9
The Bakitara (Banyoro) of Uganda and Collective Action Theory

The Nyoro State in the Nineteenth Century

In Blanton and Fargher’s cross-cultural survey, low scores on the measurements of collective action correlated with an economic emphasis on external revenues. Such external revenues, they argue, would be favored by rulers because they would then be less economically reliant upon taxes (tribute) garnished from their subjects. As a result of this, rulers would be less beholden to their subjects and could in principle maintain a stronger grip on power, assuming that bureaucratic/infrastructural means existed by which rulers could exercise their authority (Blanton and Fargher 2008:252–254). Roscoe’s (1923) ethnography of the Banyoro was the source of most of Blanton and Fargher’s data; this ethnography was based on several weeks of interviews at the royal court during the early period of British colonialism and is written in the past tense suggesting that, more often than not, Roscoe’s informants were remembering the halcyon days of the late nineteenth century under their great and last independent ruler, Kabarega (Kabalega). The same is broadly true of Beattie’s (1971) ethnography of the Nyoro kingdom based on field research in the 1950s.

Taken together these ethnographies reveal a highly despotic and increasingly centralized kingdom in the nineteenth century, during the course of which external revenues derived from exports became ever more important (see also Doyle 2006:51–52), so much so that the trade in ivory and guns was a royal prerogative. Indeed, permission to hunt elephants was granted only to a class of professional hunters controlled by the king (omukama) and the populace was forbidden to trade in ivory on pain of death (Uzoigwe 1972:446, 451). All markets, of which there were many throughout the kingdom, belonged to the king, who appointed abahoza, a term translated as “political agents/tax collectors,” to oversee them and ensure that revenue flowed to the king; even some wives of the king were abahoza (Uzoigwe 1972:450). Moreover, in Kabarega’s reign there were four different markets in the neighborhood of the palace (ibid.). Much wealth was also derived from the export of slaves in the nineteenth century; foreign slave merchants supported Kabarega’s bid for power in the war of succession following the preceding king’s death (Steinhart 1977:32).

Kabarega is also credited with military reforms that created a standing army with companies (barusura) of soldiers who were often under the leadership of foreign mercenaries appointed directly by the king. These companies were stationed throughout the kingdom and indeed expanded the kingdom through conquest, gaining their livelihood by plunder (Steinhart 1977:21–22). Thus, Kabarega instituted a potent source of authority that was mostly independent of the populace and cut across existing class and clan loyalties, further centralizing royal power and expanding his ability to exercise royal authority across the kingdom.

However, while it is clear that external revenues became increasingly significant for Bunyoro’s economy over the course of the nineteenth century and that royal control of these revenues, together with the barusura, promoted highly centralized authority, such revenues seem at first to have been supplemental to internal revenues. As Beattie (1971:139) remarked, “Everybody should give to the king.” Much, if not most, internal revenue derived from the king’s personal delegation of authority not only to territorial chiefs but also to people in more lowly political posts. “All subordinate political authority in Bunyoro was held, and was seen to be held, as the direct gift of the king himself, and at his pleasure” (Beattie 1971:147). “The grant of a chiefship by the Mukama [king] was essentially the bestowal of rights over a particular territory and the people in it . . . to be given an ‘estate’ was to be granted political authority over it” (Beattie 1960:37). It is possible that the appointment of chiefs was a practice either established or formalized by Kabarega, as one Nyoro historian has noted that Kabarega changed chiefships from hereditary to appointive (Kihumuro-Apuuli 1994:62).

From their subjects, chiefs received tribute in food, beer, and labor, much of which was passed to the king, who also received ivory, animals, and iron from his chiefs (Beattie 1971:130, 166). Similarly, the king frequently toured his kingdom, exacting food and labor from the areas he visited. Furthermore, all the women in the country belonged, at least in theory, to the king, who did not marry, since the king could not occupy the subordinate status of being somebody’s son-in-law; Beattie (1971:143) surmises that kings had “scores, if not hundreds” of royal wives, noting that Kabarega is credited with fathering 140 children.

