INTRODUCTION
A group of studies focusing on the nature of non-centralized societies, their organization, maintenance, development and continued integrity, appeared in the Journal of African History in 2001 and 2002.Footnote 1 These studies, based in Senegambia, Guinea and Burkina Faso, looked at ways in which stateless societies could maintain themselves when faced with much larger-scale neighbours, often large, powerful and centralized states, and focused on the roles of slavery and slaving in these social formations during the eighteenth and nineteenth centuries. The authors of these essays, engaging with debates on slave-raiding and the slave trade, emphasize and illuminate the sometimes tense inter- and intra-state relationships that enabled the maintenance of small-scale and/or stateless societies during the seemingly adverse conditions of the slave era. The general thrust in these pieces is in the direction of specific functional and organizational differences of these small groups in comparison with the larger groups, differences crucial to their continued existence.
My concern here is more encompassing. At one level I provide a geographical comparison, from several hundreds of kilometres southeast of Guinea, of the development and survival of a small-scale society, with powerful neighbours to the west and east. I look firstly at the construction and maintenance of this society from the late seventeenth to the late nineteenth centuries; and secondly, at its destruction and reconstruction in the last quarter of the nineteenth century – in circumstances framed not only by European economic influences, but also by Christianity and colonial rule. Whereas the earlier (JAH) essays used relationships with (either domestic or Atlantic) slaving, as a lens to focus their discussions, thus contributing to debates about ‘predatory states’, in this discussion slaving is somewhat tangential, and does not provide a differentiating lens. Similarly, I suggest that moves towards centralization in terms of borrowing and adapting chiefship structures and their attendant office(r)s was a kind of ‘optional extra’ to this particular society in the nineteenth century, and more detailed examination of political centralization is not crucial to the arguments here.Footnote 2 Instead, my findings suggest that a focus on the maintenance of difference, on notions of ur-identity, for example, particularly with respect to the economic and, more generally, ‘modernizing’,Footnote 3 must be rejected and that identity is tightly bound up with local, regional and global economic opportunities and constraints. For the society under consideration here, this is as true now as it was in the competitive milieux of the nineteenth century. This investigation of one particular society speaks to broader debates about identity in the global era and provides an opportunity to consider recent anthropological literature in a specific historical context.Footnote 4
CONTEXT
The central part of what is today Ghana's Volta Region (Map 1) is often referred to as Eυe-Dome, bounded on the west by the Volta Lake, and on the east by the border with Togo. Although Eυe speakers are in the majority, there is a variety of other ethnic/linguistic groups: there are the Guan-speaking groups of Anum and Boso to the southwest, and Nkonya to the northwest. Around the northern fringes of the area are groups speaking other distinct languages (Likpe, Lolobi, Santrokofi and Akpafu, and, farther north, Borada and Bowiri, Lefana). These groups were formerly referred to by linguists and early ethnographers as Restvölker (remnant peoples) speaking languages known as Restsprache (remnant languages), but now they are referred to as either Central Togo (CT), or Ghana-Togo Mountain, peoples and languages.Footnote 5 Avatime, at the focus of the essay, is in the centre of Eυe-Dome, together with three other CT-speaking groups, Nyangbo, Tafi and Logba, in the mountains and their western fringes.Footnote 6
of Avatime (inset shows Avatime's location in relation to contemporary towns). (Map drawn by Harry Buglass.)
In the eighteenth and nineteenth centuries Eυe-Dome had the empires of Asante to its west and Dahomey to the east. The Volta River provided some protection from Asante depredations and it also afforded an easy access to the coast (the extreme west of the former Slave Coast). However, in addition to its interstitial location, Eυe-Dome's mountainous terrain offered places to hide, recuperate and (re-)establish. The southernmost fingers of the Togo Ranges occupy much of the central part of Eυe-Dome, and to the north are more hills. The area's water-rich, steep-sided and forested valleys, and the mountains rising from the plain of the Volta, mean that Eυe-Dome provided not only a buffer zone between two major spheres of influence, but also a refuge area for the displaced, the out-of-favour, the renegades. Dahomey presumably raided the east of this interstitial area for slaves and produce, and we know that by the end of the eighteenth century Asante expected tribute, in the form of slaves and produce, from some western Eυe-Dome districts. This tribute was channelled through Asante's ally, Akwamu, re-established on the Volta after the 1730s around present-day Akosombo, Juapong and Asutsuare.Footnote 7
But eighteenth- and nineteenth-century Eυe-Dome was more than a contemporary dustbin. Although the river served as a physical barrier to Eυe-Dome's west, it also provided a channel for trade both with the coast, linking to the trans-Atlantic trade, and to the north, linking to trans-Saharan trading networks. Routes along the eastern bank of the Volta, north of Akwamu, unhindered by Asante tolls, gave Eυe-Dome peoples the chance to engage in complex networks of trade. With good land and opportunities to trade, the Eυe-Dome groups, whatever their origins, could prosper.
According to oral tradition and nineteenth-century written sources, contemporary Eυe-Dome groups claim various putative origins, from autochthony to immigrant groups claiming origins to both the east and the west. There were broad similarities among some groups, particularly the Eυe-speakers, but that did not mean that, as trade developed and prospered in Eυe-Dome before the mid nineteenth century, the area was unified in any way. There were also many rivalries and sometimes skirmishes among Eυe-Dome peoples, stemming from real or imagined slights and betrayals and their memories.
