CHAPTER
2 Modernization, State-Building, and the Hydraulic Mission in Imperial and Revolutionary Ethiopia
Pages
28–59
Abstract
This chapter examines Ethiopia’s dam-building ambitions during the Imperial era and under the military-Marxist Derg regime. The chapter examines why the two regimes’ common ambition of building dams to expand the reach of the state and transform the economy led to a pattern of dam construction dominated by Ethiopia’s least advantageous river basins, while the river basins with the most hydroelectric potential—including the Blue Nile—remained relatively untapped until after the fall of the Derg. Although Haile Selassie’s government and the Derg were inspired by sharply divergent ideologies, they shared a common desire to modernize Ethiopia, with water infrastructure playing a central symbolic and material role. Yet, both regimes faced similar technological, and international and domestic political economy constraints that limited dam building to comparatively low potential sites and, particularly, prevented exploitation of the river with the greatest hydropower potential—the Blue Nile.
Keywords: dams, modernization, Haile Selassie, Derg, Nile, Egypt, Sudan, political economy, electricity, agriculture
Subject
Environmental PoliticsInternational Relations
Collection: Oxford Scholarship Online
Ethiopia has frequently been proclaimed the ‘Water Tower of Africa’.1 This phrase is in part a recognition that Ethiopia is somewhat unusual in being the upstream country on every major river within its borders. However, the phrase also harbours resentment at Ethiopia’s inability to capitalize on these abundant water resources, despite the provision of water to all Ethiopia’s neighbours. Ever since the construction of the original Aba Samuel dam in 1912 to power the palace of Emperor Menelik II in Addis Ababa, dams have been seen as key tools for Ethiopia’s modernization and development, while electricity has been equated with hydropower. For Ethiopia’s rulers, the attraction of dams lies both in their symbolic value as manifestations of Ethiopia’s modernization and in their potential economic contribution through electrification, irrigation, and industrialization.
This chapter considers Ethiopia’s dam-building ambitions across two regimes—the Imperial regime of Haile Selassie, first as regent and then emperor (1916–1974), and the military-Marxist Derg regime under Mengistu Hailemariam (1974–1991). It was in this period that Ethiopia launched its hydraulic mission, namely ‘the strong conviction that the state should develop hydraulic infrastructure to capture as much water as possible for human uses’ (Wester et al. 2009, p. 75). The chapter examines why the two regimes’ common ambition of building dams to expand the reach of the state and transform the economy led to a pattern of dam construction dominated by Ethiopia’s least advantageous river basins. In contrast, river basins with the most hydroelectric potential—the Blue Nile and the Omo—remained relatively untapped until after the fall of the Derg. Despite much writing on the broader Nile Basin, there has been relatively little attention to Ethiopia’s dam building in this period (but see Waterbury 2002; Arsano 2007a) and no attempt to explain the spatial variation of dam construction within Ethiopia. This chapter addresses this gap.
The argument developed is that the spatial variation in Ethiopia’s historical dam building can be understood as the interplay between technical expertise and the feasibility of different engineering solutions, on the one hand, and the intersection of political processes across five levels of analysis from the global to the local, on the other. Although Haile Selassie’s government and the Derg were inspired by sharply divergent ideologies, they shared a common desire to modernize Ethiopia, with water infrastructure playing a central symbolic and material role. Yet, both regimes faced similar technological challenges, and international and domestic political economy constraints that limited dam building to comparatively low potential sites and, particularly, prevented exploitation of the river with the greatest hydropower potential—the Blue Nile.
As a late-developing country, Ethiopian dam building necessitated external financial and technological support. However, Ethiopia’s access to international support was constrained by its positioning within regional security dynamics within the Horn of Africa and its links to the Middle East. A consistent theme throughout the twentieth century was that Ethiopia was seen by the major global powers of the day as a lower geo-strategic priority than its Nile riparian, Egypt. The result was that the UK, the US, and the Union of Soviet Socialist Republics (USSR) prioritized Egyptian use of the Nile waters and ultimately withheld the financial and technical support that would have been required for Ethiopia to develop a major Nile dam. Ethiopia’s dam building therefore focused on other, less advantageous river basins that faced lesser transboundary political barriers. Sub-national politics also played an important role in shaping the pattern of dam and irrigation development. Internal political divisions undermined an early attempt to build a dam on the Blue Nile, while both Haile Selassie and the Derg tended to avoid large-scale infrastructure development that would cause major disruption to social structure in the highland core. Instead, a persistent feature of the hydraulic mission has been an intertwined process by which the state extended its territorial reach through infrastructure development and economic modernization that further marginalized those living in the periphery.