As with the ethnographic observations on external revenues, it is probable that these observations on internal revenues¹ also derive from the nineteenth century, but it seems reasonable to assume that these internal revenues had deeper historical roots. Such an assumption is perhaps supported by ethnographic data on the ways in which the king was obliged to reciprocate by giving generously to his people. Collective action theory would certainly encourage us to predict that, with less external revenue, the king would need to offer more in return to his taxpayers (Blanton and Fargher 2008:252–254), so the Nyoro ethnographic observations of the king’s “generosity” may well reflect an earlier time.

The king was expected to give generous gifts to individuals and to sponsor great public feasts. This was reflected in some of the king’s official titles, including Mwebingwa, “he to whom people run for help when in need” (Beattie 1971:141). Indeed, the king seems to have been a pivotal figure in organizing famine relief (Doyle 2006:31), while he is also credited with ensuring that traditional medical practitioners were distributed across the kingdom and seconded when necessary to areas of disease outbreaks (Doyle 2006:32). The king’s gifts were not all strictly utilitarian; his delegation of power carried with it a delegation or sharing of ritual authority or potency termed mahano (Beattie 1971:117–118). Moreover, the king was symbolically identified with his country, requiring him to stay healthy, maintain a state of ritual purity, and perform daily rituals for the good of the country (Beattie 1959, 1971:105–107). Kabarega himself also seems to have used the threat of attack by the neighboring kingdom of Buganda as a means to appeal for national unity and reconciliation (Kihumuro-Apuuli 1994:66).

The fact that the king both received tribute (tax) from his people and provided them with gifts, feasts, famine relief, and ritual support, as well as military protection, draws attention to the negotiation of power and authority between rulers and ruled that is integral to the workings of collective action. It also illustrates the essential tension between the populace’s desire for autonomy and their need for state-supported security that the historian Jan Vansina (1990:232) has argued lies at the heart of all politics in equatorial Africa (see also Doyle 2006:15). This tension or duality is recognized by the Banyoro themselves, who have reported that kingship and government stand in opposition to a set of terms expressive of community-based loyalties. Indeed, the Nyoro term for “government,” bulemi, incorporates both the idea of “ruling” and oppressive “weight,” bulemezi (Beattie 1971:6–7). Furthermore, the king was regarded as the “ruler or master, not the father of his people” (Beattie 1971:104), the king’s lineage having been “chosen long ago to rule us,” as recounted in testimony given to Beattie (1971:100). Inequality is pervasive in Bunyoro, as expressed in the proverb, “people are [only] equal in the grave” (Beattie 1971:7). It will come as no surprise to find then that the tension between state and community is evident also in the panoply of rituals associated with the kingship.

Royal Rituals

On the basis of collective action theory it has been predicted that more collective polities would be characterized by rituals associated with rulership that might serve to build trust between rulers and ruled (Blanton and Fargher 2008:203). However, rather than attempting to fathom the role of ritual per se in the negotiation of collective action, Blanton and Fargher (2008:206) attempted to measure the extent to which there was public monitoring of the behavior of rulers, as well as the extent to which rulers controlled “ideological resources” and could use these resources to render themselves in some sense sacred and, therefore, not bound by the same rules as ordinary folk. Presumably the performance of rituals in very public settings would permit public scrutiny of both the ruler’s lifestyle and his commitment to the collective (Blanton and Fargher 2008:22, 203; Golden and Scherer 2013:402). The opposite, however, might not be true of more despotic states; certainly one would expect rituals to reinforce royal authority but there would seem to be no a priori reason for assuming that such rituals would be preferentially performed in private rather than in public settings. Private rituals could presumably contribute to the mystique of power, excluding commoners from sources of creative power, but carefully orchestrated public spectacles might equally well serve to reinforce the legitimacy of royal power (see also Fleisher and Wynne-Jones 2010). However, we may also ask how the rituals themselves, not just their setting and performance, served to negotiate power. However, before I embark on an examination of Nyoro rituals of kingship, it should be noted that a distinction between the king as a political and economic agent versus the king as a nexus of state rituals is, if not entirely arbitrary, a distinction that may well be lost on the Banyoro themselves, since instrumental power, power that controls people’s actions, and creative power, the power that manipulates and invents forms of meaning (Schoenbrun 1999:139), are interdependent (see Robertshaw 2010).