AVATIME
The Eυe-speaking groups in Eυe-dome claim Notsie (today in Togo) as their immediate origin,Footnote 8 while autochthonous groups claim to have come from caves in the earth, or down from heaven on ropes. Unlike the Eυe, Avatime claim to have come originally from Ahanta, now in Ghana's Western Region, and they have been claiming this for over 100 years.Footnote 9 They say they left Ahanta because of a dispute over land or chiefship. Twentieth-century versions of Avatime history suggest that some of the proto-Avatime stayed near to Ningo, on the coast near to Accra, for a while, but were forced to leave after a quarrel with the indigenous people. This group then moved inland and settled at first near Matse, in a valley about 15 miles to the south of present-day Avatime, but again they were forced to leave because of a quarrel.Footnote 10
These proto-Avatime halted in the mountains at a place now called Oŋulusu, but before they could settle and begin to farm, they had to learn to live with or subjugate the existing inhabitants, the Baya, which they claim they did using a combination of force and trickery. The Baya who survived the amalgamation or subjugation were divided among the present Avatime settlements and today there are Baya clans (see below) in the villages of Biakpa, Fume, Dzokpe and Vane. There is also significant linguistic evidence that the Avatime adopted the language of the Baya, a CT language.Footnote 11 From Oŋulusu Avatime explored the surrounding land and eventually founded the six villages of Biakpa, Dzogbefeme, Fume, Vane, Gbadzeme and Amedzofe. The seventh village, Dzokpe (an Eυe name meaning ‘a coming together’, in this case of people), was founded later, probably as an Avatime settlement that grew up around the large market on a site on the Volta plain, about one and a half kilometres to the west of the foot of the mountains, in the nineteenth century.
But both oral histories and genealogies indicate that the present Avatime population comprises the descendants not only of the ‘proto-Avatime’ group and the Baya, but also of other individuals and groups who arrived in the area and were incorporated into Avatime well before the middle of the nineteenth century. Within the village of Amedzofe, for example, members of three clans claim common antecedents (descendants of brothers who came via Oŋulusu). But the clan recognized as the earth owners (Dopome), and thus the earliest settlers, claims no kinship with these three, and members of a fifth clan claim no original relationship with any of the other four. Clan members of all are fully recognized as Avatime from Amedzofe whose forebears settled in the area originally as a separate group – whether of immigrants or autochthones is immaterial today.
Bayas, Ahumakawos and Ohlobowos were one people whose language was the same and they were here before [Avatimes] came. It was from them they learnt this language … Akwamu chief Akoto had been fighting them at Denu Ava and followed them and they came to hide themselves at Oŋulusu.Footnote 12
And also:
Okpotobleku who was hunting in the forest found that there were some people who were already there … the settlers who came before him were the Dopome people. These first people planted the medicineFootnote 13 in the town … as the discoverers, that is why they are the earth owners … Through his hunting Okpotobleku discovered level ground at the town site … the first man to discover the town or the first to plant [the medicine] has no member of his family remaining in the town at all … Dedume family has now taken over all [their] lands, … Dedumes were strangers who came to lodge … and finally became part of Dopome and Amedzofe in general.Footnote 14
And on the founding of Dzokpe: ‘Dzokpe was the slave trade centre in the olden days and many people in [that] town were bought. People from the other six Avatime towns united together to build this town’.Footnote 15
It seems both possible and probable that the Oŋulusu group was part of the scattering of people that took place in the wake of the Akwamu invasion of the Ga and Adaŋbe areas in 1679. Sprigge traced the origins of much of the present-day population of Agotime to this event, and the general tenor of his argument is that many other groups were similarly displaced.Footnote 16 Avatime histories claim a stay in Ningo, in the Ga-Adaŋbe area, and historical association between Avatime and Agotime.
Avatime accounts of their history are obviously biased: contemporary memories can be categorized as either Avatime victories, or cataclysmic and horrific events. Whether victory or cataclysm, Avatime is given a leading and heroic role. Victories claimed by Avatime include hegemony over Logba, Avatime's immediate northern neighbour, after longstanding rivalry (manifested in a series of skirmishes);Footnote 17 Avatime's part in what is remembered there as the ‘Yao Sekyere War’;Footnote 18 and the inspirational and geographically apposite role of Avatime in making a stand against the major Asante incursion east of the Volta between roughly 1869 and 1871 (see below). The cataclysms are the events whose horror is such that they are remembered in the sacred oaths of villages: lost battles, the scurrilous murder of a chief's son, for example, and, again, the Asante invasion and ultimate razing of much of Eυe-Dome between 1869 and 1871.
By the late eighteenth century Asante claimed suzerainty over some of the peoples east of the Volta:Footnote 19 in any event, Asante expected tribute from some Eυe-Dome groups. The failure of Wusuta (Eυe) and Nkonya (Guaŋ) to send tribute to support the Asante expedition to Gyaman was the ostensible reason for the Yao Sekyere foray in 1822. The tribute was channelled through Asante's allies, Akwamu, around the Volta where there was a safe river crossing. In 1833, however, several Eυe-Dome groups allied under the nominal leadership of Peki (Eυe) and, copying Asante military formation, defeated Akwamu. In spite of Avatime's small size, it apparently wielded some political influence in the area through the nineteenth century.