While Ethiopia’s ambitions on the Blue Nile went unfulfilled in this period, downstream riparians Egypt and Sudan were more successful in leveraging their important positions on the Red Sea and in Middle East politics to secure the international finance and expertise required for their own hydraulic missions. However, their decision making was also shaped by the wave of independence following World War II. In a pattern that has continued thereafter across the basin, both Egypt and Sudan built dams that focused on maximizing control over the Nile waters within their own borders. In doing so, both countries turned their backs on basin-wide, cooperative approaches to managing the Nile waters.
The chapter begins with an overview of Ethiopia’s river basins, their potential for hydroelectricity and irrigation, and how this physical geography maps onto the history of Ethiopian state formation. The story of Ethiopian dam building, meanwhile, begins with the unrealized ambitions of British colonialists to build a dam on the Blue Nile. The failure to secure support for a dam on the Blue Nile led the Ethiopian Imperial Government to turn first to the modernization of the Awash Basin, before another stalled attempt to dam the Blue Nile with US support during the Cold War. Finally, the chapter discusses the post-revolutionary dam-building efforts. Despite major international and domestic political realignments, the Derg was no better placed to tackle the hydraulic development of the major river basins than the emperor.
Water, State-Building, and Modernization in Imperial Ethiopia
Ethiopia’s topography comprises a highland plateau in the centre of the country, which is bifurcated by the Rift Valley (see Figure 2.1). The major rivers flow from these highlands—where the vast majority of the rain falls and where most of the Ethiopian population resides—to the lowland periphery in the west, south, and east. Rainfall is far greater in the west of the country, with the result that those rivers with catchment areas in the west—the Blue Nile, or Abay as it is known in Ethiopia, Baro-Akobo, and Omo—contain the vast majority of Ethiopia’s waters (see Table 2.1). Various Nile tributaries flow from the western highlands towards the Sudanese border, of which the main ones are the Blue Nile itself, the Tekeze (which becomes the Atbara in Sudan), and the Baro-Akobo (Sobat), a tributary of the White Nile. The Omo-Gibe River flows from Jimma in the southwestern highlands southwards into Lake Turkana on the border with Kenya; and the Wabe Shabele flows from Bale and Haraghe in central and eastern Ethiopia into Somalia. The Awash is the only major internal river, since it disappears through evaporation and seepage in Afar region before crossing the border and has consequently been described as ‘the most Ethiopian’ of rivers because ‘it does not carry Ethiopian soil outside the country’ (Zewde 2008a, p. 120).2
Figure 2.1
Ethiopia’s river basins and dams pursued before 1991
Source: Author and Manchester Cartographic Unit.
Table 2.1
Ethiopia’s river basins
| Basin | Catchment area | Annual discharge | ||
|---|---|---|---|---|
| km2 | % | Billion m3 | % | |
| Blue Nile | 199,812 | 18 | 54 | 43 |
| Baro-Akobo | 74,102 | 7 | 23 | 19 |
| Omo-Gibe | 78,200 | 7 | 17 | 15 |
| Tekeze | 90,000 | 8 | 8 | 6 |
| Genale-Dawa | 171,050 | 15 | 6 | 5 |
| Rift Valley Lakes | 52,740 | 5 | 6 | 5 |
| Awash | 112,700 | 10 | 5 | 4 |
| Wabe Shebele | 200,214 | 18 | 3 | 3 |
| Danakil | 74,002 | 7 | 1 | 1 |
| Mereb | 5,900 | 1 | 1 | 0 |
| Ogaden | 77,100 | 7 | 0 | 0 |
Data source: Abtew and Dessu (2018, p. 40)
The flow of these rivers is highly seasonal, dictated by the rainy seasons in the highlands, which in most of the country comprise a short rainy season in March to April (belg) and a main rainy season in June to September (meher), with relatively little rainfall outside these periods. There is a widespread perception that these rains are becoming increasingly unreliable, with belg rains, in particular, frequently failing altogether or arriving late and merging with the meher rains.
