the personal powers of traditional rulers in the Bechuanaland Protectorate’ (quoted in Parsons 1998: 254). The long-term significance of these developments has also been noted by Gillett, who maintains that ‘under the Protectorate the Tswana chiefs [dikgosi] enjoyed almost unchallenged power’ (Gillett 1973: 180).2 Such statements reflect the British policy of leaving the dikgosi to govern their respective merafe with minimal interference until some forty-five years after the establishment of the colonial state.
The extensive practice of indirect rule over such a long period of time amply demonstrates the capacity of the dikgosi and the hierarchies of their respective merafe to keep the population in the societal fold. Nevertheless, in due course the relationship between the dikgosi and the colonial administration became increasingly ambivalent and at times riddled by serious conflicts. Although the authority of the dikgosi and the strength of their polities still enabled the British to run the protectorate at minimal administrative and financial cost, the dikgosi were increasingly perceived as dictatorial. They were seen as operating highly autocratically in relation to their subjects. The trend on the British side in relation to their possessions was, by contrast, to introduce Western principles of legal rationality and state of justice. These strongly conflicting orientations came out fully around 1930.
For example, in 1929 an important resident commissioner stated in his diary: ‘[The dikgosi] practically do as they like – punish, fine, tax and generally play pay hell. Of course their subjects hate them but daren't complain to us; if they did their lives would be made impossible’ (Parsons and Crowder 1988: 4). Other contemporary statements by representatives of the colonial administration similarly portray these rulers as rather autocratic. For example, a resident magistrate stationed in the Bangwaketse capital of Kanye reported that Kgosi Bathoen II of the Bangwaketse (r. 1928–69)
is very obstinate and inclined to be antagonistic in matters connected with the Administration…Morally, Bathoen is rather a low type of native…being legally married to an educated woman, and at the same time living openly with his concubine. He is disinterested in matters beneficial to the tribe unless he, personally, can benefit thereby, and considers everything from his own pecuniary standpoint. His councillors are young and inexperienced headmen of the Communist type; he takes little heed of his older and more reasonable me.3
Furthermore,
he is very selfish. If he requires labour for his own work, such labour is forthcoming, but if for the benefit of the Tribe he is apathetic.4
The autocratic tendencies of the dikgosi were of course very much a product of their privileged and protected position within the colonial state, reflecting indeed the British dependency upon the ways in which they had expanded their networks of power and captured vast subject communities into the structures of the colonial state (see preceding chapter). Furthermore, the range of executive powers of the rulers of the Tswana merafe selected and privileged by the British increased steadily as the political economy of the protectorate developed. Since it ‘was an explicit policy that the British Government had no interest in the country north of the Molopo…except as a road to the interior’ (Sillery 1974: 77), the Tswana were informed that they would have to bear the costs of the protectorate themselves. A hut tax system was thus enforced,5 from which the dikgosi benefited substantially since they received a 10 per cent share of whatever they collected, over and above funds received from a number of other sources.6 Significantly in the present context, the dikgosi retained their ‘customary’ privileges, including the mobilization of age regiments (mephato, singl. mophato) and draught power for the cultivation of their large fields, as well as the receipt of hunting spoils (sehuba), ‘thanksgiving’ corn after harvest from every household (dikgafela) and, most importantly, all the unclaimed stray cattle collected in their respective territories (matimela).
All the wealth thus aggregating in the royal household was customarily justified by the virtues associated with the kgosi as motswadintle (‘the one from whom good things are coming’, see Gulbrandsen 1995 and Chapter 5, this volume). By right, therefore, he should be rich – ideally as the guarantee of everybody's welfare. His wealth was also seen as a major condition for his independence and incorruptibility as a ruler. However, as the British contributed strongly to reinforcing the dikgosi's powers, the dikgosi were not dependent on extensive networks of power by dispensing cattle widely for establishing cattle clientship (cf. Chapter 1). At most, their resources were used to ensure the support of their most important political retainers in relation to the royal centre. Moreover, as labour migration to South Africa provided most ordinary families with a stable source of basic subsistence requirements (see Schapera 1947a), there was less need to dispense grain from the royal granary during times of poor harvest.
All this meant that the dikgosi retained a number of economic sources throughout the colonial era which facilitated the accumulation of wealth – mainly in the form of cattle – that progressively attained the de facto character of private property. The British became increasingly disturbed by the dikgosi's expropriation of funds considered by the colonial power to be destined for ‘tribal’ purposes. They first attempted to separate the dikgosi's ‘purse’ from that of the tribal treasury by establishing what was known as the Native Fund. In a major administrative reform dating from 1934, it was stated that ‘[the] chief could no longer impose tribal levies without written approval from the Resident Commissioner and without agreement of the tribe in the kgotla' (Colclough and McCarthy 1980: 25). Nevertheless, although this proclamation managed to a certain extent to separate public resources from those to be held by the kgosi personally, considerable uncertainty remained over what belonged to the tribe and what was the kgosi's property.3 I shall pursue the issue of the dikgosi's apparent avarice in the subsequent section.
The institution of a ‘Native Fund’ reflected a broader British concern about the dikgosi's alleged abuse of power, a concern which should be seen in the context of the colonial empire. First, after World War I, the British initiated a revision of their previous methods of indirect rule that was extended to the Bechuanaland Protectorate in the 1930s. Secondly, at this time there appeared highly critical reports in Britain attacking prolonged neglect on the part of the Protectorate Administration that had resulted in stagnation, extreme backwardness and highly autocratic chiefs. It was argued, for example, that ‘colonial authorities had failed to bring traditional institutions in Bechuanaland into conformity “with the essential requirements of a modern civilized administration”’ (Picard 1987: 49). In addition to such broader concerns, the administration had experienced difficulties with dikgosi seen to be incompetent or otherwise unsuitable for their office: the most serious of these concerned Kgosi Sebele II of the Bakwena, who was, as we shall see in the following section, dethroned by the administration and exiled to Ghanzi in the extreme west of the protectorate.
This was the beginning of a long-term process, lasting through most of the colonial era, by which dikgosi and agents of the colonial state repeatedly confronted each other over attempts to check the dikgosi's powers by means of radical changes in the government of the merafe. I shall review some of the most important ones in order to identify processes and structural conditions that explain why the dikgosi nevertheless largely maintained – even in certain respects enhanced – their position as supreme authority figures and strengthened the structures of their respective polities.
The determination of the British to make radical reforms found a major expression in two proclamations issued by the high commissioner in 1934. These proclamations were both introduced by reference to the Order-in-Council of 9 May 1891 whereby ‘the High Commissioner is empowered on His Majesty's behalf to exercise all powers and jurisdiction which Her late Majesty Queen Victoria at any time before or after the date of that Order had or might have within the territory of the Bechuanaland Protectorate’. It was conceded that proclamations issued by the high commissioner ‘shall respect any native laws and customs by which civil relations of any Native Chiefs, tribes or populations under His Majesty's protection are