1In large tracts of what is today north-western Ghana the institution of chieftaincy was introduced by the British colonial government. The 'stateless societies' of the north-west knew the office of the earth priest, the custodian of the shrine to the earthgod, which was ideally vested in the patrilineage of the first settler of the locale, and fulfilled religious as well as secular functions of conflict management. In some settlements individuals other than the earth priests also held positions of power - 'strong men' would be the translation closest to the indigenous terms. They played an important role in local politics, especially during the latter part of the nineteenth century, when Zaberma and Mande warlords like Babatu and Samori made their incursions into the region. Some of the local 'strong men' would belong to the same patrilineage as the earth priests, but many would not; they were rich farmers, outstanding warriors, petty freebooters or traders who commanded a big 'house' and had built up a large following which often included slaves. Whether the resources of the 'strong man' -and hence his position - could be transferred to the next generation of the same patrilineage or whether they were 'lost' (through, for instance, the matrilineal inheritance of the movable goods) is a question that needs further research. But there is no doubt that the position of the 'strong man', unlike that of the earth priest, was not an inheritable office. Likewise, the 'strong men' were not organized into any kind of centralized polity.

2During the first decade of this century, the British appointed 'native chiefs' who were to assist them in the administration of their newly acquired protectorate of the Northern Territories of the Gold Coast, and set up a hierarchy of village headmen, sub-chiefs and, at the top echelon, headchiefs (later called paramount chiefs). In many cases, the first subchiefs and headchiefs were recruited among the local 'strong men', who then used their networks of relatives, friends and clients to establish the chain of command that the British expected. However, while some elements of 'strongmanship' continued in the new institution, the inheritability of the chiefly office and the fixity of the hierarchy, on which the British insisted, were a radical departure from pre-colonial patterns. The colonial officials themselves were more or less conscious of this. With respect to the pre-colonial states of the Northern Territories, they were convinced that succession to chiefly office was automatic and without conflict if only the proper 'traditional' rules were respected.1 But in the case of the stateless societies, they were aware that there was no single 'tradition' on which succession to chiefly office could be modelled unambiguously. In the early years of colonial rule, the British left it more or less to the house of the deceased chief to present a suitable successor. Later, they experimented with different methods of selection, based on patrilineal succession combined with plebiscitary elements. But even these attempts at closer prescription left a wide margin for conflict and local decision-making.

3In these often conflict-ridclen incidences of succession during the colonial period the ruling families and their factions as well as the wider communities developed competing notions of chiefly 'tradition'. Today, the litigants in chieftaincy conflicts usually claim that they act in accordance with the authentic 'tradition', while their opponents' aspirations are merely backed up by 'politics', i.e. involvement in the domain of the state and party politics. Chieftaincy affairs, the general opinion holds, belong to the domain of 'tradition' and should steer clear of 'politics' - which, of course, in actual fact plays as important a role in present-day chieftaincy conflicts as in succession disputes during colonial times.

4Given the record of colonial experimentation and, more generally, of a whole century of tumultuous change and radical transformation of the stateless societies of the north-west, the notion of 'tradition' seems to be the most unlikely one to be invoked. What exactly this notion means to different groups and why and how it is being manipulated by the various local actors as well as external arbiters are some of the questions that this chapter seeks to address. The dichotomy of 'tradition' versus 'politics', which informs all post-independence chieftaincy conflicts, seems intimately linked to the continued attempts of investing the colonially created office with some measure of local legitimacy of longue durée. This chapter will therefore look at the various registers of legitimacy brought into play in chiefly succession, such as local rules of inheritance, accustomed patterns of 'bigmanship' and the authority of the office of the earth priest. I will discuss the contested production of chiefly history, and more specifically, 'family legends' that are used to back up a certain candidate in succession disputes. Finally, I will investigate the tensions between outgroup ('political') factors, which demand a universalistically legitimate candidate-emphasizing, since the 1950s, the candidate's education and his relations with the Ghanaian government-and in-group factors, which partly operate in favour of candidates who are not easily acceptable to the wider community.

5The case of Nandom, a small paramount chiefdom, which I will use to illustrate these themes, is a particularly suitable example: in the course of the past nine decades, there were six instances of succession, four during the colonial period and two after independence, providing us with a wealth of information on the patterns of decision-making and criteria of legitimacy. The history of the Nandom 'skin', as the chiefly office is now usually called (in allusion to pre-colonial northern Ghanaian kingdoms, where chiefs sat on a lion's skin), is certainly extraordinarily conflict-ridden - which is precisely the reason why local and colonial decision-making has found its way into written documents and is remembered in great detail. At the same, the conflicts are by no means untypical, but merely bring out processes that also inform the neighbouring chiefdoms.

Succession to the Nandom Skin: An Overview and Methodological Remarks

6Kyiir, the first Nandom Naa, whom the British appointed in 1903 and made headchief over twenty-nine villages in 1905,2 must have been a strong man well versed in the world beyond his natal village. According to most of my interlocutors, Kyiir was a relatively wealthy farmer and headed a large house with numerous wives and many sons. He was a trader in cloth, which he bought from the Mossi region further north and sold to the villages around Nandom; and he also knew how to sew. Some informants had him travel as far as Bobo-Dioulasso, and held that, because he returned safely from such a far-away place, he was believed to command magical powers. Others insist that Kyiir was also a prominent local freebooter and acquired a good number of slaves. A British report noted, among other things, that Kyiir was a Muslim - an observation that, if true, would underline his exceptional position in the local society of 'unbelievers'. Much controversy surrounds the question why and how he was made the first chief of Nandom. Some are convinced that it was the Nandom earth priest (belonging to a different patrician) who eventually bestowed the chiefly office on his friend Kyiir. Others maintain that the British, without any interference from the earth priest, merely recognized Kyiir's already powerful position.3 There is little doubt, however, that Kyiir looked the part and spoke various languages, which may have contributed to his colonial appointment. A most important qualification was his wealth (and the able command over the wealth of his relatives): it allowed Kyiir to 'entertain visitors', namely the British officials and their following, and to pay the fines that were levied on the chief and his subjects for the virtually inevitable breaking of some colonial commands.

