heap at Fomena and Komenda. However, none of these sites has slag and iron ore incorporated in their shrine and religion. Kormantse’s case of slag or ore shrines provide a unique evidence and probably the first of its kind, especially in a site associated with the trans-Atlantic trade. This offers an opportunity to understand this phenomenon and use archaeology to ascertain whether there was iron working activity in the past especially in the case where oral history and documentary sources does not mention Kormantse as a past iron working society. University of Ghana http://ugspace.ug.edu.gh 34 CHAPTER THREE RESEARCH METHODS 3.0 Introduction This chapter details the approaches adopted in the collection of ethnographic and archaeological data for analyses and interpretations. The methods namely library research, interview, observations, photography and archaeological excavations yielded both primary and secondary data for the study. 3.1 Forms of Data (Secondary and Primary Data) Secondary data which constitutes data collected earlier by previous researc