There is considerable oral evidence from Eυe-Dome groups that chiefs were a nineteenth-century innovation, effectively the aftermath of the adoption of Asante military formation and its offices. Welman indicated that the Peki chief (Kwadzo Dei) who led the allied Eυe-Dome forces against the Akwamu was the first Peki chief. Chiefly genealogies in Eυe-Dome, given the number of appellations attached to chiefly titles, derive from the early part of the nineteenth century.Footnote 20 Oral information from Amedzofe also indicates that chiefship on the Asante model was a nineteenth-century innovation. Accounts of early social organization imply that lineage and clan elders were effective rulers before then: names of warriors, rather than chiefs, being remembered from early struggles. In contemporary Avatime, land allocation and inheritance is still entirely in the hands of lineage elders and there is no stool land.
Asante and Akwamu dominance and influence in Eυe-Dome thus dwindled through the mid nineteenth century.Footnote 21 Not only did the Asante influence dwindle between 1833 and the last third of the nineteenth century, but chiefship, as an alien imposition, was also probably not important at this time. A central tenet of Ward's thesis is that the frequency and relative ease of destoolment of Eυe-Dome chiefs was an indication of their relative lack of significance.Footnote 22
TRADERS AND ENTREPRENEURS IN THE NINETEENTH-CENTURY VOLTA REGION
Until 1850 Eυe-Dome was deemed by European powers on the Gold Coast to be under the suzerainty of Denmark, however minimal that might have been. In 1850 Eυe-Dome was ceded to similarly minimal influence from the British when the Danes left the coast. But minimal European influence did not mean minimal development. While there are no conventional written records of early nineteenth-century Eυe-Dome, we can sketch patterns of population consolidation, trading relations and commercial networks from a range of serendipitous sources: travellers' tales in a range of languages; local oral accounts; maps deduced from accounts of markets and days' journey between them; and, from the mid-century, reports from missionaries and traders availing themselves of the area's rich natural produce.Footnote 23
Between the 1830s and 1870, trading networks and production for sale mushroomed in Eυe-Dome. Traders – local, coastal and European – bought a range of goods from local producers, and sold what was in demand. The networks were extensive, both inside Eυe-Dome and outside: to the north, up the Volta, the worlds of the trans-Saharan trade;Footnote 24 to the northeast, through Atakpame, to the Hausa states in Northern Nigeria and beyond,Footnote 25 and to the south, the Gold Coast and the Euro-Atlantic worlds.Footnote 26
Ray Kea drew a picture of the Danish-, Afro-Danish- and African-owned plantations between about 1785 and 1840 in the Akuapem and Krobo hills and the Accra plains, east to Ada, the area adjacent to the southwest of Eυe-Dome.Footnote 27 The area was fertile and while originally the Danes planned export production (cotton and coffee), plans were overtaken by post-abolition demands for maize: already it was cheaper to import African produce than to grow it locally. But maize and coffee production were replaced by more profitable trade in gold and palm oil, by the mid-century.Footnote 28 While Danish-owned plantations fell into desuetude, locally owned plantations produced for indigenous markets – the demand for foodstuffs, both old and new, from the coastal towns already being significant.
The earliest available printed source linking this network to Eυe-Dome is Thøning's map, originally drawn in 1802.Footnote 29 Versions of the map show trade routes from the coastal areas around Ada and Keta inland, each described route punctuated by named villages where there were markets. Bühler looks at Eυe-Dome as part of a wedge-shaped area he terms the ‘Volta Region’, larger than contemporary Ghana's Volta Region, narrower in the south, and stretching between the coast and the southern boundary of the Sahel, with the eastern bank of the Volta as its western boundary.Footnote 30 He also notes tributary relations with Asante, but his work suggests a more positive image of Eυe-Dome than that of Asante vassal, not only with the movement, forced and voluntary, of produce to Asante, but also with strong trading links outside the area.
The focus for slaving shifted east after 1807 and increased population in the coastal areas of the Slave Coast created a growing demand for food and other produce. Thus, settlements like Anyako and Waya grew in importance as key stages in the transport, firstly, of food and slaves, then, from the mid-century, of palm oil, from the interior as Eυe-Dome joined the export drive to Europe. While the orientation of Eυe-Dome's trade had been multi-directional in the early part of the century, by the 1850s the emphasis was much more to the south.Footnote 31
By the mid nineteenth century, then, the export of slaves from Eυe-Dome had moved east, the trades in staple foods linked Eυe-Dome to coastal settlements, and trade in cotton, coffee and palm oil linked the Eυe-Dome to the worlds outside. Although the palm oil trade from Togo was small compared with that from the Niger Delta, by the mid 1850s, according to Bühler, two German ships each made ‘semi-annual’ visits to Keta for palm oil. While the Eυe-Dome peoples had long exported foodstuffs, artisanal productsFootnote 32 and humans, what they got in return changed and expanded. In the early nineteenth century, trading seems to have been largely for arms, ammunition and spirits, but later incorporated cloth and, towards 1900, gold and Asante-style regalia, in addition to European tools and domestic wares.Footnote 33
The Bremen missionaries (North German Mission Society)Footnote 34 who established their station in Ho in 1859 were often severely short of money and supplemented their incomes through trade, including that in rubber, and, up to 1869, coffee. In addition, however, they also took part in local trade. They (or their wives) made sausages and other meat products, as well as baked goods, for sale in the local markets.Footnote 35 Trading was ubiquitous throughout Eυe-Dome, although Bonnat writes that trading in the mountains was largely in the hands of Accra men.Footnote 36 By the 1860s Eυe-Dome was producing maize and other foodstuffs for regional export, palm oil and coffee for overseas export, and had added cotton to its list of potential money-earners. Bonnat, himself a trader, was particularly interested in cotton in the Ho and Agotime areas in the 1860s, and his account, revealing his naivety and optimism in purchasing cotton even on the day of his capture by the Asante in Ho, clearly indicates how important and well-established were the networks and practices of trading in Eυe-Dome.
