[280]

Michel Verdon

Anthropologue, département d’anthropologie, Université de Montréal

“Re-defining pre-colonial ewe polities :
the case of Abutia.”

In revue Africa, vol. 50, no 3, 1980, pp 280-292.

Introduction [280]

A case of ethnographic vagueness [280]

The problem of political sovereignty [281]

The abutia body politic [282]

Criminal or martial law ? [284]

Sovereignty in alliance [285]

Village sovereignty and some of its implications [287]

Village sovereignty and the type of chiefly authority [289]

Conclusion [290]

References [291]

Résumé [292

Introduction

The Abutia Ewe formed one of the colonial administrative 'Divisions' of Eweland under a so-called 'traditional ruler', the Divisional or Paramount Chief (fiagã). These Ewe Divisions are located in the southern half of the Volta Region (Ghana) and in southern Togo. Every Division is composed of a number of villages, each headed by a 'Sub-divisional Chief (dufia). Only three villages are found in Abutia : Teti, Agove and Kloe. The fiagã or Divisional Chief comes from Teti.

In a previous article, I investigated the manner in which the Abutia 'fia-ship' (inadequately translated as 'chiefship') was articulated to the other political 'offices', namely those of stool-father (zikpi-t), heir-apparent (tsfo), linguist (tsiame) and 'public attorney' (mankrado), and I concluded that the Abutia fia displayed more of the features of a 'fetish-priest' like the Ga maηtse (Field 1940) than those of a Divine King on the Yoruba or Anlo model (Bascom 1969, Nukunya 1969). Such a research, however, only unveiled one face of the problem ; it ignored the relationship of the fia to the citizenry as a whole, thereby neglecting a proper understanding of government (Verdon 1979).

In this paper, I endeavour to broach this second line of enquiry. Knowing that the fia was the priest of a stool (zikpi—the Abutia stools were the abode of a special type of deity known as tr) certainly reveals precious little about the general distribution of power and fails to disclose anything about the type of polity in which the fia operated. To unravel the tangled question of the precolonial Abutia polity will be the task of this paper, and it will be shown to depend largely upon the answer to one question : what was the pre-colonial sovereign political entity ?

A CASE OF ETHNOGRAPHIC VAGUENESS

The problem of circumscribing political sovereignty is a difficult one ; that of defining a polity seems more formidable. There is indeed a striking lack of consensus among the students of Ewe social organization in the matter of identifying the pre-colonial type of Ewe government. The early German missionaries wrote of the Ewe Stãmme (tribes) when discussing the component sovereign groups, a usage taken over by Rattray (Spieth 1906, 1911, Westermann 1935, Rattray 1915). Subsequent English ethnographers have used 'sub-tribe' to refer to the same entities, implicitly assuming that the Ewe people as a whole formed one large 'tribe' (Manoukian 1952, Ward 1949). British administrators elected to use the more neutral 'Division' or 'traditional area'. Nukunya, who hails from Anlo (a coastal area in the south), uses alternatively 'kingdom', 'chiefdom' and 'tribe' ('Nukunya 1969). Friediander calls them 'states' (Staat) and Asamoa, who was an inland Ewe, compares the groups of villages to the Akan oman, a political group which shares many of the features of a 'chiefdom' according to Service's model (Friedlãnder 1962, Asamoa 1972, Service 1962).

[281]

Beneath this baffling diversity of labels, there is nonetheless a general agreement in treating the 'group of villages acknowledging a common fiagã, or otherwise titled Paramount Chief, as the precolonial sovereign group, a usage deeply rooted in administrative thinking since Rattray's report on the 'tribal history' of the inland Ewe (1915). However, the whole debate surrounding the report at the time it was written has unfortunately been ignored. After ousting their Akwamu oppressors in 1833 the Abutias, together with many other northern groups, acknowledged the 'overlordship' (the term is Rattray's) of the Kpekis who led them during their rebellion. In view of these historical events, the Kpeki chief claimed the role of the traditional ruler of all the inland divisions, and demanded that the British colonial officers treat him accordingly. The nature of this overlordship was very nebulous, but the claim was serious enough to raise a difficult question : which level of grouping were the administrators going to identify as the traditionally sovereign one—the Division (group of villages), or the group of divisions united under Kpeki ? Rattray mustered sufficient evidence to demonstrate that Kpeki never 'ruled' over northern Eweland, but was simply accorded precedence in a military alliance.

It was nonetheless an arbitrary step to reduce the issue to this single alternative. After all, the Germans were the first invaders and their emissaries singled out the villages as the main administrative units, directly responsible to the District (Kreis) Governor. The German colonial officers treated the village chiefs as the traditional rulers. Was this another instance of administrative expediency, or did the Germans know something which escaped their British successors ? I am inclined to believe that their first-hand contact with the pre-colonial groups had taught them one thing—namely, that the villages were the sovereign political groups in the 1880s. Although this hypothesis does depart substantially from the accepted model of Ewe pre-colonial political organization, I see it however as a more seminal notion for a proper understanding of northern Ewe, and more especially Abutia government. I will consecrate the remainder of this paper to demonstrating that this thesis is more plausible that the current administrative and ethnographic model of small chiefdoms under a Paramount ruler.