Over the past century, the idea that these rivers constitute a vital resource for national development, and that the only way of realizing this potential is through dams and the control of the waters, has become deeply engrained. This potential, however, varies between river basins. In terms of hydropower, the Blue Nile and Omo Rivers are the most important, with large rivers descending through the sort of steep, narrow gorges that attract the attention of dam engineers. While the Tekeze also has very suitable topography, its more limited water flow means that hydropower potential is more modest (Blackmore and Whittington 2008). There is considerable potential for irrigation along the various Blue Nile tributaries flowing into its source, Lake Tana, and the main river. However, along the Blue Nile’s main course steep gorges provide few options for irrigation. The Omo floodplains after the river has descended the escarpment into the lowlands, on the other hand, have attracted considerable interest in terms of irrigation potential. Hydroelectric potential is quite limited in the Awash Basin. The headwaters of the Awash where rivers descend into the Rift Valley provide the required head for hydropower, yet here water flow is limited.3 In contrast, the water flow is larger in the relatively flat Rift Valley which does not provide sufficient head. However, the Awash is perhaps the most suited to irrigation development (Halcrow 1989). Given the seasonality and annual variability of water flow in all these river basins, year-round irrigation and hydropower production require dams to regulate water flows. Historically, these rivers have also played a key role in maintaining downstream soil fertility through flooding and the deposit of sediment both in Ethiopia and beyond. However, from the perspective of dam engineers, heavy sedimentation presents a major challenge, threatening rapid reduction in water storage and hydroelectric generation capacity.
The rivers with by far the greatest international significance are those that feed into the Nile. In total it is estimated that Ethiopia provides 86 per cent of the waters of the Nile as measured at Aswan in Egypt, with the Blue Nile the most important with 59 per cent, followed by the Baro-Akobo with 14 per cent and Tekeze 13 per cent (Ethiopian Technical Experts 1996, p. 67). The Equatorial Lakes region combined only contributes 14 per cent of the Nile waters due to the vast water losses through evaporation in the Sudd.
The physical geography of Ethiopia’s rivers closely maps onto the country’s political and human geography, and the particular pattern of state-building that shaped the country (Donham 2002). The Ethiopian state originates in the north of contemporary Ethiopia—principally Tigray, Eritrea, Wollo, Begemder, and Gojjam provinces (see Figure 2.2)—several thousand years ago. Over the centuries, the territory administered by Ethiopian polities ebbed and flowed, while the centre of power gradually shifted to the south, with the expansion of the Orthodox Church and the Amharic language central to the state-building project. Land grants to the nobility were the basis of state power, with the emperor ruling through a landed nobility granted the power to extract tax and tribute from peasants accessing land through rist, a descent-based tenure system (Hoben 1973; Rahmato 1984).
Figure 2.2
Ethiopia’s topography and provinces (approximately 1963–1987)
Source: Author and Manchester Cartographic Unit.
However, it was not until the period of the European Scramble for Africa at the end of the nineteenth century that the Ethiopian state became increasingly centralized under Emperor Menelik II, enabling a rapid expansion into the southern highlands and lowland periphery, stopping only when it encountered European colonialists in Sudan, Kenya, and Somalia, and creating something like Ethiopia’s contemporary borders. Southern highlands were gradually pacified and incorporated through land grants to military leaders as reward for service, while provincial boundaries were drawn so as to administer (loosely speaking) lowland peripheries from an administrative centre in the highlands, ignoring the boundaries between ethnic groups, which were divided between provinces. Practically, the Ethiopian state’s main intervention into many lowland areas continued to be periodic raids for slaves and ivory. Meanwhile, the highland political elite discriminated against lowland populations based on their supposed inferiority in race, culture, language, and livelihoods (Young 1999; Garretson 2002; James 2002a; Lavers 2016). This gave rise to a loose distinction between a highland core in the north, a highland periphery in the southern highlands, and a lowland periphery in the west, south, and east (Markakis 2011). There were also clear differences in population density, between densely populated highlands where a landed elite extracted an agrarian surplus from the peasantry, to sparsely populated lowland peripheries in the west, south, and east where the population practised various forms of pastoralism and shifting cultivation (Figure 2.3).