7In the early years, then, the duties of a chief seem to have required substantial pre-existing material resources rather than having allowed one to accumulate wealth. It is not until the early 1920s that the chiefs were able to command their subjects to perform free labour not only on government roads and resthouses, but also on their own fields. Eventually, the most important qualifications of an influential (head) chief were no longer pre-colonially acquired wealth and a 'strong man' position, but versatility in the communication with the 'white man' and the ability to organize the village headmen and their subjects into an effective workforce.4 From the 1930s onwards, the command of English and a certain degree of formal education played an ever more important role in the discharge of the duties of a paramount chief. Since the end of colonial rule, finally, many people expect the chief, among other things, to represent his local community successfully to the outside world and to attract development projects and other resources for the benefit of his subjects.

8These changing qualifications reflect the changing roles of the paramount chief in the local society and the wider world. They informed the opinion of the colonial (and later Ghanaian) government and, partly, of the local community on chiefs and suitable candidates. They also influenced the decision-making in the chiefly family itself. But they were only one of the factors that guided the family's decision between a number of candidates who were regarded as eligible according to an altogether different criterion, namely descent. Embittered debates over who is eligible to chiefly office and who are the correct decision-makers have characterized much of the history of the Nandom skin. Before entering into details, let me start with an overview over the six instances of succession and some remarks on the methodology of research.

9The archival documents consulted for this overview (Table 5.1) clearly reveal that succession to the Nandom skin has been contested since the 1930s; it may have been so even from the very beginning. Particularly the last three transitions, in 1940, 1958 and 1984-85, were rife with conflict. Violent clashes in Nandom-allegedly between the followers of RearAdmiral l td. Kevin Dzang and Dr Charles Imoro-had occurred only two years before I started research in the area, and tensions are still simmering. Evidently, this affected my research, not only with respect to the present situation, but also the history of chieftaincy. Many of my interlocutors were, in one way or another, entangled in the conflicts, and tended to present a picture of the past that best suited their interests. Of course, I attempted to interview members of the different factions as well as apparently less involved outsiders, and to learn as much as possible about my interlocutors' engagement in current affairs. But even when I was beginning to understand who revealed what to me and for which motives and what was silenced, it was not I myself who set the terms of research, but the local elite.

10There is another effect of such a conflict-ridden situation, rife with gossip and histories, that 1 would like to mention: namely, that it seduces the observer into over-estimating the importance of the contested office. My main research topic was the local history of ethnic identities; but by virtue of my curiosity and detective instinct I was drawn into devoting much more attention to chieftaincy affairs than I had planned originally. From a more detached point of view, 1 believe that, although succession conflicts easily stir up the public imagination and although chieftaincy continues to be a vital link between local communities and the state, the institution isonly one among others that affect the north-westerners' daily lives. The Catholic Church, schools and various government agencies, as well as the many peasants 'and labour migrants' associations, are just as important, not to mention the role of kin and friends. Hence the reader needs to bear in mind that this chapter focuses on just one aspect of local history and politics in the north-west.

Table 5.1: Successions to the Nandom Skin 1903-1985

11As was to be expected, I was accused of being partisan-incidentally by each one of the parties to the 1984-85 conflict-and my first publication on the history of the Nandom skin (Lentz 1993) was harshly criticized as 'washing dirty linen in public'. But my apparently increasing knowledge of precisely the 'dirty linen' made many interlocutors volunteer additional information in order to 'correct' the 'wrong picture' that I allegedly held. Consulting the relevant archival sources helped to specify dates and other details as well as the role of the colonial officials; but the British reports left important gaps that only oral history could fill. There remain many instances in which it has not been (and probably never will be) possible to come up with any 'objective' account of the events that would go beyond mutually exclusive interested versions. Even if such 'objectivity' could be attained (and some versions can indeed be classified as non-factual), it is still of great interest to research into the politics of the historical imagination of the different actors.

12It is, however, highly problematic to put into writing and publish what has so far largely been an 'oral' history of conflict. The few documents that exist and are accessible to the litigants, such as the colonial lists of chiefs, have become ammunition in the recent succession conflict. It is certainly no coincidence that the only copy of the proceedings of the 1958 commission of enquiry, which contains verbatim records of the oral testimonies of all witnesses on the history of the Nandom skin, is kept under lock and key by the incumbent Nandom Naa. He feels that it is one of his trumps, and told me in no uncertain terms that he would only disclose it in court. It was one of his contestants who allowed me to copy the summary of the proceedings (see below) - but he also expressed his fears that anything I wrote could be used in court. There is no doubt that wrote documents, including the anthroplogist's reconstructions, have become a new source of power, especially when a case moves from the local and regional levels to the National House of Chiefs and the High Court in Accra, over which local actors have little control. I have no ready-made answer to the political and ethical questions this entails. I can only avoid presenting contentious details that are highly relevant to the local actors, but less central to the more general themes I wish to address. In what follows, I will emphasize what I see as the underlying structural dimensions of the conflicts of succession, which also allow one to compare Nandom with other cases of a colonially introduced chieftaincy.

