BLST 49 B

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to The Journal of African History.

http://www.jstor.org

Sources of Conflict in Southern Africa, C. 1800-30: The 'Mfecane' Reconsidered Author(s): Elizabeth A. Eldredge Source: The Journal of African History, Vol. 33, No. 1 (1992), pp. 1-35 Published by: Cambridge University Press Stable URL: http://www.jstor.org/stable/182273 Accessed: 19-09-2015 23:53 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

Yournal of African History, 33 (I992), pp. I-35 I Printed in Great Britain

SOURCES OF CONFLICT IN SOUTHERN AFRICA, c. 1800-30: THE 'MFECANE' RECONSIDERED*

BY ELIZABETH A. ELDREDGE

Michigan State University

During the I820S the entire region of southern Africa was affected directly or indirectly by tremendous demographic upheaval and revolutionary social and political change. The period was marked by massive migrations, sporadic raids and battles and frequent periods of privation and famine for many people in the region. This so-called 'mfecane' has been explained in many ways by historians, but never adequately. The socio-political changes and associated demographic turmoil and violence of the early nineteenth century in southern Africa were the result of a complex interaction between factors governed by the physical environment and local patterns of economic and political organization.

Increasing inequalities within and between societies coupled with a series of environmental crises at the beginning of the nineteenth century trans- formed long-standing competition over natural resources and trade in south- eastern Africa into violent struggles for dominance and survival. Trade at Delagoa Bay, involving primarily the export of ivory, had allowed some Africans to accumulate wealth and consolidate power, leading to political amalgamation at the expense of the weak. Increasing political inequality between chiefdoms and increasing socio-economic inequality within societies made weaker people, with fewer entitlements to food, more vulnerable to famine in times of drought-induced food scarcity. When the area was struck by severe droughts and other ecological problems in the first two decades of the nineteenth century, competition became keen over fertile, well-watered land, and those who had already consolidated their power prevailed over weaker groups in the open contests that emerged. The weak found themselves incorporated into the lower echelons of stronger societies, either conquered involuntarily or submitting voluntarily for the sake of survival. The stage was set for the emergence of various strong leaders, some, like Shaka, who ruled with terror and others, like Moshoeshoe, who won the voluntary devotion of their followers.

The resulting period of state formation, often involving initial stages of conquest by force, was prolonged into two decades of violence because of the activities of Europeans encroaching from both east and west. In the east, slave traders found a new supply of ready victims in the area around Delagoa Bay and fostered continued violence to meet their demands for slaves. Similarly, across the northern border of the Cape Colony, white frontiersmen supplied renegade Griqua and Korana with the guns and ammunition they needed to raid neighboring Africans for cattle and slaves which they traded

* I would like to thank James C. Armstrong, R. Hunt Davis Jr., Robert Edgar, John Mason, Alan K. Smith, and Rick Watson for their comments and suggestions on an earlier version of this paper.

1-2

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

2 ELIZABETH A. ELDREDGE

back to the farmers. With the massive migration of Dutch Boers into the interior to escape British rule in the I83os and I 840s, white intrusion became overt and direct, and any African hope of political stability and peace was lost forever.

This article uses a multidimensional approach to analyse these processes and events. There is no simple monocausal explanation for these disruptions: neither great leaders, nor environment and ecology, nor overpopulation nor trade (including the slave trade and raiding) alone set off the wars and migrations that plagued the area through these decades. The wide range of the one-dimensional interpretations which have been offered to date indicates the difficulties posed by this problematic period for historians. Each historian has furthered our understanding of the causes of these revolutionary disruptions, but all have provided only partial explanations for the events of early nineteenth-century southern Africa. Here I provide a critique and synthesis of earlier interpretations, and I reassess the controversial prominent role given European slavers in the interpretation of the 'mfecane' recently proposed by Julian Cobbing.' Recent syntheses, though useful, have failed to relate information about the environment to the economy and to socio- political change. In addition, most studies have focused on specific people, places, or periods, whereas I attempt to present a more comprehensive picture by looking at the entire region, both east and west of the Drakensberg mountains, and by identifying changes over time in the dynamics of socio- political change and the generation of competition and conflict.

Determining the causes of the upheaval of the early nineteenth century is not merely an academic exercise, as the political ramifications of interpreting it are manifest in South Africa today. It has generally been assumed that the emergence of the Zulu nation under Shaka was the key event responsible for the ensuing devastation of southern African societies. It has been convenient for white South African defenders of apartheid to blame the Zulu for the chaos and destruction, because this interpretation falsely characterizes Africans as inherently divisive and militaristic and thereby justifies the imposition and continuation of white rule in South Africa.

Rejection of the myth of Zulu culpability thus raises the question of why these conflicts arose in the first place. The recent article by Cobbing on the so-called 'mfecane' has appropriately refocused attention on the more fun- damental causes of conflict in this period, especially in the role of slaving in the conflicts. Cobbing's reinterpretation is seriously flawed, however, by a distortion of chronology and misreadings of the evidence. I first demonstrate that an extensive slave trade at Delagoa Bay did not begin until after the regional conflicts broke out in 1817. Slaving there could not have initiated this violence. In the second part of this article I challenge Cobbing's thesis that missionaries were engaging in systematic slave-raiding and slave-trading on the north-eastern Cape frontier. However, I support Cobbing's con- tention that Griqua and Korana allies of the white frontier farmers were themselves conducting an illicit slave trade and identify them, not Zulu or other Nguni, as the main sources of violence in the region throughout the I82os and I830s. In part three I examine the conflicts of the i8ios and I820S in terms of their environmental and socio-political contexts. First I outline

1J jiian Cobbing, 'The mfecane as alibi: thoughts on Dithakong and Mbolompo', Y. Afr. Elist., XXIX (I988), 487-5I9.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE 'MFECANE RECONSIDERED 3

the economic and socio-political setting in which droughts resulted in famine and consequent competition for arable land and labor among Africans east of the Drakensberg. I then use the same approach to analyze disruptions among the SeSotho- and SeTswana-speaking peoples west of the Drakensberg and to re-interpret the early migrations and raids in light of the natural environment of the region.

THE DELAGOA BAY SLAVE TRADE: A REAPPRAISAL

So startling and compelling are Cobbing's arguments that readers who have no independent basis for judgment may wonder what to make of his evidence and conclusions. Cobbing has attributed the violent disruptions of the early nineteenth century to slave trades organized by Europeans, both at Delagoa Bav and on the northern frontier of the Cape Colony, and has taken historians to task for failing to discuss this slave trade as the source of regional violence in this period.2 However, the sources regarding Delagoa Bay indicate that the slave trade there was heavy only by i 824, several years after the early migrations and wars of the 'mfecane'. They show no extensive slave trading out of Delagoa Bay before I823, while disruptions associated with the regional violence of the period began in I8I7.

Cobbing removes ultimate responsibility for the ensuing period of violence from Shaka and the Zulu by correctly pointing out that the earliest conflicts involved the Mthethwa, the Ngwane and the Ndwandwe prior to Shaka's rise to power. He draws on Hedges' work to pinpoint the attack of Zwide's Ndwandwe on Matiwane's Ngwane in I8I7 as 'initiating the Mfecane'.3 The beginning of the 'mfecane' thus predated the emergence of the Zulu under Shaka, who were then still subordinate to the Mthethwa, and this battle moves the locus of violence north away from the site of the later Zulu state to the banks of the Mzinyathi River. Ndwandwe attacks on Sobhuza's Ngwane (later known as the Swazi) on the Pongola River date from this period as well and also occurred further north.4 Cobbing identifies slave raiding as the cause of these early conflicts:

the Ngwane flight from the Mzinyathi was a response either directly to slave raiders or to secondary raiders such as the Ndwandwe, Mthethwa and Zulu who were themselves turned against each other by the compressions of the slave trade and by the prospects of profitable business.5

In order to account for the slave trade and slave raiding allegedly responsible for these initial conflicts, Cobbing relies heavily on an article in this Journal by Patrick Harries.6 However, neither the conclusions of this article nor the primary sources support Cobbing's assertion that there was an extensive slave trade from Delagoa Bay before I817.

2 Cobbing, 'Mfecane', 489. Ibid. 503-4; cf. D. W. Hedges, 'Trade and politics in southern Mozambique and

Zululand in the eighteenth and nineteenth centuries' (Ph.D. thesis, University of London, I978).

' This information also appeared in Omer-Cooper and has been repeated by both Hedges and Bonner. J. D. Omer-Cooper, The Zulu Aftermath: A Nineteenth-Centurv Re-volution in Bantui Africa (Evanston, i969), 29, 49, 86; Cobbing, 'Mfecane', 504, n. 83.

; Cobbing, 'Mfecane', 507. 6 P. Harries, 'Slavery, social incorporation and surplus extraction: the nature of free

and unfree labour in South-East Africa', Y. Afr. Hist., XXII (i980), 309-30.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

ELIZABETH A. ELDREDGE

Cbo Lake Nyasa Delgado

Mop9ambique

X ~~~~~Zambezi Riverj< ila Angoche

\ g Que~limane

BATSWANA Delagoa Bay

Rive Port Natal

CAPE COLONY 0 100 200 300 miles

Map i. Coast of south-eastern Africa, early nineteenth century

Harries generalizes about the combined exports of slaves from Lourenco Marques, as Delagoa Bay was called by the Portuguese, and from In- hambane, over 200 miles farther up the coast, without distinguishing the data from the two places. Harries notes that after treaties limited the Atlantic slave trade to the area south of the equator in I 8 I 5 and I 8 I 7, more Portuguese slavers began to buy slaves from south-east Africa. All of the data provided by Harries to demonstrate the boom in slave exports from Delagoa Bay and Inhambane refer to the later I82os and I830s, however, and none refers to the period before i 823. In addition, much of his evidence refers not

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE 'MFECANE RECONSIDERED 5

to Delagoa Bay but to Inhambane, as can be ascertained by a careful reading of the notes or his original sources. Harries concludes only that

At the height of the trade in the late I820s and early I830s, it seems likely that well over i,000 slaves were exported every year from each port.7

Though Harries never suggests that the slave trade expanded before the I820S, Cobbing nevertheless cites him to assert that 'The Portuguese trade had operated at a low level in the eighteenth century, but after I815 it took off'.8 Again, referring only to Harries' conclusions about the late I820S, Cobbing arbitrarily projects the data backwards in time to state that

Between about i8i8 and the early I83os at least a thousand, and probably twice if not three times that number of African males were exported from both Delagoa Bay and Inhambane every year.9

Having claimed that slave exports amounted to thousands annually years earlier than is shown by the evidence, Cobbing links this alleged extensive slave trade to events in the interior. He concludes:

This slaving must after c. I8I5 have dramatically heightened the previously critical, but continuing impact of the ivory and cattle trades.10

Moreover, although there is no evidence in Harries or elsewhere about the gender of the slaves, Cobbing assumes that every slave exported was male. He thus finally draws the unfounded conclusion that slave trading resulted in ' a loss of between 25 per cent and 50 per cent of the entire male population, in precisely the years of the mfecane'. 1

A survey of the primary and secondary literature on Delagoa Bay supports Alan Smith's assertion that the slave trade 'did not reach significant proportions until after the consolidation of the Zulu nation...'.2 The early trade out of Delagoa Bay did include slaves in small numbers, however, and it is therefore important to examine carefully the evidence about the numbers of slaves being exported from Delagoa Bay in the eighteenth and early nineteenth centuries in order to assess Smith's judgment about what constitutes 'significant proportions'.

The earliest reference to slaves being exported from south-eastern Africa is a reference to Port Natal. In an appendix, Bannister quotes a mention of slaves being purchased there in I7I9 by a slaver, Robert Drury:

Here we traded for slaves, with large brass rings, or rather collars, and several other commodities. In a fortnight we purchased 74 boys and girls. These are better slaves for working than those of Madagascar, being not only blacker but stronger.13

Harries, 'Slavery', 3I6. 8 Cobbing, 'Mfecane', 504. 9 Ibid. 504-5. Italics his. l0 bi'd. 5o6. "ll bid. 5o6.

12 Alan K. Smith, 'The trade of Delagoa Bay as a factor in Nguni politics i750-I835', in Leonard Thompson (ed.), African Societies in Southern Africa (London, I969), I77.

