6 Kingdoms and Associations: Copper’s Changing Political Economy during the Nineteenth Century

Historians should move beyond the gleam of ‘red gold’, however. For one, in the Copperbelt region at least, copper ore was relatively abundant, like iron, and not rare and precious, like gold and ivory. And, although a relationship between copper and power clearly existed, copper not only supported centralised, sedentary political authority, but also encouraged other forms of associational life, sometimes offering alternatives to elite accumulation. ‘With ivory, wealth only belongs to the chief’, Nkuba, the titleholder of a copper-producing polity remembered, ‘[but] with copper foundries, everyone was wealthy.’3 ‘Chez les chasseurs d’ivoire, la richesse appartient seulement au chef et à quelques hommes de métier; chez les fondeurs de cuivre, tout le monde est riche.’ Cited in Jean-Félix de Hemptinne, ‘Les “Mangeurs de Cuivre” du Katanga’, Bulletin de la Societé Belge d’Études Coloniales, 1, 3 (1926), pp. 371–403, p. 400. Copper could cut against the power of wealthy elites, contributing to uneven processes: political centralisation and decentralisation; the wealth of elites and the well-being of the poor; the establishment of sedentary political hierarchies and movements of people. This chapter seeks to disentangle how copper produced and traded in South-Central Africa inspired novel political formations, networks of wealth and opportunity, vertical hierarchies and horizontal associations during the era of the caravan trade (1750–1900) and across what would become the modern Copperbelt.

The history of the nineteenth-century Copperbelt is of interest to historians of later periods, since, similar to the colonial and postcolonial Copperbelt, South-Central Africa in the nineteenth century witnessed complex and non-linear political and economic dynamics. When European colonial exploitation of copper began, colonial powers imagined they were replacing the traditional and artisanal production of copper with a modern, industrial one. This chapter demonstrates that copper was involved in the formation of wealth and communities well before colonial rule, encouraging distinctive political, economic and social changes imbricated in international trading networks. In the 150 years prior to colonialism, the exploitation of copper involved periods of intensification, the recruitment of free and forced labour from far away, connections to regional and even global economies, and also the decline of copper production. As in the experiences of the late-colonial and postcolonial Copperbelts, the exploitation of copper did not always provide a stable foundation for the consolidation of political power, and contributed to a waxing and waning of forms of associational life.

The Copperbelt prior to colonialism also provides comparative perspectives to gauge the distinctiveness of the ‘modern’ Copperbelt political economy. An instructive example from this chapter concerns alternative patterns of labour exploitation in the context of the expansion and contraction of copper production. As with the decline of the Copperbelt in the 1970s and 1980s, copper production appears to have decreased in the latter half of the nineteenth century alongside the declining value of copper. Just prior to colonialism, however, this decline in copper production occurred in the context of an increased demand for labour for the exploitation of other exports, in particular ivory, rubber and beeswax. On the other hand, the decline of copper production in the postcolonial period was not accompanied by the rise of other forms of export-oriented production. In this sense, this longer-duration temporal comparison allows for a reformulation, or even rejection, of ideas of modernisation catalysing the exploitation of underutilised labour. To be sure, the effects of technology and productivity on such labour arrangements need to be fully appreciated. Still, the experience of the Copperbelt prior to colonialism challenges certain assumptions of modernisation theory, even before modernisation supposedly began! At the very least, the chapter shows that the history of the nineteenth century needs to be part of the history of the modern Copperbelt.

The key historiographic problem explored in this chapter concerns the relationship of commodities to the connections that constituted societies and polities.4 Inspired by Arjun Appadurai (ed.), The Social Life of Things: Commodities in a Cultural Perspective (Cambridge: Cambridge University Press, 1988), this chapter seeks to orient copper around its broader social life. Copper as a commodity was tied to various levels of production and trade – each constituting distinctive political, economic and social relationships, and leaving different evidentiary traces. During the nineteenth century, copper was first mined, then smelted, and then smithed and moulded into artifact, tool and weapon. Copper was traded, used as a form of currency, as status item, and as a form of tribute. It thus connected individuals in associations that combined different skills, concerns, and socio-economic positions. These connections, changing over time, constituted a copper network, a ‘chain of associations among actors’, as Nancy Jacobs puts it in her study of African knowledge production about birds. This chapter pivots away from a view of copper against the backdrop of kingship, detailing instead the changing associations of copper’s historical actors, including miners, smelters, coppersmiths, traders, and political titleholders.5 Building and borrowing on the notion of network developed with respect to copper by borrowing from Nikis and Smith, used effectively in Nancy Jacobs, Birders of Africa: History of a Network (New Haven and London: Yale University Press, 2016), p. 14.

An additional limitation is that few archaeological sites relate to the mining and smelting of copper itself since the richest deposits of copper ore were exploited by colonial copper mines that destroyed archaeological evidence.3 Ibid., p. 17 The best archaeological reconstruction of a mine in Kansanshi, in North-Western Zambia, gives some indications of mining communities and techniques from the fourth to the nineteenth centuries; however, broad generalisations from this single site are difficult to make.4 Michael Bisson, ‘Prehistoric Copper Mining in Northwestern Zambia’, Archaeology 27, 4 (1974), pp. 242–7. Ethnographic upstreaming instead forms a key element of historical reconstruction, particularly early- or mid-twentieth-century re-enactments by elders who remembered aspects of the process, a ritual of historical invention of tradition that continues on the Katangese Copperbelt.5 The best known is de Hemptinne, who witnessed them in 1911 and 1924, described in de Hemptinne ‘Les ‘Mangeurs’, and for summary description in English, see Herbert, Red Gold of Africa, pp. 51–6. Such sources leave the unfortunate impression of mining as traditional, with only some ‘tribal’, not temporal, variations in techniques, leaving scant indication of historical change.

