Chapter 3 Hidden in Plain Sight: The Temporal Layers of Transport Corridors in Uganda

In some ways, all this is also true of East Africa. The main modern Northern transport corridor in East Africa links the port of Mombasa to Rwanda and DRC, passing through Uganda and connecting the states of the East African Community. In fact, however, this corridor is actually mainly based on the colonial – and in some cases pre-colonial – transport system that has been used for trade and political purposes for centuries. The correspondence of the location of the corridor in the Kenyan section from Mombasa to Lake Victoria (which can be seen by simply superimposing a contemporary map on to a colonial map) is less obvious in the Ugandan section. This is in part as a result of the very limited number of available maps of the early transport system in Uganda (if any exist at all). It is this area that this chapter will analyse, focusing on south-west Uganda as the portion of the current corridor where local trading activities linking different regions (cross-border trade) were not only influenced by external events such as the arrival of Europeans, but were also closely related to, and determined by, local dynamics. Despite the very local characteristics of the trade routes we analyse here, we can see that they answered – albeit indirectly – to external demands and stimuli, as was the case in other parts of the world, such as South Asia.4 ADB, et al., 44.

Dubbed ‘untouched’ Africa by some authors due to its lesser level of involvement in the slave trade, the Great Lakes Region has long been seen as an isolated region that was not part of the trade system that connected the Swahili region globally (albeit indirectly).5 Among others see Richard Gray and David Birmingham (eds), Pre-Colonial African Trade: Essays on Trade in Central and Eastern Africa before 1900, London 1970. However, a few more recent studies have established that the Great Lakes Region was linked to the coast by an extensive and quite well-maintained mobility system.6 For example Karin Pallaver, ‘Nyamwezi Participation in Nineteenth-Century East African Long-Distance Trade: Some Evidence from Missionary Sources’, Africa: Rivista Trimestrale di Studi e Documentazione dell’Istituto Italiano per l’Africa e l’Oriente, 61:3/4, 2006, pp. 513–31. The main aim of this chapter is therefore to offer a contribution to a better understanding of the socio-economic context of this region prior to the arrival of the Europeans. Among the key conclusions of this chapter is that colonial structures and systems responded to colonial needs, but were also built on to pre-existing networks.

In many cases, modern-day transport corridors are not a new invention, just as they were not a colonial creation. In our case, we see the evolution of an earlier trade and transport system that consisted mainly of paths, but also included routes by water across Lake Victoria and Lake Tanganyika, connecting the interlacustrine kingdoms with the Swahili region.2 In this work, the author uses the term ‘interlacustrine kingdoms’ to refer to the kingdoms that developed in the territory that is now Uganda, which were broadly speaking situated between Lake Victoria, Lake Albert and Lake Edward. From east to west, they are Busoga, Buganda, Bunyoro, Toro and Ankole. This chapter is based on both archival and published materials, and focuses on the nineteenth century and the great transformations, both economic and political, that that took place during this period. The 1800s marked the beginning of the colonial era but, as we will see, it was also a time of major changes in internal equilibria among different kingdoms and peoples, when new ideas and technologies arriving from outside Africa merged with internal changes and practices that persisted, sometimes underground, despite colonial efforts to modify them.3 There are many examples of local practices that persisted despite colonial projects. One interesting example that has received only limited academic attention so far is the change in the currency used in the continent. Among others, see Karin Pallaver, ‘“The African Native Has No Pocket”: Monetary Practices and Currency Transitions in Early Colonial Uganda’, International Journal of African Historical Studies, 48:3, 2015, pp. 471–500. Owing to the breadth of the topic and the lengthy timeframe, this chapter focuses exclusively on those routes (across Uganda) and markets that later became part of the modern Northern trade corridor. Waterways, particularly those across the lakes, will not be analysed for the same reasons. Using established historical works such as those by Jan Vansina and David Newbury, it will show that this region of Africa did not diverge very much from Europe and Asia (despite the obvious differences) as regards colonial and modern political and economic attitudes towards building on existing systems to enhance or create modern infrastructures and practices.