13 Saxe Bannister, Humane Policy, or Justice to the Aborigines of New Settlements ... (London, I830; reprinted London, I968), xxxii. Bannister incorrectly cites the title as 'The Adventures of Robert Drury', whereas the book has appeared in several editions under the title Madagascar; or, Robert Drury's Journal, During Fifteen Years' Captivity on that Island and was long attributed to Daniel Defoe. A full examination of the authorship question is found in Arthur W. Secord, 'Robert Drury's Journal' and Other Studies (Urbana, I96I), I-45. This same quotation from Drury appeared in Stephen Kay, Travels and Researches in Caffraria (New York, I834), 336, but there are no other references to an early slave trade out of Natal in these sources or in Drury. Drury further

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

6 ELIZABETH A. ELDREDGE

Though Bannister concludes from this single reference that 'We find Natal about this time mentioned as a place of ordinary resort for slave-ships', no other evidence suggests that this happened more than once.14 On the contrary, Fynn wrote in the i850s that

From the time the Dutch left Natal till the arrival of the 'Salisbury' [in I 822] there is no tradition amongst the natives that any vessel put into Port Natal.15

Bannister also collected the available evidence concerning all trade out of Delagoa Bay during the period of the Dutch settlement there, I72I to I729, and collated the information into an export table. During these nine years exports included 49,574 pounds of ivory, 288 slaves, and various quantities of tin, aloes, gold dust, ambergris, honey, copper and rice.16 Because these statistics were based on official reports Bannister acknowledges that they may not be complete. However, the inclusion of slaves in this list, during an early period when there was no compelling reason to cover up the slave trade, suggests that the numbers are accurate and that the trade was indeed minimal, averaging 48 slaves per year and often amounting to fewer than twenty per year out of Delagoa Bay. It is highly unlikely that the export of people in these small numbers had a revolutionary impact on the socio- political structures of the societies where they originated.

The Dutch period in the I720S was the high point of the slave trade in the eighteenth century, however. The evidence thus supports Smith's contention that only during the Dutch occupation was there a consistent effort to obtain slaves.17 He notes that slavers avoided Delagoa Bay because it was not equipped to handle large numbers of captives, which they could obtain more easily at ports to the north.18 A Dutch trader found that no slaves were available for purchase at Delagoa Bay in I73I because, according to the Africans, there had been no recent wars to generate prisoners.19

In the I750S and I760s trade at the Bay was dominated by the English, who were seeking ivory for their trade with India.20 This orientation of trade at Delagoa Bay towards India and consequent interest in ivory persisted with the arrival of the Englishman John Bolts under the Austrian flag in the I770s.21 Smith estimates that in the late I770S annual ivory exports from Delagoa Bay may have reached over ioo ,ooo pounds.22 The Portuguese drove away the 'Austrians' in I78I, and documents relating to the creation

writes that six Africans from Delagoa Bay who had been taken aboard on a previous trip were left off at Port Natal and that the Natal slaves were left off in NMadagascar, where I30 other slaves evidently of Madagascar origins were purchased and taken aboard. Madagascar; or Robert Drury's Yournal, ed. Pasfield Oliver (London, I890; reprinted New York, I969), 304-7. 14 Ibid.

15 Mr Fynn, 'From a Fragmentary Paper...', The Annals of Natal, 1495 to 1845 (2 vols.), by the late Mr John Bird (Reprinted Cape Town, I965), i, 73.

16 Ibid. I44. Adding to the total exports the 25 slaves who were taken to the Cape in I 730 when the post was abandoned brings the total for slaves to 3 I 3. James C. Armstrong, personal communication, citing C. G. Coetzee, 'Die kompanjie se besetting van Delagoa- baai', Archives Year Book for South African History, XI (1948), 269.

17 Smith, 'Trade', I76. 18 Ibid. I 77. 19 Alan K. Smith, 'The struggle for control of southern Mo9ambique, I720-I835'

(Ph.D. thesis, University of California - Los Angeles, I970), I54. 20 Ibid. I62-5. 21 Ibid. i66-8, I76-97. 22 Ibid. I 9 I.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE MFECANE RECONSIDERED 7

of a Portuguese establishment there in the next few years refer only to the value of trading in legitimate commercial goods, especially ivory.23

One of the few relevant documents of the late eighteenth century, 'Mr. Penwell's account of Delagoa given me by Himself', states that slaves were not sold by the people living around Delagoa Bay (the 'Tembe, Mafuma, and Matoll') because 'in battle they give no quarter, consequently no slaves'. This source informs us that 'they traffick within Land for Teeth with what they get from the Europeans', confirming that the trade at the Bay involved ivory and that there was no regular traffic in slaves.24 On the other hand, a trickle of slaves nevertheless left the area. A Portuguese friar who lived in Mozambique for I7 years and resided at the Bay in I782 and 1783 reported that the chiefs there sold as slaves their enemies captured in wars.25 Lobato indicates that at this time the annual export trade from the 'two great commercial highways of the region', the Maputo and Inkomati (Manissa) Rivers, included the equivalent of six to seven boatloads of ivory, rhino horn, hippo teeth, amber, gold, copper, agricultural products and slaves from w-ars. 26

The Portuguese abandoned Delagoa Bay in 1796 after a French attack and did not reoccupy the area until I799. As Smith notes, very little evidence at all bears on Delagoa Bay for the first two decades of the nineteenth century.27

2 For example see 'Carta dos Governadores Interinos ao Secretario de Estado, em I9 cle Julho de I 785', Documento no. 21 , Alexandre Lobato, Histdria do Presi'dio de Lourenf o Marques. Vol. I, 1782-1786 (Lisboa, I949), 202-3.

-2 'I\Ir. Penwell's account of Delagoa given me by Himself', no date, no name of person to whom given; G. M. Theal (ed.), Records of South-Eastern Africa (RSEA) (9 vols.) (London, I903; reprinted Cape Town, I964), ii, 455-65. Theal estimates the document is from the late eighteenth century; the contents, which refer to specific chiefs, indicate a date from the I780s. 25 Lobato, Hist6ria, i, 2I.

21; Ibid. Zimmerman concludes that 'the years I785 to I794 mark the peak of the F1rench slave trade at M\Iocambique', which was brought to a halt by the British and never really revived. Except for the years I789 to i8oo at the single port of I\Iocambique Island, trade in foreign ships was officially illegal, so foreign trade is almost impossible to trace. The complete absence of any references to French ships buying slaves at Delagoa Bay in Zimmerman and in Mettas is therefore not conclusive. See MIatilde Zimmerman, 'The French slave trade at I\ocambique, I770-I794' (II.A. thesis, University of Wisconsin- 1\ladison, I967), I9, 2I, and passim; and Jean MIettas, Repe'rtoire des expe'ditions tiegrieres franfaises au XVIIIe siecle' (2 vols.) (Paris, I978 and I984). On the other hand, in I785 a French ship which had aided the Portuguese in a fight against Africans at Delagoa Bav subsequently went to Mocambique Island to purchase slaves, suggesting that slaves had not been available at Lourenco Niarques. Lobato, Historia, i, I27. Filliot attempted to compile a comprehensive statistical profile of the slave trade to the Mascarene Islands in the eighteenth century. He documents the trade from the Portuguese coast of NIozambique, but in his massive search he found no references to slaves from Delagoa Bay. Nevertheless this is inconclusive since he acknowledges that no use had been made bv himself or his sources of the Portuguese archives at Lourenco Marques. J.-M. Filliot, La traite des esclaves vers les Mlascareignes au XVIIIe siecle (Paris, 1974), 52. Although D)utch ships stopped occasionally at the Bay looking for lost ships and had an ongoing awareness of activities there, the Dutch sources show no evidence of significant trade in slaves from Delagoa Bay during the late eighteenth century; James C. Armstrong, personal communication. I am indebted to Armstrong for assistance with sources for this period.

2' Smith, 'Trade', I75-6. Smith's research included a search for all related materials in the Lisbon overseas archives, the Arquivo Historico Ultramarino.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

8 ELIZABETH A. ELDREDGE

The lacuna itself suggests that there may have been little trade of any kind out of Delagoa Bay in this period, although such an absence of evidence is inconclusive. The Portuguese garrison, which was supposed to maintain a monopoly of trade in ivory and other specified goods, was too small to prevent a British ship from trading at the Bay in I8o i. The next time a British East India Company ship tried to trade was not until I 8 I 5, however, and this time the Portuguese succeeded in keeping out the foreigners. In the meantime, French and English whalers had been coming to the Bay for provisions since I789, and English and American whaling ships continued to resupply their ships there after the turn of the century.28

The most significant commercial development at Delagoa Bay at this time was the Portuguese establishment of a whaling company. Their first attempt in I817 failed when their representative alienated the Tembe chief by neglecting to request permission to use Tembe land, and he and his men were killed.29 Nevertheless the whaling company was successfully established in i8i8, and to service it the Portuguese brought in both black harpooners (harpoadores negros) and specialized tools from North America.30 It is highly unlikely that the Portuguese would have been interested in setting up a whaling station there had there been an extensive and lucrative slave trade out of the Bay by i8i8. Clearly whaling activities and ivory exports still dominated the commercial scene at the Bay as late as that date. Captain Owen wrote, during his survey trip of I822-3, that

When the Sun is in the Northern Hemisphere, which is the Season for the Black Whale to Calve, this Bay is very much frequented by American and English Whalers, into which vessels many of the Natives engage themselves as Boat Crews for very trifling remuneration.3'

Far from there being an extensive Portuguese slave trade, Owen observed a thriving commerce with the Africans in legitimate goods. According to Owen,

The Portuguese shew not the shadow of pretention to interference with any of these people, and indeed have great dread of them. The commerce of all these people is similar; that is beads, brass and cottons for Elephants' Teeth, Ambergris, Rhinoceros Horn, and Hippopotamus Teeth, they also barter their Cattle, Poultry, Pigs, Goats, and Grain, as also the skins of wild animals.32

Captain Nourse, an Englishman from the H.M.S. Andromache who, as part of Britain's surveillance of slaving on the high seas, had no reason to want to hide any slave trade carried on at the Bay, mentioned only ivory when discussing the activities of the Portuguese.33

Certainly thousands of slaves were already being taken from the coast of Mozambique by then, but from areas much farther to the north. An I809 report of a British sea captain noted that French ships from the Ile de France

28 Smith, 'Struggle', 2I0-II, 225-7. 29 Ibid. 225-6. 30 Visconde de Paiva Manso, Memoria sobre Lourenfo Marques (Delagoa Bay) (Lisboa,

I 870), I 2. 31 'The Bay of Delagoa', by Captain Owen, of HMS Leven, in RSEA, ii, 475. This

article contains information not available in Owen's published journals. 32 Owen, 'The Bay of Delagoa', 474. 33 'Extracts from a letter from Commodore Joseph Nourse to J. W. Croker, esq.', I5

Jan. I823, RSEA, ix, 20.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE MFECANE RECONSIDERED

and Bourbon preyed on the Portuguese, taking almost all of the Portuguese vessels trading along the coast in 'negroes, elephants' teeth, gold-dust, and specie'; but this activity ranged as far north as Cape Delgado and cannot be pinpointed to Delagoa Bay.34 The Portuguese still had only small settlements at Correntes (the mouth of the Limpopo) and Inhambane.35

A report from the Earl of Caledon in i8io indicated that ships carrying slaves to Brazil supplied themselves at the old, established slave ports farther north at Mocambique Island and Inhambane and did not bother to stop at the smaller southern port of Delagoa Bay:

I understand there are but four ships regularly employed in the slave trade between Rio de Janeiro and the settlements of Miozambique [Island] and Inhambane and these I have reason to believe are bona fide Portuguese property ... The usual course for the slave ships after receiving the cargo at Mozambique [island] is to proceed to Rio de Janeiro.36

The Portuguese trade in slaves going to Brazil was by that time extensive, but it was still originating further north according to all the evidence. An English visitor to Mozambique in I8I2 reported that

There are regular ships come annually from Rio [de] Janeiro to MViozambique [island] with supplies, and take in return slaves, ivory, gold dust, and gum. The Americans have always been in the habit of coming there, and still continue to do so every year in the months of June and July, and smuggle slaves to the Brazils and Spanish America. A brig also, lately from Seychelles, but belonging to the Isle of France, had been there and taken away two hundred slaves.37

This British captain also noted that I2,000 slaves were exported annually from Portuguese settlements when the French had possessed the Isle of France. Thus, although there is clear evidence of an extensive slave trade from the northern coasts of Mozambique by I8I2, not until ten years later did it reach as far south as Delagoa Bay.

Owen's diary helps to pinpoint the date, late in I823, at which the slave trade began to afflict the Delagoa Bay area. On Owen's first trip to the Bay, arriving in September I822, he observed that the local people were 'accustomed to nothing but whalers and a few slave-ships... '. 38 At that time the area was thickly inhabited, and the land was cultivated 'so that the natives were enabled to live in the greatest abundance and comfort'.39 On the same trip Captain Owen wrote to George Thompson that

'Report to Captain J. Tomkinson, commanding His Majesty's sloop of war Caledon, to Vice-Admiral Albemarle Bertie', 7 June I809, RSEA, ix, i. Campbell has dealt with the early nineteenth century but has very little for the period before I820, none of which refers to Delagoa Bay. Gwyn Campbell, 'Madagascar and Mozambique in the slave trade of the western Indian Ocean i8oo-i86i ', in William Gervase Clarence-Smith (ed.), The Economics of the Indian Ocean Slave Trade in the Nineteenth Century (London, I989), I66-93.