From the eighteenth century, documentary evidence of traders, travellers and missionaries, originating from Angola, Mozambican Zambezi and Zanzibar provides additional historical details. Such European and Swahili (but not colonial) documentary evidence demonstrates the use of Katangese copper in ordinary items and as elite fashion through the Zambezi basin, Angola, and even the Great Lakes; copper trade through this region; and, in the first half of the nineteenth century, the use of copper as a regional currency. The richest early to mid-twentieth-century colonial documentary sources are historical investigations into precolonial copper mining. These investigations sought to exculpate colonial copper mining companies from remunerating Katangese for mineral production. Still, read with a critical eye and against the grain, they provide valuable historical details, including the recording of oral traditions, oral testimony about forms of production, and details of titleholders who held rights over copper.

This chapter also introduces early twentieth-century ethnographic collections of copper and copper alloy material artifacts held in museums with rich Central African collections and accompanied by provenance records of late nineteenth and early twentieth-century acquisitions. Largely unused by previous scholarship, with careful attention to when and where objects were collected, and positioned alongside other evidence, use of this material culture as historical evidence reinforces certain historical interpretations, and leads to some novel clues about widespread copper artisanry as well as trade and use of copper across the Copperbelt and its neighbouring commercial partners.

Caravan traders from the west, east and north of Katanga contributed to the expansion of copper production in the early nineteenth century. From the east, the Nyamwezi of present-day central Tanzania pioneered the nineteenth-century copper trade, using copper for regalia, status items, and exchanging it for ivory destined for the east coast.2 Andrew Roberts, ‘Nyamwezi Trade’, in Richard Gray and David Birmingham (eds), Pre-colonial African Trade: Essays on Trade in Central and Eastern Africa before 1900 (London: Oxford University Press, 1970), pp. 54–7. From the west, the Chokwe (and perhaps Lunda), as well the Ovimbundu began to seek trading opportunities in the interior.3 Chokwe and Ovimbundu trade with Katanga is well documented. John Thornton has recently argued that eighteenth and nineteenth centuries encompassed a period of Lunda expansion northwards and eastwards, instead of the usual assumption of a westward expansion towards Angola: John K. Thornton, ‘Rethinking Lunda’s Expansion: 1720–1800 from Within’, paper presented to ‘Angola in the Era of the Slave Trade’, London, 19–20 June 2019. From the north, Luba traded salt and hoes for copper from Katanga.4 Anne Wilson, ‘Long Distance Trade and the Luba Lomami Empire’, Journal of African History 13, 4 (1972), pp. 575–89, p. 580. Within Katanga, local polities, the Sanga in particular, responded to the increase in long-distance trade by producing large copper ingots.5 Wilson, ‘Long Distance Trade’, pp. 579–80, citing Jacques Nenquin, Excavations at Sanga, 1957: The Protohistoric Necropolis (Tervuren: RMCA, 1963), pp. 194, 198, 200.

The Eastern Lunda Kazembe kingdom centred on the Luapula River and Lake Mweru was at the nexus of these early nineteenth-century trade routes. A series of Portuguese expeditions describe the Kazembe polity in the late eighteenth and early nineteenth centuries organised around a wealthy, apparently all-powerful, and sometimes despotic Mwata Kazembe titleholder.6 F. J. M. de Lacerda e Almeida, ‘Explorações dos Portuguezes no Sertão da África Meridional’, Annaes Maritimos e Coloniaes 4, 7–11 (1844); 5, 1–3 (1845); F.J. Pinto, ‘Explorações dos Portuguezes no Sertão da África Meridional’, Annaes Maritimos e Coloniaes 5, 4,5,7,9,10,11,12 (1845); P. J. Baptista, ‘Explorações dos Portuguezes no Sertão da África Meridional’, Annaes Maritimos e Coloniaes III, 5–7, 9–10 (1843); Antonio C. P. Gamitto King Kazembe (2 Vols., Lisbon: Junta de Investigaçŏes do Ultramar, 1960), trans. Ian Cunnison, orig. A. C. P Gamitto O Muata Cazembe (2 Vols. Lisbon, 1937; 1st edn., 1854); J. R. Graça, ‘Expidicão ao Muatayanvua: Diaro de Joaquim Rodrigues Graça’, Boletim da Sociedade de Geographia 9, 8/9 (1890). This description of the Kazembe polity obfuscates the range of reciprocal relationships that underpinned its political and economic structure. The diversity of its local sources of wealth, including salt, fish, copper and the fertile Luapula Valley, entwined the polity in an array of local political structures difficult to discern in Portuguese accounts focused on powerful kings and lucrative ivory exports.7 For agriculture and salt, Giacomo Macola, The Kingdom of Kazembe: History and Politics in North-Eastern Zambia and Katanga to 1950 (Hamburg, LIT, 2002), 44–54; for long-distance trade, Macola, Kazembe, pp. 128–35. For fish and agriculture, David Gordon, Nachituti’s Gift: Economy, Society and Environment in Central Africa (Madison: University of Wisconsin Press, 2006), pp. 50–59. Instead, oral traditions, including those emanating from copper-producing elites, indicate webs of reciprocation.8 For an example of political reciprocation related to dishing resources, see the role of narrative in defining relationships between Lunda aristocrats and owners of the lagoon, discussed in Gordon, Nachituti’s Gift, pp. 36–50. For example, the oral traditions that refer to the earliest chiefs who controlled the copper production of Katanga, notably, the Kipimpi, Kibuye and Kasongo titleholders, relate their settlement to that of the migration story in the Eastern Lunda oral tradition of Mwata Kazembe. They, in turn, tell how they formed alliances, through marriage, with local owners of the land and copper-producing artisans. This may or may not mean that the ancestors of these titleholders migrated from Lunda with Kazembe in the eighteenth century; they do reveal the early nineteenth-century ties of trade and clientage that came to be expressed in terms of these narratives.9 Oral traditions ‘Procès-Verbal’. Also see R. Marchal, ‘Renseignements historiques relatifs à l’exploitation des mines de cuivre par les indigènes de la région de Luishia’, Bulletin des juridictions indigènes et du droit coutumier congolais BJIDCC (1936), p. 10. Copper was one way that reciprocal forms of tribute incorporated the diverse autonomous polities of Katanga into the titleholders that underpinned the Eastern Lunda Kazembe political structure, expressed in terms of relationships of fictive and perpetual kin.10 For perpetual kinship, see Ian Cunnison, ‘Perpetual Kinship: A Political Institution of the Luapula Peoples’, Rhodes-Livingstone Journal, 20 (1956), 28–48; Gordon, Nachituti’s Gift, 27–50.