Rules of Inheritance - Lines of Conflict

13Although the new chiefs were backed up by the colonial powers, they needed to invest themselves with a modicum of legitimacy according to accustomed local standards if they were to exercise their office effectively. Fear of the coercive measures that the district commissioner could inflict upon those who opposed the new dignitaries was a necessary, but not a suf ficient, basis of chiefly authority. One source of legitimacy was certainly the office of the earth priest. In many cases, the new chiefs were selected from the earth-priest families, and initially regarded as a sort of messenger of the custodian of the earth shrine. The other source of authority was, as was mentioned above, bigmanship'. There are few documents on the methods of selection of the first chiefs, but the records do witness to a process of trial-and-error during the first two decades of colonial rule. A number of chiefs-in colonial parlance 'wicked' or 'weak' ones, whose subjects repeatedly fled across the border to French territory and boycotted the colonial 'road labour' demands - were replaced by office-holders who seemed to command more respect. In some cases, these new office-holders came from a different patrilineage, and the fact that chieftaincy had first been bestowed on another house could give rise to embittered disputes in future successions (see Lentz 1998a: Chapters 3 and 5).

14Nandom Naa Kyiir, as outlined above, did not belong to the earthpriestly lineage, and drew rather upon his pre-colonial 'bigmanship'. Apparently he had no great difficulties in bringing under his authority the twenty-nine villages that still today form the nucleus of the Nandom Traditional Area, as it is called since the 1970s; and by contrast with what occurred in neighbouring chiefdoms, only a few complaints of 'lack of control' were recorded. This would not have been possible had Kyiir not built up a leadership position before the arrival of the first colonial officials.5

15Incidentally, Kyiir was the first headchief of the new Lawra District who died. His succession must have been the test case for the question whether the colonial regime had bestowed the chiefly office on an outstanding individual or on a corporate group, and if the latter was the case, how the boundaries of this group were to be defined. Clearly, the British left little doubt that they wanted the new office to be inheritable. For the local actors, however, this may not have been so obvious, particularly in cases like Kyiir's, where the chiefly office was not vested in the earth-priestly lineage. Unfortunately, we know little of the kind of debates that may have arisen after Kyiir's death. We do know, however, that the Nandom skin has always remained within the same patrician (Bekuone), and that from the 1930s onward there have been (successful) attempts to narrow the group of eligible candidates down to an ever-smaller segment of Kyiir's patrilineal offspring.

16At this juncture, I should explain briefly the rules of inheritance that are predominant among the Dagara of the Nandom area. Each person belongs to one of over forty exogamous patricians that are inherited from one's father, and to one of the seven matriclans, which are inherited from one's mother. Marriage is virilocal, and in polygamous families the sons of the head of the house usually belong to the same patrician, but different matriclans. The land, the house and the ancestral shrines are inherited patrilineally; movable goods, such as cattle, clothes, money, etc., matrilineally. Full brothers of the deceased were the preferential heirs, but if there was no patrilineal offspring with the 'correct' matriclan, wealth moved out of the house (see J. Goody 1956, 1962). There were a number of strategies to circumvent this, and particularly since the advent of the Catholic mission in the 1930s there has been a strong trend towards the patrilineal inheritance of movable goods. However, during the reign of the first three Nandom chiefs, the matriclan still played an important role for the inheritance of movable wealth.

17If chieftaincy was to be inherited, was it to be treated like a movable good? Or was it similar to the succession to the headship of the house, which followed patrilineal lines? These questions were important because, as I explained above, the duties of a chief required some wealth, and a merely patrilineal heir of the deceased chief might not have inherited the movable wealth of his predecessor. From my evidence it appears that the matriclan was indeed taken into account in the first two successions to the Nandom skin. Kyiir, Danye and Boro belonged to the same matriclan (Somda). Danye, a respected Bekuone family head in a neighbouring village, was wealthy on his own account; he was made chief because Kyiir's sons were still too young to succeed their father. As the potential spoils of chieftaincy became more evident, however, Kyiir's sons seem to have developed an interest in bringing the office back into their house.

18This is where Boro came into play. The only fact, however, on which all of my interlocutors agreed was that Boro had a strong personality and commanded some wealth-to which he is said to have greatly added during his term of office. The origin of Boro's wealth and his precise genealogical link with Kyiir are hotly contested. It is not altogether clear whether the disagreement dates back to Boro's times or whether it developed during later succession conflicts. Be that as it may, during the 1985 succession conflict members of the Boro house wrote petitions that assert, among other things, that Boro was a patrilineal 'grandson' of Kyiir, but from a different house - because Boro's father (Vana) was a son of one of Kyiir's elder brothers (Danaa). How did Boro become Nandom Naa? According to this version from the Boro house, 'old man' Danye suggested to the British that Boro should succeed him and that afterwards, in turn, Danye's son Kumbile should take over from Boro.6 This version also claims that Kyiir was merely a head of family, not a chief, and that Boro was the first substantial Nandom Naa because he was the first to receive the colonial chiefs' medallion.7

19In this account, considerations of matriclan and inheritance of wealth are not mentioned. The versions from members of Kyiir's house, on the other hand, attribute Boro's strong position - which was acknowledged by making him chief - not least to the fact that he was one of the heirs of Kyiir's wealth. They claim that Boro was an 'illegitimate' child of one of Kyiir's sisters or nieces and thus belonged to the same matriclan as Kyiir and, because children of unmarried women are adopted by their mother's paternal house, the Bekuone patrician. Boro, as one interlocutor saw it, could lay claim to Kyiir's wealth on account of his unique position in the clan system, and was supported by Kyiir's sons (of different matriclans), who thus ensured that part of Kyiir's inheritance remained in the house. It was Boro's wealth, in addition to his leadership qualities, so the argument continues, that enabled him to replace Nandom Naa Danye. The Boro-house critics of these versions insist that if Boro had actually been an 'illegitimate' child, he would never have been eligible for the skin.