A review of the literature, primarily the writings of Beattie (1959, 1971: chapter 5), on rituals of Nyoro kingship reveals that the tension between ruler and ruled to which I have already drawn the reader’s attention is present too in the panoply of rituals, a conclusion that gives credence to Beattie’s assertion that “ritual and ceremonial clustered about the Nyoro kingship because it was the centre of secular power, rather than the other way about” (Beattie 1971:107–108). The king’s right to rule was founded upon myth and history but confirmed by the coronation ceremonies performed when a new king claimed the throne—an event frequently preceded by a civil war among rival princes, which in turn served to remind people that peace was only possible with a king firmly in control of the state. The coronation ceremonies took place at a temporary royal enclosure; an initial purification ceremony was reported as being witnessed by “crowds of people” (Roscoe 1923:128), while later when the king sat on his throne, “crowds of the chiefs and better class people pressed forward, wishing him long life and congratulating him” (Roscoe 1923:130). At his accession, the king was also admonished to rule wisely and justly, and made to swear not to frighten his nation and to help without distinction both rich and poor (Beattie 1959:141, 1971:117). Thus, some of the ritual surrounding the king’s coronation appears to have been conducted in public and involved an attempt to curtail any tendency toward autocracy. In addition, even those parts of the coronation ceremonies that took place in relative seclusion in the palace were attended by various servants and assistants, who often held honorary titles; perhaps more important, each appears to have represented a particular clan (K.W. 1937). However, other parts of the coronation stressed the king’s right to rule and his ownership of the means of violence, notably the handing over to him of artifacts, including a spear, a bow and quiver, a dagger, and a stick. He also struck a hammer on an anvil to signify that he was both the head of all the blacksmiths and he himself like a hammer (Beattie 1959:140; 1971:112; K.W. 1937:296).

Once installed on the throne, many of the rituals of kingship were part and parcel of the king’s daily life (K.W. 1937:298–299; Roscoe 1923:91–107); these rituals were required of the king for him to maintain his state of ritual purity, as was appropriate given his symbolic identification with the state. The rituals included food taboos and several daily ceremonies involving cattle and milking. Similarly the king remained in relative seclusion for most of the monthly new moon ceremony and the annual ceremony in which he blessed the country (Roscoe 1923:107–112). These activities all took place within the royal enclosure (kikali), a complex of several functionally specific buildings, access to most of which was strictly controlled (Roscoe 1923:73–86). Thus, much royal ritual took place in seclusion, an observation that perhaps accords with the fact that the king was considered to be separate from and superior to everybody else; he was always addressed in the third person and a special vocabulary existed for his person and his activities. The king also possessed considerable royal regalia (K.W. 1935:160; 1936:77, 1937). Therefore, it appears that after the king had been crowned, most royal ritual took place away from the public gaze and could perhaps be interpreted within the framework of exclusionary power strategies (Blanton et al. 1996), despite the fact that some of these activities were commonly considered, at least by Roscoe’s informants at the court, to have been undertaken in order to bless the country and its people (Roscoe 1923:93).