Such interpretations of oral traditions are further supported by a close examination of oral testimony about production found in colonial documentary sources. These sources suggest that mining copper was labour intensive, even as it involved an extensive division of labour based in a differentiation of expertise, especially compared with other high risk or hyper-exploitative activities linked to the consolidation of polities, such as hunting elephants, trading slaves or collecting wild produce like rubber or wax. Copper production instead seemed akin to activities like the production and processing of fish, salt and raffia cloth, all of which were traded over long distances but involved local and rooted forms of production and expertise. Elites profited little from the production of copper in Katanga itself; instead, copper production distributed wealth across society, constituting cross-gender sodalities that included individuals of different expertise, such as miners, smelters, and those who produced energy inputs, food and other commodities essential to copper production. Different skills encouraged salaried labour, along with a relatively egalitarian form of copper rent: the final product belonged to many workers, including those who dug the mineral, who were involved in smelting, smithing, provision of wood and charcoal, and even those who fed the workers.

The most prominent position associated with copper production was a spiritual expert, a titled position that could be occupied by a man or a woman, termed fondeur in the colonial sources, or sendwe. This titled sendwe position was a mediator, an owner of the copper mine, who gave permission for others to work the mine.11 See for example the Inamfumo of Kalabi discussed by Hermenegildo Capelo and Roberto Ivens, De Angola á contra-costa: Descripcão de uma viagem através do continente africano (Lisbon: Impresa Naçional, 1886), p. 70. Nineteenth-century traveller’s accounts indicates that the sendwe owner of the mines were frequently women. In other regards, however, oral testimony suggests that a gendered division of labour was evident. Teams of men extracted the copper-bearing rocks with hoes and axes, while women and children washed and dried it. Nonetheless, the distinctive masculine culture found in the blacksmithing traditions of West-Central Africa does not seem prevalent.12 Coleen E. Kriger, Pride of Men: Ironworking in Nineteenth-Century West-Central Africa (Portsmouth NH: Heinemann, 1999).

Work began after harvest and at the commencement of the dry season. When enough mineral was acquired, the sendwe titleholder chose an appropriate termite hill for a smelting site. Malachite ore was transported for several kilometres from the site of initial processing to the smelting furnace in special baskets made out of strong and fibrous wood. Near the furnace, charcoal was produced out of hardwood, and transported to the foundry furnace. The process took several months – during which time food had to be supplied to workers.13 For wood, food, termite hills, religious and technical expertise, see ‘Procès-Verbale’, p. 14. For description of process, see de Hemptinne, ‘Les “Mangeurs”’, 380–82.

Oral evidence collected by the colonial District Administrator, R. Marchal, indicates that by the 1850s or even prior, the exploitation of these deposits had reached a veritable boom, attracting labour from across the region. Even as the sources do not indicate the exact nature of servitude, they claim that ‘slaves’ could purchase their freedom after three to four seasons of copper mining.14 For forms of slavery and manumission in this region during the nineteenth century based mostly on oral testimony, see David M. Gordon, ‘The Abolition of the Slave Trade and the Transformation of the South-Central African Interior’, William and Mary Quarterly 66, 4 (2009), pp. 915–38. An example of Catholic Missionary sources on slavery and manumission, in David M. Gordon, ‘Slavery and Redemption in the Catholic Missions of the Upper Congo, 1878–1909’, Slavery and Abolition, 38, 3 (2017), pp. 577–600. Testimonies indicate that migrant (or foreign) workers outnumbered original inhabitants. This was a cosmopolitan copper rush, with people arriving to make their fortune. The Kapururu titleholder ensured they were fed, which created the need for a food service, with those who provided food also paid in copper.

This form of labour differentiation suggests a degree of permanent settlement and political organisation linked to the production of copper. The mining camp of Lubushia (present-day Luishia), only one of several prosperous Katangese mining centres, was, according to informants in the 1930s, ‘aussi important que celui de l’Union Minière.’ An uninterrupted line of carriers transported the mineral from the mine to the foundry four kilometres away. In fact, informants remembered it being so busy that there had to be a two-way road. At the foundry, workers supplied charcoal, and they were also paid in copper. Even given potential exaggeration, these were years of prosperity. Marchal estimated that the number of miners at Lubushia mine surpassed 900, with approximately one-third migrants. According to his estimates, to feed these miners, Lubushia would need to have had four to five thousand inhabitants, an expansive economic endeavour for this region and this time period.15 Marchal, ‘Renseignements historiques’, p. 15.