20Whichever version one finds more plausible, after Boro's term of office considerations of matriclan seem to have decreased in importance. The conflicts now rather revolved around the competition between two patrilineally related houses, namely Kyiir's and Boro's, whatever the matriclans of the contestants. The competition made itself felt for the first time after Boro's death. However, there is consensus that when Gome, a brother of Boro, contested with Kyiir s son Konkuu in 1930, he did so single-handedly, while the rest of the Boro family had agreed that Konkuu - whom Boro himself allegedly nominated as his successor - should become the new Nandom Naa. The Boro family insist, on the other hand, that they only agreed to Konkuu on condition that afterwards the chiefly office would return to their house. But after Konkuu's death in 1940 Puobe Imoro, one of Konkuu's brothers' sons, challenged the candidate of the Boro family. The contest was fierce, and it is this succession that most of my informants regard as the beginning of 'outside interference' in the succession to the Nandom skin.

The Colonial Creation of 'Traditions of Customary Succession'

21Nandom Naa Konkuu's death in April 1940 occurred a few years after the introduction of 'indirect rule' in the Northern Territories. In 1934, the colonial officials had amalgamated a number of the smaller chiefdoms and established confederate 'native authorities', which consisted of various previously independent headchiefs (now called divisional chiefs) and their sub-chiefs. The 'native authorities' were responsible for the collection of the head tax, the construction of roads and schools, and - important for our context - the codification of 'native laws and customs'. The chiefdom of Nandom became one of the four divisions of the Lawra Confederacy Native Authority, which was presided over by the senior headchief, first Nandom Naa Konkuu, and, after his death, Lawra Naa J. A. Karbo.

22One of the first topics that the Lawra Confederacy chiefs discussed at length during their regular conferences was the question of succession to chiefly office. The assembled chiefs asserted that there were no longstanding 'native customs' with respect to chieftaincy, but agreed that it was necessary to create 'traditions of customary succession'.8 The first attempt to fix such'traditions'drew upon the locally dominant rules of inheritance, and decreed:

Lobi. Sucession is matrilineal. If a man dies his eldest brother by the same mother will succeed, failing this his sister's son the sister being by the same mother. Failing both these the succession goes to the maternal uncles' family in the same way....
Dagarti. Succession is patrilineal. If a man dies his eldest brother by the same father will succeed, followed by all the other brothers in order of seniority i.e. age. Failing this, the succession will go to the sons of the deceased and his brothers in order of these sons' seniority i.e. age.9

23In addition, certain 'disabilities' were to make a candidate ineligible, namely 'blindness', 'leprosy' and, interestingly, 'unpopularity'.10 Clearly, the rules were by no means unambiguous. Nandom, for instance, was regarded as 'mixed Lobi-Dagarti' and could have drawn on either 'custom'. None of the divisional chiefs who made the rules had acceded to his office in accordance with them. Lawra Naa Karbo, a 'Lobi', belonged to a different matriclan than his predecessor; Nandom Naa Konkuu, regarded as a 'Dagarti', was neither the direct brother nor the son of the late Naa Boro.

24These first codified rules remained without practical consequences. When in 1938 the Jirapa Naa, one of the divisional chiefs of the Lawra Confederacy, died, the district commissioner began by examining whether the candidates were 'eligible by birth' and not 'too old, too young or too characterless', but then proceeded to introduce an additional criterion of legitimacy, namely the 'secret ballot'. One of the candidates insisted he had been nominated by the late Jirapa Naa himself, but the district commissioner brushed this aside. He would have liked to guide himself by 'native custom', explained the British official, 'but in this district where no chiefs existed before our arrival, there can be no question of consulting any custom, except our own. And it may be said that our own custom is to see that the man appointed is congenial to the other chiefs and people, trustworthy by government and as far as possible of unblemished character.'11 In the eyes of the British, the 'secret ballot' - ten votes for each divisional chief, five for the Jirapa sub-chiefs, three for headmen, two for the Jirapa earth priest and finally one vote for each compound head of Jirapa 'town' - was the best method to ascertain the popularity of the future chief. The district commissioner was surprised, however, that the vote in Jirapa was not competitive and that the successful candidate won with only two dissenting votes.

25The unanimity in Jirapa notwithstanding, the district commissioner insisted that the 'secret ballot' be codified as 'native law'. The discussions at the Native Authority conference that preceded the agreement on the new 'customs' of succession are an instructive example of the search for some measure of local legitimacy for the new chiefly office. The chiefs emphasized their relationship with the earth priest, and eventually suggested a mixture of criteria - the veto right of the earth priest, principles of patrilineal descent, and the colonially introduced 'secret ballot':