Despite the privacy of the regular royal rituals, the king was apparently required to appease the populace after he had ruled the kingdom for nine years, nine being regarded as a very auspicious number by Banyoro. This seems to have been a remarkable ceremony or set of ceremonies, at least for Western sensibilities. Accounts of the ceremony vary (see Beattie 1971:113–114; Nyakatura 1973:205–207), but central to it was an oath of peace sworn by the king in which he pledged, inter alia, not to become angry, not to punish his people, not to kill anyone, not to wage war, and not to be ungenerous. However, what is striking about this ceremony was the massive amount of human sacrifice that either accompanied it or followed it. According to Fisher (1911:130–131), the wife of an early colonial missionary, the sacrifices included 30,000 cattle and 200 princes, the latter killed by being thrown into a large furnace, as well as a royal servant who was sacrificed instead of the king. In Bikunya’s account the king and two others climbed into a pit which was then filled up to their necks with human blood from victims sacrificed next to the pit. Once the blood in the pit reached the right level, the king had to climb on top of all the corpses and repeat the oath of peace (Bikunya 1927:52, cited by Beattie 1971:113–114). Nyakatura’s account is less gruesome but nevertheless speaks of the execution of large numbers of people, including numerous royal servants (Nyakatura 1973:205). As Beattie (1971:114) remarked, we cannot be sure that these rites were ever performed; they certainly were not witnessed by any of the authors of the accounts and probably not by any of their informants either. However, this does not detract from the fact that such a ceremony was widely believed to have existed and that the oath of peace was accompanied by a requirement, at least according to Bikunya and Nyakatura, that the king relinquish his authority to the senior members of his government.

It is tempting to view this remarkable ceremony as symptomatic of the tension between kingship and populace, with the king being required, albeit at a very auspicious moment of his reign, to publicly recognize his duties to his people, perhaps even to relinquish his authority. An analogous public ceremony, the Orun festival, took place among the Yoruba of West Africa, at which the gods decided whether or not to let the king continue to rule (Blanton and Fargher 2008:208; Trigger 2003:510–511). However, it is the scale of the human (and animal) sacrifice that accompanied the oath of peace that draws attention to the importance of the Nyoro ceremony. The clearly drawn connection between kingship and human sacrifice evident in this ceremony is also present in some other African states, including Yoruba and Buganda (as noted by Blanton and Fargher 2008:208, 213), and spectacularly so in the royal burials at Kerma in the ancient kingdom of Kush (Bonnet 1990, 1992). This can be interpreted as a reminder of the importance of “wealth-in-people” as the sine qua non of success in politics in many parts of Africa (see below), with the destruction of this wealth through sacrifice being the ultimate statement of royal power. One might think that such an awe-inspiring destruction of wealth would serve to reinforce the Nyoro king’s authority, not encourage him to give it up. Perhaps the dual nature of the Nyoro ceremony served to stress the power and importance of the kingship and the state, while reminding the king that he himself could be replaced. Be that as it may, our account of Nyoro royal rituals serves to highlight the tensions between the king and the populace. It may also suggest that for long periods of time the balance of power seemed to lie with the state, as expressed in the daily, monthly, and annual rites that were conducted in the confines of the royal enclosure, mostly hidden from public scrutiny. Nevertheless, at long intervals the king’s subjects may have had the opportunity either to curb the authority of the king or at least to remind him that his authority could rightfully be challenged.

Mobility and the Composition of Wealth-in-People

Of course, there may have been other avenues by which some of the populace might challenge the king’s authority, including allying themselves with princes at the borders of the kingdom who chose, albeit at their peril, to spurn the king and establish independent polities. People may also have chosen to emigrate, but we have very little information on this practice other than Buchanan’s study of clan traditions that focused on immigration rather than emigration (Buchanan 1974). However, some idea of how mobility worked in practice within Bunyoro is provided by a study of village composition undertaken in 1966 (Charsley 1970). The village under study, located on the periphery of the former Nyoro kingdom, was said to “possess an air of stability and permanence” that was in fact “illusory” because half of the households in the village had moved there since 1960, though the total size of the village had not increased. Mobility was made feasible by an abundance of available agricultural land, presumably a reflection of low population densities, that was neither bought nor sold, as well as by an ethos of good neighborliness (Beattie 1960:61–66). Why did households move? Sometimes there were economic incentives, such as escaping from the crop predations of elephants or moving close to a new arterial road. Sometimes people left an area where they did not get along with their neighbors or felt threatened in some way, perhaps by sorcery or vague spiritual forces (Charsley 1970:17). Where did they move? People mostly relocated to places where they had relatives, agnates or affines, from whom they might expect to receive friendship and support. However, some individuals attracted many more migrants than did others. These individuals shared a higher social standing than most that was derived from one or more sources: the traditional political system, particularly ties to the king; kinship as the head of a large family; and, finally, modern economic and occupational status. Furthermore, these individuals tended to be firmly and historically rooted in their communities (Charsley 1970).