Mining was also undertaken in a decentralised fashion, perhaps typical of older mines and foundries. In the area of a chief, called Kiembe, each village had one or two copper mining teams, consisting of ten to fifteen men, producing around 1,000 kg in a season. In these areas, migrants who had an agreement with the sendwe titleholder could as well be part of teams, or form their own teams, which seems to have been the general preference. Kiembe claimed some twenty villages of his chieftaincy were involved in copper mining, with around 300 local workers and 200 migrants working on the copper fields.16 Ibid.

Worker remuneration, according to Marchal’s informants, was fairly standard, and rendered in the form of the famous 20 kg Katanga copper crosses (fishinkoro).17 De Hemptinne describes these as milopolo. All the families from the titleholder Katanga’s area where men mined copper and the women and children furnished food, received three crosses or about 60 kg of copper for one season’s work (up to five months), or equivalent trade items, including cloth, tools and arms. Two crosses were received immediately after work and another one or two months later, when the Katanga titleholder called all chiefs and distributed refined copper and trade items. Those from away, migrant workers who received food from Katanga as part of their payment, were paid from 30 to 50 kg of copper for a complete season. Workers used the copper to purchase cloth and slaves in addition to local manufactures. Thus, copper production multiplied into other economic activities.

In terms of political structure and authority, copper tribute reinforced the titleholding system found across this region (represented in the oral traditions discussed above). At the base of a triangle, independent teams of about fifteen men producing around 1,000 kg per season, gathered around lesser titleholders. In addition to each member of the team receiving three crosses (60 kg) as payment, each team gave a tribute (mulambo) to the lesser titleholder of 100 kg. Moving up the triangle, lesser titleholders gave 60 kg as tribute to the chief sendwe titleholder. Each sendwe titleholder sent tribute to political titleholders, such as the Katanga titleholder, who in turn sent an annual tribute to regional political authorities at the apex of the triangle, such as the Kazembe titleholder. Katanga sent Kazembe around 450 kg annually, which had to be carried by 20 to 25 men. Marchal estimated that about 115,000 kg was distributed from workers through these titleholders yearly.18 Marchal, ‘Renseignements historiques’, p. 17. The Kazembe apex of this copper production was so well known in the 1860s, prior to the consolidation of Msiri’s polity, that copper was known to come from ‘Kazembe’s country’, not Katanga.19 As evidenced in collections made in by David Livingstone along the Shire in 1861 of copper wire from ‘Bazembe’s country’ (presumably Kazembe). National Museum of Scotland (henceforth NMS), Museum Reference A.762.4. Presumably the labeller used the common prefix “Ba” instead of the correct “Ka”. Personal communications, Sarah Warden and Lawrence Dritsas.

That this oral testimony makes such a clear distinction between a local, ancestral community on the one hand and new migrants on the other suggests that migration in response to copper production had increased through the nineteenth century. By the middle of the century this was an expanding industry, spurred on by greater trade opportunities and the emergence of commodities that entered central Africa via Eastern African trade routes, in particular cloth. These imported commodities, in turn, inspired the need for local manufactures, copper in particular, that could be traded regionally for commodities, chiefly ivory, that underpinned international trade. By the middle of the century, the Nyamwezi (from present-day Tanzania) were the brokers of this regional-global trade.20 Roberts, ‘Nyamwezi Trade’, p. 57. The best contemporary documentary record of Zanzibari and Nyamwezi copper traders in this period is from David Livingstone, The Last Journals of David Livingstone in Central Africa (London: John Murray, 1874), pp. 291, 298, 301, 310, 331–3, 337, 358–9. They dealt directly with the titleholders who sat at the apex of tribute triangles, and accumulated quantities of copper to exchange. One such titleholder who worked closely with the Eastern African traders was Nsama, who, at the time of his defeat by Hamed ben Muhammed (Tippu Tip), had some 700 frasilahs (24,500 lb.) of copper stored in his village (alongside 68,250 lb of ivory), which he was then trading with the Nyamwezi.21 François Bontinck (trans. and ed.), L’autobiographie de Hamed ben Mohammed el-Murjebi Tippo Tip, ca. 1840–1905 (Brussels: Académie Royale des Sciences d’Outre-Mer, 1974), pp. 56, 202–3, fn. 103.

Beyond mining and trade, copper items were produced for various uses during this period. Here, the archaeological record, in particular De Maret’s informative work, needs to be supplemented by additional sources, including documentary sources and ethnographic collections. De Maret identified a gradual move towards the proliferation of small crosses and the production of larger copper ingots, probably due to inflationary pressures on copper (see Figures 6.1–6.3).22 Pierre de Maret, ‘L’évolution monetaire du Shaba Centrale entre le 7e et le 18e siecle’, African Economic History 10 (1981), pp. 117–49. Pierre de Maret, ‘Histoires de croisettes’ in Luc de Heusch (ed.), Objets-signes d’Afrique (Tervuren, Musée royal de l’Afrique centrale, 1995), pp. 133–45. The point to appreciate, however, is also that due to the increasing availability of copper within expanding economies, finer amounts of copper were generally traded in the form of copper wire (or strips and nails) instead of crosses. (Dating the introduction of techniques of producing Katangese copper wire still requires investigation). In 1861, during his Zambezi Expedition (1858–64), David Livingstone acquired a finely smithed coil of copper wire in the Shire River (Manganja), identified as being from the country of Kazembe.23 NMS, A.762.4. The label identifies the object as .Manganja from the country of Bazembe’. Henry Stanley collected a very similar copper wire coil, as illustrated below. In 1884–85, Capelo and Ivens detail a well-established technique of wire manufacture and use:

Copper wire could be fashioned into many items and used for a range of decorative purposes, along with other manufactures such as hoes, axes, and knives. The currency function of copper crosses began to decline, as trade items began to be measured in terms of imported beads and cloth. These imports nonetheless spurred copper production, particularly wire (see Figure 6.4).