The Lawra Na related, and the meeting agreed... that chiefship was a modern innovation, and that very little custom attached to rules of succession. A very lengthy discussion then took place, in which the views of nearly every chief present were heard. The method of the secret ballot adopted in the appointment of the present Jirapa Naa was explained and discussed. It transpired that the first chiefs appointed on the European arrival were nominated to the first Commissioners by the Tingansobs [earth priests], and that in this action lay the only 'Custom' that could be said to pertain. It was agreed that in all future appointments the following principles and procedures should be carried out, and should be regarded as 'customary':

a. that a candidate for chiefship must be a member of the patrilineal family of a former chief;

b. that a candidature must be approved by the Tingansob of the area concerned;

c. that election of a candidate to the chiefship shall be by secret ballot...12

26The new rule of patrilineal descent of the candidate was both more exclusive and more flexible than the rules laid down in 1935. On the one hand, considerations of matrilineality were excluded; on the other hand, it was never defined who belonged to the 'patrilineal family of a former chief. The vagueness may have been deliberate, because at the time of the chiefs' discussions the contest over late Nandom Naa Konkuu's succession was already in full swing. Boro's son Yuori and Kyiir's grandson Imoro had presented their claims, the earth priest gave his assurance that both candidates were eligible, and the district commissioner convoked a 'secret ballot' to which he invited the compound heads of the whole of the Nandom Division, in addition to the neighbouring divisional chiefs and the Nandom sub-chiefs. Imoro won with 627 against 402 votes - a result that convinced the British official that 'the election has not been previously "fixed"'.13 Some of my interlocutors, however, insisted that Imoro won by a very narrow margin, and only because the votes of the neighbouring divisional chiefs, as a result of certain marriage relations, tipped the balance in his favour. Imoro's supporters, in turn, believed that he had won nearly unanimously. None of my informants estimated the number of votes anywhere near the above-mentioned result. All agreed, on the other hand, that the ballot was an imposition of the colonial government and went 'against custom'.

27The Lawra Conferederacy chiefs had protested against the colonial introduction of a popular vote on the chiefly successor as early as 1954, and decided that 'balloting for chiefship' was undesirable if there were 'capable sons of the deceased chief'.14 Balloting indeed no longer played a role in the Nandom succession conflict of 1958, but neither did the 1954 rule of filial succession: Yuori, who competed again, stood against Polkuu Paul, a patrilineal ‘cousin’ and not a son of Imoro. This time, the conflict was eventually decided - in favour of Polkuu - by a government-appointed commission of enquiry. Interestingly, Yuori's followers later insisted that there should have been the 'traditional' ballot.

28By the late 1950s, then, the candidates could draw on a number of different 'traditions', created by the colonial officials, the chiefs' conferences and local precedents: patrilineal descent, nomination of the candidate by the deceased chief, agreement of the family elders on a suitable successor, consultation of the earth priest, and secret ballot. Moreover, succession conflicts became an arena in which the inner-family criteria of acceptability of a candidate competed with outgroup criteria such as the new chiefs education and his political standing. In the next section, I will show how the different rules and criteria were brought into play and how party politics became an important factor in succession conflicts.

Education and Party Politics: The Succession Conflict of 1958

29The succession conflict of 1958 created a deep rift between the competing factions that has continued to affect chieftaincy affairs in Nandom until the present day. When the tension rose after each party had installed their candidate as the new Nandom Naa, the government of the now-independent Ghana saw the necessity to appoint a committee of enquiry. The committee consisted of a member of the Ministry of Local Government in Accra, one northern and one southern chief, and was 'to inquire into the correct method in tradition of selecting a Nandom-Na and to report whether Konkou Polkuu Paul or Borrow Yuori is properly the occupant of the Nandom skin'.15 The ten committee hearings developed into an arena of contest over the history of chieftaincy in Nandom. I will briefly outline some of the arguments and then explain the political context in which the conflict and the hearings took place.

30Yuori explained to the committee that after Nandom Naa Imoro's funeral he had told the Nandom earth priest that he, Ynori, wanted to be the next chief. The earth priest, Yuori asserted, 'agreed to install him as chief'.16 Polkuu, in turn, insisted that Imoro, on his deathbed, had nominated him as his successor. With the exception of the Boro section, Polkuu reported, he enjoyed the support of 'the Members of the Royal Family' and particularly those members who were 'entitled to elect a Chief'.17

31When attempting to invest their claims with historical legitimacy, Polkuu, Yuori and their witnesses presented different accounts of the origins of chieftaincy. Yuori stated that the first Nandom Naa who the British appointed had not been Kyiir, but Daga, a member of the house of the earth priest, and that later Daga asked Kyiir to take over as chief because he, Daga, 'could not perform the double duties of a chief and a priest of the land-god'. In accordance with the Lawra Confederacy chiefs' conference discussions, Yuori claimed that traditionally the earth priest 'selects a chief and presents him to the People'.18 Conversely, Polkuu asserted that Kyiir,'a brave, rich and noble man', had even in pre-colonial times been 'virtually a chief of his people,' and that, owing to specific circumstances, only Kyiir's family was present when he was officially appointed chief by the British, not the earth priest. In consequence, the earth priest had 'no official duties to perform at the installation of chiefs at all'. Rather, it was by nomination through the dying chief and later approval by 'the Royal Family' that a new Nandom Naa was 'traditionally' selected.19 Unsurprisingly, Polkuu and Yuori presented the past four accessions to the Nandom skin in a way that demonstrated how firmly the rule of succession that favoured their own claims was rooted in 'tradition'.

32In the eyes of the committee, the incumbent earth priest failed to support Yuori's case when he stated that 'after the transfer of chieftaincy to Chiir by Dagar, the chiefs were nominated by the dying chief and approved of by his [the earth priest's] predecessors and himself'. The committee took this for evidence which supported Polkuu's claims and concluded that:

1. The Correct method in tradition of selecting a Nandom-Na is by Nominationby the dying or retiring chief.

2. The second traditional method which may be used in certain cases is selection by the Members of the Chief's Family and the approval of the People which is signified by general acclamation.

