Navigation – Plan du site
Histoire, Anthropologie

The Forest and the Twis

Ivor Wilks
p. 19-75


Dans cet article Ivor Wiks revient sur le débat ouvert par lui-même en 1977 sur les conditions de l’émergence, au milieu du XVIIe siècle, des Etats akan en pleine zone forestière de l’actuel Ghana. Les locuteurs du twi se seraient organisés en matriclans, qui favoriseraient l’assimilation d’étrangers. Il conforte cette position en puisant dans les sources anciennes (archives, relations de voyage, cartes) et en s’adressant également aux sources orales, qui font d’Odumankoma l’entité incarnant la capacité d’innovation, et le fondateur du site sacré d’Adansemanso, dont le premier des Etats akan tire son origine.

Haut de page

Texte intégral

Relict landscapes of the Bron and Akan

Relict landscapes of the Bron and Akan

Part I : odomankoma as “big bang”

Revisiting the Bron and Akan

1This paper has to do with the Twi-speaking peoples of the Republic of Ghana1 In or about 1507 the Portuguese navigator Duarte Pacheco Pereira gave a list of some of the traders who, in the last decade of the preceding century, arrived on the Costa da Mina (the later Gold Coast) from the interior. Those named were “the Bremus, Atis, Hacanys, Boroes, Mandinguas, Cacres, Andeses or Souzos.” There is an order in the list. The Abrem (“Bremus”) and Etsi (“Atis”) were among the peoples of the forested country between the Costa da Mina and the region of the confluence of the Pra and Ofin rivers. The Akan (“Hacanys”), Bron (“Boroes”) and Mandekan (“Mandinguas”) were those whose operations at the time extended – in reverse order – from the northern zone of transition between forest (Twi kwaem) and grassland (serem) to the new Portuguese establishments on the coast to the south. The Akrokerri (“Cacres”) and Adanse (“Andeses”) lived in the auriferous basin of the Pra and Ofin where moist evergreen forest gave way to moist semi-deciduous2.The perplexing question of the identity of the Akan and the Bron, which continues to be of much political sensitivity in present-day Ghana, is the principal matter to be addressed here. The roots of this problem are to be found in antiquity. This paper, then, also has to do with the Adomankomasεm or Odomankoma Tales, that is, the corpus of stories about the early history of the Twi-speaking forest people3. The reader should not expect any extended discussion here of the nature of lore that has been transmitted orally for extended periods of time. For this he or she may consult Jan Vansina’s excellent 1985 work, Oral Tradition as History4. Here I shall simply put the Tales to use.

2Linguistically, Bron is regarded as a somewhat archaic dialect of Twi, but only in the sense that its rate of change was slower than that of southern dialects after Asante emerged as an imperial power in the eighteenth century (Dolphyne, 1979: 88-118). The prehistory of those who came to be known as the Bron and Akan is beyond the scope of this paper. It may, however, be noted that the distribution of the Twi language indicates, first (to follow J. M. Stewart), that it has long been spoken in its present locations and probably developed as a separate language in the area of the lower Tano basin, and second (to note a suggestion of C. Painter), that Twi may have existed as a separate language for three to four millennia5. The vexed question of the relationship between Bron and Akan is perhaps one that should be dissolved rather than resolved. During the period of Asante ascendancy there seems little doubt that the Bron began culturally to associate themselves more and more closely with the Akan, that is, with the Twi-speaking people of the southern forest country. It became, as it was, not only politically beneficial but also, perhaps, chic to be Akan rather than Bron. By the mid-twentieth century this situation had changed radically, and the rise of the Brong-Kyempim led ultimately to Kwame Nkrumah’s creation of a Brong-Ahafo Region.

3In 1969 K. Y. Daaku, to whose painstaking work I am much indebted, queried Adanse informants about the meaning of the name “Akan.” It means, he was told, “the Twi-speaking people.” Why, then, “do we use ‘Akanfo]’ (Akans) instead of ‘Twifo]’ (Twis),” Daaku asked. “Because,” he was informed, “after God created the earth, he created the Akans. Thus Akan signifies people.”6 The point of this observation is rather lost in translation. In Twi the root kan means “first.” Thus AkanfoÔ are literally “the first people.” In nineteenth century Gold Coast the term was commonly translated into English as “pioneers” and “aborigines”, both felicitously capturing different nuances of the Twi.”7 In 1970-71 D. M. Warren and K. O. Brempong recorded replies to their inquiries about the meaning of the name “Bron” (or “Brong”). “The Bono were the first people created,” they were told by one aged informant; “that is, they were the first born and this is the reason why they are called Bono, Abono woÔ, ‘first born’.”Asaman, one of the (apical) ancestors of the Bron, another aged informant said, “never found any other people on the land. Thus he concluded that he and his people were the first to occupy that land and he called his people Bono meaning ‘first born on the land’8.” The equivocation between “Bono” and “Bron” will be noted. The Portuguese rendition of the name as “Boroes,” shows that northern Twi was already diverging from southern by, in this case, the replacement of CVnV by CVrV forms, “Bono” thus becoming “Boro” in pronunciation. It also appears to have resisted the tendency in southern Twi to drop the final vowel in CVnV and CVrV words, but instead nasalised it to yield “Boron” (and hence the contemporary “Bron” or “Brong”). Both forms of the name continue in use, the older form “Bono” particularly in formal contexts9.

4I shall make use from time to time of a distinction between protohistory and history. This is for convenience only, and is not intended to have theoretical implications. I do so in order to express in English an important distinction that is made in Twi between firi tetemu and abak]sεm. The former refers to the time of the atetefo, forgotten people of a world that has long been lost. The latter refers to the time of the nananom, remembered ancestors of a world that still exists, and who continue to be commemorated in verbal and material forms – songs of praise, laments, and the like on the one hand, and carved stools, brass castings, and other such memorials on the other. The atetefo may be, and I assume are, the biological progenitors of the nananom, but they cannot themselves be nananom, for the status of being an ‘ancestor’ came into existence only when the cultural configurations of protohistory gave way to those of history. In the mid 1970s I became interested in the nature of this transition, and presented a paper on the topic at the fifth meeting of the Research Seminar on Archaeology and Related Subjects held at the Institute of Archaeology, University of London. I argued that the transition had to do with Twi-speaking people being the first to practice food crop cultivation in the high forest country or as James McCann has recently so well described it, to do with “the invention of the forest fallow cultivation system.” (McCann, 1999: 110.) I shall briefly review the history of what, for better or for worse, is referred to as the “big bang” theory10.

The Akan “big bang”

5In 1977, in an essay entitled “Land, labour, capital and the forest Kingdom of Asante,” I used documentary materials reaching back to the late fifteenth century, recensions of orally transmitted traditions, and data on forest agronomy, to suggest that,

on an hitherto quite unprecedented scale, the forestlands were explored and cleared, and new agrarian settlements established; that this represented a major transformation in the system of productive forces such that an economy with a primary dependence upon hunting and gathering gave way to one with a primary dependence upon the cultivation of food crops; that the social relations of production of the forest peoples were accordingly modified by the creation of the great exogamous matriclans and the redistribution of communities between them; that this socioeconomic revolution began, geographically, in the forest heartland of Amansie and Adanse, and chronologically, should probably be regarded as having reached its zenith in what is for the Asante the immediate protohistoric period (firi tetemu), the sixteenth century11.

6In 1982 I explored these themes further, referring to them (somewhat facetiously, so I commented at the time) as the “big bang” theory of Akan origins. My intention was to stress that the transition from foraging to cultivation was “explosive” in the sense that it made an impact on virtually all aspects of the way of life of the forest people, but not in the sense that it occurred instantly, in (as it were) a moment of time12. Akan society, I wrote, “as it is known from the historic period, came into existence as a result of a change in the mode of production which involved a transformation in the economic base, a reorganization of the social relations of production, and the appearance of new political structures.” (Wilks, 1982 (a): 87-88; 234.). In 1993, in Chapters 1 through 3 of Forests of Gold, I sketched out a rather fuller version of the “big bang” theory. At the risk of some repetition, I quote the summary offered there :

First. In the course of the fifteenth and sixteenth centuries, and in the forest country between the Ofin and the Pra, a foraging mode of production gave way to an agricultural one.... Second, that in the course of this transformation the forest people reorganized themselves in a way such that the bands appropriate to the older mode of production were replaced by the matriclans appropriate to the newer... But third, that the transformation also engendered the emergence of political structures of a new kind: the aman.... All these things, together, serve to define the Akan: sedentary agriculturalists, socially distributed between matriclans and matrilineages, politically organized into “states,” and culturally bonded by language, religion, and a common sense of shared history.

7I went on, however, to indicate the catalytic importance of one of the non-renewable products of the forest country, namely gold, in the creation of the new agrarian order. “Those,” I wrote,

who controlled the production and sale of gold were those able to procure a supply of unfree labour. Those who procured unfree labour were those able to create arable land within the forest. Those who created the arable land were those who founded the numerous early Akan polities. We see, in other words, a class of Akan entrepreneurs emerging: a class of those able to use the strength of the world bullion market, and the availability of labour locally (whether through Wangara or Portuguese suppliers), in such a way as to create the new agrarian system. (Wilks, 1993: 94, 96.)

8The following year, 1994, Norman Klein published a lengthy review essay on Forests of Gold, and devoted it to an attack on my view of Akan origins13. My suggestion that the acquisition of unfree labourers did much to facilitate the development of the forest fallow cultivation system greatly offended him. I saw these as supplementing family labour, and as perhaps in time evolving into the gyaase class that is so characteristic a feature of Akan society. Klein imagined, however, that I was arguing that the Akan developed “a mode of production based on large-scale slave labour.” I was not, and I am sure that the Akan did not! Klein also thought that I identified “the early Akan with invaders who came from the north and occupied their present home in the gold-bearing forests of southern Ghana during the fifteenth and sixteenth centuries.” Had I put forward this view it might well be asked whence such invaders came, since they would have to have been carrying with them relevant agronomic skills and the cultigens necessary to develop a forest fallow system. But I did not, pace Klein, postulate the existence of any such “invaders.” It was, I argued, the foraging peoples of the forest, the hunter-gatherers, who achieved a new level of mastery of an environment hitherto highly inimical to human settlement. They did so not only as a result of a long process of experimentation with crops, but also of a systematic programme of prospecting for gold and finding markets for it.

9His extraordinary misunderstandings of my argument aside, Klein also expressed reservations about “the very way Wilks has set up the oral histories of Akan, especially Asante, in real historical time”. He explained this. “I should point out,” Klein wrote, “that such vividly ‘remembered’ mythologized events as the Akan arrival in the forest may have occurred millennia ago, as did (?) some of the most dramatically rendered events in the Old Testament.”14 I have difficulty in following Klein’s reasoning (and his question mark) at this point, since the custodians of Akan ‘myths’ of origin – the Ôdomankoma Kyerεma or, anglicised, Odomankoma Drummers – make no reference whatsoever to “the Akan arrival in the forest.” Nor does Klein, for that matter, make any reference to the drummers! Indeed, he displays the characteristic hubris of a certain style of anthropologist in dismissing those chronologies of the origins of agriculture in the Akan forests that, as he stigmatises it, “have been put forward... by historians, relying primarily on documents.” He sees their role as superseded by that of “geneticists and archaeologists, relying primarily on the prevalence of the sickle cell trait and radiocarbon dates” (Klein, 1996: 263). I am unrepentant. In formulating the “big bang” theory my regret is that, as a historian, I did not pay even closer attention to the “documents,” and in particular to the mythopoeic accounts of origins that underpin the way that the Twi-speaking people envisage their passage from Antiquity to Present. Had I done so I would also have been obliged to give due consideration to the Bron. If the early Akan story is that of the development of agriculture, that of the early Bron is rather that of the expansion of its manufacturing base and the boom in trade and especially the gold trade. These matters will be addressed below.

Odomankoma as Force of Change

10Towns and villages exercise proprietary rights over stories about their origins. Ôbi nkÔ obi akura nkyerε n’ase is a fundamental dictate of custom: “you do not go to someone else’s town or village to tell them their past.”15 In this sense most history is local history, typically commencing with the reasons for the foundation of this or that settlement and whether as a result of overpopulation, internecine conflict, or the like. However, the stories of the oldest towns and villages – those of the “first creation” as it were – will have reference ipso facto to the very origins of the world of the Bron and Akan (as it emerged in its historic form), and in this sense should be placed in a genre not of local but of world history. They are known in Twi as Adomankomasεm, that is, Odomankoma Tales. The primary texts are played in ayan, drum language, by those who often “sign” their pieces Ôdomankoma Kyerεma se..., that is, “Odomankoma Drummer says....”16They use the highly prestigious Akan drum, the atumpan. A master drummer, so J. H. Nketia commented, “is considered the greatest of all drummers because of the breadth of his knowledge, the skill which his work demands and the role he plays as a leading musician in all ensembles in which the atumpan drums are used.” (Nketia, 1963: 153) He is “closest to the spirit of the ancestor chiefs whom he addresses,” Nketia added, and at the time of the Akwasidae festival “recounts the names of the ancestor chiefs one by one, praising each one, mentioning his accomplishments, his origin or place of domicile and so on.” But Nketia also described the Odomankoma Drummer as “close to Nature,” and added, “in accordance with the world-view held by the Akan, he conjures up the spirit of creation....”(ibid. 154).

11J. B. Danquah stressed the aspects of the drummer’s work that are of most significance to this paper. “The Ntumpan drummer, called Kyerema,” he wrote,

 is both poet and historian. His mind is a storehouse of the tribe’s traditional knowledge. He could not play a single bar on the Ntumpan or Talking Drums unless he knew what the “talking” was about, playing the appropriate stanza on the appropriate occasion, or for the appropriate person. The kyerema is from childhood trained in the poetry and lore of the people, and there is not a knowledge of nature or of man which is beyond his comprehension as the language of the talking drums testifies. (Danquah, 1944: 72-73)

12In 1923, Rattray commented on the repertoire of Asante drummers. It consisted, he wrote,

of certain holophrases which are in constant use by all drummers, [and] it will be readily understood that they become absolutely familiar with these. Should a drummer depart, however, from one of these ‘set pieces’ and strike out on his own, drumming at fancy, new phrases, i.e. new combinations of tones, etc., then, though to himself the drum would still continue to speak, yet another drummer, who heard these new combinations for the first time, could not, I am convinced, read his message with any accuracy. (Rattray, 1923: 257.).

13Rattray remarked that “a drummer who falters and ‘speaks’ a wrong word is liable to a fine of a sheep, and if persistently at fault he might, in the past, have had an ear cut off.” (ibid. 265). Over time the importance attached to preserving and transmitting the Odomankoma Tales intact has resulted in the appearance, in the verbal renditions of the drum texts, of Twi terms that have become archaic. This was especially true of “strong names” (mmerane) used to praise ancestors, and the problem of further translating the Twi holophrases into English could sometimes prove insurmountable. Texts in the speech mode of drumming might, then, be transmitted unchanged over long periods of time as their meaning became more and more obscure. “No one now seems able to interpret them,” Rattray wrote; “so they have had to be retained and rendered in the vernacular.” (ibid. 265). A proviso must therefore be made. I have sometimes had to choose between divergent Twi texts, and I have frequently offered changes in the translations. I hasten to add, however, that I recognise these are in many cases tentative (and in some, distinctly dubious!). Some phrases from the drums, however, passed into that amorphous corpus of sayings known in Twi as mmε (sing. εbε) and usually described in English as “proverbs,” and in this form they may have been lexically modernised17. Whether extant as ayan or as mmε, however, they served and serve a mnemonic purpose, prompting narrative accounts that were variously embellished for dramatic effect by local storytellers18The drummers are now popular performers in Ghana, and much of their repertoire is fortunately being recorded. Local storytellers, regrettably, are losing their appeal to generations now immersed in an electronic world far different from that of concern in this paper.

14What, then, was Odomankoma ? Firi tete, the drummers play, Ôdomankoma bÔÔ adeε, that is, “long long ago, Odomankoma made a thing.”19They frequently use another name for Odomankoma: BÔrebÔre bÔÔ adeε, “Borebore made a thing.”20 The Odomankoma Drummers are known as such because, it is claimed, Odomankoma made the first drummer along with the first herald (Twi εsεn) and the first head executioner (Ôbrafo titire). K. A. Busia famously commented that, “Akan drum language is full of riddles that conceal reflective thought and philosophy....”21 Evidently the old drummers were indulging their passion for riddles when telling of their own origin:

Hena kÔse,

Who is going to speak ?

Hena kÔse,

Who is going to speak ?

Hena kÔse,

Who is going to speak ?

Hena na ÔkÔsee ‘te.

Who is going to speak, old times ?

Maa ‘te kÔsee Ananse,

Old times, Spider is going to speak.

Maa Ananse kÔsee Ôdomankoma,

Spider is going to speak, Odomankoma,

Maa Ôdomankoma bÔÔ adeε,

Odomankoma made a thing.

ÔbÔÔ deεbεn ?

Made what ?

Ôdomankoma bÔÔ adeε,

Odomankoma made a thing.

BÔrebÔre bÔÔ adeε,

Borebore made a thing,


He made the herald.

ÔbÔÔ kyerεma.

He made the drummer.

ÔbÔÔ Kawuo Akwabrafo titire

He made Kawuo Akwabrafo chief.