THE PROBLEM OF POLITICAL SOVEREIGNTY

In order to get a clearer understanding of the Abutia pre-colonial government, I have decided to approach the question of defining a polity from the angle of sovereignty. By 'sovereignty', however, I mean something radically different from the political philosophers' and many ethnographers' notion. In another publication (Verdon 1980) I have distinguished analytically between simple or elementary groups, on the one hand, and 'aggregated groups' on the other. One gains membership of elementary groups only. These elementary groups are themselves aggregated into more inclusive units which, analytically speaking, are 'groups of groups'. Individuals may gain membership of a municipality, but not of a county ; the county is an aggregation or a merging of municipalities. Similarly, individuals may become members of a minimal lineage, but minimal lineages are aggregated into lineages. On the whole, elementary groups are aggregated on the basis of two dominant facts—(a) geographical contiguity (which gives rise to territorial groups) and (b) descent (which gives rise to descent groups). For both types of groups, however, there is a level at which the aggregation stops, beyond which [282] there is no further aggregation. I define this level as the 'sovereign political group'. Lineages, villages, or whole states may thus be sovereign ; this definition of sovereignty completely obviates the state/stateless dichotomy and provides a new basis for comparative analysis. The 'highest level of aggregation' may be a lineage ; it is completely independent from the existence of a machinery of government.

There are however other problems attendant upon a definition of sovereignty. In determining the 'largest extension of the body political' (i.e. in finding the highest level of aggregation, or the sovereign group) some 'political' activities may have to be set aside. For instance, many national states may recognize the religious authority of the Pope while retaining their political sovereignty. Sovereign nations may also associate in a military alliance (as in the case of NATO) without surrendering their autonomy. One could also find comparable associations for economic matters (the EEC being the most obvious illustration) between sovereign political groups. Briefly, groups united for military, trade and religious purposes can remain politically sovereign ; they are only aggregated with respect to these specific activities, but do not relinquish in any way their sovereignty in these coalitions.

From the point of view of individual action, no one would deny that the manipulations and machinations of individuals in the Church, the EEC or NATO are 'political' (i.e. involving the appropriation of power). From the point of view of the delineation of sovereign political groups, however, the inclusion of groups allied in military, trade and religious matters may be greatly misleading. The sovereign political group may coincide with the highest level of aggregation of the military and/or commercial and/or religious groups ; on the other hand, they may not coincide and the distinction must be analytically made. In this way, one may gain a better understanding of what happens when some of these groups coincide. I will thus restrict the notion of 'politics' to the groups formed in the activities of legislation, adjudication, administration (including executive activities), as well as in the activities related to the accession to office of individuals involved in these groups.

THE ABUTIA BODY POLITIC

If the above-mentioned hypothesis is more plausible and villages were indeed politically sovereign in Abutia in the period 1870-1890 (this is the period referred to when I use the term 'pre-colonial'—for an explanation, see Verdon 1979), it is then legitimate to expect (1) that in matters calling for legislative, administrative and judiciary action (excluding military, commercial and religious concerns), no group will be formed above the village level and (2) that the membership of these groups will be restricted to villagers.

1. In precolonial Abutia, insofar as it can be assessed from contemporary reports, civil cases were arbitrated by different courts depending upon the group of origin of the disputants. Litigants from the same fome would take their case to the fometsitsi ; if prosecutor and defendant hailed from different fomewo (sing, fome) of the same agbanu, the case was submitted to the agbanumetsitsi. A person wronged by an individual of a different agbanu but of the same sãme presented his case to the sãmefia. When a quarrel involved individuals from different sãmewo (sing, sãme) of the same du, the dufia was called in.

For the sake of convenience (a demonstration would require another paper), I will [283] describe these groups as minimal lineages (fome), lineages (agbanu), clans (sãme) and village (du) ; in these judiciary instances, they act as 'groups of reference'. To say for instance that a matter involves the clan does not mean that the whole clan, women and children included, will take part in the arbitration. It only implies that the case is taken to the authorities of such groups, which have jurisdiction in such matters ; in this instance, the clan is only a 'group of reference'.

Lineages, clans and village, moreover, are also 'aggregated groups'. In the adjudication of civil offences, however, villages are not aggregated into a group of higher degree of inclusion ; in other words, disputes never involved a group of reference above the village (du). When quarrels broke out between individuals of different villages, the cause was not presented to the fiagã (Paramount Chief). Only the parties directly involved in the conflict, together with some of the village elders on both sides, would convene to adjudge the case. Confrontation was unnecessary ; many of the village elders were kinsmen and there was always room for compromise. When discord arose between individuals of friendly neighbouring areas, it would receive a similar treatment. Between hostile areas, no settlement was possible. By definition, conflict between foes could only be resolved in homicide or warfare.