The Eυe-Dome emerging in the first half of the nineteenth century was an area of active trade networks, both in slaves and in a broad and changing range of commodities, according to local and international demand. It was an area criss-crossed by paths and interspersed with specific market places, bounded to the west by the Volta, itself a significant trade route. Its seeming accessibility, however, was contradicted by its topography. Its peoples spoke a range of languages and called themselves different names (subscribed to different ethnicities and cultural practices), and its peopling was derived from successive waves of immigrants, expelled or fleeing from elsewhere, absorbing or being absorbed into existing local groups. This was, in general terms, the state of play up to the end of the 1860s.
CATASTROPHE: … ‘A LAND DEVOID OF PEOPLE … ’
A thorough history of the Asante campaigns east of the Volta between about 1868 and 1871 has still to be written.Footnote 37 It is enough to state here that between 1869 and 1871 there were large Asante armies in Eυe-Dome and that sometime, probably in 1870, there was a significant rout of the Eυe-Dome peoples on and around the Avatime mountains. The three or so years of the Asante occupation saw destruction of villages, death, enslavement or the flight of the population out of the area or into its remotest mountains and forests, seeking refuge even within a refuge area.
Marion Johnson published what is still the base-line article on this invasion in 1965.Footnote 38 It was intended to be a preliminary account of Asante's campaigns in Eυe-Dome, effectively its last imperial, expansionist expedition, and she interpreted the causes and consequences of those campaigns from the point of view of Asante: the (re-)assertion of control over local peoples, the demonstration of strength against the British (who now had at least nominal suzerainty over Eυe-Dome), the desire for wealth, and factional politics at home.Footnote 39 When Johnson musters evidence of opposition to the invasion, it is primarily in terms of the catalytic charisma of Dompre, the Akyem military chief who, failing to persuade the Akwamu (Asante allies) to compensate Akyems for an earlier contretemps, threw in his lot with the Eυe-Dome groups in their attempts to defend themselves against Asante.Footnote 40 The British are distant, minatory and fickle dei ex machina. The Eυe-Dome peoples – Eυe, Guaŋ and Central Togo, including Avatime – hardly featured in Johnson's Asante-oriented narrative.Footnote 41 So, too, with respect to outcomes and consequences, Johnson focused on the consequences in Asante, for Asante and for the Gold Coast Colony. For the peoples of Eυe-Dome, however, the Asante incursion had revolutionary consequences which are little remarked or discussed in work on Ghana. Eυe-Dome was a refuge area in the 200 years prior to 1870, but the years immediately after the Asante departure saw a new beginning. The vast majority of the local population had been either put to flight, put to death or taken prisoner to be sold as slaves.Footnote 42
Early ethnographic work on Eυe-Dome such as Rattray's collection of village histories tends merely to record the incursion. Ward records a battle around Gemi, the highest mountain in the Avatime area,Footnote 43 and in 1924, Welman, whose work was based on British documents, notes ‘a severe defeat of the Asante’ at ‘Obodjani or Bajame’ (Gbadzeme, one of the Avatime villages) sometime shortly after 1869.Footnote 44 A formal ‘History’ of Avatime, apparently written by the paramount and other elders in the early 1930s, and deposited in the Ghana National Archives,Footnote 45 records the incursion and the battles around Gemi. In the 1970s, the Asante invasion of 1868–9 and tales of it were still widely remembered in villages throughout the Volta Region. In interviews from the 1970s, accounts of past military glory, other battles before 1868–71, and even skirmishes within (then) living memory faded into insignificance in comparison with oral narratives of the Asante invasion. This was the singular happening recounted without fail by whatever group was asked to talk about their ‘history’ and how they had come to be as they were then, in the 1970s. The Asante incursion was the most significant, destructive and disruptive set of events from the past.
When the Asante left Eυe-Dome in mid-1871,Footnote 46 they left behind a thinly populated and agriculturally bankrupt land: Avatime and other groups had their villages and crops destroyed. We also have no estimate of the numbers who were captured and killed or enslaved. Spieth, writing in 1892, states that not only were significant numbers of slaves taken from the area to both Anlo and Kumase, but also groups of people fled across the Volta to what subsequently became the northern part of the Asante colony and the Northern Territories. Johnson asserts that many Eυe-Dome people fled southwest to either Akuapem or the Accra Plains.Footnote 47 Those who had escaped the Asante without fleeing across the Volta emerged from forests, or came down from hills, to nothing. The after-effects of the Asante invasion were cataclysmic.