Another essential element of this story is markets, which were a crucial component of the infrastructures that linked different kingdoms and opened up the allegedly ‘untouched’ region to the rest of the continent. The importance of markets also lay in the extra-economic role they played: as places where people gathered, and which were often located close to courts, they had a political, legal and cultural function. They were places where people found out the latest news or gained access to the ruling powers. We have included markets in our analysis, both as places and as a principal component in economic terms because of their importance for proving that a commercial structure and trade infrastructures existed before the arrival of either the coastal traders or the Europeans.4 On the debate on the concept of ‘market’ and the origins of marketplaces in Africa see, among others: B. Turyahikayo-Rugyema, ‘Markets in Precolonial East Africa: The Case of the Bakiga’, in Current Anthropology, 17:2, 1976, pp. 286–90; G. N. Uzoigwe, ‘Precolonial Markets in Bunyoro-Kitara’, Comparative Studies in Society and History, 14:4, 1972, pp. 422–55; B. W. Hodder, ‘Some Comments on the Origins of Traditional Markets in Africa South of the Sahara’, Transactions of the Institute of British Geographers, 36, 1965, pp. 97–105; Charles M. Good, ‘Markets in Africa: a review of research themes and the question of market origins’, Cahiers d’Études Africaines, 13:52, 1973, pp. 769–80. In this sense, roads and markets are essential parts of a trading system in which local people were already actively involved in their trade with neighbouring populations and polities (whether or not they were centralised kingdoms) before outside traders and authorities arrived. This does not mean that we are denying or underestimating the external influences in any way, but simply that as they expanded, upgraded infrastructures, and reorganised trading structures, the Europeans built on existing structures that had been created by local authorities – although they naturally responded to different needs and interests, and so the structure was changed whenever it suited them.5 Olivier Pétré-Grenouilleau, ‘Long-Distance Trade and Economic Development in Europe and Black Africa (Mid-Fifteenth Century to Nineteenth Century): Some Pointers for Further Comparative Studies’, African Economic History, 29, 2001, pp. 163–96, at pp. 192–93; Alberta O. Akrong, ‘Trade, Routes Trade, and Commerce in Pre-Colonial Africa’, in N. N. Wane (ed.), Gender, Democracy and Institutional Development in Africa, Cham, 2019, pp. 67–98.

While we have an abundance of sources describing markets and trade routes in West Africa because of its long-standing links with the Mediterranean world, which have led to a far lengthier and deeper knowledge of commerce in the area, the existence of markets and trade links in the interlacustrine region was only first described by the Europeans who arrived in the area in the nineteenth century.6 Good, Markets in Africa, pp. 769–80. Most works on modern Uganda therefore focus on the two most powerful (and antagonistic) kingdoms in the area: Buganda and Bunyoro. The pre-eminence of the former during the nineteenth century is undeniable, and it was also there that the first Europeans to reach the region were accommodated, but despite its efficient centralised system and regional role, it is not the only protagonist of this chapter.7 John A. Rowe, ‘The Western Impact and the African Reaction: Buganda 1880–1900’, The Journal of Developing Areas, 1:1, 1966, pp. 55–66; Buganda is usually described as a centralised kingdom, with the Kabaka as a clear central authority, but this description overshadows the presence of different cultures and peoples within Bugandan territory, particularly regarding its 1800s acquisitions and at its periphery. The nature of the kingdom is not the focus of this chapter, but it is important to remember that these kingdoms were far more varied than they have often been described to be: Aidan Stonehouse, ‘The Bakooki in Buganda: Identity and Assimilation on the Peripheries of a Ugandan Kingdom’, Journal of Eastern African Studies, 6:3, 2012, pp. 527–43. Among the most important regional kingdoms, Bunyoro has a twofold significance, as the place where many markets were organised,8 Charles M. Good, ‘Markets in Africa’, p. 773. and as a centralised power that controlled a broad swathe of territory, including essential areas that produced two vital items for inter-kingdom trade – salt and iron. The kingdom of Bunyoro was what remained of the earlier empire of Bunyoro-Kitara, which fell in the 1800s.9 On the history of Bunyoro-Kitara and its fall, see, among others: G. N. Uzoigwe, ‘Bunyoro-Kitara Revisited: A Reevaluation of the Decline and Diminishment of an African Kingdom’, Journal of Asian and African Studies, 48:1, 2012, pp. 16–34; A. R. Dunbar, A History of Bunyoro-Kitara, Oxford, 1965. According to Uzoigwe, Bunyoro-Kitara’s economy was reasonably diversified and specialised, which led to a proliferation of markets that were later seriously affected when the kingdom of Tooro proclaimed its independence from Bunyoro in the 1830s. This meant exclusion from direct access to the trade routes to the coast and losing the salt mines of Katwe and Kibiro.10 Uzoigwe, ‘Precolonial Markets’, p. 433; Uzoigwe, ‘Bunyoo-Kitara Revisited’, p. 24.