In legislative and administrative matters, furthermore, there was no group of reference either above or below the village level. Only village authorities could take action on matters of communal concern.

2. For each and every one of these groups of reference (minimal lineage, lineage, clan and village), there was a person in a position of authority who acted as the group's representative. These individuals were known as ametsitsi (i.e. 'elder' or 'head') in the case of minimal lineages, lineages and some clans, and fia (i.e. 'chief) in the case of other clans and the village.

In judicial activities the authority of this group representative, either 'head' or 'chief, was limited to the summoning of a group which was empowered to adjudicate. The criteria of eligibility to headship were : physical and genealogical seniority, male gender and membership of a minimal lineage through one's father. In the case of '/fa-ship', recruitment proceeded on a radically different basis : chiefs had to be young, perfect in body and with a character appropriate to the task (i.e. humble and respectful of their fellow-citizens). In the case of fiawo (sing, fia), it was acceptable that their membership of a minimal lineage be traced through the mother or the father's mother.

Group 'heads' acceeded to their position automatically and without any ceremonial ; chiefs, on the other hand, were selected and 'enstooled' ceremonially. Their powers, however, were identical ; both heads and chiefs were only entitled to call a judiciary council and to preside over its deliberations. All these judiciary councils shared the same criteria of membership : they recruited elders from the group of reference, together with cognates of the litigants whose wisdom or knowledge was relevant for the arbitration of the case. Apart from a core of elders who presumably attended most of the reunions of such councils, their composition varied greatly from one occasion to the next.

The actual adjudication was not marked by any specific division of labour, save for the 'presidential' role played by the lineage or clan head. Deliberations were not ordered, despite the chairman's attempt to channel the discussion. Those who had something to say expressed their opinion freely, and naturally the most skillful orators exerted the greatest influence. The elders only passed their judgment when they had [284] reached a consensus about the verdict, and the privilege of pronouncing the sentence was reserved to the group's representative (as head or chief). Sentences were traditional, and a head who departed from traditional sentences would have aroused a flurry of further deliberations which might have become a case in themselves.

In legislative and administrative matters, the village chiefs authority was subject to the same limitations. He was only empowered to summon the villages' 'fathers' (du-two), otherwise known as the Council of Elders in the ethnographic literature. Eldership, membership of one of the villages' minimal lineages and residence in the village were the necessary criteria for membership of this group. Internally, the 'linguist' (tsiame) acted as the chiefs spokesman during the deliberations, and the mankrado played the role of public attorney. None of these title-holders had otherwise any prerogative. As with adjudication, decisions taken during legislative and administrative assemblies had to follow consensus. In all these judiciary, legislative and administrative groups, moreover, membership was reserved to villagers ; citizens of other villages were only called upon as witnesses if their testimony was relevant, but not as jury and judges.

As would be expected of sovereign villages, precolonial Abutia villages were not aggregated into any group of a higher degree of inclusion with respect to judiciary, legislative and administrative matters, and membership of the corresponding groups was restricted to villagers.

CRIMINAL OR MARTIAL LAW ?

Two major facts, however, seem to negate our starting hypothesis : (1) the fact that two offences—witchcraft and homicide—did indeed call for the presence of elders from all three villages, together with the fiagã or one of his representatives, and (2) the fact that other cases—accusations without evidence—could also be taken directly to the High Priest of the great Abutia god, Togbe Atando, by 'swearing his oath'.

1. In contrast to civil cases, witchcraft and homicide could be described as the only two criminal offences in precolonial Abutia. The penalty incurred—that of ostracism—matched their criminal status. The seriousness of the exile, however, varied according to the gravity of the offence.

I have not been able to record any special concept to express the notion of homicide. Only one elder in the whole of Abutia could recall a voluntary killing, by natural means, of one Abutia citizen by another ; in this case, the man murdered his wife's lover. All other occurrences of individuals killed by human agents, using natural weapons, were in fact treated as manslaughter. Such mishaps occurred only during communal animal hunts and were automatically construed as accidents ; the slayer's motivation does not seem to have been questioned.

In witchcraft, the means may be supernatural but the motives are clear. Deaths resulting from witchcraft are not accidental, and the penalty for such actions was found to be the most severe. The witch was completely ostracized from Abutia lands. To compound this fate, elders usually add in a confidential tone that these malefactors actually met their death in the hands of the asafo (the traditional army), in the no man's land that separates neighbouring divisions. They only escaped public execution in the village because Togbe Atando prohibited the spilling of Abutia blood on Abutia soil.