Ôbrafo titire.

Chief executioner,

Makye !

Good morning !22

15It follows that the drummers must, therefore, be afforded respect:

Ôdomankoma bÔÔadeε,

Odomankoma made a thing,

BÔrebÔre bÔÔ adeε;

Borebore made a thing,

Ôkyerεma yεyε no brebrε,

The drummer is to be treated with consideration

Ôkyerεma yεyε no gye adeεdi

The drummer is to be treated with hospitality23

16Odomankoma is, then, primarily, a maker (and perhaps, with reference to the praise name, also a searcher24). In the later nineteenth century the Basel missionary J. G. Christaller was using the Twi Ôdomankoma to denote “God, the Creator” very much in the Christian sense of the term. With seeming confidence he added that the name “seems to mark him as “the boundless, infinite, interminable, immensely rich Being, and as the author and donor of an inexhaustible abundance of things.” (Christaller, 1881: 89.). In 1916 Rattray, noting that the name did not appear to him to be Twi rejected Christaller’s interpretation of it. Instead he suggested that bÔÔ adeεapproximated in meaning to the English “inventor.” (Rattray, 1916: 19. ). Unfortunately Rattray seems to have lost interest in the issue and by 1923 was casually referring to Odomankoma as “the Creator.”25 In 1944, J. B. Danquah argued for the existence of an Akan Trinity, Onyame, Onyankopon, and Odomankoma, the last corresponding “to a conception of the Godhead as the Interminable or Infinite Being.” (Danquah, 1944: 30. ). In this context, it may be remarked that in the Twi texts Odomankoma is not strongly anthropomorphized; it is never clear, in the very structure of the language, whether references are to him, her, or it. And, to the best of my knowledge traditional storytellers very rarely treat Odomankoma as having human form26. Recently Kwasi Wiredu has taken up the matter. “But what sort of creation are we talking about?” he asks; “There are doctrinal as well as linguistic reasons for thinking that the Akan conception is one of a demiurgic fashioning out of order from a pre-existing indeterminate stuff rather than creation out of nothing.” He refers to a “cosmogonic paradox,” and formulates it thus: “Either the Creator was somewhere before creating everywhere or he was nowhere while creating everywhere. In either case, there is a contradiction. Moral: Creation can only have been a process of transformation27.” G. P. Hagan commented, “Beyond Odomankoma there is no other being to which the Akans assign any role in creation28.” The storytellers grappled with the problem with considerable confidence. The case of Adankrangya, near Bekwai, is illustrative. Odomankoma arrived to find this Bretuo town already in existence. Its chief, Opon Panin, announced, “that as Odomankoma was creating new things he wished to know what would become of his name [when] Odomankoma made things afresh in this town…. The Chief said he would allow Odomankoma to continue the creation of his things, but he wished to know what Odomankoma would do to allow his [Opon Panin’s] name to remain forever.” Odomankoma, according to the story, agreed to

do something whereby his name would remain forever. There were some efuo monkeys at Adankrangya when Odomankoma arrived there… Odomankoma then told the chief that in order that the name of Adankrangya should never be forgotten he would call any new town, which he should, make after these monkeys…. After this, Odomankoma asked if there were any other people living in these parts.29

17The Adomankomasεm, or Odomankoma Tales, are set not in mythical but in real physical space, and refer to geographical features and localities that can still be identified. The beginning of the saga depicts the gathering of the force that was Odomankoma. The drummers play,

Ôdomankoma Kyerεma se,

Odomankoma Drummer says,

OkÔ baabi a.

It [Odomankoma] goes somewhere.

Wa ma ne-ho meneso

It has caused itself to arise

AkokÔbÔn anÔpa,

[Like] the fowl crowed in early dawn,

AkokÔ tua bÔn,

The fowl arose and crowed,

Nhema hema hema.

Very early, very early, very early30

18The saga also has an end. The drummers know about it. Ôdomankoma na Ôma owuo di akane, they play: “Odomankoma made death first.”31 Death, however, was to kill its maker:

Daammere Ôdomankoma bÔÔ adeε,

Where Odomankoma made a thing.

BÔrebÔre bÔÔadeε,

Borebore made a thing,

εfiri tete yεbÔÔ nipa ne

Long ago he was making people.

Amanehunu daa daa daa daa.

Suffering always, always, always, always.

Ogyapam ne ne ntεtea firi tete.

The Ogyapam tree and its black ants,long ago

Ôdomankoma bÔÔ wuo maa

Odomankoma made

wuo kum no.

death but death killed him32

19The death of Odomankoma was a key event. The making of the new world of the Bron and Akan was complete. Ôdomankoma kÔwuiε no, Ôde n'akyi gyaa agyina, it is said, with the sense, “when death killed Odomankoma, things were left to men of affairs33.”. If the force that was Odomankoma could not, to follow Wiredu, create anything out of nothing, neither apparently was it able to change that which it had created. Odomankoma transformed one (Twi-speaking) world into another, but thereafter men and women took control of affairs. It is my contention that the Odomankoma of Akan cosmology should be read as a mythopoeic representation of the forces that totally transformed economy, society and culture in the forest zone of Ghana; that, in other words. Odomankoma and the “big bang” are two ways of conceptualising one and the same thing.

The Odomankoma Transformation: the Bron

20The first place Odomankoma created – using the term in Wiredu’s sense of transformation – was Amuowi. It lies about 35 kms northeast of the present day Takyiman. A song recorded during the annual Apo] festival in Takyiman commemorates the event:

Ye firi Amuowi oo,

We came from Amuowi O,

Ye firi Amuowi Ôdomankoma.

We came from Amuowi, Odomankoma.

Amuowi dadampowa.

Ancient Amuowi.

Ye firi Amuowi oo,

We came from Amuowi O,

Ye firi Amuowi Ôdomankoma.

We came from Amuowi, Odomankoma.

Asase kokÔÔ mma e.

Children of red earth.

Ye firi Amuowi oo,

We came from Amuowi O,

Ye firi Amuowi, Ôdomankoma.

We came from Amuowi, Odomankoma34.

21The storytellers elaborate on the matter. The ancestors, one said, came out of a hole in the ground but they “were in great numbers and therefore they found it difficult to emerge from the hole35.” Then, said another, “each family began to search for its own abode36.” A third told how the Bono leader, Asaman, established a capital. It was called, he said :

Bono-Manso, the capital for ‘the first born people.’ Bono became one of the languages of the talking drums; Bono Kyempem Duoduakwa, ‘Thousands of Bono (first born people) whom God has brought to a new land.’ They were first on the land.37

22The atumpan drummers of Takyiman play the text, Ôdomankoma bÔÔadeε, ÔboÔ Bono ansa na Ôbo adeε biara, that is, “Odomankoma made a thing, it made Bono before making anything.” (Warren, 1974: 31). Equating the old Bono with its successor, the present Takyiman, they also play,
Bono kyempim duoduakwa. Bono kyempim duoduakwa.
Domankoma bÔÔ adeε, Domankoma made a thing,
ObÔÔ deε ben ? What thing ?
ÔbÔÔ Takyiman. It made Takyiman.
Takyiman firi hen ni ? Where did Takyiman begin ?
Takyiman firi tete. Takyiman began long, long ago,
Takyiman kan kan kan. Takyiman, first, first, first38.

23It will be remarked that the drummers emphasise the status of being “first.” The storytellers take this matter up, but embellish it for dramatic effect. The Bono ancestors, the founders of lineages, arrived at Amuowi, they say, from a hole in the ground. Such stories (we shall see) feature in all comparable Bron and Akan myths of origin – myths, that is, of first creation – with the qualification that the mode of arrival is occasionally from sky rather than the ground. It has long been recognised that these are affirmations of autochthonous status39. To bring the ancestors from north, south, east or west would be to identify them as immigrants, whether invaders, refugees or whatever, and thus to compromise that status. Use of the third dimension (up/down40), however, creates a “looking glass” effect. On one side of the mirror are the atetefo, and on the other the ancestors. The latter have not come from anywhere but at the same time they have not come from nowhere. An Odomankoma transformation has occurred. E. L. R. Meyerowitz cites a maxim that may be read as a riddle having to do with the ‘up/down’ origin of the ancestors. Odomankoma, she reports, created “Ôsoro ne fam,” that is, sky and earth, “by lifting up the one and setting down the other.” (Meyerowitz, 1951: 82.).

24A second Bron town of first creation is Wenchi (or Wankyi). The storytellers relate that, preceded by an Ôkyeame, the ancestress Asaseba Ode Nseε emerged from a hole in the ground at Bonoso, near the source of the Ayaa (and about 15 kms south of the present day Wenchi). Six other families followed her. Subsequently they dispersed. Ahwene Koko, some 22 kms southwest of Bonoso, was founded, and became the first capital of the Wenchi kingdom. Sεdeε Ôdomankoma bÔÔ adeε, bÔÔWankyiman: “it is said Odomankoma made a thing, made Wenchi41.”. One further Bron settlement must be mentioned. The village of Hani lies about 16 kms west of the present Nsoko (or, commonly, Nsawkaw), which itself lies some 35 kms northwest of Wenchi. Hani and Nsoko share the ancestors, Bene and her brother Kutu, who came out of a hole together at Bonkese (ÔbÔn kεse, “big hole,”) about 5 kms southwest of Hani. After a time Kutu and Bene left Bonkese because of a shortage of water, and founded Bewo. Kutu’s drummer played (and the translation is tentative),
Abron Kyempem Duoduakwa, ]hene a Abron Kyempem Duoduakwa, O Chief.
Firi tete, firi tete, Ôdomankoma bÔÔLong, long ago Odomankoma made a adeε thing, Obi nim nahyεase. Nobody knows when it began.
(Ameyaw, 1965: 2-3. 23)

25“The old name of Nsoko was Bewo,” I was informed in Hani in 196242. Local curiosity about a name, Nsoko, that was not of Twi origin led storytellers to invent (possibly drawing upon some esoteric strand of tradition) a “mother” for Odomankoma, Aso, and to appropriate her for etymological purposes. She died in Bewo while her “son” was away. On learning of this Odomankoma made a new name for Bewo: AsÔkÔ, “Aso has gone,” hence “Nsoko43.”.

26One somewhat exotic item deserves notice:
Tete Ôdomankoma,
εme, mifir tete Ôdomankoma
Me nana na ÔbÔÔ egua
Maa mpanyimfo wÔbÔtoe.
Wofir Takyiman bÔtoo hεn.
Mankεsem Gua dadow no do.
Ancient Odomankoma,
I come from ancient Odomankoma,
My ancestor made a market
And the elders ...
You came from Takyiman ...
Mankessim Market long ago.
(Nketia, 1955: 184, 239.)

27The reference, uncertain though the reading may be, is clearly to the well-attested movement from Takyiman into the hinterland of the Gold Coast that led in time to the establishment of Mankessim. The migrants became known as the

28Borbor Fante. McCaskie tentatively dates their arrival to a period extending over the late fifteenth and first half of the sixteenth centuries, and the foundation of Mankessim on the site of an older Etsi town to sometime before 160044.

29There is a well-known drum piece, entitled “Tano,” that again has the form of a riddle:
Asuo atware Ôkwan.
Ôkwan atware asuo.
Ôpanin ne hwan ?
YεbÔÔ kwan no kÔtoo asuo no,
Asuo no firi tete.
Asuo no firi,
Ôdomankoma ÔbÔÔ adeε.
Konkon Tano....
The river crosses the road.
The road crosses the river.
Which is the elder ?
We cleared a road to the river,
The river sprang forth long ago.
The river sprang forth,
Odomankoma made a thing.
Sacred Tano....45

30The river rises near Takyiman and runs some 400 kms through the forest zone to enter the sea at the Ghana and Côte d’Ivoire frontier. Storytellers enlarge on the theme. I draw here from a version recorded from Nana Kwabena Dwomo, Tanoboasehene and Takyiman Konahene, in 197046. Population pressure forced the people who came out of the hole at Amuowi to disperse. Asaman went to Bono Manso, Kokotipe to Nkwakuru, and so forth. One of Kokotipe’s brothers, Takyi Firi, became a great hunter, “famous for his expeditions.” His relatives would join him at his hunting camp at Tanoboase. One day a spirit possessed his sister, Afua Ankomaa. She led everyone to a large stone. The spirit said that the stone had been there “before God created heaven and earth” [presumably translating the Twi firi tete Ôdomankoma bÔÔ adeεÔ, and that it “was there for the betterment of the Bono people.” It was called Taa Kora. “A river,” the spirit announced, “would take its source from the stone and drain all over the country and it would eventually join the great sea.” This happened. The god Taa Kora then “sent its children to all the Bono villages; all the Tano gods are his sons47”.

31The riddle is: which came first, the river or the road ? The answer must be the river, if for no other reason than Odomankoma “made” it. What, however, led Odomankoma specifically to “make” Tano ? The Tano river basin was a highly auriferous area48, but that seems not to be the point. It is the fecundity of Ta Kora that is at issue, its apparently unlimited divisibility. Ta Kora is the fons et origo of the atano, the “offshoots of Tano... in villages remote from that water,” as Rattray wrote. The atano were to be found dispersed throughout the Bron and Akan lands. (Rattray, 1923: 199). In 1817, for example, Bowdich commented, “the present favourite fetish of Ashantee is that of the river Tando (Bowdich, 1819: 262.). The reproductive process involved moving water from the river to new locations, and appointing priests to serve the new “sons49. The success of the operation was much increased by the use of worthy containers for the journey. The most superior of such vessels were surely the splendid imported bronze bowls of North African and Middle Eastern manufacture, a number of which have survived50.

32In 1922 Rattray visited the principal Tano shrine in Tanoboase. A priest opened the proceedings by addressing Ta Kora. Ôdomankoma obosom ankobi na huni..., he said, “Odomankoma spirit, who did not go but sees.” (Rattray, 1923: 178). The following day Rattray was taken to an older Ta Kora shrine, situated in sandstone caves. Fragments of pottery were visible. “These,” said the priest, “have been here since Odomankoma... made things.” (ibid. 190.). It may be postulated that hunters in the process of becoming cultivators have necessarily to replace gods able to appease animal spirits with ones expert in hydrological matters. As for the road of the riddle, the drummers play,
Wofiri Donkoro Nkwanta, You come from Donkoro Nkwanta,
Wowie Adanse Akorokyere, You finish at Adanse Akrokerri,
Daammere Ôdomankoma bÔÔ adeε, Where Odomankoma made a thing,
Kankyerekyere. In the very beginning.51

33Donkoro Nkwanta, quite often referred to in ayani52, is an old town located in the extreme south of the Bron country, on the old Kumase to Kintampo road about 25 kms southeast of Nkoransa53.. Akrokerri is in the Akan country, some 165 kms south of Nkoransa and deep in the forest country. Odomankoma was to move, like the river that was crossed and recrossed, in a southerly direction.

The Odomankoma Transformation : the Akan

34The present town of Akrokerri (also known in a more recently palatalised form as Akrokyere), lies in the basin of the Pra and Ofin rivers, about 6 kms north of the present mining town of Obuase. It has a pivotal role in the Odomankoma myths. Adanse atumpan drummers play :

35ÔbirempÔn, wofiri Adanse Akorokyerε, Great man, you come from Adanse Akrokerri, Daammere Ôdomankoma bÔÔ adeε. Where Odomankoma made a thing54.

36They also play, Ôdomankoma bÔÔ adeε, odε firi Adanse Akrokyere, “Odomankoma made a thing, it first made Adanse Akrokerri55. The Kokofu drummers, in a dirge commemorating the death there of Apea Kusi, traced his origins: “you hail from Akorokyere in Adanse/ Where the Creator created things/ Very early in the morning56.

37In 1969, K. Y. Daaku directed a project for recording Adanse oral traditions. By then the modernists were well established on the scene and the traditionists in some disarray. The Akrokerrihene was firmly in the former camp. Daaku was assured that the Akrokerri people had migrated southwards from Ethiopia through Kundu and Kwankyeabo to Adanse. How, then, Daaku asked, could Odomankoma have created the Akrokerri in Adanse if they came from Ethiopia? Were the Akrokerri, in other words, autochthonous to Adanse, or were they immigrants ? And who, Daaku continued, “is this Odomankoma who is said to have created Adanse?” The Akrokerrihene could only reply: “it is asserted by some to have been God himself but actually he was some great man like Christ, who created things as he went along57. Daaku pursued a parallel line of enquiry with a former Adanse Edubiasehene Appiah Dankwa. “If the Creator first created Adansi Akokerri then after his creation whom did he then make overlord ?” Daaku asked. “Nobody actually knew who the Creator was,” Appiah Dankwa replied,

38but our elders were fond of holding in awe and fear any thing that they could find no explanation for. What I know about Akrokerri is the Patakro where the deities Bonsam and Bona are to be found is the place where the whole Adanse state meets to celebrate their annual festival. On the walls of the cave of these gods are to be found, some symbols and inscriptions. Since no one knows how these inscriptions were first formed, the people relegated their origin to the realm of the supernatural claiming that as the place bore good evidence of an ancient culture it was the Creator who on creation, made the "nkronkyeree” (inscriptions). This has since been corrupted to Adansi Akrokerri.58.