In modern Uganda, both the eastern and western regions were frontier zones where there were no strong kingdoms to govern what are now borderlands. There were ‘buffer’ kingdoms that traded very actively, however, and small towns with good links to neighbouring areas and active markets. Conversely, peripheral markets and towns in regions with strong or large kingdoms such as Buganda seem to have been less important, as the central power had more control over them, leaving them with fewer opportunities for growth.8 Good, Rural Markets.

Trade in the interlacustrine region expanded considerably in the mid-1800s, due mainly to increased international demand for ivory and slaves to work in the plantations of Zanzibar.9 Pallaver, ‘Nyamwezi Participation’. This meant that coastal traders played a newly active role in regional commerce, not by creating a new trade network, however, but by using the one that was already there.10 John Tosh, ‘The Northern Interlacustrine Region’, in R. Gray and D. Birmingham (eds), Pre-Colonial African Trade: Essays on Trade in Central and Eastern Africa before 1900, London, 1970, pp. 103–118, at p. 111; Pallaver, ‘Nyamwezi Participation’, p. 514. At the time, there were three main trade routes in the region: one in the south linking Kilwa to the interior, one in the north towards the coast of Kenya, and one in the centre linking Bagamoyo with Lake Tanganyika.11 Pallaver, ‘Nyamwezi Participation’, p. 513. This last one, the central caravan route, on which coastal traders and the Nyamwezi were particularly active, is the most important for our story. This new era of regional trade was not only the consequence of new external demands for African goods, however: it was also the outcome of new balances of power within the kingdoms and people of the region, such as Buganda’s prominence and the development of a new class of Nyamwezi traders, the wandewa.12 Pallaver, ‘Nyamwezi Participation’, p. 515.

The key area in all of this is south-west Uganda (Ankole and especially Kigezi), which was the ‘frontier’ zone between or adjacent to some of the region’s most important kingdoms in the nineteenth century – Bunyoro and Buganda and Rwanda – and various economic zones, where trade offered a comparative economic advantage.13 The Baganda were not particularly active in regional trade outside their kingdom (despite some later attempts at sending expedition directly to Zanzibar, as in the case of Mutesa in 1870). Before the mid-1800s, they tended to wait for merchants to reach Buganda to trade, focusing their activities beyond their borders on demands for tributes, conquests and plundering. Despite this, Buganda undeniably played a central role in regional trade in view of their requirements of salt and iron and their production of bark cloth. The presence of a number of different kinds of terrain, production (agricultural, mineral and handcraft) and polities made the region particularly rich in terms of trade and exchanges, creating a special economic and cultural environment in which the local and regional combined and were in some way inter-dependent.14 On trade by Ankole see, among others, the always interesting Good, Rural Markets. Different people and polities met in these frontier markets, creating the lively locations that remain a dynamic borderland to this day. At the time, most of the goods that were exchanged between the interlacustrine region and the coast passed through Ankole and Kigezi, not from eastern Uganda. Goods arrived from the Swahili regions thanks to merchants who used caravan routes that mostly began in Zanzibar.15 When referring to traders from Africa’s east coast, many sources refer to Muslim traders, and others to Arab traders, in addition to Swahilis. These terms are not synonymous, and they often refer to different peoples, but considering the objectives of this chapter and to make it simpler, the author uses the generic term ‘coastal traders’, when referring to merchants from the coastal region who were active in long-distance trade. According to Edward Alpers, merchants from Zanzibar had been attracted by the possibility of exploiting the region as a source of ivory since the early 1800s, when the Sultan of Oman moved its capital to Zanzibar.16 Sayyid Said ibn Sultan, the ruler of Oman, moved its capital from Muscat to Zanzibar in 1840; Edward A. Alpers, ‘The East African Slave Trade’, in Z. A. Konczacki and J. M. Konczacki (eds), An Economic History of Tropical Africa. Volume One – The Pre-Colonial Period, London, 1977, pp. 206–15. The ivory trade was especially lucrative during this period, not only due to high demand from non-African markets, but also, and in particular, because of the large numbers of elephants in the area and the quality of their ivory, the so-called ‘soft type’, which was ideal for carving.17 R. W. Beachey, ‘The East African Ivory Trade in the Nineteenth Century’, The Journal of African History, 8:2, 1967, pp. 269–90.