A man accused of manslaughter (as women never took part in communal hunting, [285] they were never guilty of manslaughter) would be ostracized in a much milder way. He, together with the members of his minimal lineage, would have been compelled to leave the village and settle a few hundred yards away, in the 'bush'. The ban was only temporary and they would be called back to the village after a few weeks or months (I was unable to find consensus about the period of time, which varied from eight days to three months, according to the elders). His recall depended on a special ritual which was performed to close the case and seal the memory. The ritual consisted in burying a pot with its mouth towards the earth (zedzedidi), and any mention of the event after this ceremony was ipso facto litigious.

Interestingly enough, the symbolism which surrounds these two offences underlines their exceptional nature and treats them as instances of 'internal war'. The victims of both witchcraft and manslaughter were considered to have died 'accidental deaths', which were ritually treated as 'deaths in war' and were the province of the asafo. Wars are external threats which have to be 'pushed outside' and ostracism, which had to be implemented by the asafo, was the logical outcome of this internal war.

This symbolic equation suggests that there was no real notion of a criminal offence in pre-colonial Abutia. All cases which could be handled by groups up to the village level and only involved payments were regarded as civil or religious-moral offences. Criminal actions (which entailed ostracism), either voluntary or accidental, were actually construed as military offences. The council responsible for their arbitration, presided over by the fiagã, could therefore be interpreted as a martial court applying martial law. Furthermore, the case was not adjudicated in the fiagã's village ; the Paramount Chief or his representative had to go to the village where the offence was perpetrated to preside over the arbitration.

2. Where individuals were suspected of wrongdoing without any tangible evidence, the defendant could 'swear the god's oath' and thereby withdraw the case from lay courts. The procedure has parallels in medieval custom where persons who took refuge in a church escaped civil prosecution. In Abutia, this invocation was a way of calling on divine intervention where human justice was impossible because of the lack of

evidence. It was literally the equivalent of calling God as witness to one's innocence. Cases which were thus taken to the High Priest's court were dealt with by religious law in a religious court.

The three Abutia villages were thus united by a common martial and religious law, applied in martial and religious courts presided over by a representative of the whole division. However, I would still contend that, in the particular case of Abutia, these military and religious alliances did not jeopardize the sovereignty of the villages.

SOVEREIGNTY IN ALLIANCE

The main divinities of the Abutia pantheon are known as trwo (song. tr), of which there are three main types : (a) immigrant deities which have their abode in the village, select women as their earthly representatives and inhabit objects which 'hang' in the air and cannot touch the ground : (b) autochthonous divinities which dwell in the bush, select men as their earthly representatives and inhabit trees or caves (they are thus rooted in the ground ; they are 'chthonic') and (c) stools which are 'war-medicines' purchased collectively and transformed into benevolent deities ; their priests, are the fiawo (sing. fia).

[286]

Togbe Atando, on the other hand, is one of a kind ; he does not fit any of the above categories but straddles them all. He stands at the apex of the pantheon of trwo and his power far exceeds that of any other god. His prohibition are respected by all three villages. His priest, was ipso facto the High Priest of Abutia, but he acted completely independently from the fiagã. There existed no connection whatsoever between Togbe Atando and the stools, and the spiritual power of the High Priest far surpassed that of the fiagã.

The Paramount Chief resided in Teti, the High Priest in Agove, and no collaboration was needed in the execution of their functions. The two positions were not integrated and neither of the two commanded the other's subordination. The Paramount Chief was not High Priest, and lacked any authority over the latter, and this separation of the two roles would have been sufficient to hold in check any potential centralizing bid on the part of Teti.

The military coalition also left ample scope for village sovereignty. The Commander-in-chief also lived in Teti but, like the Agove High Priest, he was no subaltern of the Paramount Chief. He would never obey the fiagã's personal orders.

In warfare the Abutias formed one army, under the command of the Teti asafofia, though the three villages acted as three separate battalions, each under the more immediate leadership of its own army chief. Teti's army occupied the centre part of the military formation, Agove the right wing and Kloe the left. Every battalion seems to have operated autonomously and it is doubtful whether the asafofia even had the authority to coordinate the army's activities and movements. The asafo was an army of citizens, which nonetheless lacked the democratic phalange formation of the Greek hoplites. The ownership of a gun automatically entitled a man to membership of his village asafo, and all adult males were in possession of such a weapon. The village elders, however, did not take part in military activities.

The asafo's internal structure appears to have been rudimentary. The army's democratic recruitment also left room for individual exploits. The asafofia (army-leader) distinguished himself from the mass of citizen-soldiers by his possession of more powerful war-medicines and his ritual knowledge. In war, as in ritual performances (as, for example, when burying people who died accidentally), the asafo was thus led by a man who had been elected to his superior position because of his personal qualities and powers. The position of army-leaders had therefore to be proved on every military occasion and an army-chief could be dismissed from his post if the soldiers were dissatisfied with his performance. In actual combat, it seems that every man more or less fought for himself, although people from the same village fought close to one another.