Charsley’s study serves to remind us of the concept of wealth-in-people, which was first developed in the context of studies of African politics as a counterargument to the emphasis placed on material wealth as a basis for legitimizing power (Guyer 1993). As Vansina (1990:251) remarked in reference to a broad swathe of societies in equatorial Africa, “Wherever possible, wealth in goods was still converted into followers.” People who were able to attract others to their community not only gained access to labor, the reproduction of labor, and the products of that labor, but they could also assemble people with varied knowledge across a broad range of topics, not simply specialized technological expertise. Leaders were able to bring together and mobilize different bodies of knowledge in a process that has been referred to as knowledge “composition” rather than simply “accumulation,” a perhaps subtle but important distinction (Guyer and Belinga 1995).

Thinking about wealth-in-people and knowledge composition in the context of both collective action theory and the particular case of Bunyoro encourages the suggestion that knowledge possession combined with ease of mobility could provide individuals with considerable flexibility in terms of where to place their loyalties, a decision that would seem likely to have been made by the individual and his close kin rather than by any collective. Several or many individuals could in theory choose to settle with a leader who might provide an alternative source of authority to that of the state. Such leadership might be based on authority that ultimately derived from possession of some form of creative power. Indeed, in the nineteenth and twentieth centuries and probably in earlier times too, there were shrine sites at various places within the Nyoro kingdom whose guardians seem to have derived their authority from sources of creative power and whose ritual functions commonly invoked the spirits to bless people with fertility. The guardians of individual shrines were often members of particular clans and the authority of these guardians was acknowledged by the Nyoro kings (Robertshaw 2010).

The king and the state also attracted followers and sought to “compose” bodies of knowledge. Indeed, it seems probable that the king and his palace would generally have been the most desirable destination for migrants, provided that they could call upon preexisting connections with kin, affines, or even friends at the capital. The state’s competence in composing knowledge and attracting followers is evident from the very numerous specialized offices, duties, and occupations, most with titles, that existed at the court. Moreover, each position was usually occupied by a member of a particular clan, thus perhaps ensuring representation at the capital from a wide range of constituencies. The functions and duties of many of these royal retainers revolved around the performance of ritual acts that served to maintain the king’s state of ritual purity (Roscoe 1923). Thus, the state, too, seems to have harnessed creative power to build its wealth-in-people. Moreover, the king, at least perhaps in the nineteenth century, seems to have been sufficiently successful at attracting followers, despite the competition from the communities centered on shrine sites, that he needed to offer relatively little in terms of concessions, such as public goods or shared authority, to his people.