39Two informants from Fomena offered Daaku a more traditional version of the name. Odomankoma, they said, got to Adansi Bonaso where he found a man living on a tree. He hailed the man and asked him how he came to live in the place. The man responded that he had been there forever. Then the Creator asked him for directions and the man promptly told him where to pass. This so amazed the Creator that he exclaimed, “Truly, Adansi is the origin of all the states.” On being asked for his name, the man replied that he was called “Akoro” which has now become Akrokerri59. A large dossier of stories, principally about Akrokerri, had been compiled some twenty-five years earlier, in or about 194260. At that time, the traditional storytellers were still in the ascendancy. They appear, however, to have had a decidedly unsure knowledge of the drum texts, and it is possible that the qualified Odomankoma Drummers had been killed or dispersed during the disastrous civil wars of the 1880s. In one item in the dossier “Boadi,” that is, b]] adeε, is treated as first ruler of Akrokerri. It was however noted that, “Boadi is a mythical person and we didn’t know precisely during the date of his time…. It was he who brought the light and seat of Akan speaking people.” As a result (the “Lord”) Boadi acquired the name “Odua-Ankuma”, which became corrupted to Odomankuma. The writer had some knowledge of the drum texts having to do with Odomankoma’s death and with affairs then passing into the hands of men. “Odua-Ankuma,” he thus explained, “means the Lord had planted us and departed but [we] waited for his return, waited patiently and long, but he returned not, but that mystery remained with me...61”. The “Adansi Legend of Creation” is another item in the dossier:

It appears that Akrokerri was first created by Odumankuma the Existence of Existence. The first ancestor was called Kyewa Bondah, a woman begat Asare Adikyiye, Asare Yansa and Gyapomaa, Gyapomaa begat Ntori Benchi who on one occasion went to god Bondah and supplied him food. There, he found Bodom Akondwa (an Aggrey Bead Stool) and brought it home and sat upon it. On the second day, he went there again and found a sword. He brought it home and named it Afuna Kwaw, and placed it upon the stool. In the third day, he found three large pots (Brotua) full of gold and brought them home. At that time gold was not known, but copper called “Ehabuo” was in use. Ntori Benchi having possessed all these things proclaimed himself the first King of Adansi…. (Dossier, ca.1942 : 7).

40The “Asare Yansa” of the above text, son of the putative founding ancestress, was the subject of another story recorded by Daaku in 1969. Before the time of Asare Nyansa, the Akrokerrihene said, “our people never settled at one place for long. They stayed at different places.” During his rule, “if anyone came across somebody’s property one could keep it if one wanted.” Things were to change, for “Chief Asare found this to be a bad practice and determined to do something about it to prevent people from keeping other’s property62.” One item in the dossier of ca.1942 places the emphasis upon sedentarization. “Asare Nyansa, king of the wise,” it reads,

emphasised the point that man should hide [sic: bide] himself in a house. It was he who elected a house building and the extraordinary exhibition of his wisdom akin to that of king Solomon of old reached the length and breath [sic: breadth] of the land and the king was held in high esteem by his subjects and his name was never forgotten. (Dossier ca.1942: 40)

41The point is taken up in another item in the dossier. Asare Nyansa is highly praised:

He beautified the town of this great empire with many beautiful buildings, and his palace was a stone building. He did indeed strive to give a constitutional character to his rule and showed great wisdom in governing by calling together a regular assembly or councils called Abagua. (ibid. 19).

42Notwithstanding the hyperbole in this last item, matters of sedentarization and property rights are clearly at issue.  

43There is a saying in Twi that is axiomatic to the Akan cosmogony: Akanman piesie nnum. Twi Ôman, as in Akanman, is difficult to translate. The word now has a wide range of meanings, from “community” to “nation” to “state,” but in the present context “place” is vague enough to be appropriate. Hence the saying may best be rendered into English as “the five first-born Akan places.” The sites of the Akanman piesie nnum all lie within the Pra-Ofin basin. Each is associated with a sacred place where (as in the Bron country) apical ancestors and ancestresses are said to have arrived either from a hole in the earth or from the sky63. In 1969 the Adansehene, in the presence of his elders, addressed the matter of their identity and seniority. “In the beginning,” he said, “God created five Akan states. The eldest was Adansi. Second in prominence, might, and age, was Akyem, then came Assin as the third followed by Denkyira as the fourth. Asante was the fifth, last and smallest.” The sequence appears to have been the commonly accepted one64. Names that I heard used in the 1960s for Odomankoma’s five Akan creations were, to follow the same order, Adansemanso, Abuakwa Atwumamanso, Asenmanso, Abankeseso, and Asantemanso65. To the best of my knowledge the earliest written list of the five was that dictated by the exiled Asantehene, Agyeman Prempeh, in or about 1907: “Asante Akan noom. They mean Ashanti divided into five cities, 1) Asin 2) Adansi 3) Denkira, 4) Asanti or Ashanti 5) Akim66”. Despite the apparent authority behind this statement, it is quite obviously couched in anachronistic and Kumase-centric terms. We may be sure that the presentation of the five  “cities” as divisions of Asante – and indeed the very notion of “city” as such – was intended to boost the prestige of the fourth on the list, “Asanti or Ashanti,” and is unlikely to have been so formulated until after Asante had emerged, at the beginning of the eighteenth century, as the dominant regional power in the forest country67. There is, in fact, universal agreement among the storytellers that Adanse was the first of the five Akan aman.

44The Gâ pastor Carl Christian Reindorf learned something of these matters in the later nineteenth century. “Adanse,” he wrote, “was the first seat of the Akan nation, as they say by tradition: there God first commenced with the creation of the world.” The Basel missionary and linguist J. G. Christaller felt it necessary to insert a note in Reindorf’s work. “The expression ‘b] ade, to create (consolidate) the things,’ may refer,” he wrote, “to the beginnings not only of the world, but also of a state. Chr[istaller]68”. This rather missed the point, for it was not to the creation of the world that Reindorf referred, but to that of a world: namely, that of the Akan. Be that as it may, the pastor knew somewhat more of the stories having to do with Adanse. The very practice of matrilineality, he reported, originated there: “There was one Abu who seems to have been one of the kings of Adanse, who instituted the order of family among the Tshis; hence lineage is designated ‘abusua’ i.e. imitating Abu69”. More recently the aged Adanse headman (odekuro) of Fumso and his elders gave a very different explanation of the practice:

Tradition has it that in the olden days food was very scarce. A man unexpectedly came across some people who had yams and was told to bring a human being to exchange some of the yams. He went to ask his wife to allow him to take one of their children for the purpose but she refused but his own sister allowed him to take one of her own. Thus, it became common for the nephew to inherit his uncle instead of the son, his father70.

45Whatever the case, all the people of the Akanman belong to one or other of the seven great lineages that Odomankoma, as the traditionists have it, made “at the beginning of time71”. The issue of settlement, in the sense of sedentarization, recurs. Reindorf remarked that the very name “Adanse” means, “the foundations of the buildings i.e. the building of the Tshi nation72”. He was identifying the root of the name as the Twi adan (singular Ôdan), “houses,” so that adan-ase has the sense of “place of the houses.” The Adansehene and his elders gave an explicit account of this matter in 1969:

We the people of Adanse were the first to build houses. When the world was created people did not sleep in houses but in the spaces formed by the buttress roots of large trees and under flimsy shacks constructed of twigs and animal skin. The Adanse people built the first mud houses and became known as Adanesi fo].... It was the people of Adanse who first realised the need for and built the first mud-houses73.

46The association of Adanse with housing is well known. In the 1950s Nketia recorded a dirge sung by the Adanse Ayaasehemaa and others:


48Nana a Ôfiri Hwerebe Akwasiase Ancestor came from Hwerebe

49Akwasiase. εhÔna Ôdomankoma sii mmukyia, Where Odomankoma built a fireplace, Kankyerekyere, In the very beginning, Asabaa da ho. A pestle beside it74.

50AbutiaboÔ is probably the Abu thought by Reindorf to have been an Adanse king. Nketia reports that he is known to the storytellers as “the first male king to descend from the heavens,” that he did so at Hwerebe Akwasiase, that he belonged to the Bretuo lineage, and that he and his people subsequently founded nearby Ayaase75. Of the five Akanman piesie, the best-known stories are those having to do with Asantemanso, the name of which is of course a misnomer, since the coining of the name “Asante” occurred only in or about 160076. On the way to Asantemanso, Odomankoma, accompanied by his Ôkyeame Amoa and his treasurer [fotosanfo], stayed with Akwatea Fifire and Aberewa Mmusu. The treasurer was ill during the night and died in the morning. Odomankoma produced a hoe and buried the man, and then proceeded to Asantemanso. Arriving at Asantemanso, Odomankoma said, “Amoa bring my stick for me to create things.” Amoa had forgotten it at the place where they had buried the treasurer. He was told to recover it, which is how the place came to be called Amoafo77. There is another story to do with Odomankoma’s journey to Asantemanso. On his way, at Manso Obodom, he had encountered Asantemanso. On his way, at Manso Obodom, he had encountered Ogu Yeboafo who maintained that he had created himself and that he was Aduana. Odomankoma told him that he was intending to create another Aduana and that Ogu Yeboafo should accompany him. After a long journey they reached Asantemanso, where Ogu Yeboafo said that he was tired. Odomankoma allowed him to return home, “but first told him that in order that people might ecognise him as a chief of ancient standing, he would give [him] something whereby people would recognise his seniority.” At this point in the narrative, the storyteller displayed “a very old Gold weight of a person sitting on a stool wearing an adebereku kyew hat,” and said that when Odomankoma went to Asantemanso to create things, “a creature of which this weight is the image came out of the ground wearing an adebereku kyew hat and sitting on a stool.” This was the first Asumagyahene78.

51The origins of two of the major Akan matrilineages are in Asantemanso, namely, the Aduana and the Oyoko. For our knowledge of the former we are particularly indebted to the good relations that Rattray established in the early 1920s with Abena Firi, the Asantemansohemaa (or Asantemanso “queenmother”). She told him that on a Monday seven men and five women (all named), and a leopard and a dog (unnamed), came from the ground, and “by Wednesday they had begun to build huts” – the matter of sedentarization thus recurring. They were of the Aduana lineage. Abena Firi, if Rattray understood her correctly, then told “how Odomankoma (the Creator) on his journey about the earth ‘making things’ met these people already settled here....79” Odomankoma, she added, “took one of our [Aduana] ancestors with him as his ‘linguist’80”. Of the five women, Aberewa Samante, became first Asantemanso queenmother. One of the men, Kusi Aduaku, founded nearby Asumegya, and is still accounted its first ruler (Rattray, 1929: 131).

52There is a version of the story of the Asantemanso Oyoko that must be considered authoritative if not definitive. The exiled Asantehene Agyeman Prempeh in the Seychelles compiled it in the first decade of the twentieth century. The following does no more than briefly summarise the highly detailed account. Ankyewa Nyame, accompanied by retainers, arrived by way of a chain from the sky. Two other Oyoko families appeared from a hole in the ground, and sat to either side. Birempomaa Piesie was a daughter of Ankyewa Nyame. Three of her sons presided successively over the affairs of the Oyoko. Kwabia Anwanfi, a nephew of Birempomaa Piesie, followed. Asantemanso, he decided,

was not a convenient place for him to dwell with his families so he sent his hunter to seek for a place so that they may dwell there. This hunter in searching for a place, found a gardiner [sic] by name Kokor who had made a large garden or Affuo (the Ashanti name for a garden). So this hunter returned and told the King and so the king used to make a promenade to the garden of Mr. Kokor. When they were accustomed of the place, the king said to his families that he will go and make a dwelling at Kokor Affuo, i.e. the garden of Kokor, so they all went and dwell there. When they had settled there, they at first called the place A Kokor Affuo, and afterwards the place came to be called Kokofu. The name Kokofu which has retained up to nowadays81.

53The Agyeman Prempeh manuscript follows the story of those Kokofu who subsequently made the Kwaman Purchase and founded Kumase (Wilks, 1993: 100-103.). Rattray was told that Pinahin Panin, another daughter of Ankyewa Nyame, became first Kokofuhemaa and that her son, Kwame Gyami, was first Kokofuhene82.

54Asumegya and Kokofu were, then, respectively the premier foundations of the Aduana and Oyoko of Asantemanso. Brief reference may also be made to the Bretuo and Tana lineages of Adanse that, in the course of time, came to provide the rulers of, inter alia, the ranking division of the Asanteman after Kumase, namely, Mampon. In the early 1920s Rattray was able to record the text of one drum history, which he divided into twenty-nine stanzas83. The drummer, Osei Kwadwo, opened his performance with seven stanzas that are virtually a panegyric of Odomankoma, incorporating some of the pieces already cited in this paper. Stanza VIII invokes Asiama Toku Asare, who came from the sky84. Stanzas IX and X commemorate Boafo Anwoma Kwakye and Antiedu, who are said to be two of her sons and successive leaders of the Bretuo85Stanzas XI and XII commemorate the two next leaders, Maniampon and Adu Boahen. Both are said to have been brothers, grandsons of Asiama Toku Asare, and the latter to die in the Denkyira war86. The drummer plays,
Adu Boahen,
Boahen Kwadwo,
Wo firi he ni ?
Wo firi Mampon Akurofonso,
Daammere Ôdomankoma
bÔÔ adeε....
Adu Boahen,
Boahen Kwadwo,
From where did you come ?
You came from Mampon old place,
Where Odomankoma made thing...87

55Stories of this genre, similarly belonging to that borderland between protohistory and history that was the domain of Odomankoma, exist for the other Akanman piesie nnum. As population increased and new settlements were founded, each offshoot in turn preserved memory of its own origins in stories that revered the ancestors and, on the basis of descent, regulated marriage and asserted rights to land and office. By this time, however, the passage from firi tetemu to abÔkasεm has been made. Odomankoma has completed its work, henceforth to be commemorated in the Adomankomasεm, The Odomankoma Tales.

Part II : matters of chronology

Archaeology of the Bron Relict Landscape

56Three of the places of “first creation”, Bono Manso, Wenchi and Bewo, have been identified as (in some sense) Bron, and five, Adanse, Akyem, Asen, Denkyira and Asante as (in some sense) Akan. Excavations have been carried out at five of them. Those at Bono Manso88 and Wenchi89 were published expeditiously. Full reports on those at Bewo, Asantemanso, and Adansemanso are awaited, and any discussion of them must necessarily be tentative. In referring to places said to have been “made” by Odomankoma as ones of “first creation,” I do so in a sense that takes into account Wiredu’s principle that Odomankoma could not “make a thing” (bÔÔadeε) out of nothing. The archaeological record, in other words, should not be expected to begin with Odomankoman acts of creation. Excavations, for example, have been carried out at a site known as Bonoase. It belongs to the subsistence system that was for long labelled, perhaps rather misleadingly, the Kintampo Neolithic. Joanna Casey has recently argued that,

the Kintampo peoples may have been already starting to settle in communities and benefit from small-scale land-clearing and cultivation. The maintenance of communities and the harvest and storage of wild and some domestic foods would have enabled people to live in some areas throughout the year… As the climate continued to dry out, and people became more embedded in the economic and social commitment to a sedentary lifestyle, they may have been more amenable to the introduction of more productive crops such as millets, and possibly small livestock90.

57Bonoase was indeed a village. Its people neither made nor made use of iron; but had a considerable range of stone tools and particularly of grindstones presumably for the preparation of grains for consumption – though whether wild or cultivated grains is unclear91. The village is confidently dated to around 1500 B.C.92. As it happens the site is near the hole, Bonkese, out, of which the Bron of Bewo are said to have come. It seems prima facie likely that the inhabitants of Bonoase were indeed those whose descendants came to be, in the course of time, culturally transformed into the Bron of Bewo. At some point in time, agricultural villages were certainly established in the savanna-woodland country that lay immediately north of the high forest, though the system of subsistence was one that could not be transferred into the forest zone. The boundary between cultivators and hunter-gatherers was one that was to survive over two to three millennia93. Tradition has nothing to tell us, for the Bonoase villagers belonged to the world of protohistory, firi tetemu, and so to one that has been lost to the Bron of the historic period. We do know, however, that the belt of country along the northern forest and grassland interface has a history of manufacture that preceded the rise of Bewo and the other Bron towns. Posnansky located an iron-smelting furnace on the Bewo site (Atwetwebooso) that was radiocarbon dated to 180 ± 75 a.d.94 on the site of Bono Manso a smelting site was located by Effah-Gyami dating to 320 ± 3095. “By about the eleventh century, if not before,” Garrard writes,  

iron smelting and smithing appear to have been regularly practised in the Begho [Bewo] area, and probably also around Takyiman.... By about the fourteenth century the Akan were probably producing as much as 100 tons of iron each year and possibly more. By this time, sufficient iron was available for the techniques of blacksmithing to be widely practised by the Akan.” (Garrard, 2003: 151-53. )  

58It is unclear whether this development should be attributed to the Akan – or better, Bron – as such. There is certainly reason to believe that emigrant smiths of Malian origin, known as Numu, were instrumental in building up the iron industry in the region. Adjacent to the Bron settlement at Bewo there was a community of blacksmiths. It is remembered locally as the Tomfuor (Twi ÔtomfoÔ, “smith”) or Dwinfuor (Twi adwinfoÔ, “artificer”) section. Those dispersed people who now regard their forebears as having once lived at Bewo remember them as having been Numu96.