Coastal traders set up their main stations in Karagwe (now north-west Tanzania),18 According to Henri Médard, during the late 1800s the region south of Lake Victoria was the main route linking Unyamwezi and Zanzibar: Henri Médard, Le Royaume du Buganda au XIXe Siècle, Paris, 2007, p. 182. most of them close to the capital but some in the north near the Buganda’s border, which was proof of their preference for the southern (land) and Victoria (water) routes rather than the eastern (Kenyan) one as a way of reaching the Indian Ocean – favouring what is now the Ugandan border with Rwanda and Tanzania rather than the border with Kenya.19 Tosh, ‘The Northern Interlacustrine Region’, pp. 111–12; Rowe, ‘The Western Impact’, p. 55. Their choice was dictated by their need to be close to Buganda, which they entered in 1840s, and which became their main trading partner, thanks in part to the Kabaka’s openness to outside opportunities. However, he was always careful not to give them too much space or power, which is probably one of the reasons why they were kept close to the court, and were not free to settle wherever they wanted.20 Good, Rural Markets, p. 151. In an attempt to cut rival kingdoms out of the coastal trade, this special relationship with Buganda in the second half of the 1800s essentially became an exclusive one, especially in the case of Zanzibar. This isolated Bunyoro, which was Buganda’s principal antagonist at the time.21 Tosh, ‘The Northern Interlacustrine Region’, p. 112. Despite this attempt, some coastal traders visited Bunyoro in 1877, bypassing Buganda and opening up a new route, but they were not granted access to Rwanda until the 1880s, when the need for weapons overshadowed Rwandan suspicions.22 R. M. A. Van Zwanenberg and Anne King, An Economic History of Kenya and Uganda, 1800–1970, London, 1975, p. 170; Tosh, ‘The Northern Interlacustrine Region’, p. 113.

Coastal traders reached Buganda from their base in Karagwe in the 1840s, and Ankole as late as the 1870s.23 Good, Rural Markets, pp. 165–6. In fact, they did not cross into Ankole or Rwanda for many years. Ivory was brought to them by others, probably from Unyamwezi.24 Direct trade between Unyamwezi and Zanzibar dated back to the early 1800s, if not earlier. Sayyid Said ibn Sultan sent at least one large caravan to Unyamwezi, mainly to trade in ivory: Alpers, ‘Slave Trade’, p. 213. Unyamwezi was another key location in long-distance trade. It lay at the junction of a number of different commercial routes, and was the main place where porters were hired to take ivory to the coast.25 Pallaver, ‘Nyamwezi Participation’, p. 515. It has been estimated that half a million porters passed through Tabora every year and they were preferred to slaves for transporting such a difficult, heavy and valuable cargo.26 Beachey, ‘Ivory Trade’, p. 276. While Ankole had irregular and indirect trade contacts with Karagwe, the Baganda, in a break with their usual customs, travelled to Karagwe to trade before the coastal people arrived, and also maintained contacts with polities along the southern shore of Lake Victoria.27 Good, Rural Markets, 164; Gerald W. Hartwig, ‘The Victoria Nyanza as a Trade Route in the Nineteenth Century’, The Journal of African History, 11:4, 1970, pp. 535–52. Buganda’s traders and envoys were no strangers to using the lake to reach places such as the Speke Gulf on the southern shore of Lake Victoria.28 Hartwig, ‘Victoria Nyanza’, p. 541. Another factor that helped Buganda’s trade was its relations with the Kooki kingdom. Kooki was situated at the south-west periphery of Buganda, and had always been an important ally, but it only became a part of Buganda in the late 1800s. Buganda was its main trading partner, and it was a key to commerce because of its privileged geographical position at the junction of various different markets: Bakooki merchants traded actively with Ankole (for salt) and Tooro (for ivory), and these coastal traders played a major role in regional trade in the 1800s.29 Stonehouse, ‘Bakooki’, p. 533.