31 'Report of Captain William Fisher, of His Britannic NMajesty's Sloop of War Racehorse, to Vice-Admiral Albemarle Bertie', I7 Aug. I809, RSEA, ix, II.

36 Letter from the Earl of Caledon to the Right Honourable Nicholas Vansittart, Cape of Good Hope, 27 June i8io, RSEA, ix, II-I2.

3 Letter from Captain H. Lynne to Rear Admiral Stopford, 2I I\lav I8I2, RSEA, ix, I 6.

38 W. F. W. Owen, Narrative of Voyages to Explore the Shores of Africa, Arabia and Madagascar (2 v'ols.) (London, I833), i, 148. Ibid. i, 137, 141-

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

IO ELIZABETH A. ELDREDGE

The natives of Delagoa Bay are a timid race, and seemingly at peace with every body, but the Vatwahs treat them like a conquered people, and have lately overrun the country. They offer no objections to any one passing through their country."

Owen's testimony confirms that slaves were exported from Delagoa Bay, but the numbers were few and the local societies had not yet been disrupted, nor was slave raiding a normal, ongoing activity. As late as the ist of May 1823, Captain Owen noted that

Like all other African Nations all the countries around the Bay make Slaves of their enemies, but of the enemies only. There are however very few slaves exported from this place, and the natives have a decided aversion to the trade.41

Cobbing's argument rests on the assumption that the sources covered up a large trade before I823. Howvever, Owen made his observation that there werefew slaves exported from Delagoa Bay being fully aware that some slaves were being captured and sold, and so he did not minimize the numbers out of ignorance or deceit. He witnessed at one point the process:

... the Portuguese Commandant had persuaded him [Chief Mayetta] to seize and sell a number of his own wretched and defenceless subjects for slaves, at the price of about a dollar and a half each, paid in valueless merchandise; and he had actually taken an expedition on foot for the purpose, as the Commandant was in daily expectation of the annual ships from Mozambique to take his living cargo ... we afterwards learnt that he actually procured the number required by the Com- mandant, who exported them to Brazil.42

Owen was in fact eager to expose the extent of the slave trade from the Portuguese coast. He reported that

Quilimane is now the greatest mart for slaves on the east coast... From eleven to fourteen slave-vessels come annuallv from Rio Janeiro [sic] to this place, and return with from four to five hundred slaves each on average.43

Inhambane is by no means so rich as Quilimane, as from the small extent of its river it has not the same facilities for procuring slaves, the source of wealth to the latter place; those they do obtain are the spoils of war amongst the petty tribes... The trade of Imhamban[e] consists principally in ivory, and bees'wax, about one hundred thousand pounds of the former being annuallv conveved to l\Mozambique.44

Owen explicitly observed the change by his later visits in I823 and after:

In every succeeding visit since our first arrival at Delagoa we had observed that the natives were becoming still more unhappy; many, it appeared, had voluntarily sold themselves to slavery to avoid the miseries of starvation: for so great had been the ravages of the Hollontontes [Nguni] that even onions ... had become exceedingly scarce. It likewise appeared that the French of Bourbon had opened a trade to this Bay for slaves....

40 George Thompson, Travels antd Adventures in Southern Africa (2 vols.) (London, I827; reprinted Cape Town, I967), i, I82 n.

1' Owen, 'The Bay of Delagoa', RSEA, ii, 465-79, 42 Owen, Narrative, i, 270. 43 Ibid. i, 286-7. " Ibid. 1, 30I. 41 Ibid. ii, 2I8. This observation supports the conventional view that slaves became

readily available at Delagoa Bay as a result of the migrations from the south and that the slave trade expanded following these migrations, rather than the reverse.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE MFECANE RECONSIDERED I I

On another of his visits in late I823 or in I824 Owen learned that three French vessels had recently been to the Bay. One from Bourbon left six weeks before his arrival and carried off I30 slaves, and he believed it had marked the beginning of a slave trade between the Bay and the French of Bourbon. The other two hailed from Mauritius, and their notorious 'master' had kidnapped unsuspecting Africans whom he had plied with liquor to reduce resistance. Had there been a systematic slave trade ready to deliver slaves to ships arriving at Delagoa Bay, presumably the Europeans would not have had to resort to trickery nor would the Africans have been so trusting and so easily kidnapped. Owen concluded,

What a contrast did these poor people exhibit in their present situation with that in which we first found them [in 1822] ! The little market, the fair exchange, the busy mercantile spirit, and the ardour of speculation had vanished, leaving onlv abject beggary and want of the mere necessaries of life.46

In October I823 Owen reported that the trade from Mozambique port, in the north, included an estimated I5,000 slaves annually and that io,ooo slaves had been exported from Quelimane in the previous year. At this point the slave trade further south was still minimal relative to the northern trade, though he noted that its negative effects were beginning to be felt internally:

From Inhambane, however, the trade in slaves is very limited compared with that of 1Iozambique and Quillimane, the neighboring tribes being very averse to it, nevertheless wars are excited solely to make slaves to pay for merchandize. The same also occurs at English River [Delagoa Bay] to a still smaller extent, vet sufficiently so to keep the neighboring tribes in a ferment and continual state of warfare.47

Captain Owen exposed the extensive slave trade at Portuguese ports farther north, and he had every reason to expose it in Delagoa Bay as well, because he was determined to find an excuse for the British to take over the area from the Portuguese. Nevertheless the most damning thing he could say in I823 was not that the slave trade was then already extensive but that it might become so. In a letter dated i i October I 823, Owen wrote that

It is to be lamented that our negotiation [between Great Britain and Portugal] was so unwise as to permit the slave trade to exist even as far as Delagoa Bay, where such trade had never before existed beyond the purchase of a dozen a year, but where by this permission means will be found to keep the whole country in a state of disorder and warfare, for the purpose of having slaves in greater numbers. [emphasis added]

The port is more convenient than any other for direct communication with Brazil, and if the temptation to make slaves be permitted to be held out to the natives, by opening a market for them, they will cut one another's throats without mercy, and the whole country will be depopulated in a very few years.18

Apparently the Portuguese were eager to take advantage of local conflicts that generated captives but were not well enough established to absorb large numbers of slaves when they first became readily available. Migrants from

46 Ibid. ii, 2I8. 47 Letter from Captain WV. F. WV. Owen to J. W. Croker, 9 Oct. I823, H\AIS Leven,

Mozambique, RSEA, ix, 32. 48 Letter from Captain W. F. W. Owen to John Wilson Croker, esq., H.M.S. Leven,

AVlozambique, I Oct. I823, RSEA, ix, 37-9.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

I2 ELIZABETH A. ELDREDGE

the south raided the Delagoa Bay area, disrupting food production and precipitating famine in a time of drought-induced scarcity. Owen indicated that as of I823 the Portuguese at Delagoa Bay were still so isolated that they themselves were vulnerable to food shortages. This in turn suggests that visiting ships, which might have brought relief supplies, arrived very infrequently and that the Portuguese were still in no position to engage in an extensive slave trade as late as I823. In a letter dated io May I825 Owen described the events of I823 to the Portuguese Governor of MoSambique: It is, however, most strange that even through all this time [fighting and threats to the fort] the Commandant and officers of your fort traded with these very Vatwas [Nguni], for their spoils and slaves taken from the people of Temby and Matoll, etc. etc.... The devastation of the Vatwahs, and consequent famine, brought slaves to the fort for almost nothing; but fortunately, the fort itself was in want, and could not sell food for slaves.49

The evidence from Fynn, on his way to establish himself as a trader at Natal, also suggests that the slave trade out of Delagoa Bay was still limited in scope in mid- i 823. Fynn spent six months at Delagoa Bay, first arriving in mid-June I823. There he met the priest at the fort who '... accompanied us to the cells, where we saw about 8o slaves in irons. They had been captured in fights between neighbouring tribes, and had recently been purchased'."O Fynn observed that in I823 'the articles procurable by barter at the settlement were sea-cow and elephant ivory, also ostrich feathers. The principal objects for sale, however, were slaves'."1 Here Fynn is not indicating a large-scale slave trade, however, for he states that there was only one trading ship per year, wvhich took off all these goods at once. The message is rather that the sum total of trade, including slaves, was minimal in I823, but relative to the other goods traded, slaves had the highest value. Fynn provided some insight into the source of the slaves. According to Fynn, The Chief who is the champion of Temby is named Mlohambie ... He is the only Chief at Delagoa who takes prisoners, which are conveyed across English River to the Portuguese to whom they are sold as Slaves for trinkets and cloth which he shares with his king. These Slaves are kept in the Fort until the annual vessel comes to carry them off.52

Although according to Fynn in I823 only one ship a year picked up slaves, by the time of Owen's trips in late I823 or early I824 it appears that more ships had begun to arrive. Fynn also exposed the machinations of the Portuguese, who managed to foster warfare as a source of slaves, so he also was clearly not trying to cover up slaving activities: The various tribes in the vicinity of Delagoa, like all other native tribes of Africa, are constantly engaged in petty warfare, and, wherever there is a Portuguese settlement, these contests are encouraged, and not infrequently, one or the other of the rival parties is aided by Portuguese soldiers. The prisoners taken by each tribe are purchased by the Portuguese to become slaves. Mayetha, chief of the Tembe country at the time of my visit, had recently been defeated and many of his subjects sold into slaverv.5:

" Letter from Captain Owen to Senhor de Botelho, Gov ernor of MoSambique, Leven, IO May i825, RSEA, ix, 57.

50 James Stuart (ed.), The Diary of Henryr Francis Fnynn (Pietermaritzburg, I969), 39. 5' Ibid. 40. 52 H. F. Fynn, 'Delagoa Bay', RSEA, ix, 487. 53 Stuart (ed.), Diary, 43.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE 'MFECANE RECONSIDERED I3

It would be incorrect to assume without further evidence that the Portuguese were involved not only in slave trading in the immediate vicinity of the bay but also in damaging slave raids in the interior prior to the mid- I820S. There is contrary evidence that the outlying areas had not yet been ravaged by the slave trade and slave raiders in early I823. When the Wesleyan missionary in Tembe, William Threlfall, first arrived in mid- I 823, he observed that the people lived

all along the coast to this bay, where the population is great indeed, in every quarter; not living in large towns, but distributed in villages equally over the face of the country, for the advantage of cultivation; and the country, as far as I have seen it, is cultivated.

I found the face of the country everywhere covered with the richest pasturage, and extensive fields of caffre and Indian corn .... 54

This scene of peaceful settlement indicated that there had not yet been the widespread disruption that would have resulted from extensive slave raiding. Threlfall observed that many of the people in the neighborhood of Delagoa Bay were in distress, but the causes did not arise from slaving. According to Threlfall, the migrating Nguni from the south, called 'Vatwahs' at Delagoa Bay, had laid waste all the neighboring areas except for the southern half of the Bay, 'Inyack and Mapota' (Nyaka and Maputo). The Vatwahs were apparently raiding only for food, for there is no mention of killing, only famine. Thus the people were being raided for food at a time when drought already had created scarcity, leading Threlfall to describe the area as 'beautiful and fertile in native productions beyond description generally; but for the last two or three years, with little rain - which, with the devastations of the Vatwahs, has produced famine'. 55 Drought combined with political disruptions to create a decade of misery from i823.56 By the late I820S, the slave trade out of Delagoa Bay had reached its full proportions, and the effects were disastrous.

Since Harries has accused Portuguese colonial officials and historians of covering up Portugal's participation in the slave trade, and Cobbing accepts his case, it is worth considering whether the lack of evidence of an extensive slave trade from Delagoa Bay before I823 is in fact the product of a Portuguese conspiracy. Both the major nineteenth-century Portuguese sources on Mozambique and recent works by Portuguese historians have systematically exposed the trade in slaves. Harries himself conducted a thorough investigation and revealed the high level of the slave trade after I823, but he did not find any evidence for a heavy slave trade out of Delagoa Bay before I823. If there is in fact a conspiracy of silence, it remains to be explained why the Portuguese would have felt free to expose and condemn trading out of Mo,ambique Island, Quelimane and Inhambane to the north and yet would cover up a similar trade out of Delagoa Bay.

Harries' evidence that the modern Portuguese historian Alexandre Lobato

5W William Threlfall, Village Sleugally, Tembe, Delagoa Bay, I823, in Bannister, Humane Policy, lxvi.