59In 1975 a large smelting complex was located near Dapaa, some 3.5 kms north by northwest of Bewo. Posnansky regarded it as “clearly the result of a very intensive period of iron working.” Radiocarbon dates of 1400 ± 100 and 1480 ± 65 from one slag mound, and of 1650 ± 95 from another, led him to suggest, “that this mighty ironworking complex belongs to the period of the heyday of Begho [Bewo]97”. It is difficult to identify any market for production on this scale other than the numerous small agricultural villages lying beyond the forest limits. Garrard maintains that by the late fourteenth century imports of brass and copper from northern sources were leading to the development of a brass casting industry in the Bron country, and that this new technology began to spread south, towards the coast, in the early fifteenth century (Garrard, 2003: 154). In 1975 “a large brass foundry” was found in the Dwinfuor section of Bewo Posnansky, 1987: 17).

60It has become received wisdom that the manufacturing base in the Bron country was expanded and indeed transformed by the development of the gold industry, both in its extractive and distributive sectors. Eva Meyerowitz had much to say about this matter, having collected stories from Takyiman in the 1940s and 1950s. To the point here, she attributes the beginnings of gold mining in Bono to the third ruler Obunumankoma who, as a youth and in the company of Muslim traders, travelled in the western Sudan. He created the Bono treasury and a system of weights for use in it98. The general thrust of the tradition, identifying the time of Obunumankoma as that, in which the treasury was set up, appears acceptable, but Meyerowitz’s dating of his reign to 13631431 is not. In a statistically sophisticated paper Colin Flight reviewed her computations of the reigns, and concluded that “the formation of the Bono-Manso state appears to date from the early fifteenth century.” (Flight, 1970: 259-68). This would have the effect of dating the development of the gold industry to around the middle of that century. Nor surprisingly, this does not show up archaeologically in the appearance of gold dust or gold artefacts, but it does in that of weights.

61Garrard convincingly demonstrated that ten (undamaged) ground pottery disks found at the Bron section of Bewo were in fact weights of a standard Muslim series, corresponding to ¼, ½, ¾ and 1 ounces99. Crossland noted that such weights were found in small numbers in all parts of the town, thus indicating, “that all sections of the population were probably engaged in the gold trade100”. Effah-Gyamfi found thirteen such neatly ground pottery disks in the vicinity of the Amuowi site near Bono Manso101, and again Garrard was able to show that three of them corresponded to 2, 4 and 16 mithqals in the Muslim mithqal series, nine of them to ½, ¾, 1, 1½ and 2 ounces in the uqiya series, and only one of them was indeterminate (Garrard, 1980: 221, 224). It is difficult, granted the purchasing power represented by these weights, to envisage any transactions other than ones in gold that would, in the context of the times, require them: one presumably did not weigh iron, cloth, or beads with them. So far as is yet known, none of the Bron pottery weights seems to be based on the Portuguese ounce, suggesting that they were produced before the arrival of the Portuguese on the Costa da Mina in the late fifteenth century.

62There are other indications of Muslim involvement in Bron affairs at this time. Crucial was the growth of the section of Bewo now regarded locally as having been founded by the Kramo, that is, Muslim Juula (or Dyula, etc.) of Malian origins102. These were traders, who continued to use Malian patronymics. Spiritual authority was exercised by the Bamba (who held the imamate), and secular authority, so it seems, by the Kamaghatay (Wilks, 1982: 348.). In the mid- to late-sixteenth century two Bewo shaykhs, Isma’il Kamaghatay and his son Muhammad al-Abyad, entered the service of the ruler of Gonja kingdom to the north, and their descendants were thereafter to hold the Gonja imamates. Early eighteenth century Gonja chroniclers rendered Bewo in Arabic as “Bi‘u”, “Bighu”, and sometimes “Biku,” and in the nineteenth century the French form “Begho” was coined (and has been widely adopted by scholars who should long ago have abandoned it)103.

63Little of the Kramo section of Bewo has been excavated. Crossland pointed out that no sign was found of Mande influence on ceramic styles in the town in general, and correctly observed that the Juula who settled at Bewo usually married local women, thus not engendering change in existing vessel forms (Crossland, 1989: 104). However, he also noted that some earthenware shapes were clearly influenced by the Arabic-inscribed brass bowls imported from North Africa and the Middle East by way of the Sahara, or by brass copies of them made by western Sudanese and in time local Bron or Akan smiths104. A piece of Chinese porcelain found in the Bron section of Bewo and dated to the late sixteenth or early seventeenth century had probably passed along the same caravan roads105.

64The people of a number of communities in the Bron region still speak Juula and Ligbi (including the Numu of the blacksmiths), languages closely related to or dialects of Mandekan (Dakubu, 1976: passim). In 1962, I drew attention to the existence of a Mandekan loan element in Twi as such, and remarked that most of the borrowings had to do with commerce106. Some thirty years later M. Kropp Dakubu generously reviewed my suggestions, and accepted, as “well-founded,” twelve out of the sixteen loan words I had labelled “certainties.” The main linguistic impact of Mandekan on Twi, she wrote, “most probably reflects the cultural impact of the Mande in the north-west…. that preceded the rise of Asante. The words involved are therefore relatively old and well-integrated linguistically and culturally.” (Dakubu, 2001: 273-91.). Dakubu, moreover, significantly expanded my list, and her view that the Twi sika, “gold, money”, was a borrowing will be the subject of comment later. She also pointed out that Twi burÔn, “street, lane”, was a direct borrowing from the Mandekan with the same meaning (Dakubu, 2001: 283). An observation of Rattray’s on the names of the southern Akan for their mmusua keseε, or major lineages – Aduana, Oyoko, Bretuo, and so forth – is particularly germane to this. He expressed his surprise on discovering,  

that the pure Brong are apparently wholly ignorant of these Ashanti and Fanti clan names, and that their exogamous divisions seem based upon an entirely different model. In Tekiman one does not ask, ‘To what abusua (totemic clan) do you belong?’ but ‘In what street (abrono) do you reside?’ There, such exogamous divisions as appear to exist apparently take their names from streets or quarters in the towns. (Rattray, 1929: 64.)  

65On the face of it the mborÔn or streets were associations not so much for the management of land but rather for that of trade and perhaps ironworking. The Mandekan origin of the term is surely a pointer in this direction.

66The three Bron places of first creation are, as one might anticipate, shown on the earliest map of the inland region that is extant. Hans Propheet produced this in 1629 telling us, that he made enquiries from “the most experienced Blacks107”. The furthermost country about which he heard was Bewo, though under its alternative name, Nsoko (“Insoco”). “There is no gold or trade from there,” Propheet wrote, presumably implying that the Dutch had not yet secured access to the major Bron entrepôt. He learned, however, that the Nsoko “have very fine goods,” and that they wove cloths “like carpets” that were favoured by the Akan (“Acanists”), a reference perhaps to the famed kente. Propheet also noted, inter alia, that the Nsoko had horses108. Wenchi (“Wanquie”) and Bono (“Bonnoe”) are shown, correctly enough, as lying south of Bewo. The former, Propheet writes, “have gold, and they are merchants”, but of the latter he notes, “Simple people. No forest.” Some four decades later Olfert Dapper was to improve on Propheet’s descriptions, but not always from sources that can be identified. Bono remained, apparently, terra incognita: “there is,” Dapper wrote, “little trade in this area.” Either the merchants of Bono were required to conceal the extent of their activities or some monumental confusion of names had occurred109. But Dapper was quite effusive about Wenchi (“Wanki”) and even more so about Nsoko (“Insoko”). “The former,” he wrote, “has gold and beautiful cloth which the inhabitants make very artistically and sell to the Akanisten in the country.” The Akanisten (or Akanier), however, “very rarely go to Insoko because the road is made very unsafe by robbers”. It was, however, famed for its manufacture. “The inhabitants make very fine cloths with which they can make a good profit if they can get them through. They are bought for pieces of eight and other silver and also for several European cloths.” The more profitable gold extractive industry, and indeed brass working, remained protected by silence. “The inhabitants, according to the Akanier,” Dapper added, “know no gold or copper, and do not trade in them.”110

67The destruction of Ahwene Koko in the course of an Asante campaign is referred to in contemporary sources; it occurred in or about 1715 when the capital was moved to the present Wenchi111. On the basis of contemporary documentary sources, Bono Manso is believed to have been abandoned, and refounded as Takyiman, after the Asante invasion of 1722/23112. In this period Bewo, or Nsoko, was drawn into the ambit of Asante and its trade was increasingly controlled by the Kumase market. Having lost its pre-eminence in the northern trade, Bewo’s prosperity declined. In 1773-74, having joined Banda in rebellion against Asantehene Osei Kwadwo, it was invaded and seems never to have recovered. Thereafter internecine struggles between Muslim and non-Muslim sections of the community led to the town’s abandonment113. A terminus ad quem for each of the three Bron towns of “first creation” can, then, be established. But what can be said of their “first creation”, in the sense of their Odomankoman transformation?

68Interpretation of the radiocarbon dates obtained from the relevant excavations must necessarily be tentative, since their number (or at least of those known to me) is small. Iron smelting sites as such are starred. Otherwise the charcoal samples were obtained, for the most part, from house sites.

a.d. 180 ± 75


Posnansky and McIntosh, 1976: 165, 189

a.d. 320 ± 30

*Bono Manso/Abam

Calvocoressi and David, 1979: 12, 22

a.d. 440 ± 70

Bono Manso/Ammuowi

Posnansky and McIntosh, 1976: 166

a.d. 750 ±  90


Posnansky, 1976: 62

a.d. 1005 ± 85


Posnansky, 1976: 62

a.d. 1045 ± 80


Posnansky and McIntosh, 1976: 166

a.d. 1120 ± 80


Posnansky and McIntosh, 1976: 166

a.d. 1235 ± 85

Bono Manso

Posnansky and McIntosh, 1976: 166

a.d. 1370 ± 75

Bono Manso

Effah-Gyamfi, 1985: 204

a.d. 1400 ± 100


Posnansky, 1976: 61-62

a.d. 1430 ± 100.


Flight, 1973: 548

a.d. 1450 ± 100.


Flight, 1973: 548

a.d. 1480 ± 65.


Posnansky, 1976: 61-62

a.d. 1520 ± 75.

Bono Manso

Effah-Gyamfi, 1985: 204

a.d. 1520 ± 75.


Posnansky, 1976: 62

a.d. 1530 ± 30.

Bono Manso

Effah-Gyamfi, 1985: 204

a.d. 1540 ± 75.


Flight, 1973: 548

a.d. 1550 ± 100.


Flight, 1973: 548

a.d. 1585 ± 70.

Bono Manso

Effah-Gyamfi, 1985: 204

a.d. 1595 ± 80.

Wenchi/Ahwene Koko

Posnansky, 1976: 62

a.d. 1595 ± 60.


Posnansky, 1976: 62

a.d. 1610 ± 80.

Bono Manso/Ammuowi

Posnansky and McIntosh, 1976: 166-67

a.d. 1615 ± 70.

Bono Manso

Effah-Gyamfi, 1985: 204

a.d. 1635 ± 30..

Bono Manso

Effah-Gyamfi, 1985: 204

a.d. 1650 ± 95.


Posnansky, 1976: 61-62

a.d. 1665 ± 100.


Flight, 1973: 548

a.d. 1710 ± 75.

Bono Manso

Effah-Gyamfi, 1985: 204

a.d. 1710 ± 100.


Flight, 1973: 548

69On the basis of this list it is difficult not to see the era of Bron prosperity as extending over three centuries, the fifteenth through seventeenth. The Bron world that Odomankoma made is not remembered in the stories as having been a “Golden Age” but as one of all-encompassing change. If it was not, however, a “Golden Age”, it was certainly an age of gold.  

Archaeology of the Akan Relict Landscape

70Garrard has recently reiterated his view (using “Akan” to include the Bron), that “there seems to be no evidence, archaeological or historical, from which a trade in Akan gold can be confirmed before the late 14 century”115. I agree. The Juula settlers in Bewo played a crucial role in organising flows of gold from the Bron country to the entrepôts of the western Sudan and beyond. They controlled the distributive trade. The seemingly insatiable world demand for the metal led them to extend their operations further south. There were no roads and the nature of the terrain obliged them to follow watercourses116. Topography dictated that the earliest sorties proceeded by way of the Tano, since that river rose in the Bron country117. Be that as it may, access to the most productive goldfields was achieved when traders crossed the Mpesaso Hills that separated the Tano valley from the upper reaches of the Ofin. I have noted elsewhere that at least one writer, Alvise da Cà da Mosto, heard that some gold traders had even arrived at the coast. This was in the mid-fifteenth century (Wilks, 1982 (a): 340-42.). It is likely that they followed the Pra south from its confluence with the Ofin. They would have reached the coast at Shama, where the Pra enters the sea. There is reason to believe not only that they could have done so but also, we shall see below, that they did.

71Many scholars have been reluctant to recognise the crucial role played by Malian traders in the fifteenth century gold boom in the forest country. As recently as 1994 P. E. H. Hair could refer to a comment of mine that appeared in 1962, to the effect that “itinerant Mande…. were already in the Elmina area when the Portuguese arrived,” and summarily dismiss it118. Yet in 1980 Garrard had remarked that, “by 1471, when Europeans reached the Gold Coast, Mande traders may have been active in the coastal villages for many years,” and offered a great deal of new evidence in support of this view. “Begho and other northern market towns,” he wrote,  

served as bases from which, probably by the end of the fourteenth century, groups of Mande began to come down the forest paths as far as the coast. These traders may have been responsible for introducing weights to the Akan coastal areas. The southern Akan were now drawn into the gold trade, which came to involve a wide area of the forest belt stretching 300 miles along the coast east of the Bandama River and extending for about 250 miles inland.” (Garrard, 1980: 25)

72The Juula organised the distributive trade, and created the busy frontier market at Bewo. The Twi-speaking people of the northern forest country, the Bron, built up the extractive industry. Their long involvement in iron working presumably facilitated the economic initiatives that resulted in the emergence of the boomtowns of Bono Manso and Wenchi. The Juula-Bron consortium (so to speak) was reconstructed as a Juula-Bron-Akan one when operations were extended into the southern forest country. This is the sequence of events that is commemorated by the Odomankoma Drummers and the storytellers in texts having to do with the “making” of Akrokerri and of the Akanman piesie nnum: Adansemanso, Abuakwa Atwumamanso, Asenmanso, Abankeseso, and Asantemanso. Peter Shinnie and Brian Vivian have carried out excavations at two of these, Adansemanso and Asantemanso.

73The site of Adansemanso lies 1 km west of Old Edubiase and about 8 kms east of Akrokerri119. The storytellers regard it as the first settlement “made” by Odomankoma after his departure from Akrokerri, the first, that is, of the Akanman piesie nnum. Ewurade Basa, they say, emerged from a hole in the ground. Accompanied by his Ôkyeame or spokesman, Ôkwawe Nfrafo, he brought with him the sacred sword known as Afena Kwa. From his settlement at Adansemanso, he ruled over Adanse. His tyrannical behaviour led his people to criticise him, but only indirectly. This gave rise to the saying : Yεkasa Nfrafo, nso yεne Awirade, “we reproach Nfrafo, but we mean Awirade120”. On his death authority became decentralised, and many small, autonomous Adanse aman arose, Ayaase, Dompoase, Edubiase, Fomena and so forth. This is commemorated in the well-known refrain, nkotoa nkotoa, literally, “small crabs, small crabs” – each of which has its own hole.  

74Mounds of 100 metres or more in length were excavated and found to be the remains of rows of houses, aligned in streets. Only small gaps separated one house from the next, and many of them had an earthen larder dug against one of the external walls121. Shinnie estimated that occupation debris covered an area of more than one square km. There is evidence of iron production. Artefacts of other than local production were rare: two geometric gold weights, a cuprous bell, one smoking pipe, and two pieces of pottery possibly, but not certainly, of European origin. Fourteen small terracotta funerary heads were found, all resembling those from nearby Ahensan. Shinnie suggests that Adansemanso was abandoned before becoming heavily involved in trade with the European coastal establishments. Nine radiocarbon dates were obtained that seem to indicate a fifteenth century floruit :

a.d. 590 ± 160

a.d. 1410 ± 50

a.d. 1460 ± 80

a.d. 1370 ± 160

a.d. 1420 ± 150

a.d. 1520 ± 90

a.d. 1380 ± 50

a.d. 1440 ± 60

a.d. 1640 ± 30122. 122

75Further interpretation of the results must await the appearance of the full excavation report. A coincidence, however, may be noted. On the one hand the spade reveals the existence of rows of houses resting on undisturbed soil. On the other hand, storytellers give much attention to Asare Nyansa, who is credited with the introduction of an innovative building style in Adanse and to the promulgation of laws intended to protect property. I assume that the development of food crop cultivation is the underlying issue; that is, as hunter-gatherers began to make the transition to agriculture, they wrested farms from the forest, built permanent dwellings, and claimed personal property rights. The akuraa, or village, of the farmer, thus superseded the nnanso, or camp, of the hunting band. Tumi nyina wa asase so, power passed to the land.