3. That Kunkuu Polkuu Paul has been made Chief by these two methods combined and therefore Kunkuu Polkuu Paul is properly the occupant of the Nandom Skin.20

33The formulation here of the second 'method' in particular was ambiguous, because it could be interpreted as reintroducing the plebiscitary element once established by the colonial officials and later dismissed by the Lawra Confederacy chiefs. Consequently, when Yuori, supported by nineteen village headmen, petitioned against the committee's decision he no longer insisted on the earth priest's prerogatives but on the 'tradition' of the ballot. However, Yuori not only appealed to 'tradition'; he also drew on the secular radical-democratic (and partly anti-chief) discourse that was en vogue particularly in the circles of Kwame Nkrumah's GPP (Convention People's Party), the party that had led Ghana into independence:

So long as there is rivalry there should be ballot that there is no record in the pass [sic!] in our tradition and customs that empowers or obligates any Royal Family to elect a Naa for the people. Why? Because he is not going to rule only the Royal Family but all the people and therefore we the masses are at the liberty to give our wish which must be respected... Sir, the tradition in all respect must be recognise [sic!] or honoured, as long as there are two claimants to the vacant 'skin' we must go on ballot to elect one; that we want every compound owner of the Nandom Division to cast a vote each, and that we shall accept whosoever is unanimously elected per the majority votes.21

34The Governor-General dismissed the petition on the grounds that no right of appeal existed once government had accepted a committee of enquiry's findings, and gazetted the new chief.

35It is not clear if the committee of enquiry actually had a popular ballot in mind when stating that the new chief needed 'general acclamation'. But by claiming that the family's candidate should be acceptable to 'the People', it certainly pointed to the complicated relationship between ingronp and outgroup factors of chiefly legitimacy. Tensions between these factors played an important part in the 1958 conflict. With respect to the support of a wider local constituency, namely the village headmen and chiefs of the Nandom Division, Polkuu and Yuori seem to have been on a par. But there were other extra-family criteria. By the 1950s, some measure of formal education had become an important prerequisite for the discharge of chiefly duties, particularly at the level of the paramount chiefs, not least because they continued to play a part in the increasingly literate local government institutions. Moreover, educated chiefs were more likely to become successful representatives of their communities in the wider political arena. In this respect, Polkuu had clear advantages over Yuori: while Yuori was basically an illiterate farmer, Polkuu was a trained teacher who, due to his education in the Northern Territories' capital Tamale, could benefit from close connections with many members of the regional political and chiefly elite. Even my interlocutors from Boro's house admitted that with regard to the 'changing times' of the 1950s and 1960s, Polkuu was the more suitable candidate. The Boro house, too, explained one of them, would have put up an educated candidate, had not Yuori insisted that his seniority prevailed over the aspirations of younger (but more highly educated) members of his family. However, members of the Boro family insist that Polkuu won the contest not only because of his better education, but also due to the 'manoeuvres of government' and 'through politics'.

36"The time when politics came', as local informants refer to it, is the time of the 1950s and early 1960s, when party politics and local conflicts became deeply entangled. After the Second World War the colonial government had replaced the 'native authorities' by a dual system of State Councils, which were to concern themselves exclusively with chieftaincy affairs, and District and Local Councils, which were responsible for all local administrative matters and whose members were one-third appointed by chiefs and two-thirds elected by general vote. In addition, the north was finally granted representation in the colony-wide Legislative Assembly. It was in this context that political parties such as the UGCC (United Gold Coast Convention), from which the CPP later broke away, started to proselytize also in the north, particularly among the small educated elite. Many northern chiefs, however, supported the oppositional NPP (Northern People's Party). Programmatically, the NPP was formed in order to safeguard northern interests in the new political union of the Gold Coast Colony, Asante and the Northern Territories, which in 1957 amalgamated into independent Ghana. In practical terms, however, it was mainly personal networks, created through chieftaincy and patrician ties, to which the educated NPP activists-just like the GPP followers-appealed when canvassing for votes.22

37Lawra Naa J. A. Karbo was among the founding fathers of the NPP in the Lawra District. His son Abeyifaa, then teacher at the Lawra middle school, won the Legislative Assembly seat of the Lawra-Nandom constituency in the elections of 1954 and 1956. In Nandom, the most active NPP member came from the Gaamuo house, a section of the earth-priest family - W. K. Dibaar, a teacher and a colleague as well as a friend of Abeyifaa Karbo. Dibaar never stood in elections, but won his extended family over to the NPP. Initially, Nandom Naa Imoro also supported the NPP. But owing to long-standing rivalries between the Nandom and Lawra chiefs, when Imoro 'became aware that the NPP candidate would be Abayifaa Karbo..., he changed his mind, preferring to support a candidate from his own town regardless of party' (Ladouceur 1979: 120-1). In the 1954 elections Imoro supported a candidate not only from his own town, but his own family, namely Polkuu Paul, who stood as an independent candidate-but an 'independent CPP', as my informants explained. In 1956, Imoro came out openly in favour of the GPP candidate, and at the same time petitioned that the Nandom Division should secede from the Lawra Confederacy.

38Soon afterwards, Imoro was told in no uncertain terms by the Lawra Confederacy State Council that he should consider himself 'removed... from his office as a chief.23 It was a coalition of the NPP-supporting paramount chiefs of the Lawra Confederacy, the Nandom earth-priest family and a number of chiefs and earth priests from other villages of the Nandom Division who joined in the attempt to destool Imoro. Some of Imoro's local opponents had old grudges against him, such as, for instance, the Nandom earth-priest family, who had quarrelled with Imoro over land rights and the allocation of rents; others seem to have mainly taken offence at Imoro's change of mind from NPP to GPP and his attempt to secede from the Confederacy. But because of the constitutional separation between local and party politics on the one hand and chieftaincy affairs on the other, the party-political motivations of the destoolment attempt were never voiced officially. Before the Lawra Government Agent-until mid-1958 still a British official-Imoro's critics claimed that Imoro had gone against 'the customary law of our land and the rules of our chieftaincy' and therefore had to be destooled by the very Lawra Confederacy chiefs who had voted him into office in 1940 (see Lentz 1998a: Chapter 13).24