5 Threlfall, Portuguese Fort, Delagoa Bay, 29 Nov. I823, in Bannister, Humane Policy, lxvii.

56 M. D. D. Newitt, 'Drought in iMiozambique, I823-I83I ',J. Southern Afr. Studies, xv (I988), I4-35.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

I4 ELIZABETH A. ELDREDGE

deliberately concealed extensive slaving out of Delagoa Bay rests on Lobato's statement that 'LourenSo Marques was one of the rare ports of Mozambique that did not export slaves'." Harries neglects to mention, however, that in this passage Lobato is referring to the time period covered by the second volume of his collection, I787-99; in volume i, which covers I782-86, Lobato reports the export of slaves from Delagoa Bay in those years, and he does not imply that no extensive slave trade out of Delagoa Bay ever developed in the nineteenth century. In addition, Lobato is not covering up Portuguese slaving at all in the sentence cited, as it confirms that the Portuguese slave trade was extensive out of other Portuguese ports. According to official figures collected by Lobato, some 46,893 slaves were exported from Mozambique between 178I and 1790, approximately half on Portuguese ships and half on foreign vessels. Lobato assumes that these official figures are low, noting that the illicit trade must have been large since the clandestine export of slaves was so easy along the coast north of Quelimane and especially north of Angoche.58

Neither can early Portuguese sources be rightly accused of systematically covering up the Portuguese slave trade out of Mozambique. Sebastiao Xavier Botelho was governor of Mozambique from I825 to I829, and in his book of I835, the major primary source for the period, he exposed and denounced the slave trade out of Mozambique.59 Botelho attributes the poverty of Quelimane by the I 830S to slave trading, as compared to its prosperity before the slave trade increased there in about i8io. He also readily acknowledges that I4 to I 8 ships a year arrived in the northern Mozambique ports from Rio de Janeiro to transport slaves beginning in i8io. He also condemns later activities of public officials at Lourenco Marques, where governors

were not content to collect the fruits of spontaneous disagreements among the Caffres but who fomented them themselves, beginning wars among them in order to offer kindness, capture them and sell them ... It is up to me to say that all these crimes have advanced here not only unsanctioned but also authorized because when vice begins with those in power, far from being strange or abhorrent, it is taken as an example and used to excuse other crimes.60

Other scholars studying the slave trade from Mozambique have tended to confirm that Delagoa Bay was late in entering the slave trade on a large scale. Capela and Medeiros trace the slave trade from Mozambique to the islands in the Indian Ocean between I 720 and 1902. They note the rise of Quelimane as the dominant port for slave exports after ships from that port became exempt from customs at MoSambique (island) in I814, but they did not find evidence of an extensive trade from LourenSo Marques until early i827.61

57 A. Lobato, Historia do Presidio de Lourenfo Marquies. Vol. II, 1 787-I799 (Lisbon, I960), 356, quoted in Harries, 'Slavery', 31I.

; Lobato, Hist6ria, ii, 356-7. Lobato does not elucidate the period I800-30 since his work ends in 1799, but his exposure of the slave trade in the earlier period indicates that he was not part of a conspiracy of silence about the Portuguese slav,e trade out of Mozambique.

51 Sebastiao Xavier Botelho, Memoria estatistica sobre os dominios portuguezes na Africa oriental (Lisboa, I835). 60 Botelho, Memoria, 92. Translation mine.

61 Jose Capela and Eduardo Medeiros, 0 trdfico de escravos de iIIofambiqaue para as ilhas do Indico II720-902 (Maputo, i987), 32-41. Their ship lists are far from complete, and these data are supplemented with evidence from a broad range of archival sources.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE MFECANE RECONSIDERED I5

We have ample evidence, then, from both Portuguese sources and outsiders about the Mozambican slave trade up and down the northern coasts. It is extremely unlikely that an extensive slave trade was being carried on, completely unrecorded, at Delagoa Bay. English, Dutch, French and Portuguese ships all frequented the coast, and all had conflicting interests in the area, which argues against a conspiracy of silence concerning slaving activities at Delagoa Bay. Sources which do expose the slave trade elsewhere but consistently portray Delagoa Bay as primarily a site of ivory exports and as a whaling station can be taken as implying the lack of extensive slave trading and slave raiding in the Delagoa Bay region prior to I823. There is thus no evidence for Cobbing's assertion that the slave trade at Delagoa Bay reached the proportions of i,ooo slaves exported annually from 'about i 8 i 8', when all sources suggest that Owen was correct when he estimated the numbers prior to I823 at about a dozen a year. The numbers of slaves who were exported from Delagoa Bay before I822 or I823 were therefore far too few to disrupt significantly the societies from which they hailed. The slave trade and slave raiding cannot have been responsible for demographic upheavals among the northern Nguni between I817 and I822. Although the slave trade out of Delagoa Bay prolonged and intensified regional violence after I823, it could not have originated dislocation far inland as early as I817-22, and the explanation for early conflict there must lie elsewvhere.

THE CAPE FRONTIER SLAVE TRADE AND VIOLENCE IN THE INTERIOR

Sporadic fighting and demographic dislocation in southern Africa persisted into the I830s, and the Zulu have long served as convenient scape-goats for the continuing violence. In fact neither the Zulu nor such related breakaway groups as the Ndebele were ultimately responsible for most of the conflict in the interior of southern Africa. After an initial period of migration and localized struggles between various African chiefdoms, slave-raiding by Europeans and their agents played the primary role in fostering and exacerbating conflicts throughout the region.

XV'hat was the nature of slave-raiding in the interior, and who was responsible for it? Cobbing argues that most of the turmoil farther west, that is north-east of the Cape Colony, was caused by white missionaries engaging in slave-raiding in that area. However, it was European settlers and their Griqua allies and not the missionaries named by Cobbing who raided for slaves. The Griqua were the main marauders in the area of the Orange River from the time they arrived there, and they had been a major cause of demographic dislocation among the BaTswana before the Zulu kingdom was ever formed.

The white colonists were both directly and indirectly responsible for the activities of the Griqua. The slavery of the Cape spawned unified groups of people of Khoi and mixed-race descent who moved beyond the border and became known by various names: Bastaards, Kora, Korana and eventually Griqua and Bergenaars.62 But traders in the Cape Colony armed the Griqua

62 See Martin Legassick, 'The northern frontier to c. I840: the rise and decline of the Griqua people', in Richard Elphick and Herman Giliomee (eds.), The Shaping of South African Society, I652-I840 (2nd rev. ed., Middletown, i988), 358-420.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

i6 ELIZABETH A. ELDREDGE

BAROLONG B Delagy

Lithakong * X ~ BATLHAPING Pon,Jl / |

*Ku ru mang

Griquatown

COLONY XHOSA /

Grahamstown * 0 50 100 miles

Map 2. South-eastern Africa, c. I830

and wanted them to supply legitimate goods acquired by hunting, such as ivory, skins and ostrich feathers. Colonial officials also wanted the Griqua to provide security on the northern frontier, eventually leading to an I843 treaty. For both of these reasons the white frontiersmen and the colonial government were happy to continue supplying the Griqua with guns, which the Griqua used to attack and raid their neighbors ruthlessly both for cattle and for the illicit item of commerce, slaves.

Cobbing is therefore correct to identify Europeans and their allies as the main instigators of violence across the Cape Colony's frontier. However, Cobbing also accuses certain missionaries from the London Missionary

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE 'MFECANE RECONSIDERED 17

Society of masterminding the slave trade in this area and of concealing their involvement in it, in spite of the fact that they actively tried to expose and stop the trade. Cobbing's accusations that white missionaries were system- atically enslaving Africans for sale to the Cape Colony rest on inadequate evidence, and there are data to suggest that his charges are false. Numerous missionaries condemned the immoral conduct of the white settlers, and Cobbing himself relies on their own writings for evidence of the slave trade. With competing mission groups vying for influence in the region and all seeking support from home, a conspiracy of silence covering up a missionary role in the slave trade is hard to credit. So many BaSotho who had worked in the Cape Colony returned to Lesotho in the I83os and I840s that had the missionaries been involved in enslaving them, and had they suffered direct enslavement, surely there would be recorded reports of it. The early missionaries in Lesotho were Protestants who hailed from France and had no direct connection with the missionaries from the London Missionary Society serving the Griqua nearby, and their copious reports beginning with their arrival in I833 would certainly have revealed a scandal had one existed on a significant scale.

Cobbing begins his indictment of the missionaries as slavers by presenting a famous battle, a confrontation of various people at the BaTlhaping town of Lithakong, as a missionary-led slave raid. Cobbing builds his case against the missionaries by claiming the supposed raid to be merely one of many. He depicts the battle as pre-planned by missionaries on the basis of only part of one European eye-witness account, disregarding contrary evidence from this same account and from other European and African witnesses. A survey of the historiography surrounding the Lithakong battle, to present the accepted version of events in this context, allows us to analyze Cobbing's use of the sources and to assess the plausibility of his reinterpretation.

The complicated confrontation at Lithakong occurred in the context of considerable demographic turmoil in the interior of southern Africa. The background events of I822-3 have been fairly well described by scholars. Large-scale migrations in the interior began when the AmaHlubi of Mpan- gazita crossed the Drakensberg from the east and attacked the BaTlokoa under the leadership of their queen regent, 'MaNthatisi, in early I822. The SeSotho-speaking BaTlokoa of 'MaNthatisi raided others for food, both cattle and standing crops, as they fled the AmaHlubi and eventually settled west of the Caledon River about I 824. Subsequently they came into occasional conflict with their SeSotho-speaking neighbors, who emerged after I824 as the BaSotho nation of Moshoeshoe. In I853 they were finally defeated by Moshoeshoe and incorporated into his nation.

The AmaHlubi and AmaNgwane, under Matiwane, fought a decisive battle in I825, and in the aftermath the AmaHlubi chief Mpangazita was killed and his followers dispersed. This left the AmaNgwane dominant in the area of the Caledon River until they were disbanded in I827. In the meantime their raids had dislodged various SeSotho and SeTswana-speaking groups further north, including the BaFokeng under Sebetoane, the MaPhuting under Tsooane and the BaHlakoana under Nkarahanve. These were the three groups involved in the most famous battle of the time, the battle at the just-abandoned BaThlaping town of Lithakong (Lattakoo) on 24 June I823. On that date, two Europeans, John Melvill and the London

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

i8 ELIZABETH A. ELDREDGE

missionary Robert Moffat, accompanied the BaThlaping group of BaTswana and a force of mounted Griqua armed with guns to defend the town. They managed to rout the BaFokeng, the MaPhuting and the BaHlakoana, killing the chiefs Tsooane and Nkarahanye. The BaTlokoa of 'MaNthatisi were not present.

This battle has come under close scrutiny by scholars in the past because of a dispute over the identity of the participants. European observers, including the traveller George Thompson, reported that rumours circulated widely beforehand among the BaThlaping BaTswana at Kuruman that an infamous group of so-called 'Mantatees' was invading the area. Subse- quently this appellation stuck to both the 'invaders' and the refugees from the battle, leading later historians to assume that the 'attackers' at Lithakong had been the BaTlokoa of 'MaNthatisi. Cobbing takes this confusion over the identity of the Africans at Lithakong, though it has long since been resolved, to indicate contradictions arising from a cover-up constructed around a battle that had not in fact occurred. On this false premise he reinterprets the events at Lithakong instead as a slave raid instigated by John Melvill and Robert Moffat, with the BaThlaping and the Griqua as their allies.

The identity of the participants has been adequately dealt with by Marion How and subsequently by William Lye, both of whom used a variety of independent sources, African and European.63 Marion How corrected the version of the battle put forward by her grandfather D. F. Ellenberger, who had mistakenly described the encounter as two separate battles instead of one and placed the BaTlokoa of 'MaNthatisi at the scene. Cobbing considers Ellenberger's mistakes to be sufficient reason to dismiss him as a source altogether. This ignores the fact that Ellenberger's work was based on oral testimony from dozens of African informants, whose information he sys- tematically cross-checked whenever possible with his colleague and trans- lator (and son-in-law) J. C. MacGregor, who collected his own version of Basuto Traditions.64

To assume that this battle involving three identified African chiefs and

63 Marion How, 'An alibi for Mantatisi', Afr. Studies, XIII (I954), 65-76; and William F. Lve, 'The lifaqane: the mfecane in the Southern Sotho area, I822-24', Y. Afr. Hist., ViI (I967), I07-3 I. Historians have long since recognized that the colonists misidentified many Africans, indiscriminately calling otherwise unknown groups wvho originated from the north-east 'Mantatees'. The term 'Mantatees' was a catch-all name used for all non- Xhosa Africans of whom Cape Colonists became aware during this period, either through hearsay or as incoming laborers. Cobbing's treatment of this issue illustrates why his work is so difficult for a non-specialist to read and criticize. Cobbing uses the term Mlantatees beginning on page 492 to refer to captured, enslaved laborers sold into the colony, without initially explaining this use to the reader. He does not trace the origins of the term and its misuse in the sources until pages 5I4-I5, and he never acknowledges that the distinction between so-called Mantatees and the people of 'MaNtatisi was recognized even wvhen the term Mlantatee was first in use and is widely understood by historians today. It is not until the reader reaches the discussion and footnotes on page 5 i6 that we learn why Cobbing rejects the accepted view of events at Lithakong, found in the work of How and Lye. Even here he never addresses the evidence on which the accepted view is based, making it necessary to review this evidence briefly here.