76The sacred sites at Asantemanso, from where the Aduana and Oyoko ancestors take their origin, lie some 40 kms south of Kumase. They are thickly wooded, and all forms of cultivation are forbidden123. In 1986, 1987, and 198991 Shinnie was permitted to dig in the surrounding areas where occupation debris was abundant. He distinguished three adjacent areas of settlement and dates these, on “a rough assessment,” to 800-1200, 1200-1500, and 1500 to present (Shinnie, 1996: 198-200). The average depth of occupation was about one metre, and the maximum two metres124. There is evidence of ironworking in levels of about 1000 A.D. or even before. Fragments of European pottery and glass, and many locally made smoking pipes, were present in the later (post1500) levels. There was, however, one remarkable feature. Earlier levels of occupation revealed a completely different style of pottery from later ones. (Shinnie and Shinnie, 1995: 6-15). The second was obviously ancestral to present day (traditional) pottery, and I shall refer to it simply as Akan-ware. The first, which was far more ornate and heavily flanged, appears to have become extinct (except, perhaps, as specialised funerary vessels). As long ago as 1937 Robert Wild, an Inspector of Mines in the Gold Coast administration, had recognised its characteristics. He consulted modern potters, who always insisted that it was the work of the atetefo, the people of long ago (Wild, 1937: 94-99.). I shall refer to it, therefore, as Atetefo-ware.  

77The Atetefo-ware was the subject of an important paper published by J. Bellis the year before the Asantemanso excavations began (Bellis, 1987: passim). He was able to show that Atetefo-ware (and I trust that he will accept the term) was turning up at virtually every site of a requisite age throughout the forest zone of Ghana, and always at levels beneath those of the Akan-ware. At his own excavations at Twifo Heman Bellis thought that Akan-ware did not begin to appear until about 1600. He read D. Kiyaga-Mulindwa as indicating, on the basis of his excavations at the earthworks in the Birim valley, that two radiocarbon dates establish a time bracket for the change : 1510 ± 80 and 1740 ± 110125. “At some time before A.D. 1600,” Bellis wrote,

a ceramic tradition which had been present throughout the forest zone seemed suddenly to disappear.Though we cannot yet say with confidence how long the tradition had been there, its broad distribution and its great stylistic homogeneity suggests that it may be considerably pre-European in its origins. At some time between A.D. 1500 and 1600, this early tradition was replaced with a wholly new constellation of ceramic wares, which also seemed to be just as widespread in the entire forest zone. (Bellis, 1987: 47-48)

78Bellis asked how we explain this. He rejected, and rightly so, an answer that required some form of massive population displacement, but recognised that if an evolutionary model is accepted, then a proto-Akan culture changed “at an almost explosive rate” into an Akan one. “This would require,” Bellis observed, “that we somehow explain why a culture which has moved along for perhaps centuries with little apparent change, suddenly hits an evolutionary burst of speed that is totally unprecedented.”

79We have a truly Odomankoman style riddle. The answer does not have to do with gold. Bellis showed that the metal was being extracted, and weighed for sale, on tributaries of the Ofin in the vicinity of the present day Dunkwa, and that the workings were associated not with Akan- but with Atetefo-ware (Bellis, 1987: 37-38.). The answer surely has to do with the development of cultivation, and the different kinds of utility pottery required by more mobile and less mobile populations126.

80Shinnie obtained some forty radiocarbon dates from Asantemanso127:

a.d. 660 ± 80

a.d. 1280 ± 80

a.d. 1550 ± 80

a.d. 880 ± 80

a.d. 1310 ± 80

a.d. 1550 ± 80

a.d. 930 ± 130

a.d. 1320 ± 80

a.d. 1550 ± 100

a.d. 970 ± 80

a.d. 1350 ± 80

a.d. 1640 ± 80

a.d. 1040 ± 80

a.d. 1360 ± 50

a.d. 1650 ± 50

a.d. 1100 ± 110

a.d. 1370 ± 80

a.d. 1670 ± 80

a.d. 1140 ± 80

a.d. 1400 ± 80

a.d. 1670 ± 80

a.d. 1220 ± 100

a.d. 1420 ± 90

a.d. 1710 ± 70

a.d. 1230 ± 60

a.d. 1450 ± 80

a.d. 1720 ± 80

a.d. 1240 ± 80

a.d. 1477 ± 40

a.d. 1740 ± 80

a.d. 1250 ± 80

a.d. 1480 ± 50

a.d. 1740 ± 80

a.d. 1260 ± 80

a.d. 1510 ± 80

a.d. 1760 ± 80

a.d. 1270 ± 100

a.d. 1530 ± 70

a.d. 1280 ± 80

a.d. 1540 ± 80

81No direct evidence was found either of agriculture, or of house building other than a few possible floors (Shinnie, 1996: 199-200). However, the lower levels in the excavations yielded Atetefo-ware and the higher levels, Akanware. Shinnie thought that my inclination to date the beginning of “settled life” in the forest to the fifteenth and sixteenth centuries was unacceptable. “Though the evidence from Asantemanso strongly implies an earlier date than that proposed by Wilks,” he wrote, “there is nothing to suggest that the mechanism by which agriculture and the subsequent political and social developments of the proto-Akan was other than that which he proposes....” (Shinnie, 1996: 201) As at Adansemanso, however, sometime that transition from nnanso to akuraa must have occurred at Asantemanso. It is my impression that this has to do with the shift of the population from Shinnie’s 1200-1500 to his 1500-present locations. The Nampansa site to south and west of the Aduana sacred forest contained Atetefo-ware at radiocarbon dated levels of the 13 to 15centuries128. The full excavation report should enable the chronology to be more firmly established.

82Neither the documentary nor the archaeological records for the Bron and Akan countries are as full as might be wished. There is, nevertheless, a rather impressive degree of agreement in the inferences that may be made from them. Inter alia, they attest unambiguously to the extraordinary growth of the gold extractive industry in the fifteenth century, at a time when the market for the commodity was a northern one. First the Bron and then the Akan became peripheral, but none the less important, parts of the Mande world and ultimately, therefore, of the Mediterranean economy.

Part III: Arrivals

Exotic Europeans

83BrÔfo de nyansa na εforo po, a Twi saying has it : “knowledge enabled the white men to sail the sea129”. Portuguese navigators reached the stretch of coast that they first came to call A Mina de Ouro, “The Mine of Gold,” in January 1471. Whatever the source of their information, they sailed unerringly for Shama130. It was the ultima Thule for the inland traders, lying (it will be remembered) at the mouth of the Pra river that was joined by the Ofin only some 50 kms from Akrokerri. “The first traffic of gold done in that land,” wrote Joâo de Barros, “was at a village called Samm< [Shama], which then had about five hundred inhabitants; and later on it was done farther down. This spot is called by our men Aldea das duas partes131”. In late 1479 the Fleming, Eustache Delafosse, was aboard a Spanish ship that arrived on what soon became better known as the Costa da Mina. He was familiar enough with the region to know that slaves could profitably be traded for gold. There were, he wrote, two ports on “la Mine d’Or,” the first named Shama (“Chama”) and the second, six leagues distant, known as “l’Aldee duos partz.” The merchantman anchored off Shama. Its officers landed, and secured the protection of, as Delafosse has it, “manse et caremanse, qui sont le roy et vice-roy.” The next day the ship continued to “the village of the two parts,” where business was set up and trade done with Abrem merchants (“berrenbues”) from the hinterland (Escudier, 1992: 26-27). It was here that the Portuguese began construction of a fortress in 1482. This was the Castelo de Sâo Jorge da Mina, later and better known as Elmina.  

84Delafosse’s “manse” must have been the ruler of Shama, for all transactions at “the village of two parts” were done, Ruy de Pina noted, with “the lord of the place, who was called Caramansa…”132 In his lengthy account of Elmina, De Barros repeatedly referred to “Caramança” as “king,” “lord,” or “prince.” (Crone, 1937 : 114-119). In the early 1960s I suggested, very tentatively, that these sources attest to the early borrowing into Twi of the Mandekan mansa, “ruler.”133 P. E. H. Hair was later to accept the proposed etymology for Delafosse’s “manse,” but summarily dismissed the suggestion that Mande traders had introduced the word into Twi. He preferred to assign that role to the Portuguese, (Hair, 1994: 55) which seems unlikely in view of the fact that Duarte Pacheco Pereira, as we have seen, indicated the presence at Elmina, in the late fifteenth century, of traders known as “Mandinguas.” They must certainly have been Mandekan speakers and were most probably from Bewo. Their association with imperial (albeit declining) Mali, and with its imperial religion, Islam, afforded them a prestige that led local rulers with whom they did business to adopt Malian titles. On the model, then, of Mandekan karamoko, from karama, “respect,” and moko, “person,” I suggest karamansa, from karama and mansa, with the sense of “respected ruler.”134 It is to the point in this context that, by 1479, the word sika had already been borrowed from Manekan into Twi. L’or s’appelle chocqua. “gold is called chocqua,” Delafosse wrote (and for the pronunciation of which see his enchou for Twi nsu, “water”, and coucque for Twi akokÔ, “hen”)135.

85The administrative accounts kept by the Portuguese at Elmina throw considerable light on affairs in the interior. In 1971 Vogt found an uncatalogued ledger in the Portuguese National Archives that records all official business transacted at Elmina in the period 1529 to 1531. Fifty percent of the gold taken was traded against metal hardware, that is, brass manillas, shaving pots, chamber pots, and pans of various kinds. Twenty-five percent went in the purchase of woven fabric, both ready-made garments (of Maghribi styles) and undyed cotton cloth. Nineteen percent was accounted for by the sale of slaves brought to Elmina principally, at this time, from Benin. Only six percent of the gold was spent on miscellaneous trinkets for which long distance traders seem to have displayed little interest (Vogt, 1973: 93-94).  

86Garrard has commented on the strong demand for brassware. He argues that the first brass weights in the Bron country “were probably the products of the early Mande-influenced casting industry.” He thinks that this began to develop towards the end of the fourteenth century and to flourish in the fifteenth, though growth was constrained by the shortage of cuprous metal. However, as Garrard points out,

the relatively large imports of European cuprous metal from the end of the 15th century stimulated brass-casting among the southern Akan. By the time this secondary casting tradition emerged, the Akan casters would have been familiar with all aspects of casting technology. They abandoned the Islamic-inspired designs of the northern school, and created distinctively Akan forms which included figurative castings.  (Garrard, 2003: 174-77)

87The sheer weight of brassware and cloth despatched inland created a demand for carriers, to satisfy which the Portuguese had to procure slave labour from Benin and the Niger Delta and increasingly, from the early 1510s, from the Kingdom of Kongo (Elbl, 1997: 44-46, 72). It rapidly became apparent that inland traders would purchase labourers beyond those required simply for porterage purposes. It is difficult to ascertain the numbers involved, not least because private traders and interlopers competed with the Crown. In 1529 a consortium of merchants headed by a German financier agreed to supply Elmina with at least 500 slaves annually, but lost its lease in 1532 for failing to meet the quota136. It is also, of course, impossible to make even an informed guess about the number of slaves that were being purchased from the Juula merchants in the north. It does seem quite clear, however, that the demand for slaves in the Akan country exceeded the supply available at Elmina. In 1977 I drew attention to the large amount of labour that was a “necessary precondition” of clearing farms in the forest, and argued that family labour was inadequate for the purpose. Hence my view, that “the influx of [unfree] labour made possible massive forest clearances, such that a society in which the dominant mode of production was hunting and gathering transformed itself into one in which the dominant mode was food crop cultivation.” (Wilks, 1977: 52021). But I also suggested, using Asante as the example, that one of the major social directions of the period “was towards prevention of the consolidation of a slave caste : those of unfree origins were assimilated as rapidly as possible into the class of free Asante commoners and their acquired status afforded full protection of the law.” (ibid.: 523). This is surely misrepresented by Klein, who comments, “the presence of foreign slave traders, by itself, is not evidence of a slave society – that is, one dominated by a mode of production driven by labor – of the sort Wilks suggests.”137I do, however, agree with Klein that it is extremely difficult to assess the impact on the forest economy of the imported slaves who were resold to inland traders. First, we do not know their numbers from either European or northern sources with any degree of precision, and can only hazard a guess that the addition to the labour force was at most a few thousand a year. And second, it is extremely difficult to assess the impact of an influx of labour, on however modest a scale, on a presumably thin (granted the nature of the forest environment) population of hunter-gatherers in process of becoming cultivators.  

88Fortunately the relatively good figures for gold exports provide a more satisfactory window into the interior. Garrard made a number of bold estimates of the amounts of gold being exported from the Akan and Bron lands. Those for the late fifteenth and early sixteenth centuries are of relevance here. He calculated that around 500,000 troy ounces were consigned to the Lisbon mint in the period 1471 to 1500, and guessed that interlopers, Castilian and others, took something of the order of 50,000 ounces. Thus he estimated the export to Europe for 1471 to 1500 at an annual average of 18,300 ounces. For the shorter period 1501 to 1521 Garrard arrived at a total of about 300,000 ounces, or some 14,000 ounces a year (Garrard, 1980: 149-51). These figures take no note of the northern, or overland, trade, for which Garrard offered what were no more than informed hunches. He guessed a total export of 250,000 ounces, or an average of 5,000 annually, for the period 1400-1450; of 500,000 ounces or 10,000 annually for 1451-1500; and of 600,000 ounces, or 12,000 annually, for 15011551 (ibid. : 160-63). Even allowing for a considerable margin of error the figures show beyond doubt that the gold extractive industry was well developed before the entry of merchants from Europe into the trade for it.  

89The Portuguese had, naturally, a strong interest in the interior mining districts, and the reports of their agents who travelled to them were closely guarded. Much information was compiled in an unsigned report of 29 September 1572 (henceforth, the 1572 Report). It was written in Elmina for despatch to the king of Portugal. “I am certain that there is much gold,” the king was assured;

One of the reasons which persuade me is that in almost every locality, in which the blacks have set out to look for gold, they have found it; and if they have not discovered more mines than now exist, it is because they have not sought them, lacking the industry, diligence, and the will. For it is said that, when the blacks of the interior want to pay for anything they have bought, do they then go and extract the gold from the mines138.

90Thirty years later another Fleming, Pieter de Marees, discussed the early days of the gold trade : It is difficult to know what knowledge they had of Gold long ago, before our time: according to what the Negroes say about it themselves, they knew only a little about it and it was not much valued among them, because there was nobody who came to take it from them.  

91He confessed his ignorance “about the Mines of Gold, whether they are situated here or there,” and remarked that “the Mines are kept secret and well hidden by the owners.” (Van Dantzig and Jones, 1987: 188-89). In fact there was an element of dissemblance in this, for he apparently intended to include in his Beschryvinge ende Historische verhael, to be published in 1602, a map prepared by the Portuguese cartographer Luiz Texera, one purpose of which was to show the location of the major inland mining centres139.Seven of these he marked with crosses. It may be that it was one of these that had become known by 1551 to the Portuguese king, João III, when it was believed to be about 50 leagues from Elmina. It was “a business to investigate,” the king informed his Chancellor of the Exchequer, but noted “the secrecy which you see the business requires.”140 No investigation is known to have followed.

92The writer of the 1572 reports complained about the matter141. “Their disregard,” he wrote, “can be seen in that almost every man, when asked about the gold which comes to this land and where the mines of most gold are found, replies that the gold is from Elefante Grande.142” He knew, however, much more than that. He knew something of the environment in the mining districts, “The land,” he wrote, “does not enjoy the benefit of the sun on account of the great thickness of the trees, into which animals cannot penetrate, let alone the blacks.” But obviously there were miners, for he remarked, “they do not open mines or seek gold anywhere that is at all far from the water to wash the earth from which they remove the gold.”143 The implication of the 1572 Report is that at such sites hunter-gatherers were both prospecting and carrying out mining operations144.The writer saw that the spread of agriculture was essential to the development of the extractive industry. “Once the hills have become opened, and the land cultivated,” he predicted, “thereafter what is produced will be enormously increased145….”He made a recommendation having to do with expediting the flow of gold from the mines in the interior. “In view of the mines that are already here,” he observed, “it seems to me not unreasonable that when a mine is discovered there should be built in that place a fortress or lodge suitable for white people.” This was to function as a trading post. The work force of “blacks” comprised those who dug for gold and those who produced food. These latter would till the soil and plant, among other crops, milho, in order “to sustain themselves and those that dig.” The writer clearly referred to maize. “The blacks,” he remarked, “sow milho grande which they call Bruy and which in Castile is called “Indies corn146...”Bruy is, of course, the Twi aburo. In the event Corisseno de Aliphante Grande, that is, “District of the Great Elephant,” duly appeared on Luiz Texera’s map of 1602. It was shown west by southwest of Acanes Grande and west by northwest of Acanes Pequenos, that is, of the “Big Akan” and “Little Akan.” I have suggested elsewhere that it lay in the highly auriferous district around the present Manso Nkwanta, then part of a large hunting ground known (according to dialect) as Meranson or Menason. (Wilks, 1994 (b): 6-7)

93In 1520 Elmina Governor Duarte Pacheco Pereira had ordered presents to be given to the “king” of the Abrem in order that he should facilitate the passage to Elmina of the “king of the Acames.” For the latter rather more lavish presents were to be assembled147. Pereira clearly believed that the Akan people (“Acames”) belonged to a single kingdom. It may be that he was simply misinformed. He was, however, a man of much experience, and it may be – just may be - that his apparently erroneous description in fact echoes the time when Adanse Akrokerri was regarded as the first Akan creation (“Great man, you come from Adanse Akrokerri / Where Odomankoma made a thing”), and when the akanman piesie nnum, “the five first-born Akan places,” were only beginning to take shape. However that may be, it must have been traders visiting Elmina who identified themselves as Akan, “first people.”148

94Around the turn of the fifteenth century Portuguese merchants made a distinction between Acanes Castelhanos, “Castilian Akan,” and Acanes Portugueses, “Portuguese Akan.” The terms probably indicated no more than that some of the traders had formerly done business with the rivals of the Portuguese149. Whatever the case, these terms became defunct after the collapse of Spain’s involvement in the Guinea trade. The documents begin instead to refer to the Acanes Grandes and Acanes Pequenos. As the century wore on the Portuguese became involved in bitter competition with the northern merchants for the supplies of gold. While this struggle is beyond the scope of the present paper, it did however have repercussions on the extractive industry in the forest country where there was considerable turmoil as groups vied with each other for control of local gold resources150.