The asafofia ultimately derived his authority from the Council of Elders, of which he was a member. He was not empowered to initiate any military action on his own, but would only act on the elders' recommendation. Since Paramount Chief and village chiefs wielded no special authority within the village councils, they lacked any control over the army-leaders. The asafofia was in no way subordinate to the fiagã, and only acted on the advice of the village leaders. Wars, like trials or legislative processes, were matters of consensus, and in these no more than other activities could an individual leader take the initiative.

Finally, the three villages were also united economically through their involvement in a market system which rotated between neighbouring areas. The Abutia market [287] place was then located in Kloe, the elders say. This system of multiple military, religious and commerical associations, with their foci in different villages (the fiagã in Teti, the High Priest in Agove and the market in Kloe) succeeded in disconnecting the different sources of power and operated in favour of village sovereignty in a loose confederacy. It gave the village protection against slave-raiders but, at the same time, allowed them to retain their sovereignty.

This hypothesis of village sovereignty thus seems to account for some features of the political system, but the unreliability of some of the contemporary statements about precolonial institutions makes it necessary to find more corroborating evidence. The more ethnographic facts one can derive from it, the more plausible it will be. In dealing with precolonial institutions on the basis of oral traditions, one can simply hope for a high degree of plausibility ; there is no pre-packaged truth which only awaits to be collected. Our starting hypothesis will therefore appear as a more likely representation of reality if many of the observable features of traditional Abutia can be derived from it. This is what is attempted in the following section.

VILLAGE SOVEREIGNTY
AND SOME OF ITS IMPLICATIONS

1. If membership of a minimal lineage is mostly gained through patrifiliation, and matrifiliation in specific circumstances (the situation which prevailed, and still does, in Abutia), and granted also that villages are sovereign, it will follow that a person is only a citizen in his or her father's locality, and that of his or her mother only in special situations. Outside these localities, people are 'strangers' (amedzro), people without any citizenship, without anybody to stand for them in times of trouble.

This corollary applies well enough to pre-colonial Abutia, although the matter is complicated by the political inequality of the sexes. Women were indeed excluded from membership of legislative, administrative, military and most judicial groups, whereas all men could hope to join these groups if they lived long enough. Women's political 'status' (defined in terms of the membership of groups to which they are entitled) thus rated lower than men. This fact made it easier for them (not psychologically, but politically) to follow their husbands when marriages took place across villages.

Including these instances of village out-marriage, a male child's domicile and citizenship could be determined in one of four ways only : (a) in cases of village in-marriage, a child could not elect domicile outside his native village ; he was a citizen of that village only, (b) Where the parents came from different villages but the children had been brought up in their father's village, they (the male children, let us remember) would be welcome guests in their mother's village but would nevertheless derive their citizenship from their paternal village, (c) In cases of village out-marriage where the children were brought up in their mother's village, they enjoyed dual citizenship. The father could certainly 'claim' his children back as they reached adolescence, but the latter were welcome to remain in their mother's village if they so chose, (d) Children without an acknowledged genitor would automatically belong to their mother's father's minimal lineage and would only be citizens in their mother's village.

The situation was identical for female children, except in case (b). If brought up in their father's village, women might in fact be married back into their mother's ; a woman if reared in her mother's village would have little incentive to move back to her [288] father's when he claimed her. Women also refused to be married in villages where they had no matrilateral kin ; for, as complete strangers, they could only have counted on their husband's support in case of disputes, and this uncomfortable situation made such marriages almost impossible. To force a girl into a union in a completely alien village would have been tantamount to selling her into slavery.

Not only the past, but also the present pattern of village domicile tallies with the notion of a pre-colonial village sovereignty. Teti indeed never acted as a capital to which individuals would flock in the hope of sharing in the windfalls of power. Whatever mobility between the villages there was and is (when it comes to settlement and domicile) only results from village out-marriage ; apart from this connubial exhange, extremely few individuals have settled outside their father's or mother's village.

2. The above corollary also influenced the citizens' relationship to their dufia (village chief). Indeed, a man's domicile in a village was an intrinsic corollary of his birth into one of the village's minimal lineages. Although the village's various localized clans were aggregated into a du on the basis of territoriality, the dufia cannot be construed as a 'territorial ruler' ; by being domiciled in a village, a man did not express subjection to the head of an administrative unit, nor was it meant to be a personal bond to the chief or headman of the village, as one often finds in some African chiefdoms (Barnes, Gluckmann and Mitchell 1949, Colson 1951a, 1951b, Ruel 1969, Brain 1972). Nor did a man's settlement in a locality express kinship preferences, as obtained in Nuer localities (Evans-Pritchard 1951). In the great majority of cases, a man's father's village was the only place in the world whence he could derive his citizenship,
so that his domicile did not spring from any type of personal commitment.