Ethnographic collections of objects from the late nineteenth century also indicate that artifacts, many of them made with copper wire, strips or nails, proliferated through this period. Around the Kazembe kingdom, copper was used as a form of status, decorating weaponry as well as adorning bodies of men and women. Artifacts collected near Lake Mweru by Emil Torday in 1904, but which appear to have been used for decades before that, include status items: a snuff box inlaid with copper and brass strips and a large knife, engraved with two triangles on each side, with a handle of wood inlaid with copper and brass wire, with copper/brass studs and a conical copper cap. A third object collected by Torday in the same Lake Mweru area, identified as a ‘Luba Charm’, is a remarkable small figure of hard black wood studded with copper nails, probably indicating the status of the spiritual figure or ancestor depicted.25 British Museum (hereafter BM), Snuff box, Copper with brass, Asset #351644001, Reg. #Af1904,0611.12; Knife, Asset #1613039012, Reg. #Af1904,0611.31.a; Luba Charm: Asset #1613057240, Reg. #Af1904,0611.20. (On the knife and the ‘charm’, the studs here are labelled as copper, even though they are usually brass; I have not been able to inspect the items closely.)

By the middle of the century, Katanga, then referred to as the land of Kazembe, was known across Central Africa as the place of copper and associated opportunities. Leaders like Katanga Kapururu encouraged in-migration, benefitting from their labour and exploitation of copper, and facilitating relationships from migrants and local masters of country. They also served as intermediaries between migrants, autochthonous titleholders, the Mwata Kazembe titleholder, and the growing number of Swahili and Nyamwezi traders.26 Marchal, ‘Renseignements historiques’, p. 15; for family exploitation, also see ‘Procès-Verbale’, p. 14. Oral traditions expressed political relationships between these new migrant workers, long-standing owners, traders, and the titleholders of the Eastern Lunda, defining the rules of access to resources, systems of trade and clientage, and hence, when examined closely, an entire political economy.

The major political change of this era was a reshaping of older forms of triangulated titleholders through which copper had been previously distributed, and especially the replacement of those at the apex of the triangles. Networks remained; rulers were more transient. The historiography of Msiri’s polity, like the Kazembe kingdom, is conventionally oriented towards the persona of the ruler himself, and not the network of relationships in which the polity was entwined. In part, this is because most outside observers, especially those connected with the Congo Free State expeditions, described Msiri as a cruel despot. Those who remained at his court for longer periods, such as Arnot, paint a different portrait, describing many individuals with significant power and authority. Arnot was particularly taken with the ‘remarkable’ rights of women: ‘Women are allowed to attend the court, and to have a voice equally with the men.’4 Arnot, Garanganze, p. 241. Oral traditions, like those of the Kazembe kingdom (in fact some of the narratives seem to be borrowed from the Kazembe kingdom), portray Msiri’s rule as rooted in local forms of legitimacy.5 A point emphasised by the most recent historical account of Msiri, which argues against describing Msiri’s polity as ‘warlord’: Hugues Legros, Chasseurs d’ivoire: Une histoire du royaume yeke du Shaba (Zaïre) (Brussels: Editions de l’Université Libre de Bruxelles, 1996). Copper was one of the principal mechanisms for ensuring that Msiri’s polity was entwined with these local networks.

Oral traditions, first recorded by Arnot, have Msiri gaining his knowledge and experience in Katanga from his father, Kalasa, who was primarily a trader in copper with the Sanga.6 Arnot, Garanganze, p. 231; Legros, Chasseurs, p. 40. (A popular version of the oral tradition has Msiri ‘following an elephant’ to Katanga; this is probably a later emphasis in an oral tradition that valorised Msiri’s ivory trade above copper). Msiri then extended his influence by using his guns to aid the Sanga to defeat the Luba. The presence of Msiri further contributed towards the breakdown of the Katanga copper trade with the Luba polities to the immediate north.7 Wilson, ‘Long Distance Trade’, p. 586. Msiri’s initial alliance with Sanga typified the type of powerbrokering and integration that characterised his polity.8 Legros, Chasserurs, pp. 39–57. With regard to copper production, documentary evidence indicates that Msiri relied on a network of local families to produce copper. For example, Arnot wrote of a type of trade caste involved in the mining of copper, where the ‘business is handed down from father to son, and the instructions of the forefathers are followed with the greatest accuracy.’9 Arnot, Garanganze, p. 238. Although based within families, outsiders associated with Msiri’s Yeke were incorporated into these castes, making them into more expansive associations.

Early Belgian colonial enquiries portray a different picture of Msiri to Arnot, perhaps because their informants were not as closely aligned with Msiri. (Colonial accounts were also interested in demonstrating Msiri’s illegitimacy in Katanga, as they did not want to remunerate Yeke chiefs for mineral rights). Most testimonies claim that Msiri, by insisting on centralisation and control of the copper market, discouraged the egalitarian and associative nature of copper mining, leading to its decline in popularity. In some cases, such as the Kasonta mine of Kasongo, Msiri was said to have prohibited the mining of copper under penalty of death except by those who declared direct allegiance to him.10 ‘Procès-Verbal’, p. 12.