The Materiality of Nyoro Kingship

The attraction of followers and the harnessing of creative power were intimately bound together with a material component. The materiality of Nyoro kingship illustrated once more the tension between the state and local communities. Regalia are frequently mentioned in the ethnographic and historical literature of Bunyoro as being intimately entangled with royal authority. For example, when Rukidi, the first king of the Bito dynasty, to which Kabalega also belonged, came to power, he sent for the regalia left behind by the previous, Cwezi dynasty. The messenger sent by Rukidi carried one of the two drums that were part of this regalia to the new capital, while the other drum was said to have followed of its own accord. Moreover, when Rukidi beat the great drum at his accession, it made a suitably impressive noise, proving that Rukidi was not an imposter (Fisher 1911:120; see also Beattie 1971:53–54). Thus, the behavior of the regalia validated Rukidi’s claim to the throne. In similar vein, if the king was too ill to carry out his duties, he would be represented by the royal spear, which had its own name (Roscoe 1923:118). The list of regalia is impressively long (K.W. 1935:160, 1936:77 for what may be only a partial list; see also Nyakatura 1973:181–185, 188–190) and an impression of the pervasiveness of the regalia can be obtained from a visit nowadays to Kabalega’s tomb. As mentioned earlier, some of the regalia, such as spears and knives, symbolized the king’s coercive authority but the regalia as a whole was by no means under the sole possession of the king; clearly the regalia also spoke to what Blanton and Fargher have termed “principal control” (see also Blanton 1998a; Levi 1988). Numerous individuals had named positions involving regalia, from a “head regalia-man” (Mujaguzi) to persons with very specific tasks, such as the man who looked after the king’s personal drum. Many of these and other offices were linked to particular clans (Beattie 1971:124). Beattie (1971:125) explicitly recognized that the myriad duties of the specialized establishment of the palace, including the regalia,

However, as Lane Fargher (personal communication, 2012) has noted, it is not clear from this description whether clans were corporately organized. Did the clan membership appoint or recommend one of their number for the particular palace office vested in their clan, or was the individual chosen by the king, at whose whim he may then have served? The available literature offers no clear answer to this question, despite the frequency with which the assertion is repeated that particular offices were hereditary within particular clans. What seems more certain is that the shrine sites scattered through the kingdom (see above) were “intimately connected with particular clans” (Buchanan 1974:228) and that the “tensions inherent in this development of priestly power worked as a check to limit the power of the Mukama [king]” (ibid.:227). Moreover, the Nyoro proverb, “The Mukama rules the people, but the clan rules the land” also indicates that, at least prior to Kabarega’s usurping of the right to appoint chiefs, the clans mitigated the king’s power beyond the capital (see also Uzoigwe 2013:22).

While the regalia served to legitimize royal authority while simultaneously negotiating that authority with the clans at the palace, other items of material culture were part and parcel of both the assumption and delegation of royal authority in other economic and political spheres. In particular, during the coronation ceremonies, a man said to represent foreigners presents the king with an elephant tusk and two copper bracelets (K.W. 1937:293), which may be interpreted as signifying the king’s control of the ivory trade and of the import of copper, the wearing of which appears to be associated with high status.² Although this connection has not yet been investigated, it is perhaps supported by the observation that members of the king’s clan who were clearly related to the king could claim from him a brass ankle-ring (Beattie 1971:99). When the king delegated political authority, he also conferred both its associated spiritual power (mahano) and artifacts that signified that authority, notably crowns, spears, and knives (Beattie 1971:102, 118). Finally, when visited by foreign dignitaries, the king gave gifts of salt and iron hoes, the two most valuable commodities produced in his country.

If the regalia and other artifacts of the Nyoro kingdom played a role in negotiating the tension between state and populace, the palace itself seems to have served first and foremost as an agent of exclusionary power strategies. The royal enclosure (kikali) was always located in such a way that it was highly visible, in addition to being by far the largest in the country. The six-foot-high fence of elephant grass surrounding the royal enclosure was said to have a circumference of about two miles and enclose perhaps hundreds of buildings. Prominent among these was the court-house (kaluzika), which incorporated the throne-room (hamulyango), reported as being often 40 yards across and 80 feet high at its apex. However, it was not just the size of the enclosure and buildings that impressed; the whole complex of royal houses and rooms was governed by a plethora of rules and restrictions to access that served to highlight the king’s power. Only the king, it appears, could enter every room through every door, though daily rituals even prescribed where the king should be within the palace at set times in order to maintain his ritual purity (Roscoe 1923:73–86; Nyakatura 1973:202–204). Moreover, the plan of the royal enclosure (Roscoe 1923:86–87) shows numerous fences and screens that restricted both the access to and the visibility of many houses and courtyards (see the Kuba kingdom of central Africa for a comparable example [Vansina 1978a:137]). It would be intriguing to discover whether the massive royal enclosure described by Roscoe, which must have entailed the efforts of a large labor force, was only a nineteenth-century phenomenon financed primarily by external revenues or whether it was also typical of palaces in earlier times. It is also worth bearing in mind that every new king built a new capital and that the capital may have been relocated on several occasions during each reign (Roscoe 1923:73–74). As yet none of the capital sites of the kings of the Bito dynasty has been explored archaeologically, though the general locations of several of them are known (see Nyakatura 1973).