64 J. C. MacGregor, Basuto Traditions (Cape Town, 1905). For further discussion of Ellenberger and MacGregor see Elizabeth A. Eldredge, 'Land, politics and censorship: the historiographv of nineteenth-century Lesotho', History in Africa, xv (I988), 191-209.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE 'MFECANE RECONSIDERED I9

their followers never took place just because Ellenberger was partially incorrect is to deny a voice to the African participants themselves, who left behind the oral traditions of the event that appear in MacGregor's work as well as Ellenberger's. Marion How and then Lye tied together the oral traditions and the written evidence, including that of the missionaries, to make a convincing case as to who was present.65 In the written sources, we find that Moffat clearly identifies those present: the 'Maputee' (MaPhuting) of 'Chuane' (Tsooane) and the 'Batclaquan' (BaHlakoana) of 'Carrahanye' or 'Karaganye' (Nkarahanye), and that he specified that Tsooane's people had stayed in the town while the BaHlakoana engaged in fighting the Griqua-BaThlaping force. Thompson referred to the presence of the 'Bacloqueeni' (BaHlakoana) and of 'Mahallogani' (Nkarahanye).66 The European witnesses distinguished only two and not three groups by sight, but in the confusion this is not surprising: battlefield eye-witnesses are notoriously unreliable. There is no reason to assume that Sebetoane's BaFokeng were not present, as Cobbing does, noting Smith's implausible explanation that they were hidden from view.67 Both Sebetoane and one of his retainers told Livingstone of their presence (leading the BaFokeng) at the battle, and Nkarahanye's son Setaki was one of Ellenberger's informants. Since many MaPhuthing and BaHlakoana ended up in Lesotho and their descendants there are alive and well, they are hardly 'shadowy Caledon groups', as Cobbing calls them, questioning their very existence.68

It is in this context that Cobbing's conclusion that no battle occurred must be rejected. Cobbing presents his rendition of events without reference to previous analyses, making the novelty of his premises difficult for non- specialists to perceive, and he does not present the evidence directly for the reader's consideration. Hence the reader is asked to accept at face value that the confrontation at Lithakong was an organized, pre-conceived missionary slave raid. Cobbing asserts that

the 'battle' of Dithakong was one such slave (and cattle) raid, unprovoked, on a still unidentified 'enemy', who became immortalized as mantatees. This is clear from the writings of Moffat, Melvill and Thompson, the former two of whom, both missionaries, were the instigators and organizers both of the raid and the disposal of the prisoners. This emerges best in Melvill's account.69

To build his general case against missionaries Cobbing mistakenly identifies Melvill as a missionary here. Melvill was still a government agent in I 823 and did not become a missionary until four years later. Cobbing further depicts the actions of these men in terms which suggest a conspiracy:

65 Unfortunately the renditions of How and Lye, like that of Ellenberger before them, read much into the material which they have onlIy surmised and w hich is highlN questionable, such as the attitudes and intentions of the participants. l\Many of the pejorative adjectives from the missionary sources which appeared in Ellenberger also appear in How and in Lye.

Lye, 'Lifaqane', 127-8. The difficulties of transcribing SeTswana sounds account for the many variations of name spellings in early sources. For example, a sound which many foreigners have difficulty pronouncing can be rendered as 'hl' or 'tl' or 'cl'; another guttural consonant is rendered alternatively as 'r' or 'g' or 'h'.

"' Cobbing, 'Mfecane', 5I6. " Ibid. 5 I 6n. In the text Cobbing refers to 'the Hlakwana and Phuthing, whoever they

were'. 69 Ibid. 492.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

20 ELIZABETH A. ELDREDGE

It was Mr Melvill who brought the three most feared Griqua leaders of their generation - Waterboer, Adam Kok and Barend Barends - together, and organized the arms and powder. It was Moffat and Thompson who spied out the positions of the victims; and Moffat and Melvill who guided the army into 'battle' on 25 and 26 June.70

Melvill and Moffat estimated that 400 to 500 of their opponents were killed, but Cobbing lowers this estimate to 200-300. The Europeans noted that the retreating Maphuting and BaHlakoana burned the town of Litha- kong, but since Cobbing does not accept that it was these outsiders who were attacked, he accuses the Europeans of claiming that the victims burned their own villages and asserts that instead the forces with the missionaries burned some villages.71 Then Cobbing makes it plain that he believes that Melvill and Moffat were motivated solely by the prospect of profits:

Thirty-three cattle were given to Melvill 'according to the custom of the country'. Moffat, Melvill and a mission labourer named Hamilton used armed Griqua to round up the women and children who were not dead or had not been able to escape. Over ninety prisoners were taken back to Kuruman on 25-27 June. There a squabble broke out between the missionaries and the Tlhaping chief, Mothibi, over their disposal. Griqua guns decided the issue in favour of the missionaries. During the next few days Melvill scoured the countryside and captured at least fifty more women and children. He avoided the men. Women and young males were what the Cape market preferred. Melvill immediately despatched fifteen Manatees for sale to Graaff Reinet in the north-eastern Cape, for which he received payment in ammunition. At least thirty remained with the Griqua in Griqua Town. Moffat kept several at Kuruman, and took one boy as a personal servant who was 'affectionately domesticated in the family of his benefactor [sic].' Others, including five women - who fortunately 'indicated nothing of cannibal ferocity' - and a 'fine boy', Moffat took with him for distribution in Cape Town in January i824 to the applause of the local press. Almost certainly the rest were sent to various destinations in the Cape.72

Cobbing's case against Melvill and Moffat must therefore be interpreted in terms of the outcome - were these people enslaved ? - as well as in light of their original intentions - did they plan and execute a slave raid ? I conclude below that many of the people brought back after the battle did indeed end up as captive workers who could be called slaves, but this was an unintended outcome and Moffat and Melville never planned this expedition or any other as a slave raid.

Since Cobbing's case against the missionaries rests on a piece of evidence from Melvill, it is important to assess Melvill's credibility. John Melvill was alternatively known in the sources as Melville, leading to possible confusion. Melvill was appointed government agent in Griqualand on 2I March i822; he was accepted as a missionary by the London Missionary Society in I827 and remained stationed among the Griqua. Melvill was held in high regard by John Philip, who said that if any government agent had to be appointed to live among the Griqua, Melvill was 'the fittest man the government could

70 Ibid. 492. 71 Ibid. 492. Cobbing mistakenly says that Lithakong was the residence of the 'Maida',

when in fact it had long been the main BaTihaping town, and the Maill lived among the BaTlhaping near Lithakong. 72 Ibid. 493.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE MFECANE RECONSIDERED 2I

have selected to fill the office '.7 Cobbing's main source of information on Melvill and the slave trade was George Thompson, a businessman and traveller, and as Thompson never set up business in this area and was not connected to missionaries, he was perhaps an objective witness.74 Thompson had a high opinion of Melvill as well:

No one who is acquainted with Mr. Melvill personally, can for a moment doubt the benevolence and disinterestedness of his character. Indeed his being here at all is a sufficient proof of these qualities. He formerly held an easy and respectable situation under Government in Cape Town, namely, that of Inspector of Public Buildings, etc. with an income of about 7000 rix-dollars per annum; but being a religious man, and zealous for the civilization and conversion of the heathen, he applied to the Government for his present appointment, and voluntarily resigned it for his lucrative situation, with the benevolent purpose of promoting Missionary

75 operations.

Thompson was not entirely convinced by Melvill's account of himself and his activities among the Griqua but respected his intentions nevertheless:

How far Mr. Melvill justly estimated his own qualifications for the arduous task of influencing a semi-barbarous people, may well be questioned, on witnessing the unhappy results of his interference with the affairs of the Griquas; but his praiseworthy motives and generous self-devotion must ever be respected. He now receives, as the Government agent here, a salary of only iooo rix-dollars (751); besides which he occupies a small house belonging to the London Missionary Society. 76

Dismissing Melvill's and Moffat's explanations of their own activities, Cobbing suggests that instead of helping victims of the battle they were capturing slaves:

The missionaries tried to depict themselves as succouring the prisoners, and rescuing them from their evil chiefs and starvation (with all those cattle ?). There is no doubt, however, that they were fully and consciously engrossed in what they were doing, i.e. collecting slaves, and that the cover of hypocrisy was intended to deflect the certain censure from the government in London and from their seniors in the London Missionary Society if it had leaked out that they were selling people into slavery.77

There is no evidence to support this particular missionary slave-trading conspiracy, and the only evidence Cobbing cites to support his general argument comes from Melvill himself. But why would Melvill raise possible questions by implicating himself at all when he could have avoided all mention of receiving cattle or of 'disposing' of the captives ? Melvill's account seems credible under the circumstances. This also is the only 'evidence' used by Cobbing to indict Moffat:

The next morning a party of men were sent to bring them [the captured women and children] along, and most of them were then distributed among the Griquas to become their servants, which was considered to be the best way of getting them taken care of, and provided with food. With the apprehension, however, that the

7 John Philip, Researches in South Africa (2 Vols.) (London, I828; reprinted Newv York, I969), ii, 79. " Thompson, Travels. 75 Ibid. i, 74.

76 Ibid. i, 74.77 Cobbing, 'Mfecane', 493.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

22 ELIZABETH A. ELDREDGE

providing of victuals for these poor creatures might, after all, fall exclusively upon me, I applied for a share of the captured cattle, for myself and the Missionaries, on account of our having furnished the commando with ammunition. By this means I secured a supply of provisions for them, in any emergency, or for any other prisoners who might hereafter be taken. I had allotted to me thirty-three head of cattle, not choosing to receive any more than a regular share, according to the custom of the country, in order to prevent the Griquas from murmuring; at the same time I expressly stated to the chiefs, that I designed the cattle for the subsistence of the Mantatee prisoners.78

Cobbing questions Melvill's veracity by wondering how the women and children from the enemy could have been starving, as he described them, when there were so many cattle at hand.79 Some of these women may have only recently joined these chiefdoms, however, and in times of insecurity and scarcity men often ate while depriving or casting off women and children. The traditional productive role of women was less important in mobile fighting groups, and their access to food was extremely limited. Men had a strong incentiv,e to control women in the context of ongoing threats to life and the need for women's reproductive capacities, but they had no incentive to treat women well. When wealth and food was in the form of cattle, women had no guaranteed access to this food and were often left to fend for themselves, living on gathered grasses and seeds. It is therefore not inconceivable that this group of 'Mantatees' with large herds included women and children who were starving.

As for the disposal of these prisoners, their fates varied, but it is doubtful whether Melvill or Moffat profited. Watson discredits Cobbing's account of Moffat as supposedly 'distributing' six refugees in Cape Town by clarifying the import of his newspaper source, the South African Commercial Advertiser. The newspaper indicated that Moffat was accompanied by seven Africans, identified two of them, and discussed them at greater length without referring to their distribution or sale.8" In fact the group that accompanied Moffat to Cape Town included Chief Mothibi's counsellor, Thaiso, and son, Phetlu. Thompson repeats the newspaper account from the South African

Yournal about their trip, and they appear in several places in his diary of his travels; Phetlu died of anthrax in March i825 after returning to Kuruman from Cape Town, apparently free to do so. Cobbing's insinuation that these people were sold into slavery by 1'vloffat is unfounded.8'

The essence of Cobbing's case is thus to argue that the missionaries deliberately planned and executed a slave raid (the battle at Lithakong), to imply without providing evidence of other slave raids that this event was part of a general pattern of missionary-organised raids and to misconstrue newspaper accounts to imply that these missionaries regularly sold the captives into slavery for their own gain. It is true that Africans were continually being captured and sold into slavery and that some of the people who returned with Melvill and Moffat from the battle suffered such a fate.

78 'Melvill's narrative of transactions after the battle, and of his excursion to rescue the women and children of the invaders', in Thompson, Travels, i, 153.

79 Cobbing, 'Nlfecane', 493. 80 R. L. Watson, The Slave Question: Liberty and Property in South Africa (Middle-

town, i990), 19-20, 234n. 8 Thompson, Travels, i, 86, 88, 92-3, 112, 115, i65-6; ii, 48n, 223-4, 235.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE MFECANE RECONSIDERED 23

However, these instances are not proof that this battle was planned as a slave raid by these men nor that these missionaries or any others were sys- tematically engaged in slave-raiding.