To complicate the picture even further, the rivalry between Buganda and Bunyoro played a relatively important role in the coastal traders’ inability to expand their activities beyond Buganda.30 Good, Rural Markets, p. 165. This proved not to be an impediment to regional trade, however, because, while coastal merchants were involved, they did not control it. Their activity consisted in trading with locals, changing their business partners if needed, and taking their goods to the coast. Despite the fact that there is little evidence so far that there were any full-time professional local traders (most were also involved in agriculture or in producing the goods they were exchanging), locals did not give up their roles, and managed the organisation of regional trade, apart from the transport of goods for long distances.31 Tosh, ‘The Northern Interlacustrine Region’.

As we have seen, the network of regional trade formed the basis for, or a small segment of, long-distance trade routes.32 Andrew Roberts, ‘Nyamwezi Trade’, in Pre-Colonial African Trade, pp. 39–74. In the case of interlacustrine trade (in modern Uganda) one of the main links to the outside world, heading south through modern Rwanda, was the Nyamwezi and their trade network. The Nyamwezi had long traded goods such as iron, salt, slaves, cattle and goats, but they became particularly well-known on the east coast of Africa (particularly Zanzibar) in the 1800s as a link between non-African economies (Europe, India and America) and the source of ivory in central Uganda.33 Roberts, ‘Nyamwezi Trade’, pp. 47–9. The Nyamwezi mainly used two trade routes to reach the coastal region, one through Sangu to Lake Tanganyika, and one through Unyanyembe (Tabora) to Lake Tanganyika, or heading north through Rusubi and Busambiro.34 Roberts, ‘Nyamwezi Trade’, p. 50; Beachey, ‘ivory trade’, 218; The former is more or less where the modern central railway line from Dar es Salaam to Tabora now runs.

While ivory became the principal commodity in long-distance trade (its extraction was mainly a royal monopoly), particularly from the time of Kabaka Semakokiro’s kingdom (1787–1814), because of external demand, the two key goods in regional trans-kingdoms trade were historically salt and iron. Salt was an essential and strategic commodity because of its scarcity (not many areas produced a high-quality version) and the location of the main production area, Kigezi (in south-west Uganda), which was a frontier zone between two kingdoms that fought constantly for control of the salt fields. The salt trade was therefore of particular importance for the region, above all as regards the salt produced in Katwe, which linked the Lake Edward town to Congo, Rwanda, Ankole and Buganda.4 Despite the significance of salt production for Katwe, it is important to point out that it was not its exclusive activity. Because of fluctuating production levels, people in Katwe engaged in other activities such as agriculture to ensure they could make a living: Kathryn Barrett-Gaines, ‘The Katwe Salt Industry: A Niche in the Great Lakes Regional Economy’, African Economic History, 32, 2004, pp. 15–49. Other sources were Kasenyi, on the shores of Lake George, and Kibiro, on the eastern shore of Lake Albert, which was closer to Bunyoro’s centre of power than Katwe, and more oriented towards the north and east.5 Graham Connah, ‘The Salt of Bunyoro: Seeking the Origins of an African Kingdom’, Antiquity, 65:248, 1991, pp. 479–94; Charles M. Good, ‘Salt, Trade, and Disease: Aspects of Development in Africa’s Northern Great Lakes Region’, The International Journal of African Historical Studies, 5:4, 1972, pp. 543–86, at p. 565. The main salt producers areas were therefore in territories controlled by Bunyoro-Kitara, which had a monopoly of both salt production and trade routes until the early to mid-1800s.6 Uzoigwe, ‘Precolonial Markets’, pp. 442–45.

A thriving salt trade was common to the entire western region, but the role of Katwe is particularly important for the purposes of this chapter because it was mainly the salt from there that went south, especially to modern-day Rwanda, through Kigezi. According to Charles Good,7 Good, ‘Salt, Trade, and Disease’, pp. 558–59. the main land route from Katwe crossed Ankole, ‘ultimately linking up with Kigezi, Ruanda and Karagwe’. The feature of the route through Kigezi was that the trade in salt was supported by exchanges with other goods through the intervention of part-time traders (who were also involved in agricultural and pastoralist activities), making the route particularly important (and sustainable) for the economy of the region. On the other hand, Kasenyi’s salt became key in the late 1800s, when the long-distance trade route from the coast crossing Tabora reached Kitagwenda, east of Lake George, where some of Kasenyi’s salt was traded.8 Good, ‘Salt, Trade, and Disease’, pp. 561–62. The situation changed in the mid to late 1800s, when a trade shift took place in the region for geopolitical reasons: Tanganyika became more important, and the salt route shifted south.9 Newbury, ‘Lake Kivu’, p. 20. In some ways, there was already competition between the North and Central trade corridors, as there is today.