Msiri and the Yeke are also credited with the introduction of mobile furnaces that were temporary, destroyed after each copper smelting operation.11 De Hemptinne, ‘Les “Mangeurs”’, pp. 379–80. Perhaps such operations were well suited to nineteenth-century warlordism, since they were not re-used and thus did not require major sedentary investments. They could have discouraged the established norms and rituals of copper exploitation, replacing the ties of between copper mining families, societies and titleholders with a type of ad hoc exploitation. The Yeke mobile furnaces relied on extensive labour and energy inputs, such that copper could only be produced in the dry season, once agricultural work was over and there was plentiful wood. The final product of these Yeke mobile furnaces was reputedly not as fine as copper produced by the Sanga. Perhaps the Yeke inability to smelt copper as efficiently as the Sanga and their reliance on the Sanga for finer copper smithing, in addition to the changing trade and other factors detailed below, further contributed to Msiri’s abandoning attempts to control the copper trade from the 1880s.

To recap, there are two different emphases in the evidence of the relationship between Msiri’s polity and copper production. First, early versions of the oral traditions and Arnot have Msiri relying on and working with local copper producers. Second, however, are the colonial-era sources that indicate Msiri abandoning copper and clamping down on independent production, in favour of ivory.

The incongruous evidence of the relationship of Msiri’s polity to copper can be reconciled in part by late nineteenth-century changes in Msiri’s approach to copper mining. As he abandoned attempts to control production and trade, autonomous polities, the Sanga in particular, continued small-scale copper production. Thus, while contemporary witnesses such as Arnot describe the trade in copper, it does not appear to have been monopolised by Msiri. This explains why oral informants like Nkuba drew such a stark contrast between a trade in copper for ordinary people and the trade in ivory for the wealthy and powerful. Msiri, never able to control copper production and fearful of providing trading opportunities to his rivals, became intent on having it closed down.

Copper provided the necessity and opportunity for Msiri to root himself in local forms of production, to ally with local leaders, and to rely on local miners and artisans. Motivating Msiri’s migration to and settlement in Katanga, copper contributed to the extension of trade and other networks that constituted the Yeke polity, even as it disrupted older flows of copper production, tribute and trade. Msiri was never able to fully control the political economy of copper however. His failure to do so led to an alternative orientation around ivory, along with an entrenchment of some of the despotic and war-like features of the Yeke polity.

Part of the reason for the declining investment of Msiri in copper production was a reorientation of Msiri’s trade from east to west. Trade in copper northwards and eastwards had been disrupted. The Eastern Lunda Mwata Kazembe titleholder had previously received an estimated tribute of 450 kg of in copper from the Sanga per year. The increase in Swahili trade with the Yeke bypassed Kazembe’s control, leading to conflict. In one telling and impactful incident in 1868, Mwata Kazembe Muongo confiscated 1,700 kg of copper from the Zanzibari trader Mohammad Bughari, who had purchased it from Msiri. Perhaps this was an attempt by Kazembe to combat the erosion of his Katanga political networks by Msiri’s Katanga alliances and the Swahili-Nyamwezi copper trade. This incident, in addition to the usually cited reluctance of the Zanzibari to sell firearms to Msiri, contributed to Msiri’s decision to turn away from the eastward trade, and in particular the trade in copper.1 Legros, Chasseurs, p. 115; Macola, The Kingdom of Kazembe, p. 141; Bontick, L’autobiographie, p. 202, fn. 102; Livingstone, Last Journals, pp. 276, 297.

Instead, to bypass Kazembe and avoid reliance on Swahili traders, Msiri turned to the west, in particular Ovimbundu caravans financed by Silva Porto. They did not demand copper, in part because cheaper brass, used for similar purposes, was available from Angolan Atlantic ports, as further discussed below. Instead, the Ovimbundu encouraged the collection and sale of rubber, beeswax and ivory.2 According to informants of Marchal, ‘Renseignments historiques’, p. 13 Msiri, initially attracted by the copper fields, thus turned from copper to ivory and other exports. Since copper production wasted local labour inputs compared with other more profitable exports, Msiri discouraged it.3 As concluded by Marchal, ‘Renseignments historiques’, pp. 12–13

Heavy and haphazard exploitation of the richest deposits of the prior decades also contributed to decline. The mines most exploited over the previous decades, which were open-pit and sometimes underground, dug in vertical shafts of 15 to 20 feet, had become dangerous. On at least two recorded occasions, at Kambove and the Sanga mine of Kalabi, they collapsed.4 Capelo and Ivens, De Angola á Contre-costa, pp. 69–70; de Hemptinne ‘Les “Mangeurs”’, p. 377; Cornet’s report on Kambove in his ms, and in J. Cornet, ‘Mines de cuivre du Katanga’, Le Mouvement Geographique 12, 1 (1895), describes the extent of the deposits and excavations at Kambove. At Kalabi, a collapse of the mine in 1882 had closed the mine. The owner of Kalabi, a woman with a royal title Inamfumo, was allegedly waiting for a sign, perhaps in a dream, that the mine was safe to be opened once again.5 Capelo and Ivens, De Angola á Contra-costa, pp. 69–70. The mine collapsed two years prior to their 1884–85 expedition.

Imported European copper and brass began to replace and undermine the profitability of Katangese copper. With the influx of foreign copper and brass, artisanship – the smithing of copper – was valued above the raw material, confirming earlier trends for the preference of wire over crosses. Copper crosses were still produced, to be sure, but once copper was transformed into wire or bullets, it attained a higher value. Sanga informants claimed that they far preferred to work the copper into wire (mitako or mitaga) and bullets than to sell it in ingot form. The weight of the ingot was far greater than the equivalent weight of wire of the same value. Copper bullets (chipolopolo), which, as Zambian soccer fans know, move at high speed, were the most valued.6 De Hemptinne ‘Les “Mangeurs”, pp. 396–7, 401–2 In other words, people paid for the labour of artisans who could convert copper into forms that were easily used – bullets and wire, in particular. Like iron blacksmithing in this period, coppersmithing and craftmanship were valued above smelting and the production of the metal itself, at least on the Copperbelt.7 For the argument of the value of blacksmithing above smelting, see Kriger, Pride of Men.

