Judging from the arguments he employs in support of his third proposition, Cobbing is no post-modern. His contention that labour-raiding to serve the Cape and Mozambican economies was the root cause of commotions on the highveld and in Natal is accompanied by a plethora of categorical statements. 'There is no doubt' that missionaries at Dithakong 'were fully and consciously engrossed in what they were doing, i.e. collecting slaves.'5 'Everyone knew what Somerset meant.'6 'The conclusion is inescapable' . . . 'there is a single, unavoidable, explanation' . . . 'it is clear, then, that slaving was having a dramatic impact'.7 When he writes in these terms Cobbing does indeed rush upon his readers like a 'terrorist of truth'. Historians of all tendencies are likely to react with scepticism, if not hostility, when assaulted in this way. The vehemence of the counter-attack could have been expected. The pity is that it appears to many scholars to have cast doubt on all three of Cobbing's central propositions.
In order to see why this is not the case, it is necessary to recognise the difference between the method of argumentation used to discredit the Zulucentric mfecane and the quite different method employed to assert the vital importance of slaving as a cause of turmoil among southern African societies. The first argument proceeds, as we have seen, by falsification. It shows that certain state formations and events which had previously been attributed to the impact of Zulu raiding following the rise of the Shakan kingdom could not have been the result of Zulu activity. The second argument proceeds by inference: if the Zulu could not have produced the turmoil attributed to them, what did? Documents indicate that Griquas, Bergenaar and others were engaged in labour raiding on the highveld. Documents also say that slaves were sold in Mozambique. Raiding and slaving are known to have produced great turmoil in other parts of Africa. Therefore we may infer that the turmoil of the early nineteenth century was caused by slaving. Only Cobbing does not say he is making an inference. He says he is revealing the truth.
The argument is scantily clad. Having attacked the notion of a Zulu-generated mfecane because researchers thus far have failed to show the mechanism by which the rise of the Zulu state was linked to trade, drought, population pressure or any of the other phenomena cited by previous scholars,8 Cobbing should acknowledge the same problem with his own argument based on slaving. What amount of raiding is required to produce reactive militarised state formations? Why do the most formidable states not appear closer to the centres of raiding? Why is people-raiding more disruptive than raiding for cattle or some other item? He can only respond to these queries by putting forward hypotheses as untestable as those he has attacked. Worse still, he lets himself be drawn on to terrain where he should not be fighting. If the Zulu were not responsible for the round of state formation and population movement that occurred in the first half of the nineteenth century, the Zulu kingdom ceases to be an object of special interest. The reasons for its appearance are no more nor less interesting than the reasons for the appearance of any other African state in the region. Scholars can give up the elusive search for a first cause which set the process of Zulu state formation into motion. While the study of rival hypotheses will continue to be a useful way of introducing undergraduate students to the study of African history, it ceases to be the 'revolution in Bantu Africa'.
However, by putting so much emphasis on slave trading at Delagoa Bay, Cobbing gives the appearance of offering yet another explanation for the rise of the Zulu kingdom.9 He should be able to share Elizabeth Eldredge's doubts about the extent of the Delagoa Bay trade. After all, Delagoa Bay is only important if one accepts the key role of the Zulu state. Similarly Cobbing should not be greatly worried by Carolyn Hamilton's view that 'Shaka the monster' was as much a creation of the Zulus' black neighbours as of Port Natal freebooters. Reviving the psychopathic image cannot restore Shaka's status as 'the black Napoleon'. The available texts on both the slave trade and the character of Shaka are capable of enough different readings to keep critical theorists happy for years to come. Cobbing need not have tied his colours to specific interpretations. The work brought together in Parsons's essay, 'Prelude to Difaqane in the Interior of Southern Africa' pushes the onset of conflict, population mobility and state formation so far back that the Shakan state begins to look like a relative late-comer.
Conclusion
This introduction to Part One has attempted to combine a setting for the historiographical essays with an assessment of argumentation and 'truth' in the mfecane debate. In an era more than usually suspicious of truth, the historian's task is especially burdensome. We must be more than usually careful to bear in mind that most of our knowledge about the past is provisional. The kinds of evidence available to us carry most weight when they are used to falsify propositions about the past. When, on the other hand, we argue inductively, our inferences are uncertain and always subject to correction. That is why the case against the Zulu-generated mfecane reads more convincingly than the arguments that the mfecane was the creation of 'settler history' and that slaving was the ultimate source of terror and turmoil in the South African interior.
Leonard Thompson can thus be seen to be only partly right when he writes in his new History of South Africa, that Cobbing's argument 'cannot be substantiated'. Those parts based on inference cannot be substantiated, but the parts based on falsification can be. To judge how far this is so, the reader needs to do no more than appraise how much revision is now required in Thompson's own text on the mfecane:
Until the late eighteenth century the Bantu-speaking mixed farmers south of the Limpopo River lived in small chiefdoms . . . The nucleus of change lay in northern Nguni country . . . This transformation was in essence an internal process within the mixed farming society in southeastern Africa. By the last quarter of the eighteenth century the relation between the population level and the environment was changing . . . Previously, the society had been expansive and the scale of political organization had remained small . . . Gradually, however, the population of the region had been increasing to a level where that expansive process was no longer possible . . . Zulu impis (regiments) and bands composed of people who had been driven from their homes by the Zulu . . . had created the widespread havoc throughout southeastern Africa that became known as the Mfecane.10
Should there be another edition of Thompson's book, virtually the whole of that section would require rewriting. To that extent, at least, we can revise 'historical truth'.
1.P. M. Rosenau, Postmodernism and the Social Sciences: Insights, Inroads and Intrusions, Princeton, 1992, 7 8 .
2.J. Caplan, 'Postmodernism, Poststructuralism, and Deconstruction: Notes for Historians', Central European History, 22 (1989), 272–3.
3.See N. Etherington, 'The Great Trek in Relation to the Mfecane: A Reassessment', South African Historical Journal, 25 (1991), 12–13.
4.The closest he comes is implying that Eric Walker invented the word in 1928. More thesis writers can follow in the footsteps of Jeff Peires and Neil Parsons by spotting earlier uses of the word or its Sotho equivalent, difaqane.
5.J. Cobbing, 'The Mfecane as Alibi: Thoughts on Dithakong and Mbolompo', Journal of African History, 29(1988), 493.
6.Cobbing, 'The Mfecane as Alibi', 503.
7.Cobbing, 'The Mfecane as Alibi', 509; 'Rethinking the Roots of Violence in Southern Africa 1740–1840', paper presented to the colloquium on the Mfecane Aftermath, University of the Witwatersrand, Johannesburg, 1991, 21, 15.
8.Cobbing, The Mfecane as Alibi', 488.
9.Or, more precisely, to be reviving an explanation which attracted W.M. Macmillan.
10.L.M. Thompson, A History of South Africa, New Haven, 1990, 80–5.
1
Pre-Cobbing