3. If the village was the political group of largest extension, the village chief was therefore only a citizen in his own village, and his chiefly functions were confined within the boundaries of his locality.

This situation also prevailed in precolonial Abutia. No dufia had any jurisdiction outside his own village, and none could vie with chiefs of other villages for political allegiance. Any competition between chiefs of different villages was thereby precluded. Every office or position of authority was closed in on itself, and did not lead to a higher one. The Abutia polity thus precluded any kind of mobility between political offices. A clan chief was indeed confined forever to clan chiefship ; he could never aspire to become village chief, as only the 'chiefly clan' gave the village its dufia. No village chief could ever nurse any hopes of being enstooled as fiaga ; the title simply belonged to the Teti dufia. The offices were all 'terminal' ; there was no upward mobility from office to office through the interplay of political allegiances and manipulations. The size of any political group, therefore, depended entirely upon natural increase ; movements of population between villages were practically ruled out. During the time of fieldwork, the resident population of the three villages was approximately the same (over 1,000 each) and, as far back as the censuses go, their relative sizes has been approximately equal. To put it briefly, the organization of titled offices in precolonial Abutia seemed eminently static. It ought to be remembered, however, that this system of offices was superimposed on a prior polity in which there were no 'chiefs' (it is unfortunately impossible, in the present state of my knowledge, to state with some assurance what this earlier polity was), and that it only operated without drastic outside interference for twenty years, from 1870 to 1890. This external organization (it was borrowed from the Akan neighbours) imposed on the earlier political organization [289] could afford to be so static because very little power was .vested in it. It is within the different judiciary, legislative and administrative groups that the real struggle for power took place, as individuals tried to assert their influence by attempting to sway the general opinion in their favour.

VILLAGE SOVEREIGNTY
AND THE TYPE OF CHIEFLY AUTHORITY

These severe limitations of movement also influenced the type of political authority wielded by the fiawo. As he was a stranger outside his own village, the Paramount Chief was forbidden to interfere in the politics of other villages. When his presence was required he would actually come in person to the village where the case was tried, instead of convening a council in his own village. These and all the other facts suggest that Abutia can hardly be described as a 'chiefdom' ; the very label 'chief to translate fia is thus highly questionable in the case of Abutia and possibly all Ewe-dom divisions.

In most polities which have been dubbed 'chiefdom', individuals are not generally bound to one locality. Their freedom of movement jeopardizes the local chiefs' authority, but it reinforces that of the Paramount Chief, whose capital usually boasts of a larger population. Such inter-village mobility is not a feature of Abutia. The villagers' confinement to their native village, on the other hand, has not favoured either local despotism.

Powerful chiefs stand at the apex of what Service has labelled 'redistributional economies' (Service 1962 : 44). Such economies operate over a territory which is ex definitione larger than one single locality, and which is unified through this particular mechanism. Abutia clans, however, were not united through accumulation and redistribution, since they all equally shared in government. Neither the fiagã nor any of the dufiawo was ever in a position to attract anybody to his own village, nor were they authorized to eject citizens. Only witches could be ostracized, and they only on the authority of the Council of Elders. The chief could not, a fortiori, emerge at the apex of a redistributional economy. This would explain why no tribute of any kind was ever paid to the Paramount Chief or any of the village chiefs. None of them could accumulate wealth by virtue of their position. As a corollary, the fiagã did not have any obligation of hospitality towards his people, nor did he have any trading or economic privileges, in contrast to the Anlo Awoamefia who could exact corvee labour and was also expected to provide hospitality (Nukunya 1969 : 11).

Incapable of exacting tribute, the Abutia fiagã or dufiawo thus lacked the means of entertaining a court or a retinue. They could never build a 'palace' nor afford any special retainers, and their houses and households were indistinguishable from those of others, both in architecture and in size. They had no special right to more wives or to special brides. Like any of their fellow-citizens, they simply married whom they could afford to marry, whoever was willing to marry them. The fiagã or dufiawo did not own slaves by virtue of their position, and no special status was attached to their lineage or clan ; no special aristocratic privileges or prestige was ever a result of being the chiefs kin. The chiefly clans never constituted the embryo of a rank society, as is reported in the case of Polynesian chiefdoms.

The village was also the highest level of grouping (or aggregation) where offices were defined with respect to someone else's stool. The chiefs of Kloe and Agove never defined their respective positions with respect to the Teti stool. Only office-holders [290] within the Paramount Chiefs own village did so (his own tsiame and mankrado). The village chiefs did not behave as subordinates or functionaries of the Paramount ; nor did in fact the fiagã's own tsiame (linguist) and mankrado ('village attorney'). The latter two office-holders defined their titles with respect to the dufia’s stool but still behaved as independent figures in their own right. Within their own village, the chiefs were even excluded from the affairs of lower councils unless they were directly involved or invited. Chiefs were also liable to prosecution in the same way as all other citizens.