Apart from the immediate loss of life, enslavement and destruction, the socioeconomic networks underpinning the former development of the area were destroyed. Although commercial centres, like Kpandu on the Volta, began to be re-built fairly quickly,Footnote 48 it took longer for the survivors of other groups to come back, to rebuild and re-establish settlements, to plant and harvest, to begin to live and reproduce, let alone to re-establish the trading networks of the pre-war era. When missionaries from the Bremen Mission returned to Ho in 1875 they reported: ‘The Bremen Ho station, or what remained of it, was on land devoid of people’.Footnote 49 Müller wrote that the people of Ho began to rebuild three of their villages only by the end of 1877,Footnote 50 and Bonnat that the Peki chief, a major player in the conflict, was still in Sokode (8 km west of Ho), in 1875, where he had taken refuge.Footnote 51
In 1876, Gouldsbury, a British official, reported that there was serious distrust among the Eυe-Dome groups resulting from different allegiances during the invasion, that there was ‘anarchy, confusion and insecurity’.Footnote 52 Once farming was re-established, the drive to produce for trade and export began again. By the late 1870s and early 1880s, the missionaries in Ho had re-established their coffee and fruit farms, and agents from European firms were stationed in the area, travelling to buy maize and cotton.Footnote 53 Once re-establishment was underway, trade routes to the coast were also quickly re-habilitated.Footnote 54
NEW BEGINNINGS I: (RE-)ESTABLISHING AMEDZOFE
The extent of destruction of Avatime after the battles around Gemi in 1870–1 is not recorded and there are no descriptions of when and how resettlement occurred. Gbadzeme was certainly ‘there’ by 1876,Footnote 55 as was Dzokpe,Footnote 56 and probably also Amedzofe. While those who fled from Gemi and hid might have begun rebuilding, those who were captured and not killed were taken out of Eυe-Dome, either south to Anlo (Asante allies) or to Accra and sold into slavery. Once the Asante retreated and people began to move back to their villages and farms, repopulation was helped significantly by the British proclamation of December 1874 abolishing the status of ‘slave’. Not until then was it possible for ex-slaves, whether escaped or redeemed, to move back to wherever they had come from, and to live freely among their kin, if that was what they wanted.Footnote 57
But how was Amedzofe ‘constituted’ let alone re-constituted? In the mid twentieth century when I first went to Avatime formal social organization was focused on loosely patrilineally structured, localized ‘clan’ groups, most of whose forebears were believed to have migrated to the area. Each clan was composed of patrilineal kin groups, ikun∊ (sing: oku), localized within the clan. The Avatime word that I have glossed by the classical anthropological ‘clan’ is lεkp ɔlε (pl.: akp ɔla), a word that can also mean ‘knot’ or ‘lump’ in other contexts. Avatime clans are perhaps better described as ‘knots’ or lump-like groups of ikune, rather than neatly structured organizations, one generation following from the next and all descending from an apical ancestor. Clan and oku membership are important as they confer rights to land, to living space in the village, to support in times of adversity and during life-crisis rituals, as well as to inheritance of titles and positions if the group is eligible to hold any. In addition, clan and oku membership also confer obligations and duties: to give support, to give advice, to be involved in life-crisis rituals of their members.Footnote 58
The five contemporary Amedzofe clans do not claim a single origin. We saw earlier that three of them claim descent from ‘brothers’; the founders of the other two were unrelated, but wholly integrated, settlers from well before the nineteenth century, possibly predating the fraternal clans. Whatever their claimed original kin relations, contemporary Amedzofe people believe that members of these five clans intermarried and thus developed new affinal and kin links among themselves, the whole network constituting Amedzofe.
Ikun∊, in theory, as patrilineal groups within clans, might effectively nestle within an overarching clan structure. They may do, but often do not. Genealogies from Amedzofe clans suggest there were members of ikun∊ who traced their ancestry back over six or seven generations and who claimed the clan founder (the ‘apex’ of the clan structure) as their ultimate progenitor. Elders of these ikun∊ could remember the names and origins of in-marrying wives and husbands of ikun∊ women sometimes four or five generations above the oldest living people. However, there were members of other ikunɪ who traced their clan progenitor only as far as three or four generations at most. For members of these ikun∊ the link to the clan was simply asserted rather than narrated; names and origins of wives, names and destinations of sisters were rarely given. But elders from lengthier ikunɪ in the same clan remembered things differently. They remembered a woman or sometimes women, daughters of their ikun∊ members, who married a man/men whose name and origin they might not know.Footnote 59
Where ikun∊ should nestle neatly and in orderly fashion, but do not, it is more useful to think of the clan as a ‘knot’, as its Avatime name suggests. Clans could perhaps better be thought of as an untidy knot, or dense cluster of knots. The array of clan forebears constitute the solid mass of the knots while ends of thread emerging from the central knot represent the descent lines of the ikun∊. Some of these lines seem to have a firm and sequential connection with the central knot, others are more tentatively connected. The uncertainty about connection to the central ‘knot’, in part at least, stems from the way that the villages were reconstituted in the last quarter of the nineteenth century. This is the 25-year period – the resettlement after the Asante invasion – in which genealogies fill out, to contain details of in-marrying wives and their origins, details of polygamous marriages, serial monogamy or widow inheritance.