Copper smithing and craftsmanship could, however, be applied to other, imported copper alloys. In those places that had been significant consumers of Katangese copper, a range of artisanal practices linked to copper alloy artisanry proliferated by the late nineteenth century. The replacement of Katangese copper by imported brass is best appreciated by considering the objects, including insignia of political and religious power, collected in the late nineteenth or early twentieth century, in a vast arc, from Lakes Mweru and Tanganyika in the east, to Angolan societies around the Kwango river. In this region, the ubiquity of copper is evident. Throughout the Luba region and in related polities to the north and west, which had in the early nineteenth century imported copper from Katanga, late nineteenth-century metal daggers were inlaid, decorated or developed handles with copper. Distinctive styles emerged. For example, in the north of the region, mobile warrior Tetela workshops decorated the handles of the daggers and knifes with pointed or conical copper ends. The function of these pointed ends seems unclear – potentially to attach to a wooden shaft, or as an additional weapon through a back-handed strike. The grips of the handles of these knives were also inlaid with copper wire and studded with brass. Perhaps these knives were meant more as decoration and indicators of power than as functioning weapons.8 These include many collected in American Natural History Museum (AMNH); and BM Tetela knife with copper wire handle, Ass# 1613050585, Reg. #Af1909, Ty.782; Tetela knife also with slightly pointed end, but not as sharp Reg. – Ass.# 1613050061 – #Af1907,0528.351. The status element of copper-decorated weaponry was certainly evident for the Songye, immediately east of the Tetela, and with a similar reputation – artists inscribed iron and copper axes with up to 40 faces, perhaps depicting the followers of prominent leaders.9 BM Songye (Zappo-Zap) Ornamental Axe with 40 faces made of iron wood and copper, Ass. # 1613050840, Reg. #Af1907,0528.398; Songye (Nsapo) Asset #34314001, Reg. #Af1909, Ty.976. Also published in C. J. Spring, African Arms and Armour (London: British Museum Press, 1993). Also see AMNH collection. Warriors of the Songye of Zappo Zap (Nsapu-Nsapu) coated a stool, presumably for an important man, with copper.10 Zappo Zap stool – copper overlaid on wood. BM Asset #1613229315, Reg. #Af1907,0528.402. Miniature Pende initiation badges, in the form of their renowned masks, often made in ivory or bone, were also moulded in copper.11 Pende initiation badge – mask made of copper, BM Asset #1613047273, Ref. # Af1910,0420.457.

By the latter half of the nineteenth century brass began to replace copper in the western regions, where authority had passed from the Lunda of Mwaant Yav to Chokwe traders and hunters. As men of the gun, their profession demanded that the Chokwe were masters in metal work. The guns that proliferated through this region were low-quality lazarinas or lazerinos, flintlock muzzle-loaders, which historian Giacomo Macola argues had a profound impact on warfare and political economy on the Angolan plateau after 1850, coinciding with the rising power of the Chokwe.12 Giacomo Macola, The Gun in Central Africa (Athens: Ohio University Press, 2016), pp. 53–73. In addition to marksmanship, Chokwe required expert smithing skills to repair these low-quality guns, and to manufacture bullets. These smithing skills underpinned their craftsmanship in metal, copper, brass and iron. They celebrated their metal-working and the importance of metal to their prosperity by ornamenting themselves and their guns with copper and brass. They inlaid their gunstock with copper and brass wire and studs, to help with grip but also for decoration.13 See extensive collection held in National Museum of Ethnology, Lisbon, Portugal. The most striking and distinctive element of Chokwe material culture was their ibenye or yipenye head ornamentation (or diadems) made by flattening brass wire into stylised blades. Yipenye were most typically made for the head (cipenya mutwe), but also referred to other forms of Chokwe-made brass pageantry.14 Henrique Augusto Dias de Carvalho, Ethnographia e Historia Tradicional dos povos da Lunda (Lisbon: Impresa Nacional, 1890), 340–1; Marie-Louise Bastin, Statuettes Tshokwe du héros civilisateur ‘Tshibinda Ilunga’ (Arnouville: Arts d’Afrique Noire, 1978), pp. 100–1. In his 1904 expedition, Porto adventurer and anthropologist, Fonseca Cardoso photographed a Chokwe chief with an elaborate brass headdress. These yipenye became widespread fashion, representing distinction and elegance, worn even by the daughter of the Lunda paramount titleholder, Mwaant Yav.15 P. Pogge, Im Reiche des Muata Jamwo (Berlin: Verlag von Dietrich Reimer, 1880), p. 195, with an illustration on p. 194. Elites sought Chokwe-fashioned brass ornamentation throughout this region.