39Before any definitive decision was reached, Imoro died. The factions in the ensuing succession conflict continued the alliances of the destoolment attempt. Polkuu was backed up by the descendants of Kyiir and by the CPP supporters on the local as well as on the national level: Yuori counted on the Boro house and the Lawra Naa and his son Abayifaa, as well as local NPP followers, particularly from the Nandom earth-priest family. When discussing the details of these alignments with my interlocutors, however, partisan points of view made themselves felt. Some claimed that Yuori had initially supported the CPP and changed to the NPP only after his defeat in the chieftaincy conflict; others asserted that he had already had a hand in the destoolment attempt, and joined the NPP in order to further his own claims to office. Similarly, some stated that Imoro as well as Polkuu Paul had supported the CPP throughout, while others stressed aspects of 'political gamble' and pointed to the CPP government's tactics of backing those candidates in chieftaincy conflicts who would in return promote CPP interests. And, unsurprisingly, Polkuu's opponents claimed that he was supported by the commission of enquiry only because of his CPP affiliation, while Polkuu himself emphasized his 'traditional' legitimation.

40In any case, there is no doubt that from the mid-1950s onwards chieftaincy affairs and (party) politics became inextricably entangled, not only in Nandom, but in the whole of Ghana (see Robertson 1973; Staniland 1973; Arhin 1985: 112-14 and Wilks 1989, for some examples). While constitutionally chiefs are no longer accorded any official political role, political parties do use the influence of chiefs to canvass for votes, and military governments to secure a modicum of rural support and to explain government politics to the people. Aspirants for chieftaincy, on the other hand, trade their support of a political party or the government of the day for the latter's assistance in securing the desired office. Nevertheless, Ghanaians insist on distinguishing between the sphere of 'politics' and the sphere of 'tradition', and while they acknowledge that in practical terms it is necessary to use 'politics' in order to become chief, the legitimacy of the chiefly incumbent is ultimately seen to rest on 'tradition'. In the final section, I want to explore these notions of 'politics' and 'tradition', using the example of the most recent succession conflict in Nandom.

'Tradition' Versus 'Politics': The Succession Conflict of 1985

41Predictably, when the Nkrumah government was overthrown by a military coup in 1966, Yuori attempted to redress his defeat. He sent a petition to the Chieftaincy Secretariat in Accra that complained that only because they had been 'staunch supporters of the Kwame Nkrumah rigime [sic]' Polkuu Paul and his section 'unlawfully claimed also to be entitled to the Nandom skin25 However, it seems that Yuori did not enjoy much support and that Nandom Naa Polkuu was popular enough to ride out the attack. It was only after Polkuu's death in 1984 that the old factions re-emerged with renewed vigour. This time, Yuori's junior brother Gbeckature, then working as regional stores inspector, put forward the claims of the Boro 'gate', as he now called it, and secured support from Gaamuo Le-ib, who claimed to be the legitimate earth priest of Nandom. As for Kyiir's house, matters were more complex. Nandom Naa Konkuu, Imoro and Polkuu had all come from a single section of the Kyiir family, who now maintained that only they were entitled to the Nandom skin.26 Disagreement within this section, however, resulted in the presentation of two competing candidates: C. Y. Dery, a brother of the late chief Imoro, contested in the name of the house elders and on the principle of his own seniority; Imoro s son, Dr Charles Imoro, based his claim, in continuation of Polkuu's arguments in 1958, on an alleged will of his predecessor. In addition, Dr E. N. Delle and Rear-Admiral (rtd) Kevin Dzang, who come from other sections of the Kyiir family, put forward their claims and argued that the Imoro-Polkuu section had monopolized the chiefly office for far too long.

42Shortly after the final funeral rites for the late Polkuu, and while family consultations were still going on, Dr Imoro's supporters organized his installation ceremony, which seemed to receive the tacit support of the Rawlings government. This provoked a flurry of petitions by Dzang and Gbeckature to the Upper Regional House of Chiefs and to the Chieftaincy Secretariat in Accra. When the petitions remained without response, Dzang seems to have wanted to create a fact of his own, and he also had himself installed as Nandom Naa. A few months later, however, the Rawlings government officially recognized Dr Imoro as the new Nandom Naa.27 The defeated parties continued to write petitions and letters of complaint; but when they were finally due to be heard before the House of Chiefs in September 1992, the oncoming session was postponed, and eventually indefinitely adjourned. Shortly before the presidential elections of November 1992, a decree by the Rawlings government abated several pending chieftaincy suits, including that of the Nandom Naa, and declared the disputed office-holders chiefs for life. Rawlings's electoral victories in 1992 and 1996 and Dr Imoro's high political offices-he has been elected into the Council of State and was made vice-president of the National House of Chiefs-seem to have taken the wind out of his opponents' sails.28 Nevertheless, a future change of government could give a fresh impetus to their aspirations. Moreover, the new constitution finally abolished the procedure of government 'gazetting' of chiefs, which the British had introduced, and leaves the settlement of chieftaincy disputes entirely in the hands of the houses of chiefs and their judicial committees. Consequently, the contestants for the Nandom skin disagree whether the abatement decree of November 1992, proclaimed under a military regime, is still valid under the new democratic constitution.