NEW BEGINNINGS II: EXPANDING AMEDZOFE
After the abolition of the slave status, some of those captured and sold in the south returned to Avatime, asserted kinship with already (re-)settled groups, and became truly ‘free’ in the paradoxical fashion of West Africa: they had kin, they had rights and obligations as members of a kin group.Footnote 60 Although Amedzofe's gradual re-establishment may have been begun by its former inhabitants returning from refuge or slavery, its growth and development, infrastructure and reputation in the last quarter of the nineteenth century were also significantly shaped by Christianity and colonial rule.
While Avatime had been known to the missionaries before the Asante incursion, there were no Avatime Christians then: the first were people who were part of the general repopulation of the area after the Asante retreat, largely those who had formerly been enslaved. Some of these had been sold in the Gold Coast, but redeemed by Basel missionaries in Akuapem, converted and lived around the Basel stations of Maiera or Abokobi and nearer Accra.Footnote 61 The Bremen missionaries saw in these incomers a kind of call to arms. Müller writes of ‘the Eυe refugees returned from the Gold Coast to their home, most urgently requiring/demanding of spiritual care’.Footnote 62
In the early 1880s the Bremen missionaries began to extend their work in Eυe-Dome, but it took until September 1887 to decide to build a new station, school and chapel in Amedzofe, and until 1889 before they had funds to begin the building.Footnote 63 The Bremen Mission's station in Amedzofe proved to be a magnet for new people. Christians came to enable them to maintain and strengthen their faith; craft workers, apprentices and labourers came for the building of the mission house, the chapel and the school buildings; and parents sent their (male) children and adolescents to the new school there.Footnote 64 This influx received a further boost when the Bremen Mission Seminary was moved from Ho to Amedzofe in 1894.Footnote 65 By the time Missionary Seeger wrote his first Annual Report from Amedzofe in December 1889, there were, he estimated, between 200 and 400 people in the village.Footnote 66
After the Berlin Conference, Avatime became part of the German colony of Togo, and, in spite of protests both from local people and from the British, remained under German rule in the subsequent border wrangling. The early colonial period saw Eυe-Dome groups competing with each other for access to the colonial powers, their ‘ear’, presumably with the intention of benefiting from access to trade and broader opportunities for economic and political development thought to be at their disposal.Footnote 67 This was also a period when local groups were struggling to renegotiate both local/internal statuses and their external relationships in the aftermath of the Asante invasion.
In general, the German colonial record in Africa is harsh and repressive.Footnote 68 In Togo, however, the picture is not so overtly violent: German merchants, mainly from the Hanseatic towns, had, by the time of partition, established strong interests in Togo and, coupled with the presence of the Bremen missionaries, effectively prevented the worst excesses of German colonial rule from emerging there.Footnote 69 While the coastal towns and their immediate environs were, for the most part, under ‘Direct Rule’, lack of manpower in relation to the size of administrative districts in the centre and north of the colony allowed a form of ‘Indirect Rule’ to develop.Footnote 70 Misahöhe, close to Kpalimé (now in Togo), was the district centre for Avatime, about a 25-kilometre walk away as the crow flies, and, although the name and – perhaps apocryphal – deeds of the first long-term administrator, Dr Grüner, were well remembered in Avatime in the 1970s, German interference in local affairs was relatively small. In spite of this, however, since chiefs were recognized as the ‘leaders’ of local communities and received remuneration for various tasks, chiefship and its attendant offices grew in importance after the 1880s.
Those who came to Amedzofe as ‘new strangers’ had to be incorporated into the social organization of the village if they wanted land to farm and a network of support there. When a ‘stranger’, that is, anyone who was not Avatime, arrived in Amedzofe, to work and possibly to convert, or as a result of marrying an Amedzofe man or woman, they were expected to affiliate themselves formally to an oku and, hence, a clan. The process is referred to as ‘choosing a father’.Footnote 71Ikunɪ, as the smallest kin groups in Amedzofe, were responsible for both the moral and economic affiliation and socialization of people, while descent was formally reckoned patrilineally. Thus by choosing a father, a stranger gained access both to land for building and farming and to support in case of life-crisis rituals, disputes or quarrels: s/he assumed Avatime social personhood. Reciprocally, as a formal member of the oku the incomer had obligations to contribute to the clan and oku in whatever way possible, to contribute money for life-crisis rituals, to contribute advice/help at meetings, and to contribute labour if asked. The same procedures are still mandatory.
While in the late twentieth century, a teacher, researcher or agricultural extension officer, any stranger to Amedzofe, chose a father for the duration of his/her posting in the village, it is obvious from a close study of genealogies that there were a number of incomers, whether Christians, artisans, school pupils, traders or simply migrants, in the wake of the resettlement of the area in the late nineteenth century, who came to the village, affiliated to ikun∊, and stayed. They might marry Amedzofe women, or have children who grew up in the village and married from there. Using the knot and thread analogy again, the uncertain, tentative links between the threads and the knots are effectively these formal affiliations, the ‘choosing of fathers’, or, in the subsequent generation, kin ties through women rather than men. To all intents and purposes, in everyday terms, the lives and livelihoods of these affiliated incomers were no different from genealogically correct Amedzofe people.