Kigezi therefore played a special role. It formed a kind of buffer zone between the centralised kingdoms, with a long history of migration from the south (Rwanda) and to central and southern Uganda.10 Grace Carswell, ‘Soil Conservation Policies in Colonial Kigezi, Uganda: Successful Implementation and an Absence of Resistance’, in W. Beinart and J. McGregor (eds), Social History and African Environments, Oxford, 2003, pp. 131–54. Because theirs was a segmentary society, Bakiga markets were not politically controlled, unlike in neighbouring Bunyoro and Buganda, where the king controlled the main markets, usually keeping them close to the court and demanding a fee.11 B. Turyahikayo-Rugyema, ‘Markets in Precolonial East Africa’, p. 288. Despite the lack of a central authority, ‘prominent people in society such as the elders, medicine men and rain-makers were particularly important as the rulers of the Basigi, especially before their migration from Rwanda in the 1800s. The role of the rain-makers and medicine men and Nyabingi priests becomes meaningful in the context of their function as economic managers of their society’.12 Turyahikayo-Rugyema, ‘Markets in Precolonial East Africa’, pp. 288–89. The physical location of markets also depended on the availability of goods (mainly the food and commodities needed for everyday life), which made them particularly active during droughts or other times of shortages and important for their role as a space where people could meet, come together and find out the latest news.13 Turyahikayo-Rugyema, ‘Markets in Precolonial East Africa’, p. 288. The economy of the area was mainly based on agriculture and trade: the Bakiga produced a surplus of food crops (primarily peas and beans) that was traded in neighbouring areas, proving that the dichotomy between food and cash crops does not always apply.14 Grace Carswell, ‘Food Crops as Cash Crops: The Case of Colonial Kigezi, Uganda’, Journal of Agrarian Change, 3:4, 2003: pp. 521–51, at p. 523. Its location was particularly favourable for this purpose, due to the diverse agro-ecological neighbouring zones, the flexibility of its agricultural system, and the important trading route (mainly for Katwe salt) that passed through its territory.15 Carswell, ‘Food Crops’, p. 525.

Kigezi was formally acquired by the British in 1912 as a consequence of the Anglo-German-Belgian Boundary Commission of 1911, which settled the claim by the three parties that wished to gain control of Kigezi.16 Good, ‘Salt, Trade, and Disease’, p. 568. When the British found that there was no recognisable or usable system of government, they established their own administration using Baganda agents, as they had in other regions such as the Eastern Province, which altered the social balance.17 Carswell, ‘Food Crops’, p. 526; Carswell, ‘Soil Conservation Policies’, p. 133. Despite their poor understanding of the size, shape and characteristics of the region, the British recognised that Katwe’s salt and position was key to their situation, and so, on his arrival in 1891, Lord Lugard built a fort (Fort George) in Katwe, which had always been disputed by various powers.18 Good, ‘Salt, Trade, and Disease’, p. 568. Katwe’s importance was later confirmed by the British decision to guard it as a means of combating Kabarega (the defiant King of Bunyoro) in order to prevent his invasion of Tooro (which controlled Katwe at that time), and at the same time seeking to obtain revenues for the government through trade.19 Barrett-Gaines, ‘The Katwe Salt Industry’, p. 22. The British also maintained a defence force in Tooro in the early 1900s to ‘keep a watch on the salt lakes and the Belgian frontier’.20 Reported in Good, ‘Salt, Trade, and Disease’, p. 569. The British underestimated the local production of food and cash crops, however, and unsuccessfully attempted to introduce new cash crops for export, particularly coffee, flax and tobacco.21 On the reasons behind this failure, see Carswell, ‘Food Crops’, p. 533. This colonial policy was also impacted by a marketing strategy, because the newly-introduced cash crops had to be sold through the colonial authorities, while other (pre-colonial) cash crops were sold directly in markets. This stimulated the smuggling of coffee to the Belgian Congo, where sellers could obtain higher prices and avoid colonial taxation.22 Carswell, ‘Food Crops’, p. 536. Famers preferred to produce goods they could maintain control over for marketing and consumption purposes, not least because the new cash crops could not be eaten in the event of a poor harvest. In a nutshell, one might say that the strength of local market and cash crop production made a powerful contribution to the failure of colonial policies, a failure inspired by a lack of knowledge of the local systems of production and trade.