The Kuba, who had long traded with the Chokwe, also used Katanga copper, and in the nineteenth century began to value brass. Emil Torday claimed that when brass came from the west, ‘it was so rare that it had a value equal to our gold and the king alone was allowed to possess or use it.’16 Quoted in Herbert, Red Gold, p. 177. Its rarity was linked to it being in high demand by the neighbouring Chokwe, who were also the purveyors of brass and promoted it as a fashionable item to signal elite status across the region. The Kuba, as evidenced by Torday’s collections, combined copper and brass, in their varied manufactures: whistles, combs, pipes and razors, were either made or inlaid with copper, in addition to their famous ikul knives with copper handles or blades, or both.17 Kuba (Bangongo) smoking pipe, copper wire, BM, Asset #1613237540. Reg. Af1907C7.175; Kuba Bangongo razor made of copper, Ass. #1613235461; Kuba whistle with copper wire, Asset #1613243250, Reg. #Af1908, Ty.205; Kuba ikul knife. Found in many other collections. Copper wire and brass studs were attached to all items to increase decoration and indicate wealth. The prominent polities and traders of the west, from the Chokwe to the Kuba, valued copper, but, as indicated in their material culture, began to replace it with imported European metals, brass in particular.

It is important to nuance the impression left by the art historical record here: gauging by the amount and prominence of nineteenth-century artisanship and use, copper and copper alloy items were first produced and sold by mobile warriors who used them as status items; only then, did the sedentary Lunda and Kuba begin to adopt them. Viewing the relationship of copper to political formation through the production and use of copper by sedentary states misses the key role of mobile associations in the history of copper.

In central Katanga during the 1890s, copper was still widely used and traded among all, even as Msiri did not invest greatly in its actual production. Le Marinel noted plentiful copper around Msiri’s settlement, with a copper cross worth three small pieces of cloth (mouchoirs). Still, even given the relatively low value that Le Marinel reported, copper adorned precious guns, and a royal execution lance.18 P. Le Marinel, Carnets de route dans l’Etat Indépendant du Congo de 1887 à 1910 (Brussels: Progress, 1991), pp. 160, 185. Bonchamps, on the Stairs expedition that ended Msiri’s rule, reports that Msiri wore heavy rings of ivory and copper on his legs.19 R. de Pont-Jest, ‘L’Expedition du Katanga, d’après les notes du Marquis Christian de Bonshamps’, Le Tour du Monde (1893), pp. 257–72, p. 262. Even as it remained a status item, Msiri concentrated on the production of ivory, rubber and beeswax for export: more profit was to be made from other commodities and westward-oriented trade routes. Copper was still traded to be sure; however, Msiri only insisted on, or could only enforce, his own trade monopoly over ivory.20 Arnot, Garangaze, pp. 234–5. Indeed, so key was ivory to his polity that the Yeke oral tradition came to celebrate Msiri’s alleged following of an elephant to Katanga as the founding migration myth. Copper was the trade item of ordinary people; ivory that of the global trader and warlord.

Yet this late-nineteenth-century impression of Msiri’s Yeke can be misleading, since the polity was in fact a product of copper-based networks of production and trade. Msiri did not ‘follow an elephant’ as the myth proclaims, but rather built on the copper-trading relationships of his father. The Yeke allied with local rulers and employed their mining experts and technologies. At least some Yeke were initiated into their mining and smithing fraternities. It was only with the collapse of these copper relationships by the late 1880s, when subject peoples near the Yeke heartland, the Sanga, rebelled against Msiri, that he invested in the arms and ammunition to suppress them. From this period some of the least flattering portrayals of Msiri as a foreign and illegitimate ivory and slave-trading despot emerges. These depictions, partly alibis for invasion, justified the action of King Léopold’s mercenaries. In 1891, in the midst of negotiating with Congo Free State agents over guns and ammunition, Omer Bodson assassinated Msiri.

European colonial regimes succeeded the trader-warlords of Katanga. At first uninterested in copper, for them as well, indigenous copper mining represented wasted labour. The first administrator of the only mine still producing copper in Lukafu, ‘Bwana Capitaine’ (Commandante Gor), secured a good portion of the approximate two tons produced in Lukushi in 1903, and then sent soldiers to ensure the closure of the mine.21 Marchal, ‘Renseignements historiques’, p. 13 In the context of various demands on labour, for hunting, for war, for agriculture, and, for the collection of tribute by predatory regimes like Msiri’s and the Congo Free State, local leaders could not sustain the labour demands of a copper-producing economy. Rulers did not deem it profitable for them to do so: there were more effective ways to capture profit from South-Central African labour, in the short term at least. Copper’s political networks, from Kazembe through the Yeke to the Chokwe, ended, only to be resurrected some two decades later.

By the middle of the century, imports of cloth from the east spurred the production of copper. In addition to the currency use of copper crosses and wire, artisans crafted a range of manufactured copper items that signified wealth, power and fashion. As guns became part of this globalised trade, copper bullets were also in demand. With, however, the inability to control local copper production and trade, the increasing globalisation of trade, the influence of Angola traders from the west, along with imports of cheaper brass, the mining of copper declined, and was discouraged by those leaders who prospered through international trade, such as Msiri. Metal-working artisanry and use that had been encouraged by the spread of Katangese copper were now applied to imported European metals, brass in particular, as evidenced by networks of Chokwe brass artisanship. With the declining influence of copper’s associations in Katanga, centralised and sometimes violent rule came to predominate.

Across the region and through the nineteenth-century Katangese copper had constituted communities and political networks. Copper contributed to diverse forms of associational life, to the wealth and well-being of many. The political economy of copper in the nineteenth century can be viewed as precursor to the expanding labour arrangements and forms of associational life on the colonial Copperbelt. Nonetheless, it is rarely viewed as an early modern phenomenon, for, by the time the modern and industrial copper mining arose in the early twentieth century, the dynamism of copper mining, production and trade during the nineteenth century appeared part of a distant past, relegated to the study of archaeology, often cast in the idiom of tradition and sedentary kingship, not part of a modern political economy and civil society. This chapter has pointed out the ways in which the history of the nineteenth-century Copperbelt is part of the history of the modern twentieth-century Copperbelt.