But affiliated incomers could never aspire to any formal political or ritual offices in the village: they could never become ‘big men’ in late nineteenth-century village terms, priests or clan or village chiefs. What happened instead was that the incomer groups, those who stayed and founded their own ikun∊, tended to look for advantage and improvement, big men status, in the novelties of Christianity, education and craft skills. The majority of Amedzofe names in the early baptism and confirmation records of the Church belong to such families. So, while some of the connections between the knots (the clans) and the threads (the ‘lines’ of the ikun∊) are continuous, there is a dislocation where other patrilineal threads (ikun∊) reach the knot, where lineages are attached, probably through ‘choosing a father,’ to the knots or clans.Footnote 72 Some of these attached ikun∊ are still flourishing and based in Amedzofe, indistinguishable as far as day-to-day events go, from ‘original’ ikun∊. Others were an intrinsic part of the village for a while and then, perhaps, their members left or produced no sons to continue the family name. These post-resettlement attached groups are the ikun∊ that have no localized residential base in the village proper: their residential base is entirely in the Christian area, nearer to the Mission buildings and outside the formal boundaries of the village.Footnote 73
There are currently at least six ikun∊ whose family houses are in the Christian part of the village rather than the village proper, and whose oku and clan connections are uncertain in this sense. But these people did stay in Amedzofe. Their families are there today, they and their children speak Avatime, they have land there and have built houses there, and there are members of all of these families, or their daughters' children, still living in the village. They are Avatime from Amedzofe. More ‘detached’ are the families of Mission employees, posted to the village over the years. The best example here is the family of Stephen (Stefano) Kwami, one of the Bremen Mission's earliest local converts. He was from the north, by origin, and was manumitted by the Mission in Ho, stayed with the missionaries, went to school and became a Christian. One of Kwami's sons, Robert, was educated initially in Togo, but sent to the seminary in Westheim (Germany) for further training in 1894, when he was 15. He returned to Togo in 1897 and was posted to Amedzofe from then until 1900 initially, and later from 1909 until his death in 1945. He was ordained during the First World War. Kwami was appointed Synod Clerk of the (then) Eυe PresbyterianFootnote 74 Church in 1923, a post that he held until his death.Footnote 75
When Robert Kwami was posted to Amedzofe, he ‘chose a father’ in the Tsikɔɪ clan, and several of his children were born, baptized and confirmed in Amedzofe.Footnote 76 Like other Christian incomers, he did not build a house in the Tsikɔɪ clan area in the village proper, but built instead in the Christian area, on the side of a hill, just down from the Mission House. The house still stands today and commands a fine view to the Volta Lake. Although some of the Kwami children were born and initially went to school in Amedzofe, and some married people from the village, there is no ‘Kwami oku’ in Amedzofe in the early twenty-first century. Few of the Kwamis spoke Avatime well, preferring instead to speak Eυe (the language of the Church) and, after various successes in educational and business fields, most chose, unlike other Amedzofe people of their generation, not to retire to the village. Their children may have links with the village, especially if their other parent were from Amedzofe, but, while there were older Kwamis in the village until the mid-1990s, today there are none: members of the older generation have died and the younger generations do not know the village as ‘home’, the place where they grew up. Now there is no one left there for them to visit. It is probably fair to say now that the Kwamis view themselves as residing in / coming from Ho, where Stephen Kwami originally settled, rather than Amedzofe.Footnote 77 But ‘detached’ as they have become, the Kwamis are among the most attached of the Mission workers who came to Amedzofe over the years. Others, even if they built a house in the village, are now dead and their children, having no kin connections in the village, do not visit, leaving their houses empty, or occupied by clan ‘relatives’.
Throughout the twentieth century, however, descendants of ex-slaves and refugees who did not return to Amedzofe or the other Avatime villages on emancipation have made contact with ‘home’. Close and more traditional webs of kinship in the Volta Region are often overlaid with complex networks of distant kin relationships and recognized connections and obligations deriving from the Asante invasion and its aftermath. An account from Amedzofe, written in the early 1930s, serves as an example:
Ayisawɔɪ had Nyadzi, Wofemenya and Alomenu at Atɔkɔ-Anlo. It happened at the base of Dodome mountain during the Ashanti war that the child fell from the mother's back and the Ashantis picked her and gave her to the Anlos. This family is at Agottɔɪ till today – Kodjo Adetsi and the others.
Kosi Elewosi [a Christian name for Ayisawɔɪ] was taken to Anlo, a town known as Agottɔɪ. It was there that she had Wofemenya. Kosi was telling the child that she wasn't from that town. It was the Ashantis who took her from her parents, she wasn't sold. Her father's name was Klu from Amedzofe. This thing remained in her children's mind when the mother died. So when Wofemenya also began to give birth he was telling the children …
… when he grew up, he tried and traced the family until he found them. They were insulting them there that they weren't from that town. Through that, if they gave birth the child did not stay alive. Fetish priests told them that their grandmother's god was troubling them.