As we have seen, long-distance trade was not an invention of the colonial rulers, and they did not organise it. The Great Lakes Region was already linked to the Swahili region through a well-organised trade system that included the active participation of local traders. There was also a reasonably good network of paths connecting the various areas of the Great Lakes Region and facilitating the movement of people and traders all over the area. The local economies and local trade in the area were also flexible and reactive to external stimuli, proving African initiative and ‘continuity from earlier processes of economic change’.4 Roberts, ‘Nyamwezi Trade’, p. 74. Finally – and this was probably the main change – there was the marginalisation of African traders from the system. The other change was, at least in part, the subsequent diversion of the trade route from Tanganyika to Kenya, which inspired the more recent competition between the Northern and Central corridors. All this happened outside Uganda, however, and within Uganda itself colonial rule did not totally alter the pattern of trade and route networks. In 1953, in its Report of Enquiry into Road Transport Conditions in the Protectorate of Uganda, which analysed the condition of the ‘Kampala–Masaka–Mbarara–Kabale–Kisoro–Ruhengeri (Ruanda)’ route, the East Africa Royal Commission stressed that ‘the route between Kampala and Kabale must in the future become one of the most important in the Protectorate’, even though it was ‘clear that this route will not be served by railway in the foreseeable future and therefore every encouragement is present for the setting up of a reliable and adequate service for passengers, mail and freight’.5 London, The National Archives, Commercial and Industrial Development, CO 892/15/10.

Notwithstanding the initially good impression the British had of Ugandan roads and its trade network, they soon realised that they were not suited to their requirements and that they should encourage improvements, introducing reforms to adjust the trade system to their needs. After a few years, however, it became clear that the colonial investment had been inadequate and, in the 1950s, on the eve of independence, the British administration deemed the trade system to be unsatisfactory for the development of the protectorate. Upgrading the routes to roads suitable for motor vehicles was particularly problematic, and Indian enterprises were favoured over African companies because they had more assets and greater access to cash and lorries. As reported in the Kigezi District Annual Report for 1930 ‘there is no doubt that Indians are capturing much of the trade which was formerly in the hands of natives, e.g. Katwe salt and locally-caught fish’.6 Good, ‘Salt, Trade, and Disease’, p. 580. For the British administrators, the scarce participation by African traders was particularly worrying:

The communication routes of Buganda, Bunyoro and other kingdoms formed the basis for controlling the commercial network, which was kings’ main goal for gaining wealth and prestige. They even promoted the development of a fleet of large canoes that was capable of crossing the lake, which was a target of expansionist ambitions that were not limited just to the mainland.3 Richard Reid, Political Power in Pre-Colonial Buganda. Economy, Society & Warfare in the Nineteenth Century, Woodbridge, 2002. The backbone of the trade and transport system in the area remained its land routes, however. The pre-colonial routes, and the markets alongside them, were proof of how dynamic the population of the area was in terms of exchanging goods and their interest in an active trade environment in the 1800s. South-west Uganda is of special interest here because of its dynamic trade network, its frontier characteristics, and the major changes it underwent over the course of a few decades. It was an established trade hub controlled by local populations during the 1800s, and it was during the colonial period that Africans lost control of the system, especially to Indian enterprises. This created an economic and social disruption rather than a real change in the location of trade paths, however.

In conclusion, what is generally seen as a modern solution to boost African economies and regional integration is not a recent invention at all. The colonial powers did not bring integrated transport systems to East Africa: a fairly detailed and widespread transport system (using water and land routes) already connected the eastern coast of the continent to its interior. Indeed, the well-known Swahili trade network linked up with a well-maintained network of paths in the west to Congo, using the lake trade network, which was controlled by local traders and connected by coastal traders. Another element that should not be forgotten is the relationship between economic and political influences in maintaining an efficient communications network, which was confirmed by the positions adopted by Buganda and Bunyoro in the 1800s. Political opportunities and geopolitical views are important factors for deciding the fate of modern transport corridors today, but this also applied to their historical precursors. Despite the differences in time and space between the experience of the Roman Empire in Europe and the road networks in East Africa in the 1800s, we have seen that they both responded to economic/trade and political/military needs (as Bunyoro’s attempts to control the salt trade demonstrate), reminding us that if we only take the economic requirements of a given region into account when we study and analyse transport corridors, and ignore history, we can only ever have an incomplete picture.