95António de Brito was governor of Elmina around 1545. He received a visitor who was described as “the son of the King of the Acanes Grandes” and “his oldest son and heir.” He was converted and given the governor’s name. There was discord inland, between the Acanes Grandes and Acanes Pequenos, and the roads to Elmina were blocked. In or about 1556 Acting Governor Afonso Gonçalves de Botafogo sent an envoy to the interior to resolve the dispute and have the roads reopened. He remained there for eight months and negotiated a settlement. Both parties agreed to send a son to Elmina. António de Brito (of the darker hue) arrived to represent the Acanes Grandes. Shortly thereafter his wife joined him – and was promptly baptised to become, in honour of the Queen of Portugal, Dona Caterina (with the governor and commander of the fleet as godparents). But even this was not quite the end of the story. António de Brito continued to visit Elmina regularly. He was there in or about 1572. Described as “a respectable black, of good bearing,” he was accompanied by his son who was about twenty years of age and “speaks Portuguese reasonably well through having been here with his father.” Advantage was taken of the young man’s language ability to question him. His grandfather, he said, was still alive. He was very rich, but much of his gold came from Tafo, a further five days’ journey inland. He had himself seen nuggets there the size of hazelnuts – rising to ones the size of his head – or perhaps fist !151

96It was for long received opinion that the Acanes Grandes were those who were to become the Akyem Abuakwa. This identification was based primarily on the Dutch map of 1629. Hans Propheet showed the “Akim or Great Acanij,” described them as “very delicate people, and rich in slaves,” and located them to the northeast of Elmina and northwest of Accra”152. It has become received opinion that the “Groot Acanij” of seventeenth century Dutch accounts are to be equated with the “Acanes Grande” of the sixteenth century Portuguese ones. The problem is, however, that the António de Brito described in 1557 as “son of the King of Acanes Grandes” is described in 1572 as “son of the King of Asaẽes Grandes.” (Da Mota and Hair, 1988: 64, 86). Either this was an error, which seems unlikely in the context, or the implication is that the two descriptions were compatible at the time. In fact Luiz Texera’s map of 1602 shows a large region, “Assas,” that extended eastwards from the upper reaches of the Ankobra and northwards perhaps to embrace the Acanes Pequenos and beyond, the Acanes Grandes. The present day Asen (or Assin) live for the most part south of the Pra, but in some communities stories are still told linking their origins with Odomankoma and the ‘making’ of Adanse Akrokerri153.

97R. Kea has used Texera’s map to argue convincingly that in the sixteenth and seventeenth centuries the Asen dominated trade between the interior and the coast, and were the principals in a complex but effective “Akani trading organization.” (Kea, 1982: 85-94, 248-87) It has been compared to the Hansas of late medieval Europe, and suggested that, like them, it collapsed when confronted by the rising power of new nation states. (Van Dantzig, 1990: 205, 214). The parallel has much merit. The documentary record for the seventeenth century attests to the considerable turmoil that resulted as petty kingdoms took shape and competition for land and other resources intensified154. Stories of the appearance of lineages and of the “towns of first creation,” belong to the Adomankomasεm, the Odomankoma Tales. Stories of the development of chiefdoms, petty kingdoms and in some cases imperial states, do not.  

98One of the akanman piesie nnum – “the five first Akan places” - is reputedly the birthplace of the Asona lineage. The apical ancestress was Nana AsÔ BÔadeε, who led her people out of a hole in the ground at Adaboye – also known as Abuakwa Atwumamanso – near the confluence of the Ofin and Oda rivers in Adanse155.Another of the “first Akan places” claims to be birthplace of the Agona. The apical ancestress was Ayekra BÔadeε. She arrived with her people at Banso, otherwise known as Abankeseso (and only a short distance from Abuakwa Atwumamanso)156. It will be noted that the strong name of Odomankoma, ÔbÔÔ adeε, is attached to both ancestors. The storytellers had and have different ways of expressing Odomankoma’s relations with the lineages. One, Abakomahene Kusi Wiamoase of Dormaa Ahenkro (“custodian of the history and traditions of the Dormaa and royal prince”), related the following to the British district commissioner A. W. Norris in 1914. “Eduana,” he claimed,  

was the first Clan ever made by Great Odummankoma when he made the Eight Clans…. In those days Odummankoma lived among them (with all the Clans he made) but let it be known to all that he was departing from them to go to another place only he knew…. he called together the Great Clans and said he would make each of them a present before he went on his great journey.

99Odomankoma presented each mmusua kεseε (“Great Clan”) with gifts. The storyteller then took up a theme from drummers’ texts cited above. Odomankoma, he continued,

had no wife to leave behind him but he had a child called Death. He offered Death as a present to each of the Eight Clans in turn but no one wanted to take it. So Odummankoma was left over with Death (‘death remains’, they say) by his side after he gave away everything else. Death was the only thing that was left…. Odummankoma had a face that no one was allowed to see but Death looked into it and killed him (his own father) when he did so. All the Peoples wailed when they saw what Death had done157.

100Thenceforth, Ôde n'akyi gyaa agyina, “things were left to men of affairs158”. I have offered elsewhere a speculative model for the process I referred to as “From Estate to State,” that is, the emergence of chieftaincies, petty kingdoms, centralising kingdoms, and in some cases imperial states. (Wilks, 1982 (a): 241-44; 1993: 91-103). The Asona of Abuakwa Atwumamanso were in time to provide the ruling dynasty of, inter alia, Akyem Abuakwa, and the Agona of Abankeseso that of, inter alia, Denkyira. The “Acanes” of sixteenth century Portuguese writers re-emerged in seventeenth century Dutch and English ones transformed into the Akyem, Denkyira, and the like. The traditional drum and horn texts on the one hand, and the records of the European trading companies on the other, attest to the struggles between the new polities that resulted in the emergence of Denkyira and Akwamu as regional powers in the seventeenth century, and of imperial Asante in the course of the eighteenth. These events, however, have no place in the Odomankoma Tales, which terminate with the rejection of Death by the eight lineages and its acceptance by Odomankoma.

Exotic seeds

101T. E. Bowdich spent four months in the Asante capital in 1817. He was a highly gifted individual, of an inquiring disposition. He was led to understand that the forest people “were originally comprehended in twelve tribes or families,” and may well have gleaned this from Fante guides eager to proclaim their affinities with the Asante. The topic interested Bowdich, however, and he took “some pains” to pursue it. Traditions to do with the origins of lineages are the stuff of the Adomankomasεm, and Bowdich’s own curiosity about etymologies was more than matched by the determination of the storytellers to explain, albeit in the absence of any discernable philological principles, the meaning of any and every name. (Bowdich, 1819: 229-31.)

102There were, Bowdich learned, “four patriarchal families,” namely, Aquonna (Buffalo), Essonna (Bush Cat), Intchwa (Dog), and Tchweedam (Panther)159. The four, Bowdich wrote, “appear to record the first race of men living on hunting.” Two more families, the Abrootoo (Corn stalk) and Abbradi (Plantain) – see Twi aburo, “maize,” and borÔdeε, “plantain,” – are readily recognisable as the present day Bretuo and Abradi. Bowdich saw them as the “immediate descendants” of the “four patriarchal families,” and as marking “the introduction of planting and agriculture.” He grouped with them the Yoko (Red Earth), the Agoona (Palm oil), and the Appiadie (servant race), and associated the first of these – now usually Oyoko – with “the origin and improvement of architecture.” There can be little doubt that here Bowdich was accessing the body of tradition, to which reference has been made above, having to do with the transition from the camps of hunters to the villages of cultivators160. His comments on the Agoona – now Agona – are perhaps even more intriguing. “Agoona,” he wrote, “signifies a place where palm oil is collected,” and associated it with (“probably”) the origin and improvement of commerce. “Indeed,” he added, “the natives have included the Portuguese, the first foreign traders they knew, in that family161….”

103Bowdich conceded that he had achieved only “imperfect success” in his explanation of lineage names, (Bowdich, 1819: 229-31). We have hinted, however, at the ‘Odomankomaesque’ characteristics of his account. The emphasis is upon beginnings, a signature tune as it were, of the Odomankoma drummers and storytellers. In particular the Bretuo (maize) and Abradi (plantain) lineages are associated with the beginnings of agriculture, the Oyoko with sedentarization, the Agona with the development of long-distance trade -and the Appiadie with, perhaps, the procurement of unfree labour. Most intriguing is the trinity of maize, plantain, and Portuguese. In 1961 Raymond Mauny put forward his considered opinion that maize – Zea mays, a Central American domesticate - could not have been brought to Elmina before 1493, but was growing there by 1502162.In 1540 a Portuguese pilot could refer to the grain called “Subaru, or in the West Indies mehiz,” that is, maize, that grew in the Cape Verde Islands and along the West Africa coast, and was “the chief food of the people163.”In much the same period, the Portuguese introduced plantains – Musa spp, of Asian origin – to the Costa da Mina from the Kingdom of Kongo164. The author of the anonymous Portuguese report of 1572 referred to the cultivation of “bananas, which in the Spanish Indies they call platanos.” On the way to Agri, he wrote, “I saw it growing in the forests, where I think that no-one can have planted it165.” In 1602 de Marees described a type of plantain known as “Congo banana” because the Portuguese originally brought it from the Kingdom of Kongo (Van Dantzig and Jones, 1987: 162-63). Both maize and plantain were part of that post-1492 global movement of seed that has become known as the Columbian Exchange166.

104In the late sixteenth century it was, of course, still remembered that the Portuguese had introduced maize into West Africa. Writing of Elmina in 1602, de Marees commented,

This Corn Maize is called Maize by the Indians; the Portuguese or Spanish call it Indian Wheat and the Italians Corn of Turkey. It is a grain, which is known virtually throughout the world, and was brought from the West Indies to St. Thome. The [Portuguese] people of St. Thome brought it here (after establishing their strongholds and Castles) to supply their needs, and sowed it here in these Lands; for before the arrival of the Portuguese the inhabitants of these Countries did not know it. Thus the Portuguese sowed it there and distributed it amongst the Savages, so that the country is now full of it and at present it grows in abundance.  (Van Dantzig and Jones, 1987: 113).

105Demand for the imported seed was initially high. “All of these,” de Marees added,

Were great and strange novelties for the Negroes, who gave a lot of money for some Fruits, when they first came out; because of the appetite they had for them. For instance, for Pineapple they paid not less than the Weight of a Quienta in Gold, or, according to our [Dutch] reckoning three Angels of Gold and they paid similar prices for other things, but gradually all these things have so much multiplied that they are nowadays much less esteemed and are plentiful throughout the Country. (ibid. : 216).

106On the matter of propagation, he reported that, “on some plants I have counted no less than five hundred and fifty and more grains, all of which grew out of a single Maize grain167.”Corn was among the foodstuffs brought to the coastal markets by “peasants” from farms inland. There it was used in brewing a beer, pito (“poitou”), and was baked to produce a very acceptable bread that remained edible for three or four months and was used by the Portuguese in Elmina to provision their ships168. C. C. Reindorf had an interesting comment on the maize trade in somewhat later times:  

Our people, after the Europeans had established themselves in the country, did not carry on the trade in slaves only, but most of them turned their attention to the cultivation of the soil. Yams, rice and corn were so plentiful that slave ships were supplied with corn, in peaceful times 1000 stalks for 5 shillings, but in times of war 1000 stalks for one ounce of gold. From Asabu and the Fante countries about 100 canoes were daily laden with corn and yams, and potatoes at Mowure for Axim and Accra for sale.  (Reindorf, 1895: 274)

107James McCann has remarked that “the documentary record of maize is often maddeningly elusive….” He points out that the plant is “highly sensitive to deprivation of water, sunlight, and nitrogen,” and that “maize monocultures are extremely vulnerable to environmental shocks, especially drought169.”Clearly, however, conditions for its spread into the forest in the sixteenth century were highly favourable. “The grains of this Maize,” so de Marees noted of whatever variety had been spread inland in the sixteenth century, “grow well in hot, moist earth and produce Corn twice a year.” (Van Dantzig and Jones, 1987: 114). Studies of Bosomtwe, the lake in the central part of the Pra and Ofin basin, have shown that a humid period from about 800 to 1300 A.D. was followed by a much drier one of about 1300 to 1800170.The sharpest fall in the level of the lake can be dated to 1520-1580, but it is uncertain whether the change in rainfall would have had any major impact on vegetation171. Is it possible that wet and moist evergreen forest slowly gave way to moist semi-deciduous, so facilitating the development of the forest fallow system and in due course the introduction of maize ? And conversely, in the north, did moist give way to dry semi-deciduous forest as rainfall fell? Whatever the case, maize seems to have been introduced to the Bron country later than to the Akan. Corn cob rouletted pottery occurred in the excavations at Ahwene Koko and Bono Manso, but seemingly in seventeenth rather than sixteenth century contexts. It was absent from Bewo172.

108The semi-deciduous forest country is inhospitable to human settlement not because of the lack of proteins, sufficiently available from fish, snails, and game, but from the insufficiency of carbohydrates. The creation of a forest fallow system requires the opening, on an appropriate scale, of gaps in the canopy, and the removal of the understory of smaller trees, bushes, and vines. McCann accurately enough summarises my general argument. “As Wilks pointed out,” he writes,

the Akan people were in the process of expanding their population and labor capacity at about the same point that they developed a social system that encouraged population growth. Gold extracted from forest mines provided the means to attract labor and link forest ecology to the expanding Atlantic world.  (McCann, 1999: 118)

109Arguing from radiocarbon dates and genetics, critics have suggested that my setting for the forest transformation is too recent173. McCann points out, however, that there is no necessary inconsistency between my thesis and the recognition that “existing forest populations began from a low population density foraging economy that grew rapidly and underwent a social transformation in the half millennium before the birth of the Atlantic economy.” I have no problem in accepting this. Indeed, I have long argued that the Bron and Akan economies were shaped by the constraints of the Mediterranean economy. I did, however, stress time and time again that my concern was “not with the origins of cultivation as such, but with the emergence of a fully formed agrarian order174.”McCann reformulates the matter in terms of the Columbian Exchange.“Arguments for extensive pre-1500 forest settlements,” he writes,

nevertheless ignore an ingredient that points directly to a take-off point of around 1500: the arrival from another world tropical forest biome – the forest zone of Central and South America – of new food crops ideally suited to provisioning and expanding set of forest polities. New World food crops – cassava, cocoyam, cowpeas, and above all, maize – brought by Portuguese ships to provision their coastal fortresses spread quickly beyond the fortress gardens and created and sparked an agricultural carbohydrate revolution….The confirmation of Wilks’s hypothesis on the fifteenth century “big bang” in the forest of central Ghana is the fundamental role of cassava and maize as the frontal edge of a human assault on the forest landscape using “forest fallow,” an accumulated repertoire of knowledge and revolutionary new plant germ plasms, to convert the forest’s biomass and energy into carbohydrate calories.

(McCann, 1999 : 119 ).

Currencies and class

110There is a Twi εbε or ‘proverb’ that refers to the monetisation of gold. Wonni sika a, anka wÔfrε no anwea kwa, it is said: “if you cannot use gold then it would just be called sand175.” In 1962 I suggested that the Twi sika might be a loan word ultimately from the Arabic, though from which root and by what route I did not guess176. Dakubu has convincingly shown that the root is the Arabic razaqa, with the sense of “endowed with the means of livelihood,” and thus arzaq, sing. rizq, “wealth”, “possessions”, etc.177. We have seen that the word had taken its place in Twi, presumably via Mandekan, before the end of the fifteenth century. We have also seen that the Bono ruler, Obunumankoma, is said in tradition to have developed the treasury in gold (“and a system of weights for use with it”), probably in the mid-fifteenth century. The corresponding innovation seems to have been much later in the southern forest country. The early Akrokerri leader Asare Nyansa, remembered in tradition for his concern with the rule of law and the protection of property, is said to have “instituted the use of Nnabuo, iron melted [sic: smelted] from stones, which were normally used to buy things178.”In 1895 Reindorf recorded the story he had heard earlier in the century, to the effect that in the time of Kwabia Anwanfi, Oyoko leader at Asantemanso, “gold was not known, the currency was pieces of iron.” (Reindorf, 1895: 47). In one of the short historical pieces written in the Seychelles in the early twentieth century a similar comment was made with reference to Kwabia Anwanfi’s successor, Oti Akenten, that in his reign “gold dust was not known for money, but nah-boo a sort of brass was used.” (Prempeh I, 2002: 97). The floruit of Oti Akenten was a mid-seventeenth century one. Some insight into the nature of the situation is provided by the 1572 Report to which references have been made above. At that time those engaged in the extractive industry produced to order, that is, they dug in response to the specific demands of traders. Gold, however, was not accumulated. “None of the other blacks,” the writer claimed, “amasses treasure because of the little trust that there is among them, for, when he so wishes, the most powerful removes from the weakest what he owns. Also it is because of Wars, which frequently set some villages against others179.”