So Wofemenya went to Amedzofe and traced his mother's family's descendants: ‘They stayed for 3 days and went back so that they might bring the others. Elewosi died a long time ago, only the children are still alive’.Footnote 78
The visit, which took place probably in the 1920s, was recorded by one of the earliest Christians in Amedzofe, a man who was Ayisawɔɪ's brother's son, S. A. But Ayisawɔɪ's brother, S. A.'s father, also had links with returning Avatime. From the same source:
L. A. [S. A.'s father] learned carpentry with the Mission whites. At that time Odɔkosi and the mother came back from Anlo. They became Christians immediately … Odɔkosi had the name Lisete. This Lisete worked for the missionaries. It was there that LA saw her and asked to marry her from (the mother) and the brother … The children that Lisete had with LA were … Lisete died on September 11th, 1912, L. A. died 1931Footnote 79
So, L. A. learned carpentry from the missionaries in the 1890s.Footnote 80 Odɔkosi's mother had been captured by the Asante and sold in Anlo after 1871. Some twenty years later, during the 1890s, she returned, as a free woman, with her daughter to Amedzofe, the daughter's name indicating the association with slavery.Footnote 81 We do not know Odɔkosi's father's origins, but her mother's and mother's brother's names are remembered in genealogies, and, in the absence of a father, Odɔkosi was affiliated to her mother's clan, whose members still recognize the affinal link between them and LA's clan. The male line from Odɔkosi's mother's brother has subsequently died out, but was remembered by clan elders from other ikun∊ in the same clan in 1973.
These networks of old relationships are not confined to Amedzofe among the Avatime villages. An Anlo-Eυe colleague has similar connections with Fume and, in 2006, her mother planned a visit to a ‘family’ funeral in Fume, recognizing the kin connections and ensuring a new layer in the warp and weft of relationships.
CONCLUSION
Minimally, this essay describes the construction, destruction and re-construction of a small ethnic group in Ghana's Volta Region and advances the historical knowledge of a small part of what is now Ghana. The essay spans the precolonial/colonial divide and suggests what influences Christianity and colonial rule came to have in the late nineteenth century. It shows how embedded the area had become in trade networks and how it experimented with new products and new networks. Although experimentation with Akan military forms and offices proved successful in the 1830s, the success could not be repeated in 1868–9, at which point the peoples of Eυe-Dome employed guerrilla tactics in their fight against the Asante.Footnote 82
Beyond the minimal, the essay looked at local structures that enabled social formations and ideologies to emerge in the construction, reconstruction and subsequent development of Avatime. The pre-nineteenth-century movements of people into this refuge area were complemented by the circulation of people – both voluntary and forced – through trading, and, within largely domestic networks, slavery and pawnage.Footnote 83 Avatime incorporated slavery, slaves and relationships of slavery (and pawnage) rather than using them in specific ways to maintain an identity separate from and apart from surrounding groups, as was seen in the earlier group of papers referred to at the start of this article. Avatime participated both in domestic slaving and in the positive and innovative trading and productive enterprises widespread in the area. While being Avatime in cultural senses, perhaps, they were already regional citizens in many respects.
The Avatime experience also begs some questions about identity and ethnicity, both concepts that have relevance in much wider contexts in the twenty-first century. Through the years I have been working with Avatime, I have never supposed that, apart from linguistic competence and a willingness to subscribe to local norms,Footnote 84 there is some pure essence of Avatime-ness claimed as Avatime identity or ethnicity. Avatime-ness, Avatime identity, consists of sites of cultural (including linguistic) density, but there are no hard-and-fast rules about in-groups and out-groups, sameness and difference. Identity and ethnicity here are above all malleable concepts.Footnote 85
Under colonial rule, chiefs and senior elders gained respect, authority and some access to cash, but the late-comers/new-comers, settling in Avatime and Amedzofe in the wake of the re-distribution of population after the Asante invasion, had no access to these forms of wealth. They tended to be those who saw in Christianity (and literacy), whatever the spiritual pull, the chance of a new life: new ways of living, education, a new era. Thus, even today, Amedzofe, although decidedly remote in practical terms, has a reputation, at least in the Volta Region, as a place of learning and pilgrimage. Some of its ‘newer’ citizens, both men and women, have taken advantage of the opportunities available to them as educated and Christian, and become prominent in the development of both the Volta Region and elsewhere. Others have practised and passed on the artisanal skills acquired by their grandparents. Even today in Amedzofe, although the overwhelming majority of the population is literate, the ‘new big man status’ acquired by early educated Christians is not forgotten and family respect lingers.
As the content, meaning and significance of more traditional statuses wax or wane, so the relationships among the constituent knots and threads of social organization alter. People leave and may return: new beginnings happen and innovations arrive as the villages still provide a focal point as ‘home’ for some migrants. Village development committees, under whatever name or government, work in conjunction with whichever people or organizations have money available for projects and plans for change and development, for keeping up and being a part of the 21st-century world(s). Chiefs may have what amount to local government roles, in addition to their village-based, chiefship-derived roles.Footnote 86 Regular transport, electricity and mobile phones have arrived, bringing with them chances of incorporating and practising new technologies, new forms of relationships and visions of elsewhere, films and football. Contemporary Avatime are found all over Ghana, and all over the world, making new connections and maintaining older-established ones. Their actual and individual relationships with Amedzofe, and the other Avatime villages, have changed, fluctuating with the local – wherever ‘their’ local is – real-politics and real-economics, and the twentieth and twenty-first centuries' versions of global–local relations. But the structures of incorporation, deriving from before the nineteenth century, and so important during the last quarter of the nineteenth century in enabling Avatime access to a basic livelihood, support and freedom within their kin nexus, remain and are entirely relevant in the twenty-first century.