43It would take us too far to outline all the chiefly candidates' arguments in detail. They repeat the arguments of the 1958 dispute, but also bring new elements into [day. Dr Delle's and Rear-Admiral (rtd) Dzang's argument, for instance, that not only a single section, but all sections of Kyiir's house have a right to the Nandom skin provoked a protracted debate on the question whether some of Kyiir's wives, namely Delle's and Dzang's grandmothers, had been 'slaves', and, if so, whether the descendants of 'slaves' were eligible as chiefs. Delle's and Dzang's claims forced the Boro house, on the other hand, to find additional arguments for its aspirations. On the basis of the Chieftaincy Act of 1971 and the models of chiefly succession that prevail in pre-colonial chiefdoms such as Dagbon or Gonja, Gbeckature Boro argued that only persons whose fathers had been chiefs were eligible - an argument that would exclude C. Y. Dery, Dzang and Delle from the list of contestants. Secondly, Gbeckature claimed that considerations of seniority should decide between the eligible candidates - which would exclude Dr Imoro. And thirdly, he insisted that the passage of the chiefly office from Boro to Konkuu in 1930 was meant to initiate a system of rotation between these two houses or 'gates'.

44Just like the witnesses before the 1958 commission of enquiry, the petitions of the chiefly candidates foreground exclusively the 'traditionalist' arguments of 'customary succession'. The exact definition of this 'custom' is contested, and different family legends are constructed in order to give weight to the candidates' own arguments. But none of the parties questions that chieftaincy needs to be based on 'tradition'. Outside this official register, however, the 'political' aspects of the succession conflict are debated extensively. On the one hand, political considerations influence the evaluation of the suitability of a candidate. No one denies, for instance, that the potential political clout of the chief-to-be plays an important part in the considerations of whom the family sections should put forward as a candidate. Who will best be able to rule the Traditional Area and to represent Nandom in the wider world? One of the reasons why so many candidates presented their claims in 1985 and the contest became so fierce is precisely that for the first time in the history of the Nandom skin, there were several candidates with a higher level of education and a certain political experience who could be expected to link the local community successfully with state institutions. In addition to such universalistically framed arguments, particularistic expectations of tangible benefits from the new chief - and the better his position in the outside world, the more can be expected in return - also played a role (and all candidates complained that the others had 'bribed' the family elders).

45On the other hand, all my interlocutors were keenly aware and highly critical of what they perceived as 'political interference' in Nandom chieftaincy affairs. In particular, the question why the Rawlings government backed Dr Imoro and not Rear-Admiral (rtd) Kevin Dzang, who had served as its Secretary for Defence in 1983-84, was the subject of much debate and speculation. During the elections of 1992 and 1996, Dr Imoro is said to have pulled his weight in supporting Rawlings's NDC (National Democratic Congress), while Dzang kept somewhat aloof from party politics and Gbeckature openly canvassed for the NPP (New Patriotic Party), the current reincarnation of the NPP-UP tradition.29 Some commentators believed that in return for Dr Imoro's support for the NDC the parliamentarians of the area supported his candidature to the Council of State. The seat in the Council of State guarantees not only access to the corridors of power but also quite lucrative allowances, which, in turn, can be invested in generosity towards sub-chiefs and followers, splendid festivals and ostentatious hospitality, building up the chiefs prestige.

46Nobody in Ghana would deny that these observations bear witness to the general and in practical terms unavoidable interpénétration of chieftaincy affairs and 'politics'. On a normative level, however, all agree that crude material and secular political interests should have no influence in chieftaincy affairs, and that succession conflicts should be regulated exclusively by reference to 'tradition'. Even the most trenchant comments on the increasing corruption of chieftaincy by 'politics' do not suggest abolishing the institution altogether, but a return to an unspoiled, 'traditional' conduct of office. When in early 1996 the Rawlings government proposed an amendment of the constitution, intended to allow chiefs to participate actively in party politics, there was a public outcry that the move would be 'inimical to the prestige, honour and the very survival of the institution of chieftaincy'.30 Chiefs, it was argued, must guarantee the unity of their respective communities, and should not dabble in party politics. Their status is regarded as guaranteed by 'tradition', not by the grace of the government of the day. A candidate to chiefly office would therefore certainly impute self-interested motives to his opponents, but present his own aspirations as a service and a sacrifice undertaken in the interest of the family, the local community and'tradition'. At first sight all this might appear to be merely a transparent rhetorical trick. But the normative separation of 'tradition' and 'polities', which dates back to the colonial period, has become entrenched in popular concepts of legitimate power.

47That the notion of 'tradition' now also influences chieftaincy debates in previously stateless societies, where chiefly 'traditions' are admittedly of recent date, shows how successfully the political organization of these societies was transformed by the colonial regime. In Nandom and the neighbouring chiefdoms, the notion of 'tradition' comes to stand for some measure of locally derived legitimacy for the new institution, even if the exact content of the 'tradition' is contested. By contrast, 'politics' stands for a potentially illegitimate external interference in local affairs. The appeal to 'tradition' serves as a backdrop for a critique of the abuses of power; 'tradition' speaks to the quest for continuity and stability in a local world that has experienced a century of tumultuous change through 'outside' interference. There is, then, a systematic denial of the 'external' origins of the chiefly office and the rules of succession, particularly by the 'royal family' itself. But other registers of discourse and patterns of legitimacy co-exist. Among ordinary villagers, for instance, and particularly when the language of debate changes from English to the vernacular, one can often find that chieftaincy is called the 'white man's chieftaincy' (nasa naalu, in Dagara), and thus distinguished from other forms of authority such as the earth-priestly office or the power of the family elders. Nevertheless, these villagers, too, agree that 'chieftaincy has come to stay', because it forms a vital link between the local community and the state that none of the other more 'autochthonous' forms of social organization could provide, and they too share the critique of 'political' opportunism in chieftaincy affairs.