111Ôdomankoma bÔÔ adefoÔna ÔbÔÔ ahiafoÔ+ it is said: “Odomankoma made the rich and he made the poor180.”This item from the Adomankomasεm has to do with procedures introduced in order to encourage, and protect, the accumulation of wealth by, to use the words of Pieter de Marees, the “Nobleman or great Monsieur, whom they call Brenipono” – that is, Twi obirεmpÔn (plural, abirεmpÔn), “great man.” Among the privileges conferred upon those joining the ranks of the abirεmpÔn was to be “allowed to buy slaves and trade in other things which they are not allowed to do before that time.” (Van Dantzig and Jones, 1987: 170) De Marees made several sketches illustrative of Akan coastal society in the later sixteenth century. He mentioned the “Captain or Headman… with his Council,” and such officials as the executioner. He was most interested, however, in the abirεmpÔn who paraded the streets with their distinctive bonnets and linen cloths caps and did business with the “Batafou”, that is, merchants who came, accompanied by their interpreters and slaves, “from distant Lands to trade on the seashore.” One of the sketches showed “with what kind of triumphal ceremony they make a Nobleman. The postulant, de Marees reported, “(when he is promoted to this status) gives away all his goods to the people in order to be a Nobleman, which is something they very much hope to achieve from their youth onwards;” indeed, “right from their childhood they begin to make savings in order to spend them on their Ennoblement.” (ibid. : 167, 168). De Marees remarked on the haughtiness of the abirεmpÔn. “If spoken to by someone of lower standing, they will not answer, but will address him in an unfriendly way and show him an angry face, saying ‘Be silent and don’t speak to me,’ lest they be humiliated or degraded by speaking to someone of lower rank.” (ibid. :31-35, 167-70). A current saying makes the point : Osikani ne hiani goru a, εnsÔ, “if a rich man and a poor man are friendly, it cannot last181.”The ceremony establishing the division between rich and poor was institutionalised in Akan society and, in the guise of the award of the elephant’s tail, has been recorded from eighteenth and nineteenth century Asante182.

112The Adomankomasεm may be seen as constructed around a complex riddle (aborÔme) that, in typical Akan fashion, requires the listener to guess at the answer and preferably, such is the nature of the genre, to guess wrongly. The master riddle has to do with the death of Odomankoma. Why was Odomankoma unable to give death away, and so was killed by its own creation ? We have to guess. This, we may think, was the point at which responsibility for the world that Odomankoma had made passed into the hands of people, of men and women. The force that had radically transformed society was a spent one. Karl Marx’s famous observation, that people make their own history but not under conditions of their own choice, is reversed. With the death of Odomankoma people come to make their own history under conditions of their own choice. There were, however, some who knew that the revolution – Odomankoma – was dead, but bid the revolution long life. Paradoxically, and this is an aspect of the riddle, they were traditionalists who wished to conserve the world that Odomankoma had created. I have cited elsewhere one manifestation of this tendency in the movement, referred to as “Domankama,

113the Creator” and as “Abonsamkom,” that gathered strength in Asante in 1879 and, in early 1880, was responsible for an attack on Asantehene Mensa Bonsu. I suggested that the movement was a reaction “against the departure, by Mensa Bonsu’s government, from the traditionally accepted canons and prescriptions of political behaviour183.” I have been concerned in this paper with constructions that a historian may place upon the Adomankomasεm. I have not ventured into the realm of religion, and have particularly avoided – as irrelevant to my present purposes – any consideration of efforts to assimilate Odomankoma into a framework of Christian beliefs184. I end nevertheless with a verse sung at the Apo festival in Wenchi in 1946, and recorded by Eva Meyerowitz.  

114Ôdomankoma a ÔbÔÔ adeε,
Ôbra redan yεne, O!
Nsan bεbÔyεne, O !
Odomankoma made a thing,
Life is changing for us, O!
Come back and remake us, O!  
(Meyerowitz, 1951: 158-59)
The piece should perhaps be seen as incorporating an Odomankoman style riddle! If Odomankoma is “God, the Creator” of Christian theology, how then could it have been killed by that which it had created ?