The flag2 was neither High-Priest nor Commander-in-chief, neither wealthy nor despotic ; he was simply primus inter pares among the citizens of the three villages, what one might term the 'First Citizen' of Abutia. His claim to precedence springs from the assertion that his clan's apical ancestor led the Abutias during their migration to their present territory. This ascendancy did not place him at the top of a hierarchy ; neither fiagã nor dufiawo had any subalterns. Other office-holders were meant to instruct the chiefs rather than to be instructed by them. These First Citizens were selected and elected by their own clansmen, and also liable to be 'destooled' if their behaviour became arrogant and disrespectful. Rather than 'chiefs', the fiagã and dufiawo would be more appropriately described as 'ritual custodians'.

CONCLUSION

Labels such as 'tribe', 'chiefdom', 'state' or 'kingdom' would give an inaccurate description of precolonial Abutia. 'Townships' as they existed among the Tswana or Tonu Ewe suggest a division by ward and a degree of centralization unknown in Abutia. The Igbo 'village-groups' would provide a more plausible model, were it not for the fact that their villages are composed of only one clan, and are organized on the basis of age sets (Uchendu 1969). Fifth-century Greece suggests an interesting parallel. Its cities were sovereign but many of them allied for defence against the Persian threat. The association here became known as the Delian League. On this model, one could represent the alliance of autonomous and sovereign Abutia villages as a 'Village League', some kind of village confederacy formed for defence against slave-raiding by the more powerful neighbours to the south and west.

In the Delian League Athens gained military command, and with it came the expectation of taxes levied for military purposes, and the religious supremacy of Athenian deities. As Athens became the focus of this military, economic and religious association, it also employed funds and its newly-acquired power to promote its own interests. The League thus served as a stepping-stone for the Athenians in the creation of an Athenian empire.

These historical developments may also shed light on some of the differences between southern (especially Anlo) and northern village leagues. Religious and military command were united in the same town (Anloga), presumably because a slave-trading post was erected there. This gave Anloga a privileged position in the slave-trade and, as a result, it emerged as the capital of a 'kingdom' of formerly sovereign and autonomous villages. As far as I know, it does not seem to have achieved this through military conquest ; whichever way the association was implemented it was promoted by Anlo's dispersed clans. Northern Ewe localized clans could not be manipulated to bring together a large number of villages and, in this respect, Anlo's dispersed clans may have promoted the development of a centralized polity.

[291]

Because of its direct association with the European slave-traders, its large size and its greater degree of centralization, Anlo developed a more aggressive political style. Like the Ashanti, its slave-raiding operations extended to the northern Ewe groups. Northern villages also leagued themselves, but for purely defensive purposes. Their polity was rather a contracting one, and the decentralization of military, religious and economic power thwarted the growth of a capital which would have used the village league to foster its more limited interests. As a result, northern villages seem to have retained their sovereignty.

Until 1833, however, Abutia was directly ruled by Akwamu. Until 1870, they had no institution of fai-ship. The period of village sovereignty in alliance probably dates from the 1830s, but the association in the framework of a system of titled offices is much more recent. The 'precolonial polity' discussed in this paper thus represents a short lapse of twenty years before the German take-over (1890 approximately), but it represents the substratum upon which present-day definitions and manipulations of 'traditional' institutions rest.

The pervasive ethnographic representation of small Ewe 'chiefdoms' is therefore highly questionable in the case of Abutia and other neighbouring groups. The evidence gathered in Abutia seems to confirm the hypothesis of village sovereignty within the framework of a village confederacy or league. From my knowledge of other northern groups, I would suspect these findings to be of general application to much of the central part of Eweland, excluding the groups of Anlo origin settled in the northern part and those groups bordering the Volta such as Kpandu and Kpeki. The 'chiefdom' or 'kingdom' model distorts the northern Ewe reality ; it would only seem to apply to Anlo (coastal) or Tonu (riverain) groups.

Furthermore, the Abutia case underlines the dubious validity of the distinction between states and stateless societies. The Abutia adopted a system of titled offices, in imitation of either Akwamu or Ashanti, presumably without changing much to the previous distribution of power. Had the conditions changed, however—especially their proximity to an important trade route, or their access to European forts—they might have developed a notably different, more 'centralized' polity, as did their southern brethren (and as did other polities ; see Horton 1971). The manner in which I have defined descent (Verdon 1980) and political sovereignty represents a step in a new direction ; by delineating the group of highest level of aggregation in the framework of judiciary, legislation and administrative activities, instead of identifying a 'political community' and its political hierarchy, I have been able to derive a number of properties which indirectly circumscribe the extent of office-holders' authority and thereby provides a new way of defining polities outside the stifling and misleading dichotomies between states and stateless political systems.