Bunyoro Before the Nineteenth Century: The Archaeological Evidence

I have argued in this chapter that the despotic nature of the Nyoro state in the nineteenth century, reflected in the scores on the collective action measures devised by Blanton and Fargher (2008), was the product of a historical moment in which the state took full advantage of its position toward the periphery of an economically booming world system to generate substantial external revenues and consolidate royal power and authority, a point on which Blanton and Fargher (2008:46) concur. However, the kingdom already possessed a well-established system of generating revenues that was inevitably entangled within a dynamic negotiation of power and authority between the state and the populace. This negotiation took place in a variety of arenas, of which I have highlighted here the composition of ritual, wealth-in-people, and materiality.

Of course, negotiations over power and authority began long before the Nyoro state was formed. In earlier publications, I proposed that it was a shortage of people, particularly a shortage of female labor, that was the engine of efforts toward political centralization in Bunyoro in the early second millennium AD (Robertshaw 1999a, 1999b, 2003). This proposal rested on the premise that precolonial African kingdoms and chiefdoms exhibited low population densities, in contrast to the high population densities commonly encountered among acephalous societies (Shipton 1984). I argued, following the ideas of Rich Blanton and colleagues (1996), that emerging leaders used exclusionary power strategies to accumulate and compose wealth-in-people. By about the mid-fifteenth century, there had emerged several larger polities wherein corporate power strategies, based on staple finance, became instrumental in the construction of large earthworks at several sites (Robertshaw 2001, 2002). These earthworks appear to have been abandoned around the end of the seventeenth century, with a concomitant shift to the kind of peripatetic Nyoro capitals recorded in later ethnographies; a shift that might reasonably be associated with the establishment of the Bito dynasty (Robertshaw 1999a:60; 2003:163) and that must also be considered within the context of climate change (Robertshaw and Taylor 2000; Robertshaw et al. 2004; Lejju et al. 2005).

This reconstruction can now perhaps be profitably revisited in the light of insights provided by collective action theory. The evidence for exclusionary power strategies, prior to the fifteenth century, comprises the monopoly of prestige goods, notably ornaments of glass beads and metal, perhaps especially copper; the real or symbolic control of iron production; and the harnessing of creative power, as reflected perhaps in the initial occupation of shrine sites such as Mubende Hill (Lanning 1966; Robertshaw 1994:108). On reflection I may have given too much interpretive weight to the relatively rare discoveries of glass beads and metal ornaments, particularly in light of the fact that at Munsa they are mostly associated with female, sometimes juvenile, human skeletons (Robertshaw, Murphy, and Ambrose 2012), which either calls into question their status as prestige goods or prompts us to reconsider the age and gender distribution of elite power. However, the sample of burials, particularly of men, from this period is small and hence injects a note of caution, while the occurrence of these ornaments at several major sites reasserts their importance. When it comes to the shrine sites, however, these could equally be seen as the harnessing of creative power as a form of collective action rather than as an exclusionary elite strategy, since they may not have been under elite control, as indeed I have argued for the use of these sites in later times (Robertshaw 2010). Finally for this period, the question of the control of iron production merits further study, particularly since Iles (2010) has drawn attention to the variation that existed in iron-working technology style across the Kitara region, including the probable introduction, tentatively dated to the fifteenth century, of new technology involving the use of an additional manganese-rich ore in smelting. Iles’s results suggest, at least to me, that elite control of iron production could only have been symbolic, as was described ethnographically.