Haut de page


1 In conversations, some brief and some extended, several scholars have much stimulated my thinking on the topics of this paper. I should especially thank Drs. Kojo S. Amanor, James Fairhead, Nancy Lawler, T. C. McCaskie, and Albert and Christiane Owusu-Sarpong. The last has also given invaluable advice on linguistic matters. I do not, of course, wish to imply that they will agree with all or even most of my arguments.
2 A caveat must be entered here. The boundary between forest and savanna zones did, of course, vary over time, for which see Fairhead and Leach, 1998: 60-93.
3 In using the plural form, Adomankomasεm, I have followed Akuffo, 1950: 1, 2.
4 Vansina, 1985: passim. For epistemological problems with particular reference to Akan tradition, see McCaskie, 1989: passim.
5 Stewart, 1966: 54-58, and Painter, 1966: 64-65. I have cited from this source since both papers, pioneering new approaches in the context of the times, had been presented to a Seminar on the Akan held at Legon in 1965.
6 Daaku, 1969: 364. The informants on this occasion were the very aged Nana Kwaku Afoakwa II of Sodua and ]kyeame Yaw Tawiah.
7 See e.g. Mohr, 1909: 142.
8 Warren and Brempong, 1971: 115, Tuobodom Adontenhene Yaw Appiah; and 236, Tanoboasehene and Takyiman Konahene Kwabena Dwomo.
9 See Dolphyne, 1979: 88-118.
10 See most recently Pescheux, 2001: vol 1, 117-29, “Wilks et le ‘Big Bang’.”
11 Wilks, 1977: 515-16. This essay was subsequently reprinted, with some changes, in Wilks, 1993: 41-90.
12 See, e.g, Boni, 2001: 143-44, who quite understandably mistakes my meaning.
13 Klein, 1994: 627-56, and see Wilks, 1994 (a): 657-65.
14 Klein, 1994: 637-38. Klein repeats this observation word for word in his 1996 article: 253.
15 See Rattray, 1914: 177, citing Christaller, 1879: item 204.
16 Nketia, 1963: 153, comments, “... the most important is the drummer of the talking drums, who alone may drum certain portions of the Akan drum language.”
17 For some relevant account of which see Gyekye, 1995: 13-24.
18 For this process in the old bardic traditions of Mali and Wales see Wilks, 1999: 25-7.
19 See e.g. Owusu-Sarpong, 2000: 56. Hornblowers who play for chiefs may use the refrain, Firi tete, firi tete, Odomankoma firi tete, “Long ago, long ago, Odomankoma, long ago,” see Arthur, 1965: 166. For the translation, “made a thing,” compare Danquah, 1944: 64.
20 See, e.g, the piece by Okyerema Kwaku Pon, chief atumpan drummer of the Asantehene’s court and “ranked among the best living exponants of the drum language,” Nketia, 1966: introduction and p. 41.
21 Busia, 1952: 11. For comments on Busia’s view, see Gyekye, 1995: 14, and Wiredu, 996: 119-20
22 I have used the texts recorded in Akuffo, 1950: 74; Danquah, 1944: 44, 70-71; Nketia, 1966: 12-13; Nketia, 1963: 154, 192. See also Rattray, 1923: 272 text XV and 282, text XV, for a Mampon version. The variant in Akuffo was apparently recorded in Akuropon, Akuapem, and extends the list of functionaries “made” by Odomankoma to include the chief (]hene) and further named persons, see further Akuffo, 1950: 122, 164, and xi.
23 See Owusu-Sarpong, 2000: 257.
24 See the Twi b]re, “to search out,” but compare Danquah, 1944: 58.
25 See. e.g. Rattray, 1923 : 124.
26 Compare Danquah, 1944: 63.
27 Wiredu, 1996: 120-21. The pleasure I have derived from this book is all the more so since Wiredu was, at the University College of Ghana in 1953, a member of a class of five students that was the first I taught in my academic career.  
28 G. P. Hagan, “Black Civilization and Philosophy: African traditions of Philosophy,” paper presented at FESTAC, Lagos, 1976, cited by Wiredu, 1996: 120.
29 Story told by Adankrangyahene Osafu Agyeman II, in presence of the elders and queenmother, 15 September 1938; see Ashanti Social Survey Mss, ex-Robert Steel, now in possession of T. C. McCaskie to whom I am grateful for making this text available. The monkeys also figured in the story recorded by Steel from Manso Abodom on 15 January 1938. Odomankoma met Ogu Yeboafo at Manso Abodom and decided to kill him. Ogu Yeboafo’s dog barked, however, whereupon Odomankoma placed his spear on the ground and took Ogu Yeboafo as his ally.
30 See Rattray, 1923: 266-77, items I, II, III, IV, V, VIII. The drumming was recorded at Mampon. Compare Danquah, 1944: 54.
31 Christaller, 1879: item 966. For useful but not always acceptable translations of these proverbs, see Lange, 1990.  
32 Nketia, 1966: 40-41, and compare Christaller, 1879: item 964 (and Lange, 1990: 81); Rattray, 1916: 28. Compare Owusu-Sarpong, 2000: 100-102, 243, and Wiredu, 1996: 119-20.  
33 Christaller, 1879: item 967 (and Lange, 1990: 81); Owusu-Sarpong, 2000: 102. Compare Danquah, 1944: 58-77.
34 See Effah-Gyamfi, 1974 (a): 41, and see Boachie-Ansah, 1986: 25-26.
35 Warren and Brempong, 1979: 115. Information from Tuobodom Adontenhene Yaw Appiah, February 1971.
36 Warren and Brempong, 1979: 102. Information from Tachiman Twafohene Kwaku Gyan Ababio, 12 April 1970.
37 Warren and Brempong, 1979: 236. Information from Tanoboasehene and Takyiman Konahene Kwabena Dwumo, 8 February 1970.
38 See Effah-Gyamfi, 1985: 9-10.
39 See e.g, Goody, 1965: 2.
40 For this utilisation of a three-dimensional space see Wilks, 1996: 15-17.
41 Owusu-Sarpong, 1999: 187-88. See further Owusu-Sarpong, 2000: 91-92; Boachie-Ansah, 1986: 20-21, 26-27.  
42 Ivor Wilks, “Conversations about the Past, mainly from Ghana. 1956-1996,” henceforth “Conversations”] Vol. I: Hanihene Kofi Ampofo II and elders, 9 February 1962.  
43 Ameyaw. 1965: 4; 1975: 149-50. The name “Nsoko” probably originates from the Arabic suq, “market,” and reached the Bron country through a dialect of Mande.  
44 McCaskie, 1990: 134. In 1966 Edmund Collins argued, carefully but not I think, convincingly, that the intrusion of the Borbor Fante occurred in the first half of the eighteenth century, see “Borbor Fanti,” paper presented to the Historical Society of Ghana, 4 April 1966. See further Fynn, 1976: 54-59. It is interesting that in 1817 Bowdich mentioned “Takima... whence the Fantees are reported, by tradition, to have emigrated, and there is yet but little differences in the languages,” see Bowdich, 1819: 171.
45 Owusu-Sarpong, 2000 : 60 and compare 243; Rattray, 1923 : 11, 102, 208; Akuffo, 1950: 10­11; Nketia, 1963: 47, 139, 185, 191; Nketia, 1966: 15.
46 Warren and Brempong, 1971, Pt 1, No 3: 236-37.
47 A comparison of this version of the story with that narrated by Takyiman Krontihene Yiadom Boakye in the same year, see Warren and Brempong, 1971, Pt 1, No 3 : 212-13, reveals something of the nature of the embellishments, at this level, that may be made to a story in the interests of lively performance.
48 For the early Tano river basin, see Kea, 1982: 74-78.
49 For an informative account of the moving of gods, see Daaku, 1969: 319-21, interview with Kwame Anowuo, Bonahene of Patakro, et al.  
50 For these imports, and the copies made of them by Bron and Akan brassworkers, see Silverman, 1983 (a): 10-29; and more generally, Silverman, 1987: 272-88.
51 See Owusu-Sarpong, 1999: 118; Akuffo, 1950: 123 and compare 68-69. 53
52 E.g., Nketia, 1966: 36.
53 See Effah-Gyamfi, 1985: 21, 26, 34. In the imperial period an Asante official was stationed at Donkoro Nkwanta to oversee trade, see Royal Anthropological Institute, London, Papers of R. S. Rattray, MS Bundle 107, notebook 3: 1872.  
54 Nketia, 1963: 154, 12. Compare Akuffo, 1950: 67.
55 Daaku, 1969: 373, interview with Opanin Abubekr and Opanin Kwame Bio, Fomena. Compare Daaku, 1969 (b): 22, Assin Nyankumasi.  
56 Nketia, 1955: 40. I have followed Nketia’s translation.  
57 Daaku, 1969 (b): 75-78, interview with Akrokerrihene Owusu Sampa III and elders.  
58 Daaku,1969 (b): 380-81, interview with ex-Edubiasehene Appiah Dankwah.
59 Daaku, 1969 (b), 366, interview with Opanin Abubekr and Opanin Kwame Bio, Fomena, and also 373, “while creating, the Creator met Okro there. It was this Okro who showed him where to pass on his way. Thus we name the place Okoro kyere (literally ‘one directs’) which has been corrupted to Akrokerri. The whole nation originated from there.”
60 In the 1940s Nana Agyeman Prempeh II chaired a committee responsible for producing “a definitive” history of Asante. Submissions were requested from traditional authorities throughout the country. Among them was a 92-page typescript misleadingly entitled, “History of the Immigrants from Takyiman.” Much of the dossier is concerned with Adanse and particularly Akrokerri. The identity of the compiler is unknown, though a prefatory page is notated, “Son of Tekyimanhene.” Page 4 of the typescript is headed, “History of Ex-Krontihene Kwasi Kyei on 2nd July, 1942,” and he may have been one source of information. In the event, Nana Agyeman Prempeh II rejected the dossier, and it remained among his personal papers. A copy is in possession of the author. Henceforth: Dossier ca.1942.  
61 Dossier, ca.1942: 40. I do not intend to endorse this particular explanation of the meaning of the name “Odomankoma;” several others equally plausible are offered.
62 Daaku, 1969 (b) : 75, interview with Akrokerrihene Owusu Sampa III and elders.
63 For an interesting discussion of such sacred places, see Sarfo-Mensah, 2002: passim.
64 Daaku, 1969 (b): 4, interview with Adansehene Kwantwi II et al., and compare 363, interview with Soduahene Kwaku Afoakwa and Yaw Tawiah, and 366, interview with Opanin Abubekr and Opanin Kwame Bio.  
65 See Wilks, 1993: 91.
66 Prempeh I, 2002: 115.
67 I have argued that various Akan groups combined together in the late seventeenth century and, having destroyed the power of Denkyira, only then began to call themselves “Asante,” see Wilks, 1993 : 111. For observations on this matter, see Darkwah, 1999: 57-79.
68 Reindorf, 1895: 44 and note. For Reindorf’s access to informed Asante sources, see McCaskie, 1998: 141-44.
69 Reindorf, 1895: 44-45. Twi abusua, “family.”  
70 Daaku, 1969 (b): 266, interview with Odikro Kwasi Konadu et al. of Fumso.  
71 Daaku, 1969 (b): 17, interview with Adansehene Kwantwi II and elders, Fomena.
72 Reindorf, 1895: 43.
73 Interview with Adansehene Nana Kwantwi II and elders, Fomena, July 1969, see Daaku, 1969 (b): 4, and compare 366-67, interview with Opanin Abubekr and Opanin Kwame Bio who also spoke of people having once lived in the “buttress roots of trees.  
74 Nketia, 1955: 151, 218. Nketia describes the asabaa as a “beating stick” used for preparing a house polish from plantain stalks.
75 Nketia, 1955: 151 n. 3, 4. For useful comments on the story, see Sekyi-Baidoo, 1999: 119-21, 124.  
76 See Wilks, 1993: 111.
77 Story told by Kojo Kuru, aged Amoafo Ankobeahene, in presence of Amoafohene and elders, 6 May 1939, see Ashanti Social Survey Mss, ex-Robert Steel, now in possession of T. C. McCaskie to whom I am grateful for making this text available. An early nineteenth century traveller was somewhat more accurate in referring to it as “where in olden times Koemassie [Kumase] town stood, according to the reports of the Assaintijns.” See Wilks, 1993: 71.
78 Story told by chief of Manso Obodom, 15 January 1938, see Ashanti Social Survey Mss, ex-Robert Steel, now in possession of T. C. McCaskie to whom I am grateful for making this text available.
79 We may be reasonably certain that at this point the Asantemansohemaa used the expression, ]domankoma b]] adeε.
80 Rattray, 1923: 123-24. If Rattray did understand Abena Firi correctly, then she was claiming that the Aduana were, so to speak, more autochthonous than the Oyoko! Over the centuries and even recently, Aduana claims to the Golden Stool have been unsuccessfully advanced.
81 See Prempeh I, 2002: 89-91. A more recent version of much the same story was recorded from the Kokofuhene in 1953. Two interesting passages may be noted. One has it that “... the present area of Kokofu belonged to a certain farmer, by name Kokoo, who was more or less a hunter,” and the other has the wife of “Kokoo” persuading the descendants of Ankyewa Nyame “to accompany her to her husband’s area, which was the present area of Kokofu, since they were hemmed in at Asantemanso.” See J. Agyeman-Duah, Kokofu Stool History, Institute of African Studies, University of Ghana, AS/144, 1963.  
82 Rattray, 1929: 198-99, 217. Rattray’s material on Kokofu must be used with caution. At least one of his informants was an advocate of Aduana claims to the Golden Stool, see 199, n 1. This matter is one of much interest but is not of direct relevance to this paper.  
83 Rattray, 1923: 266-86, and see his comment, 280, n. 2, that “the first seven stanzas... precede every drum ‘piece’.”
84 Rattray, 1923: 280 n. 3, notes that she is said to have arrived at Ahensan, about 5 Kms north of Fomena in Adanse. Other accounts refer to her by different praise names, Asiama Nyankopon Guahyia; see e.g, Agyeman-Duah, 1960: 21.  
85 See Agyeman-Duah, 1960: 21, who refers to them as Kwakye Panyin and Bafuor Antiedu.
86 Agyeman-Duah, 1960: 22 refers to Adu Boahen as Boahen Antuo (“Boahinantuo”). If correct, the story of his death would place it in or about 1700.  
87 See Rattray, 1923: 271, 281.
88 Effah-Gyamfi, 1974 (a): passim; 1974 (b); 217-27; 1985: passim.  
89 Boachie-Ansah, 1986.
90 See Shinnie, 2003: 218-19.
91 For a useful list of the plant domesticates from the West African savanna zones, see Carney, 2001: 379. Only a few of these would presumably have been available in the late-Kintampo or early post-Kintampo periods.
92 Dombrowski, 1976: 64-71.
93 Such a phenomenon is not unknown in other contexts, see Gregg, 1988: passim, for a prehistoric central European example.
94 Posnansky and McIntosh, 1976: 165-66, 189. Posnansky remarked that the site “provides the earliest date for iron working so far [1976] obtained west of Nigeria…” See, however, Calvocoressi and David, 1979: 12, 22, who give the slightly earlier date 13080.
95 Calvocoressi and David, 1979: 12, 22.
96 Goody, 1964: 195-97; Wilks, 1971: 359.  
97 Posnansky, 1976: 61-62; and 1987: 17.
98 Meyerowitz, 1958: 107-108; 1951: 202-205. See more generally, Wilks, 1971, vol. I: 357-59; Garrard, 1980: 46.  
99 Garrard, 1980: 221, and see also 41.
100 Crossland, 1989: 105, and see also 45-46. See also Posnansky, 1987 : 19-20.
101 Effah-Gyanfi, 1974(a): 196-99. There was one similar disk, but of stone.
102 I have always assumed the Twi kramo to be borrowed, through the Mandekan, ultimately from the Arabic root qara’a, “to read.” Dakubu, 2001: 278, accepts it as a borrowing but offers a different derivation.
103 Wilks, Levtzion and Haight, 1986: 91-93, 109-10.
104 Crossland, 1989: 104-105. For the locally made kuduo, see Silverman 1983 (b): passim; Garrard, 2003: 153-54, 174-6.
105 See Insoll, 1997: 29, referring to a sherd of Chinese pottery of the sixteenth to seventeenth century found on a tell in Gao.
106 See Wilks, 1962(a) and Stewart, 1962: 26-28, 28.
107 Algemeen Rijksarchief, The Hague, Collectie Leupe 743, dated Mouri, 25 December 1629.
108 For the importance of the horse in Bono Manso, see Effah-Gyamfi, 1985: 36-37, 211. Many of the bridle paths are still visible. They are known as aponkokwan, from Twi kwan, “path,” and ÔpÔnkÔ, “horse,” the latter a loan word from the Mandekan ponko, “black horse.
109 The seventeenth century descriptions of Bono apply accurately enough to the extensive grasslands that lay to its east, extending beyond the present town of Atebubu to the Volta River. This is referred to by Bowdich, 1819: 172-73 and map, as “Booroom,” and by Dupuis, 1824: xxvi-xxviii, xxxiii-xxxiv, and map, as “Bouromy.” It is possible that Propheet heard of these grasslands and thought that they applied to Bono (or Bron).
110 Dapper, 1670. I have relied on the translations in Ghana Notes and Queries, 9, 1966: 15-18, and see Garrard, 1980: 44.
111 Daaku, 1968: 40-43, and Boachie-Ansah, 1986: 48-51.
112 This event is recorded (in Arabic) by a contemporary annalist, see Wilks, Levtzion, and Haight, 1986: 98, 123, 141. For contemporary Dutch references to the invasion see Fynn, 1971: 62- 63.
113 See Garrard, 1980: 45-46; for Osei Kwadwo’s campaign, see Yarak, 1979: 58.
114 But see Calvocoressi and David, 1979: 12, 22, who give the date 130 ± 80.
115 Garrard, 2003 : 169-72. Garrard allows, however, that pierced nuggets may have been used for personal adornment, and thinks that corpses may have been powdered with gold dust. We may note that living dignitaries might also decorate themselves with gold dust. See the case of the Elmina Caramansa below, and for the Asantehene Opoku Ware (ca.1720-1750), see Winsnes, 2000: 157, “His servants anoint his (Opoku Ware’s) entire body twice a day with tallow, and then strew fine gold dust on it…”
116 Even in the early to mid-eighteenth century those proceeding from Accra inland to Akyem and Asante had, so the Danish factor L. F. Rκmer reported, “to go most of the way along the banks of small fresh-water rivers, otherwise they could not possibly make their way through the bush,” see Winsnes, 2000:110. For road building, see Wilks, 1975: 1-14; 1993 (a): 191-95.
117 R. Kea provides a useful account of the early development of the gold industry in the Tano valley and in that of the Ankobra that runs parallel to it for much of its course, see Kea, 1982: 74-85.
118 Hair, 1994: 53, and see Wilks, 1962 (b): 339.
119 I am grateful to Shinnie and Vivian for giving me a guided tour of the site, see Wilks, “Conversations.” Vol. IV: Field Diary, 19-21 March 1995.
120 There are many versions, differing in detail, of these stories. That here is based on Daaku, 1969 (b) : 42-43, interview with Edubiase Gyaasehene Kwame Brewu II et al ; 3-4, interview with Adansehene Kwantwi II et al; and Reindorf, 1895 : 44.
121 I do not know whether the excavators accept that the pits are larders, but for the practice and its probable antiquity, see Effah-Gyamfi, 1974 (a): 165-66.
122 Shinnie and Shinnie, 1995: 16-17. I am grateful to Peter Shinnie for kindly communicating the radiocarbon dates to me, letter of 25 August 1996. See further Vivian, 1992.
123 For an evocative description of Asantemanso around 1920, see Rattray, 1923: 123-132.
124 Brian Vivian, personal communication, 7 December 1992.
125 Bellis, 1987: 44-47, and see Kiyaga-Mulindwa, 1982: passim.
126 I am suggesting, very hesitatingly, that pots that have to be carried over long distances need larger rims, for stringing purposes, than pots that remain relatively unmoved in kitchens.
127 Shinnie, 1996: 202-203. An earlier and somewhat more detailed list, with calibrated dates, appeared in “Erosion as a Factor in West African Settlement Patterns,” paper presented by Brian Vivian at the 24th Annual Chacmool Conference, Calgary, Alberta, Canada, 1991.
128 Shinnie, personal communication, n.d. but 1988.
129 Christaller, 1879: item 645 (and see Lange, 1990: 55).
130 For the situation of Shama, see Hair, 1994: 61, 77, 85.
131 Crone, 1937: 109. For the location of the “two parts,” see Decorse, 2001: 47-49, 86-89. Van Dantzig 1990: 206, suggests that the ‘part’ on the tip of the Elmina peninsular may have been “a largely Mande trading town,” and the other ‘part’ further west a fishing and salt making place. In view of what is known of Juula practice in their trade networks, Van Dantzig’s speculation has merit.
132 See Blake, 1942: I, 73.
133 See Wilks, 1961: 10 n.3; 1962 (b): 339. For a more recent and fuller treatment, see Wilks, 1993: 5-7.
134 Karamansa apparently did not succumb to attempts to convert him to Christianity, for which see Hair, 1994: 23, 97. Had he done so, he would then presumably have adopted a new and appropriate style as several of his younger contemporaries did, for example, Don João, Don Afonso, and the like, see e.g. Blake, 1942: I, 112-14.
135 Dakubu, 2001: 280. For the 1479 forms, see Escudier, 1992: 28-29.
136 Vogt, 1973: 94. See also Birmingham, 1970: 4-5.
137 Klein, 1994: 650 n.24. In a later paper, Klein, 1996: 264 n. 1, refers to “Wilks’s ‘big bang’ myth,” insists that I argue that “slave labor animates the whole story,” and complains about my (supposed) view that “foreign slaves generated a food-producing revolution…”
138 Anonymous report on Mina, 29 September 1572, see da Mota and Hair, 1988 : 85-86.
139 Van Dantzig and Jones, 1987: xxiii-xxvi. See further Kea, 1982: 24.
140 João III to Count of Castanheira, 5 February 1551, in Blake, 1942: I, 178.
141 Anonymous Report on Mina, 22 September 1572, see da Mota and Hair, 1988: 87.
142 For the association of the elephant and wealth in Akan culture, see Wilks, 1979 : passim; McCaskie, 1983 : passim.
143 Anonymous Report on Mina, 22 September 1572, see da Mota and Hair, 1988: 86.
144 Posnansky, 1987 : 16, suggests that exploitation of the very visible alluvial gold deposits may long have pre-dated mining operations.
145 Anonymous Report, 1572, see da Mota and Hair, 1988: 86.
146 See da Mota and Hair, 1988: 76, 81, and compare Kea, 1982: 201-205.
147 Pereira to Factor de Figueiredo, 8 August 1520, see Blake, 1942: I, 130-31.
148 Boahen, 1973 : 110, was surely correct in arguing that in the sixteenth through eighteenth centuries the terms “Arcanny,” “Akani,” “Accany,” “Haccany,” “Arcania,” “Accanists” and the like, were used to refer to “any Akan trader in general and to an Akan trader in gold in particular.”
149 Van Dantzig, 1990: 206, suggested that the Acanes Castelhanos were those who maintained a connection with Spain via the trans-Saharan trade. This is interesting but unlikely. It is simplest to assume that the nomenclature was carried over from the period of Portuguese and Castilian rivalry, which effectively ended in 1480.
150 But see Garrard, 1980: 160-63; Wilks, 1982 (b): 463-72.
151 Acting Governor to Queen of Portugal, 18 April 1557, and anonymous Report on Mina, 29 September 1572 in da Mota and Hair, 1988: 64, 72 n. 29, 86.
152 Ghana Notes and Queries, 9, 1996: end piece map.
153 Daaku, 1969 (a): interview with Nana Kwaku Aputae II, Assin-Nyankumasi.
154 See, for example, Daaku, 1970:144-148.
155 A. C. Denteh, “Asona,” unpublished MS, 1973: 2. Copy in writer’s possession. See also “History of Ejisu,” dictated by Edwesohene Kwabena Wusu, Edwesohemaa Afua Ampobin, and elders, 27 June 1933. Typescript in author’s possession, copied from original among papers of Asantehene Sir Osei Agyeman Prempeh II. Attobrah, 1976: 1, provides the Asona with a spurious northern ancestry, having As] (“So”) arrive not from the hole but from “around the Niger Bend roughly between Djenne and Timbuktu…” He thus creates a problem about Odomankoma, and so presents him as “a shrine or deity” otherwise known as Bona. He is followed in this by Addo-Fening, 1997: 2.
156 C. E. Aidoo, “History of the Denkyiras,” compiled with the cooperation of Denkyirahene Owusu Bori II, copy in writer’s possession, n.d. but ca. 1960. See also Kumah, 1966: 34. Both authors found it necessary to attempt a reconstruction of pre-Odomankoma ‘history,’ thus departing from the tradition of the Adomankomasεm. For Abankeseso, see Ameyaw, 1966: passim.
157 National Archives of Ghana, 158/1/11, Commissioner, Western Province of Ashanti, to Chief Commissioner, Ashanti, 23 March 1914. I am indebted to T. C. McCaskie for drawing my attention to this document.
158 Christaller, 1879: item 967 (and Lange, 1990: 81); Owusu-Sarpong, 2000: 102. Compare Danquah, 1944: 58-77.
159 Aquonna and Essonna are the modern Ekoona and Asona. Intchwa is Atwea, associated with Aduana. The identification of Tchweedam is unsure. The explanation is usually but not always obvious, for example, Twi εkoÔ, “buffalo.”
160 “Yoko is the red earth used to paint the lower parts of the houses in the interior,” Bowdich was told, and he must have been aware, from personal observation, of the vast amount of (women’s) labour that went into distinguishing a house from a hut by the application of red ochre reliefs. The reader will recollect the reference above, in the tale of the creation of Amuowi, to “children of red earth.”
161 Bowdich adds that his informants observed that the “their long and more intimate intercourse with the blacks, had made the present race a mixture of the African and Portuguese.”
162 Mauny, 1951: 238-42, where a convincing argument for the date of 1502 is made. For a lengthy if much outdated consideration of maize in Africa, see Miracle: 1966: passim.
163 McCann, 2001: 250, citing M. D. W. Jeffreys.
164 Van Dantzig and Jones, 1987: 162-63, and see further, J. D. La Fleur, “Atlantic-era experimentation with plantains on the Gold Coast,” paper delivered at the African Studies Association Annual Meeting, Washington D.C., 6 December 2002. As of the time of writing, La Fleur’s University of Virginia dissertation is awaited.
165 Anonymous Report on Mina, Da Mota and Hair, 1988: 81. Kea, 1982: 81, identified Agri as Mampa, on the lower Ankobra River. I suspect the sense of the Portuguese is: … “Where it is hard to think that anyone had planted it.”
166 Crosby, 1972: passim. Carney, 2001: 381- 84.
167 Van Dantzig and Jones, 1987: 114. See the Akan saying, Burofua na εyεsan, “one grain of maize can fill a hut,” Christaller, 1879, item 678 (and Lange, 1990: 50).
168 Van Dantzig and Jones, 1987: 40-43, 63. Pito occurs in many West African languages, e.g. Hausa fito.
169 McCann, 2001: 248-49.
170 Fairhead and Leach, 1996: 49-51. Posnansky, 1987: 17.
171 I draw here upon a personal communication from James Fairhead, 26 September 2002.
172 Effah-Gyamfi, 1974 (a): 201; 1985: 205. Calvocoressi and David, 1979: 17. Crossland, 1989: 95, 106. Boachie-Ansah, 1986: 290.
173 See e.g, Shinnie, 1996: 201; Klein, 1996: passim.
174 See e.g. Wilks, 1994 (a): 659.
175 Rattray. 1914: 152; Christaller, 1879, item 917 (and Lange, 1990: 77).
176 Wilks, 1962 (a): 28, where I misguidedly referred to an Arabic sikkat, “coinage.”
177 Dakubu, 2001: 280. It is surprising that the Twi names of the gold weights appear not to be borrowings, see Garrard, 1973: passim, but see Garrard, 1980: 24-25 for other loan words to do with weighing. Dakubu, 2001: 281 suggests, however, that Twi abrammu,”gold weight,” is a Twi/Mande hybrid.
178 Daaku, 1969 (b): 75, interview with Akrokerrihene Owusu Sampa III and elders.
179 0Anonymous Report on Mina, 22 September 1572, see da Mota and Hair, 1988: 86.
180 Christaller, 1879: item 963 (and Lange, 1990: 81).
181 Christaller, 1879, item 2959 (and Lange, 1990: 244).
182 Wilks, 1979: 12-16, and see generally, McCaskie, 1983.
183 Wilks, 1975: 521. See further McCaskie, 1981: 129-31, who argues for the movement’s anti-witchcraft character. With regard to present-day concepts of the nature of Odomankana, I can testify personally to the utter consternation occasioned when a dust-devil passed through a room in which I was conversing on the topic with two Asante! They literally fled on the appearance of such a “force”.
184 The classic in this field is undoubtedly Danquah’s The Akan Doctrine of God, and see particularly 1944: 59, “Let us now do the work which [J. G.] Christaller, obviously, must have left to us, his students, to carry on where he left off.”
Haut de page

Table des illustrations

Titre Relict landscapes of the Bron and Akan
Fichier image/png, 17k
Haut de page

Pour citer cet article

Référence papier

Ivor Wilks, « The Forest and the Twis », Journal des africanistes, 75-1 | 2005, 19-75.

Référence électronique

Ivor Wilks, « The Forest and the Twis », Journal des africanistes [En ligne], 75-1 | 2005, mis en ligne le 15 septembre 2008, consulté le 27 mai 2020. URL :

Haut de page


Ivor Wilks

Professeur émérite, Northwestern University, USA

Haut de page

Droits d’auteur

Société des africanistes

Haut de page
  • Logo Société des africanistes
  • OpenEdition Journals