REFERENCES

Asamoa, Ansa 1971 Die Gesellschaftlichen Verhältnisse der Ewe-Bevölkerung in Südost-Ghana. Berlin : Akademie Verlag.

Barnes, J. A., Gluckman, M. and J. C. Mitchell 1949 'The village headman in British Central Africa,' Africa, 19 : 89-106.

Bascom, W. 1969 The Yoruba of Southwestern Nigeria. New York : Holt, Rinehart and Winston.

Brain, Robert 1972 Bangwa Kinship and Marriage. Cambridge : Cambridge University Press.

Colson, Elizabeth 1951a, 'The Plateau Tonga of Northern Rhodesia,' in Colson, E. and M.

[292]

Gluckman (eds.), Seven Tribes of British Central Africa. Manchester : Manchester University Press.

_____, 1951b 'Residence and village stability among the Plateau Tonga,' Human Problems in British Central Africa. (Rhodes-Livingstone Journal), 12 : 41-67.

Evans-Pritchard, E. E. 1951 Kinship and Marriage among the Nuer. London : Oxford University Press.

Field, M. J. 1940 Social Organization of the Ga people. London and Accra : Crown Agents for the Colonies.

Friedlãnder, Marianne 1962 Zur Frage der Klassenverhaltnisse der Ewe unter dent Einfluss der Kolonisation. Ph.D. thesis, Berlin University.

Horton, Robin 1971 'Stateless societies in the history of West Africa,' in Ajayi, J. F. A. and Michael Crowder (eds.), History of West Africa, vol. I. London : Longman, pp.78-120,

Manoukian, Madeline 1952 The Ewe-speaking People of Togoland and the Gold Coast. London : International African Institute (Ethnographic Survey of Africa, part VI).

Nukunya, G. K. 1969 Kinship and Marriage among theAnlo Ewe. London : Athlone Press.

Rattray, R. S. 1915 Togoland : a History of the Tribal Divisions of the District of Misahoe and of the sub-districts of Ho and Kpandu. Unpublished report.

Service, Elman R. 1962 Primitive Social Organization : an Evolutionary Perspective. New York : Random House.

Spieth, Jakob 1906 Die Ewe-Stämme. Berlin : Dietrich Reimer.

_____, 1911 Die Religion der Eweer in Sud-Togo. Leipzig : Dieterisch's Verlag.

Verdon, Michel 1979 'The structure of titled offices among the Abutia Ewe,' Africa 49 : 159-171.

_____, 1980 'Descent : an operational view,' Man, 15 : 129-150.

Ward, Barbara E. 1949 The Social Organization of the Ewe-speaking people. M.A. thesis, University of London.

Westermann, Dietrich H. 1935 Die Glidyi-Ewe in Togo. Berlin : de Gruyter, XV.

Résumé

Une redefinition des systèmes politiques pré-coloniaux Ewe :
le cas Abutia

Les auteurs qui se sont intéressés à l'organisation politique Ewe n'ont pas reussi à presenter un tableau cohérent des systèmes politiques de ces populations, en ce sens qu'ils ont, sans discrimination, qualifié les groupes politiques souverains Ewe de sous-tribus, tribus, chefferies, royaumes ou états. Tous s'accordent cependant pour assurer que cette entité politique souveraine pré-coloniale était composée d'un 'groupe de villages reconnaissant l'autorité d'un chef suprême'. L'auteur conteste ce postulat, tout au moins en ce qui concerne un groupe de l’intérieur du pays, les Ewe Abutia. II soutient que ceux-ci sont representatifs des groupes Ewe du nord, et que le postulat couramment accepté ne s'applique qu'aux groupes côtiers (Anlo) et riverains (Tonu).

Pour étayer son propos, l'auteur démontre qu'à l'époque pré-coloniale, c'était les villages Abutia qui étaient des groupes politiques souverains. Cette demonstration s'appuie sur une nouvelle définition de la souveraineté, celle-ci fêtant 'le plus haut niveau de rassemblement de groupes pour les questions législatives, juridiques et administratives, en dehors des sphères d'activité militaires, commerciales et religieuses.' Si nous admettons que les villages étaient souverains, il découle de cette definition un certain nombre de corollaires, et ce que nous observons du recrutement, à l'époque pré-coloniale, des groupes législatifs, judiciaires et administratifs, ainsi que de l'origine de la citoyenneté, des relations des citoyens avec leurs 'chefs', des relations entre les différents chefs de village, coïncide parfaitement avec les termes de cette nouvelle hypothèse. Par conséquent, il serait plus exact de définir les systèmes politiques des Ewe septentrionaux comme fêtant des alliances entre villages, des conféderations assez lâches, ou des 'ligues de villages'. Enfin, cette nouvelle formulation suggère une nouvelle manière d'opposer les systèmes politiques du nord à ceux du sud ou la souveraineté semble avoir négligé l'échelon villageois.