The presence of a forced labor system at the Cape which received captives from the frontier meant that many of the women and children brought back from the battle found themselves introduced into the colony as unwilling workers. Others remained, helpless, as servants to the Griqua. The ac- ceptance of client status by people who lost their herds was routine, so that those who remained among the Griqua in this status would not have considered this form of oppression unusual. It is not inappropriate to characterize all of these helpless victims as slaves: whether working for the Griqua or in the Colony, they were captive and had no control over their own fates. However, those who went to the Cape Colony did not necessarily end up as chattel slaves, as argued by Cobbing. Thompson explains:

Within the last two years upwards of iooo fugitives, mostly in a state of extreme destitution, have taken refuge in the Colony, - a circumstance wholly unpre- cedented in any former period. These refugees have been, by the direction of the Home Government, indentured as servants for seven years to such of the Colonists in the eastern districts as are not slave owvners, and precautions have been adopted, (efficient ones, I trust,) to prevent any of those poor exiles from being ill-treated, or from hereafter merging into a state of slavery.82

The children captured for sale to the colonists by Griqua and Bergenaars must have been vulnerable to permanent enslavement, and no doubt that is why they were preferred. Adults, such as the women sent by the missionaries, would have remembered their original homes and would not have been as easy to retain permanently. The BaSotho who entered into colonial service during these years did indeed return to Lesotho in the late I83os and earlv I840s, bringing back with them the herds they had accumulated during their periods of service. This seems to indicate that the terms of indentured servitude described by Thompson may have indeed governed the relationship between some 'Mantatee' BaSotho and their employers.

One reason that it is so surprising that Cobbing accuses Melvill of covering up his activities as a slave trader is that Melvill deliberately exposed the slave trade in his correspondence. Among his early records is a lengthy letter written when he was still a government agent, in i 824; parts of its were published by the missionary John Philip in his famous I828 book criticizing Cape Colony policy, Researches in South Africa.83 Subsequently Melvill's letter was published almost verbatim in an I835 government report.84 He exposed the violent raiding committed by the Bergenaars, who had broken away from the Griqua in I82I and who captured cattle and slaves to sell to white settlers. Melvill distinguished carefully between those renegade

82 Ibid. i, I87. " Philip, Researches, ii, 79-84. Original letter from John Mlelvill to Sir Richard Plasket,

Secretary of Government, Cape Town, 17 Dec. I824. 84 ' Extract from a Report by Mr. Nlelville [sic], Government Agent for Griqua Town,

dated December I824, relative to the State of the Griquas, addressed to the Colonial Secretary', in Papers Relative to the Condition and Treatment of the Native Inhabitants of Southern Africa within the Coloiny of the Cape of Good Hope or Bevyond the Frontier of that Colony, Part I: Hottentots and Bosjesmen; Caffres; Griquas, in Great Britain, Par- liamentary Papers, I835 (50), XXXIX, 2I2-I9.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

24. ELIZABETH A. ELDREDGE

Bergenaar Griquas, who were guilty of destructive raiding, and the Griqua under Waterboer at Griquatown and under Cornelius and Adam Kok, who were maintaining the interests of the Colonial officials. Melvill explained that the Bergenaars had attracted followers because they avoided colonial restric- tions on commerce and thereby derived huge profits, while those who co- operated with the colonial authorities and their representatives, i.e. the Griqua under Waterboer, suffered the disadvantages of restricted access to trade. Melvill specifically identified the colonists as complicit in fostering and promoting the destructive activities of the Bergenaars by selling firearms to them.85 Melvill also explained the use of these guns by the Bergenaars for raiding neighboring Africans such as the BaSotho. In one instance some BaSotho who were attacked by the Bergenaars did not fight, deciding it was useless, and many of them were enslaved:

... four white men, different from the rest of the plunderers, joined the party, and having collected all the boys carried them off.86

Apparently not only boys were enslaved, for 'one woman, however, resisted when one of the band attempted to drag her away'.87

Melvill took a tour with a fellow missionary, Kolbe, in i828, with a view to discovering more about the neighboring 'Bashutoos'. On this trip Melvill recorded more instances of Bergenaar depredations against the BaSotho.88 Melvill ended his journal notes from this trip with clear allusions to the complicity of the white settlers in this slaving, though these were deleted in publication. Melvill accused the 'haughty and unfeeling farmer' of 'un- christian-like acts according to his own selfish views', but the adjective 'unchristian-like' was edited out. More significantly, he asked, 'And why must these boers as if a superior race of beings deprive their poor fellow creatures of what God has given them?' The question was printed, but Melvill's next sentence was censored and can be found only in the original letter:

We hesitate not to say that they [the Boers] are robbers of the oppressed. We are all pleading for the emancipation of slavery, - a christian-like work indeed! - but in Africa christians are daily making slaves.89

Again he emphasized a few sentences later:

Govt is continually making regulations agreeable to justice, to punish the Bushmen and other plunderers of Cattle, but we trust the same Govt will prevent the farmers from plundering - we say plundering - the bushmen & Bassutoos.90

Melvill's evidence supports Cobbing's description of the Bergenaars as slavers, except that it does not implicate Melvill himself. Of course both Melvill and Philip, who quoted Melvill extensively, were interested in

85 Ibid. 2I5-I6. 86 Ibid. 2 I 7. 87 Ibid. 2 I 7. 88 Extracts from the Journal of Mssrs Melvill and Kolbe, addressed to the Rev

Richard Miles, Superintendent of the Society's Missions in Africa, Philippolis, 25 Nov. I828, Council for World Missions (London Missionary Society) Archives (hereafter CWM). An edited version of this was published as 'Missionary tour through the country of the Bashutoos', Transactions of the Missionary Society, II (Oct. I829), I23-8. Other extracts from Melvill appear in this journal.

89 Extracts from the Journal of Mssrs Melvill and Kolbe, 25 Nov. I828, CWM Archives (unpublished original letter). 90 Ibid.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE MFECANE RECONSIDERED

promoting the interests of 'their' Griqua under Waterboer and were not necessarily friendly to the renegade Bergenaars; however, there is no reason to suspect that their descriptions of them as slavers are incorrect. It is therefore surprising, given Melvill's role in exposing the slave trade, that Cobbing accused him of involvement in it. Cobbing has not provided any evidence that the missionaries condoned or participated actively in the slave raiding in this area; on the contrary, they did not hesitate to condemn colonists engaged in illicit activities, which included the enslavement of Africans. It would be a mistake, however, to dismiss Cobbing's emphasis on slaving on the northern Cape Frontier because of his error in assigning responsibility for it. Although Cobbing's case against Melvill and the missionaries may be unsupportable, his case against the colonists stands.

ENVIRONMENT AND SOCIETY: SOURCES OF COMPETITION AND

CONFLI CT

Since there is compelling evidence that extensive slave trading at Delagoa Bay did not precede the conflicts associated with political amalgamation and state formation in south-eastern Africa, this leaves open the question of what processes did generate the dramatic socio-political changes of the early nineteenth century. A number of theories have been put forward, some of which offer useful insights into the root causes for change in this period. Gluckman first proposed the possibility of over-population as a source of stress and an explanation for political consolidation and the emergence of the Zulu nation.9' Not long after Alan Smith demonstrated that political change in south-eastern Africa was linked to the ivory trade out of Delagoa Bay.92

More recently, Jeff Guy pursued the idea of environmental influences by seeking to explain the rise of the Zulu kingdom in terms of 'stock-keeping and the physical environment of Zululand'.9 He argued that in the area that became the Zulu state cattle needed access to varying types of grasses found in different areas, according to topography and rainfall, during the changing seasons of the year. The need for access to the whole range of pasture types governed the location of settlement sites and created an incentive for bringing more land under Zulu control.94 According to Guy, deterioration of

91 Max Gluckman, 'The rise of the Zulu Empire', Scientific American, ccii (April 1960), I57-68. Omer-Cooper subsequently focused on the age-regiment system of Dingiswayo and argued that it was newly borrowed and adapted from the customs of his SeSotho-speaking neighbors, but recent research indicates this was a false assumption. Omer-Cooper, Zulu Aftermath. In the context of the historiography of South Africa in the 1960s, Omer-Cooper broke new ground in giving primary attention to the internal dynamics of African history in his book. Unfortunately, the very title of the book perpetuates the myth of Zulu responsibility for the violence of the period. By focussing only on the internal dynamics of these societies, Omer-Cooper ignored the fact that they were affected by external trade and other forms of European influence. Implicit in Omer- Cooper's book, in addition, is a 'great man' approach to history, as his focus tends to be on the character and innovations of leaders.

92 Smith, 'Trade'; and Alan K. Smith, 'Delagoa Bay and the trade of south-eastern Africa', in Richard Gray and David Birmingham (eds.), Pre-Colonial African Trade (London, I970), 265-89.

9 Jeff Guy, 'Ecological factors in the rise of Shaka and the Zulu kingdom', in Shula Marks and Anthony Atmore (eds.), Economy and Society in Pre-Industrial South Africa (London, I980), I02-I9. 94 Ibid. I05-I2.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

26 ELIZABETH A. ELDREDGE

the land from overuse, combined with drought, produced famines in the late eighteenth century, which brought about conflict and political change and eventually led to state formation.

Guy's thesis of overpopulation and environmental degradation in Zululand is problematic. The carrying capacity of land in terms of people is determined by the use that people make of the land. Guy attempted to assess the relationship of northern Nguni to their environment in terms of land use, but he looked at only the pastoral side of the production process and ignored cultivation. On this basis Guy proposed the hypothesis that 'by the end of the eighteenth century an imbalance had arisen between population density and the resources of the region and that this contributed to the radical social changes which took place '." There is no evidence that environmental degradation had occurred, and Guy jumped to unfounded conclusions when he assumed that 'by the end of the eighteenth century, the physical resources were breaking down under existing systems of exploitation'.96 The seasonal rotation he emphasizes argues against unconscious misuse of pastures.

Furthermore, there is evidence that arable land was being used efficiently through cultivation practices that allowed for soil regeneration. To Guy's discussion of stock-keeping and pasture use must therefore be added an analysis of cultivation and arable land use. Cultivation represents intensified land use relative to pastoralism and produces much greater quantities of food in proportion to the amount of land used. Solving a problem of 'over- population' by increasing food production can therefore be accomplished either by acquiring more land for extensive pastoral use or by converting land already available to more intensive use through cultivation. Cultivation was as important to the people of south-eastern Africa as was stock-keeping, and grains, both sorghum and maize, were a critical component of their diet.97 Because of the warm climate crops matured in only three months, and on well-watered fields with proper soils two or even three crops could be grown in a single field each year.98 The Zulu were careful to ensure that the fertility of the soil was renewed: the stalks from the harvested grain were collected and burned, and the ashes scattered to serve as fertilizer in the fields.99 Zulu recognition of the critical value of food crops was evident in the severity of the law under Shaka that no one was allowed to pick even a single ear of corn or head of sorghum without Shaka's permission, on pain of death.'00 In order to support a growing population with food, the acquisition of good, well-watered arable land with rich soil was critical.

Guy's interpretation stressing overpopulation, environmental degrada- tion and the desire for pasture land is therefore incomplete. Had there been generalized overpopulation as postulated by Gluckman and Omer-Cooper, it is hard to explain in terms of production needs why the fertile tract of land

Ibid. I 03. 9" Ibid. I03.

Guy acknowledged the importance of cultivation in another article, but he did not attempt to revise his earlier thesis based on the recognition that 'cereal production was not only fundamental to the existence of these societies, but absorbed massive amounts of labour time dominating not only the productive processes, but profoundly affecting social life generally'. Jeff GuN, 'Analysing pre-capitalist societies in southern Africa', Y. Southern Afr. Stuidies, XIV (I987), 29.

98 Nathaniel Isaacs, Travels and Adventures in Eastern Africa (2 vols.) (London, I836; reprinted Cape Town, I936), ii, I27.

9 Ibid. ii, I27. 1 Isaacs, Travels, i, 46, and ii, 241.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE 'MFECANE RECONSIDERED 27

in Natal south of Zululand, which was available for herding, was left unutilized by Shaka. Furthermore, if the population had increased to the point where it could no longer be supported by a primarily pastoral economy, it was still possible to grow more food by using the most fertile land more intensively in cultivation.