While we need more fieldwork to generate new archaeological data and hence to shed light on the power strategies of the early centuries of the second millennium, it is perhaps easier to infer the existence of tension in the negotiation of power and authority between rulers and populace from consideration of the earthworks of the mid-millennium. Previously, I interpreted the earthworks at Munsa as monuments that “not only expressed group solidarity in material form but also encircled and metaphorically captured the power and legitimacy of earlier elites” (Robertshaw 2003:161).

However, rather than seeing the earthworks solely as the product of corporate power strategies funded by agricultural surpluses collected from taxpayers, as I did previously, they can also be seen as materializing the essential tension between rulers and populace. On the one hand, the very long outer trenches at this and other sites indicate collective action, in a very real sense, likely aimed at keeping elephants out of agricultural fields (Robertshaw 2001); on the other hand, the symbolism of the innermost trench and what lay within its circumference may be reinterpreted in terms of my earlier discussion of the Nyoro palace and could be seen as a place of seclusion, likely the locus of royal rituals, separating rulers from ruled while also drawing attention to itself through its central location on a hill. This perspective aligns well with the oral tradition recorded in the 1950s describing the death of Munsa’s ruler, Kateboha, who was killed by his people after they had employed subterfuge to gain access to this inner sanctum. Kateboha is said to have built the earthworks to protect himself, his property, and his daughter from the populace, for whom he was a hard taskmaster, requiring them to cultivate the land and executing any man who slacked off (Lanning 1959; see also Robertshaw 2001:28). The abandonment of the earthworks, probably around the end of the seventeenth century, seems to have ushered in a suite of changes that resulted in the establishment of the Nyoro state, aspects of which were explored earlier in this chapter. Unfortunately there is a dearth of archaeological evidence for this crucial period.

Conclusion

In this chapter, I have argued that the highly despotic nature of the Nyoro state in the nineteenth century was a product of a historical moment in which external revenues fueled the economy. The importance of internal revenues in earlier centuries probably provided a stronger basis for the negotiation of power between the king and the populace than existed in the nineteenth century. However, even for the nineteenth century evidence indicates an essential tension between rulers and ruled. While this tension was notably expressed in ritual, it was also both materialized and negotiated in regalia and in the location and architecture of the royal enclosure. My brief examination of the archaeological evidence for earlier centuries suggests that Munsa and other earthworks represent a material expression of the tension between royal authority and collective action.

While the balance of power between state and populace varied over time, in part at least as a result of the relative economic importance of internal and external revenues, the fact that collective action was still employed when state power was at its peak in Bunyoro suggests that the negotiation of authority may have been a dynamic process in all states. Therefore, it seems equally improbable that either total state power or its opposite, a victory of collective action, could have held sway, at least for more than a brief historical moment. If that is indeed the case, then perhaps we should challenge claims of the existence of states without rulers, such as that of the Inland Niger Delta (IND), in which heterarchy has metaphorically vanquished hierarchy (e.g., McIntosh 2005:189). It has become a widely accepted tenet of African archaeology that the IND represents complexity without hierarchy (e.g., LaViolette and Fleisher 2005:333–336). While my analysis of Bunyoro may offer theoretical challenge to the heterarchical IND, recent field research in Mali has provided a model of state-generated landscapes, manifest in later times, that may explain both why hierarchical states could have easily been overlooked and why trading cities like Jenné-Jeno and Dia show little or no evidence of state organization (MacDonald and Camara 2012). Looking beyond Africa, I am struck by the similarities between the Nyoro state under Kabarega’s despotic, albeit perhaps beneficent, rule in the nineteenth century and the political structure found in the Mixteca Alta region of Oaxaca, Mexico, during the Postclassic (Fargher, Heredia Espinoza, and Blanton 2011:317ff.). Such comparisons reinforce the fact that the analytical tools engendered by collective action theory and allied concepts have much to offer in our efforts to elucidate past political systems.