In light of these reservations, other arguments that Guy puts forward in support of environmental stress need to be reassessed. According to Guy, because of overpopulation the marriage of women was deliberately post- poned, using female age-sets linked to male regiments in order to limit human reproduction and restrict the rate of demographic expansion. He argued that the Zulu king delayed marriage for both men and women to limit the rate of biological reproduction as well as the reproduction of homesteads through marriage, in order to solve the supposed problem of overpopu- lation.'0' In a later article Guy went so far as to assert that 'reproductive sexual relations however could only take place with the king's permission', assuming both popular compliance with the king's orders and a desire on the part of the king to limit reproduction.'02

The evidence does not support this emphasis on limiting reproduction. Although the marriage age of men was postponed under Shaka, the marriage age of women, the only factor relevant to birth rates, was not. Evidence referring to women remaining unmarried to the age of thirty used by Guy dates from a much later period and can in no way be convincingly projected backwards.'03 Although girls could not marry until released by Shaka, there is no evidence that he made them wait beyond the normal age of I 5. The only contemporary reference to the age of marriage of women from the reign of Shaka indicated that women married ('were allowed to become wives') at the age of I4 or I 5.104 The young women in age-sets attached to military barracks were displeased when they were forced to marry older men instead of the men closer to their own age.105 Men who were released to marry could take as many wives as they could afford, that is, those for whom they could pay bridewealth.106 Bridewealth at the time was very low outside of the royal family: one or two head of cattle only.'07 Observers noted that married men were all polygamous, and it is unlikely that many girls remained unmarried for long. Shaka may have wanted to retain control over young men through the military system, but reproductive rates would have been reduced only if

10 Guy, 'Ecological', ii6. 102 Guy, 'Pre-capitalist societies', 32. 103 Krige sayTs marriage was also delayed for women in Shaka's time, but she has no

source or evidence for this assertion, and she elsewhere tends to collapse evidence from different periods, a mistake Guy makes here as well by using Shepstone's evidence concerning delayed marriage of women in i873. Eileen J. Krige, The Social System of the Zulus (3rd ed., London, I936; reprinted Pietermaritzburg, I957), 38; Guy, 'Ecological', i i6.

1014 Captain J. S. King, 'Some account of Mr. Farewell's settlement at Port Natal, and of a visit to Chaka, King of the Zoolas, etc.', in Thompson, Travels, ii, 25I.

1'o Krige, Social System, 38. 106 Lunguza ka Mpukane, informant in C. de B. Webb and J. B. Wright (eds.), James

Stuart Archi7ves of Recorded Oral Evidence Relating to the Historv of the Zulu and Neighboring Peoples (JSA) (4 vols.) (Pietermaritzburg and Durban, I976-86), i, 3I7.

107 Lunguza in YSA, 1, 3 I 7. Isaacs indicates that bridewealth was seldom more than ten cows: Travels, ii, 237. Gardiner estimates it was four to six cows, but twenty to one hundred for a chief's daughter: Allen F. Gardiner, NarratiVe of a Journey to the Zoolu Coluntry in South Africa (London, i836), 98.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

28 ELIZABETH A. ELDREDGE

the marriage age of women rose, which apparently did not happen. Competition for people remained a driving incentive for chiefdoms and nations in southern Africa throughout the early decades of the nineteenth century, and there is no reason to think that biological reproduction was limited deliberately.

Using dendroclimatology, Martin Hall reassessed timing in the en- vironmental factor when he demonstrated that droughts in south-eastern Africa had long occurred in twenty-year cycles. He suggested that the carrying capacity of the land was increased by a steady rise in rainfall during the second half of the eighteenth century, which encouraged the use of formerly marginal land. This time of plenty was followed by a severe drought at the beginning of the nineteenth century, which brought about a crisis as the habitable lands decreased, food production fell and famine ensued.108

Bonner accepted Hall's argument that drought could partly account for the revolutionary socio-political changes of the period even though cyclical droughts had recurred many times previously, and he offered an explanation for why those of the early nineteenth century were more radical than had been earlier, similar but 'less formidable' strains.109 He concluded that dominant groups who exercised control through the new amabutho (age- regiments) had then emerged and that when drought and famine on this occasion led to shortages of manpower and cattle the new, more powerful aristocracy harnessed their control over both through use of the age- regiments. Bonner raised the question of why the process of political amalgamation occurred where it did as well as at the given point in time, but he did not pursue regional geographic variations, and his focus was limited to the AmaNgwane of Sobhuza.

Bonner's interpretation is a useful starting point for a successful reinter- pretation of revolutionary demographic and political change in the early nineteenth century. In the late eighteenth century and probably considerably earlier, African societies in southern Africa were characterized by increasing socio-economic stratification and correlated political amalgamation.110 The Delagoa Bay ivory trade significantly accelerated the process of economic growth and political consolidation in the area. As Smith notes, there is compelling evidence that 'until the second or third decade of the i gth century ivory was the most sought after, consistent, and important com-

108 Martin Hall, 'Dendroclimatology, rainfall and human adaptation in the later Iron Age of Natal and Zululand', Annals of the Natal Museum, xxii (I976), 693-703. Wright and Hamilton have asserted that 'the "environmental" argument is speculative and cannot by itself explain why conflict over resources should have begun when and where it did, nor why it should have produced the particular political effects that it did'. This led them to emphasize an explanation based on trade, but they have admitted that trade also can account for state formation in the region only partially, since 'why these processes should have begun in these particular chiefdoms and not in others is more difficult to explain'. John Wright and Carolyn Hamilton, 'Traditions and transformations: the Phongolo-Mzimkhulu region in the late eighteenth and early nineteenth centuries', in Andrew Duminy and Bill Guest (eds.), Natal and Zululand From Earliest Times to I9I0: A New History (Durban, I989), 6i.

109 Philip Bonner, Kings, Commoners and Concessionaires: the Evolution and Dissolution of the Nineteenth-Century Swazi State (Cambridge, I982), 2I.

110 Bonner summarizes the interpretation of this process found in D. W. Hedges, 'Trade and politics'; Bonner, Kings, I0-23.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE MFECANE RECONSIDERED 29

modity to be exported from Delagoa Bay'.111 The export of ivory generated a surplus of imported valuables which was expropriated by chiefs, who used their new-found wealth to consolidate their power, and wealth and power became mutually reinforcing. The problem for the beneficiaries, however, lay in the limited nature of the herds on which the ivory trade drew. With extraction of large quantities of ivory the elephant population eventually declined, and exports became scarce. The decline of the elephant population must have created a crisis in the economy and the polity: chiefs who had established themselves on the basis of wealth derived from the ivory trade had adopted systems of taxation and expropriation, but the trade had not promoted the intensification of land use or the generation of a surplus based on agriculture to support them from local production. As a result, local economies could not support high levels of extraction indefinitely, and chiefs must have found themselves competing for declining supplies.112

Neither the export of ivory nor the import of cloth and beads transformed local industries or the production of food. However, this trade was important because it served as a source of wealth which powerful people were able to control and manipulate to increase their own sway. Hence trade stimulated the emergence of socio-economic inequalities. Socio-economic inequalities fostered the unequal distribution of food in times of scarcity, both within societies and between chiefdoms, resulting in famine for those lacking access to productive resources.113 Famine in turn provoked competition over scarce resources necessary for producing food (land and people), and the resulting conflicts led to political consolidation.

Droughts reduced the carrying capacity of both arable land and pastures, and increased competition over resources in the late eighteenth and early nineteenth centuries. Both written and oral sources confirm that south-east Africa suffered major droughts between i8oo and I803, in I8I 2 and in I8I6-I8.114 Some areas suffered also from the destruction of crops by crop

... Smith, 'Trade', I76; see also Portuguese sources, e.g. Manso, Memoria, II-I3, 66-9, IOO-I, ii8, I2I, 13I.

112 See also Alan Smith, 'The Indian Ocean zone', in David Birmingham and Phyllis M. Martin (eds.), History of Central Africa (2 vols.) (London, i983), i, 233-6.

113 It is not sufficient to assert that drought must somehow have induced social stress without explaining the precise causal relationship: in well-managed systems of food storage and distribution through any form of social security system, drought need not cause food scarcity, famine, or social distress. As Amartya Sen has shown, it is politics and the distribution of food entitlements which determine the social impact of drought, i.e. whether or not drought leads to famine. Amartya Sen, Poverty and Famines: an Essay on Entitlement and Deprivation (Oxford, i98 I). For an explanation of Sen and an analysis of drought and famine in nineteenth-century southern Africa using Sen's approach, see Elizabeth A. Eldredge, 'Drought, famine, and disease in nineteenth-century Lesotho', Afr. Economic Hist., XVI (I987), 6 I-93.

114 The great drought of i8oo to I803 affected all of southern Africa. Europeans in the Cape Colony lost so many cattle from drought in i8oo that an expedition was sent north to acquire cattle from trade with the BaThlaping (BaTswana). John Barrow, An Account of Travels into the Interior of Southern Africa (2 Vols.) (London, i8oi; reprinted New York, I968), ii, 55. Numerous oral traditions about this drought and the associated famine have been recorded from the BaTswana, BaSotho, and AmaZulu. 'Mabokoboko, ou une page d'histoire', Journal des Missions Evangeliques, i884, 420; H. Laydevant, OMI, 'La misere au Basutoland', Les Missions Catholiques, I934, 333-7; Nehemiah Sekhonyana Moshoeshoe, 'A little light from Basutoland', Cape Monthly Magazine, 3rd series, ii, part IO (April I88o), 221-33, and ii, part I I (May I88o), 280-92; Almanaka ea

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

30 ELIZABETH A. ELDREDGE

rust and from a cattle epizootic from i8i6 to i8I8. Given a situation of ongoing competition for land and cattle, these droughts explain the timing of overt conflicts between the Xhosa and white settlers in i So i, I8 iI-I2, and i8I8-I9, and they correspond chronologically with periods of demographic and political turmoil among the northern Nguni, first under Dingiswayo and later under Shaka.

Increasing political amalgamation and socio-economic stratification pro- duced inequalities in access to productive resources between and within chiefdoms, resulting in famine in times of drought and provoking the political revolution among the Mthethwa and the Zulu in times of drought in the early nineteenth century. The changed socio-political context explains why strategies of famine resistance had broken down so that the drought of iSo to I803 had severe consequences, including widespread famine, even though dry periods had occurred in the past without similar consequences. In its earliest stages drought compelled herders from neighboring societies to compete over pastures because the normally sufficient carrying capacity of the area for their herds declined dramatically. Without water or forage, animals died quickly, and stored grain provided the most secure food supply for people.1"5 The production of surplus grain for storage in case of scarcity required access to fertile and well-watered land, and in extended droughts crop failure brought people into overt conflict either over grain stores or over riverine lands which supported minimal crops even in dry years and in normal years produced surpluses.

During droughts the wealthy and powerful took advantage of their greater access to food resources while the poor suffered disproportionately, pro- voking conflict. Drought provided an incentive and opportunity for the powerful to consolidate further their control over fertile and well-watered arable land at the expense of weaker neighbors. Furthermore, droughts made evident the extreme vulnerability of the poor and weak to food scarcity and increased their incentive to submit to political authority, even when ex- tremely oppressive, because such submission promised security from star- vation.

There is considerable evidence that early conflicts involving the Mthethwa and Zulu were related to competition over arable lands. Both the Ngwane under Sobhuza and the Ndwandwe under Zwide consolidated and expanded their areas of control on the banks of the Pongola River, and the major struggle between them arose over arable lands there.116 Among the regional soils, many were poor and unsuitable for cultivation, and fertile arable land that was well-watered and drought-resistant was relatively scarce.117 Zulu

Basotho, Selemo sa I894, Khatiso ea A. Mabille (Morija, I894); Litaba tsa Lilemo (Mlorija, I93I). See also Charles Ballard, 'Drought and economic distress: South Africa in the i 8oos', J. Interdisciplinary Hist., XVII (I986), 359-78; Eldredge, 'Drought'; Guy, 'Ecological'.

115 The Nguni had the capacitv to store large quantities of food for up to seven years in order to prevent famine in times of drought. Contrary to what Hall states, both grain pits and grain baskets were used by the Nguni for storing huge quantities of grain, and although some would always be lost to rot and pests, the incentives for storing grain to avoid famine were very great in the nineteenth century. Hall, 'Dendroclimatology'.

116 Omer-Cooper, Zulu Aftermath, 49; J. B. I\I. Daniel, 'A geographical study of pre- Shakan Zululand', South African Geographical Yournal, xv (I973), 29.

117 Isaacs, Travels, i, 85, io6, I49-52, I59, i6i.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE 'MFECANE RECONSIDERED 3I

military kraals were located in very fertile areas, good for cultivation as well as herding."18

Intensive use of these fertile areas for cultivation necessitated mobilization of greater amounts of labor, provoking competition for people as well as land. The process of state formation under Dingiswayo and during Shaka's early years was a process of incorporation of people, not extermination. Women in particular were highly valued both for reproduction and because they were the primary agricultural producers. During Shaka's reign a Zulu chief explained that he spent much time searching for refugees from Shaka, 'for the purpose of enriching himself by adding their females to his establishment (who were a valuable property and disposable at the will of the possessor) '.1"9

The conflicts generated during political consolidation under the Zulu must be seen in the context of this ongoing competition for both land and people. Although Zulu aggression under Shaka contributed significantly to the violence in the immediate area, the majority of the Zulu people were victims of their circumstances and cannot be held responsible for the violence which plagued the region. Growing vulnerability to famine in times of food scarcity drew people to any strong leader who could provide them with a livelihood and protection. The expansion of arable production was the best means of achieving food security in the long run, but gaining the necessary labor and opening new lands was a long and tedious, even painful, process. Hence there arose the natural impulse to accumulate wealth more quickly and directly. The employment of women in agriculture left men free to acquire surplus directly through raiding and the accumulation of booty. The booty from raids provided an easy way for chiefs to satisfy the needs and demands of their people in order to keep followers from falling away. Some people were motivated to support Shaka by the attraction of a share of the booty, as he kept only a portion of captured cattle for his own herds and distributed the rest among his people.120 The raiders accumulated livestock much more rapidly than they could have through natural increase, and military expedi- tions became the quickest route to wealth for the state and the soldiers. As the Zulu military became important in generating state revenues, over time the age-regiments became institutionalized and took on their own imperative, fighting to support themselves as an end in itself and not merely for defensive purposes. Systematic raiding in turn fostered hostile neighbors and made protection a major incentive for further population concentration and political consolidation.

The sporadic conflicts that continued in the interior of southern Africa beyond the period of state formation in the years following I822, associated with large-scale migrations and considerable human misery, also had environmental sources. The damaging European-armed raids of the Griqua can only partly account for these events, since massive migrations like those that converged at Lithakong in I823 clearly resulted from other factors. In fact these conflicts on the highveld can also best be understood with reference to the appearance of drought at a time when political events undermined traditional strategies of avoiding famine. However, there, because the land was much drier and less fertile than in areas east of the Drakensberg, soils and vegetation appear to have suffered permanent damage from human settlement over time.

118 Ibid. ii,

171. 119 Ibid. i, iIo. 120 Ibid. i, i8o, 283-

2 AFH 33

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

ELIZABETH A. ELDREDGE

There is evidence of gradual long-term environmental degradation in the southern African interior prior to the arrival of the missionaries in the early nineteenth century. Moffat reported that when they first arrived in i820 (years of drought had been severely felt', prompting the BaThlaping both to blame the missionaries and to seek a famous rainmaker from the BaHurutsi 200 miles to the northeast.12' The BaTlhaping told Moffat their oral traditions of 'incessant showers', 'giant trees and forests which once studded the brows of the Hamhana hills and neighboring plains' and impassable torrents in the Kuruman and other rivers in order to demonstrate that the land had been gradually drying up. Moffat noted the physical evidence that supported this: ... the dry seasons had commenced at a period long anterior to the arrival of the missionaries. Independent of this fact being handed down by their forefathers, they had before their eyes the fragments of more fruitful years in the immense number of stumps and roots of enormous trunks of acacia giraffe, when now scarcely one is to be seen raising its stately head above the shrubs; while the sloping sides of hills, and the ancient beds of rivers, plainly evinced that they were denuded of the herbage which once clothed their surface. Indeed, the whole country north of the Orange River lying east of the Kalagare desert, presented to the eye of an European something like an old neglected garden or field.122 John Campbell, a fellow missionary from the London Missionary Society, also heard that the Kuruman River had been drying up over time. In I820 he wrote that All the elderly people at this meeting asserted, that in their young days the Krooman [sic] was a great river, and as a proof of this, said that, sometimes it rose and continued high for so long a time, that women who happened to be on the other side of the river, frequently lost all hope of being able to recross it, and married other men, they also asserted that great quantities of reeds grew in it. Much water they said used to come down from the Moloppo, which formed a junction near them, and from another river called the Mesaree, but that the Krooman does not now receive any supply from them.... No stream now ever flows within its banks, but large pools are formed in its bed in times of much rain.'32

Moffat attributed the environmental degradation of the area to human agency, since the BaThlaping and other BaTswana with whom he was acquainted moved their huge towns periodically, and, when they chose a new site, they levelled every tree in sight in order to clear spaces for homes and fields and to acquire timber and poles for houses and fences. Moffat assumed that deforestation caused by people seeking building materials in turn had caused the drying up of both springs and rivers, since rain evaporates quickly without vegetation and does not feed the rivers and underground reservoirs for springwater.124 Over time the loss of vegetation and diminishing water supplies would have decreased the carrying capacity of the land in terms of both livestock and people. Although Moffat's causation cannot be es- tablished, these early missionary observations offer evidence that significant environmental deterioration had been occurring in the region.

In this gradually deteriorating environment the SeSotho and SeTswana- speakers of the interior, like their neighbors near the coast, experienced

121 Moffat, Missionary Labours, 307. 122 Ibid. 329-30. 123 John Campbell, Travels in South Africa (2 vols.) (London, i822; reprinted New

York, i967), ii, 93-4. 124 Ibid. 332-3.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE MFECANE RECONSIDERED 33

both migration and famine during the drought of I800-3, and they subse- quently faced crop rust and a cattle epizootic that destroyed both their crops and their herds between i8i6 and i8 i8.125 These disasters devastated people throughout the region, including the BaTlokoa and other SeSotho and SeTswana-speakers, leaving them without stored grain or large herds. Drought was still evident when Moffat arrived in I820. With little time to recover, the BaTlokoa may already have been on the brink of famine in I 822, when their migration was first provoked by the arrival of the AmaHlubi. Evidence indicates that drought had been ongoing in the area for several years prior to I824 and that it persisted into I826, precisely the years in which the massive migrations of the interior were provoking such fear among the Griqua and Europeans on the border of the Cape Colony.126 Before the rain finally came in I 826 Moffat observed that '... the cattle were dying from want of pasture, and hundreds of living skeletons were seen going to the fields in quest of unwholesome roots and reptiles, while many were dying with hunger '127

An uninhabitable landscape, then, was the context in which the com- petition for cattle and standing crops occurred on the highveld after I822. Because the land to the west of the Drakensberg mountains is much drier than the land to the east, settlement patterns on the highveld have always tended to be highly concentrated around the rare sources of water. Hence the various BaTswana towns found by the Europeans at the turn of the nineteenth century were huge. Lithakong had io,ooo to I5,ooo BaThlaping inhabitants in i8oi and remained a huge and thriving town even after the chief moved some of his people eighty miles south-west to Kuruman, the new capital where the missionaries set up their mission in I820. Kuruman itself had an estimated 8,ooo to I0,000 inhabitants in I823. These towns were of typical size for the area, so it stands to reason that were any single town routed by an invader, the refugees would constitute a huge group themselves. The entire populations of BaFokeng, MaPhuting and BaHlakoana ap- parently converged at Lithakong. Thompson reported a conversation with Melvill who referred to the rumors about 'an immense horde, or nation', which was approaching from the north-east.128 The term 'nation', perhaps borrowed here by Melvill from his BaTswana informants, more accurately describes these groups, which included non-combatants, and reflects the fact that they had been complete societies prior to being uprooted. The numerical estimates of the size of groups involved in the confrontation at Lithakong do

125 'Mabokoboko'; Almanaka ea Basotho (I894); Litaba tsa Lilemo (I931). 126 In August I824 Thompson encountered some Korannas who were literally starving

to death and noted that they had been reduced to this state because of extreme drought. Thompson, Travels, ii, 30-3. Similarly Moffat noted that there had been 'several successive years of drought, during which water had not been seen to flow upon the ground; and in that climate, if rain does not fall continuously and in considerable quantities, it is all exhaled in a couple of hours'. This drought lasted several years prior to early I826, when finally rain came, indicating that the drought dated back at least three years to I823 if not earlier. Moffat, Missionary Labours, 315, 447. BaSotho oral traditions also indicate that there was a severe drought during these years. 'Liketso tse etsagetseng Lesotho I820-I870' [Events in Lesotho I820-I870], Leselinyana la Lesotho (newspaper of the Paris Evangelical Missionary Society in Lesotho), October I871, 73-7; Litaba tsa Lilemo (1931); N. S. Moshoeshoe, 'A little light from Basutoland'.

127 Moffat, Missionary Labours, 316. 128 Thompson, Travels, i, 79.

2-2

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

34 ELIZABETH A. ELDREDGE

not seem so high in this context: if I0,000 to I5,000 BaFokeng men, women and children were joined by similar numbers of MaPhuting and BaHlakoana, the total would have amounted to the 40,000 to 50,000 people estimated by Melvill and Moffat.

The otherwise mystifying convergence of three nations at Lithakong in I823 must be viewed in this context. The generally dry environment of this region left its people very vulnerable to food scarcity and famine in times of drought. Lithakong was different, however: the town was located near the Kuruman spring, which was reportedly the most abundant spring in South Africa and next to the 'Lattakoo' or Mashowing River."29 The site was no doubt originally chosen for its capacity to support crops even during droughts, and its reputation for successful agriculture must have been widespread. Khoi accompanying the London missionary Campbell to 'Lattakoo' were 'amazed at the extent of land under cultivation, having never seen so much before in one place'130 The old BaTlhaping capital also had a great reputation for artisan industries, and success in producing agricultural surpluses on a regular basis would have been a precondition for the emergence of craft specialization. Messengers sent to warn the BaThlaping were told that the invading people wanted crops as well as cattle.131 Apparently even the people at Kuruman did not have good crops that year, for Thompson noted the superiority of the millet crop at Lithakong, which was flourishing because of the settlement's proximity to the spring and the river, although at that point even the Mashowing River 'was only a chain of pools, 132 In the context of drought-induced food scarcity exacerbated by an unstable political climate, the search for food helps to explain the migration of entire chiefdoms and the confrontation at Lithakong.

A succession of distinct factors perpetuated violence among Africans west of the Drakensberg, like their neighbors to the east. The first disruptions associated with the migrations of Africans in i 822 from across the Drakens- berg were apparently prolonged by pre-existing problems of food scarcity caused by drought and a cattle epizootic. These migrations and related conflicts ended in the I820S, however, and it was the Griqua and Koranna raiders, capturing cattle and slaves for their white frontier neighbors, who bore responsibility for intensifying the violence in the I82os and prolonging it well into the I 830s. Their raiding activities predated this period but clearly intensified in the I820S, as the white settler frontier moved outwards and provided a growing market for stolen cattle and captured children.

CONCLUSIONS

Trade-intensified competition over productive resources in an increasingly strained natural environment generated conflict among southern African societies in the early nineteenth century. The white colonists in the Cape and, after I823, slave traders at Delagoa Bay fostered further conflict and violence by providing a market stimulus to open warfare and enslavement. Cobbing is right to absolve the Zulu from the responsibility for most of the regional conflicts of the I 82os and I 83os, and this is his abiding contribution.

129 Ibid. i, IOO 107. 130 Campbell, Travels, i, 64. 131' Thompson, Travels, i, 11. 132 Ibid. i, 107.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions

THE 'MFECANE RECONSIDERED 35

These disruptions were associated with attempts by Africans to cope with their environment, and they involved and were caused by Europeans as well as Africans.

History is political, and it has always been systematically distorted in South Africa to suit the ends of continued white domination, most notably in the historiography of the so-called 'mfecane'. We must continue our search for a more sophisticated understanding of the historical processes involved and recognize the dilemmas that confronted all Africans when faced with the European presence in the region. It is a mistake, however, to assume that historical change in southern African societies was merely a reaction to the European presence. In discrediting the old 'Afrocentric' interpretations of the 'mfecane', Cobbing argues that there was no self-generated internal revolution in the societies of south-eastern Africa and that the Zulu state developed only as a defensive reaction to the Delagoa Bay slave trade.133 Not only does he reverse his chronology, but he also rejects generally the notion that dynamic internal forces generated change in African societies. Historians working throughout Africa have spent the last three decades challenging the racist assumption of exclusive European agency in African historical change and demonstrating the complexity of internally generated change over time in Africa. It is a step backwards to depict African history in South Africa again as a wholly reactive process and to deny to Africans their own history. Events and dev elopments in this period in southern African history must be interpreted not only as African attempts to contain and resist the aggression of the Europeans but also as a reflection of the complex interplay of environment, society, and economy.

SUMMARY

The so-called 'mfecane' has been explained in many ways by historians, but never adequately. Julian Cobbing has absolved the Zulu of culpability for ongoing regional conflicts, but his work is severely flawed in its use of evidence. Cobbing is incorrect to argue that the Delagoa Bay slave trade existed on a large scale prior to the disruptions beginning in I8I7, and European slaving therefore cannot have been a root cause of political turmoil and change, as he claims. Cobbing correctly identifies European-sponsored slave-raiding as a major cause of violence across the north-eastern Cape Frontier, but his accusations of missionary involvement are false. Jeff Guv's interpretation of the rise of the Zulu kingdom based on environmental factors is inadequate because he examined only stock-keeping and not arable land use, which led him to false conclusions about demography and politics. In this paper I argue that the socio-political changes and associated demographic turmoil and violence of the early nineteenth century in southern Africa were the result of a complex interaction between factors governed by the physical environment and local patterns of economic and political organization. Increasing inequalities within and between societies coupled with a series of environmental crises transformed long-standing competition over natural re- sources and trade in south-eastern Africa into violent struggles.

133 Cobbing, 'Mfecane', 517.

This content downloaded from 128.111.94.109 on Sat, 19 Sep 2015 23:53:46 UTC All use subject to JSTOR Terms and Conditions