Figure 2.3
Population density by wereda
Data source: CSA (2008a). Note: No data are available for Somali region.
The threat of European colonialism and the growing recognition that Ethiopia had fallen behind the most advanced economic powers prompted Menelik and then Haile Selassie to pursue projects to modernize Ethiopia through political reform, state-building, and economic development. However, as with any mediated state, dependent on a landed elite, the Imperial regime faced the inherent challenge that modernization risked undermining its own political power (Skocpol 1979; Waldner 1999). Following the brief Italian occupation (1936–1941), Haile Selassie created a state bureaucracy, largely by co-opting the landed elite into positions as ministers and civil servants. At this time, there was also a push to modernize the economy by attracting foreign investment in large-scale agriculture and some industry. However, the emperor’s unwillingness to pursue land reform, which would directly threaten the political and economic basis of the landed elite and the Imperial state, meant directing agricultural investment to the lowland periphery, where the land use of pastoralists and shifting cultivators could more easily be ignored (Ashami 1985).
As a result of both hydrology—with the main dam sites located in the gorges descending from the highlands—and politics—with the need to avoid disrupting the existing political order in the highlands—the main focus of dam building for hydropower and irrigation was on the sparsely populated lowland periphery. The result is that populations that were only incorporated into Ethiopia relatively recently at the end of the nineteenth century, and who have tended to be marginalized and stigmatized by Ethiopia’s state-building project, tend to be those most directly affected by dams, reservoirs, and associated plantation agriculture.
Colonial Ambitions on the Blue Nile
The first attempts to build dams in Ethiopia were largely undertaken by British colonialists in the early twentieth century. British engineers were focused on controlling the entirety of the Nile waters for the benefit of irrigated cotton agriculture in Egypt and Sudan, ignoring Ethiopian sovereignty or interests. In response, Ethiopia under Haile Selassie began to explore options to build its own Blue Nile dam with US support. However, these efforts were undermined by internal rivalries and, just when a negotiated agreement with Egypt and Sudan seemed possible, the Italian invasion in 1936. Ultimately the failure to secure alignment between political dynamics at global, river basin, national, and sub-national levels meant that these dam ambitions went unfulfilled before World War II.
The Nile is Egypt’s only renewable source of freshwater.4 Although the Nile’s exact hydrology was unknown until relatively recently, Egypt and Ethiopia have long been aware that Ethiopia is the main source of the Nile waters on which Egypt is so dependent. Moreover, as Erlich (2002) lays out, Egypt and Ethiopia have been connected since the fourth century AD by the ties between the Ethiopian Orthodox Church and the Egyptian Coptic Church of Alexandria. The Coptic Church provided the first Egyptian abun—the head of the Ethiopian Church—in 333 AD and continued to do so until the 1950s, with the result that Ethiopia became reliant on the Egyptian Coptic Church for its Christian state-building project (Erlich 2002). Consequently, the Nile and Church links became intertwined in relations between the two countries. It was widely believed in Egypt that Ethiopia had the ability to divert the Nile, and periodic droughts leading to reduced flow of the Nile were often attributed to Ethiopian interference (Collins 1990; Erlich 2002). Indeed, this was a myth that the Ethiopians and the Egyptian Copts were happy to propagate as a means of increasing their leverage with Egypt’s Islamic leaders (McCann 1981). For Egyptians, the Nile was considered a gift, albeit dependence on the Nile waters created a set of ‘anxieties and myths that, in themselves, went to the core of the Egyptian soul’ (Erlich 2002, p. 4). In contrast, for Ethiopia the river offered little potential for irrigation and presented a major barrier to transportation and communication. Instead, for the Ethiopians the Nile was mainly of strategic use as ‘their best card in their desire to retain their most important connection with the Middle East’ (Erlich 2002, p. 8).
From the nineteenth century there were various attempts by Egypt and Britain, on Egypt’s behalf, to control the Blue Nile for irrigated cotton production in Egypt. Under Muhammed Ali (1805–1849), Egypt began to limit its historical dependence on the Nile floods by constructing a network of dams, barrages, and canals, enabling the introduction of cotton as a cash crop. Yet, Egypt’s dependence on the Nile led to an expansion of its desire to control the river’s flow. In 1820–1821 Muhammed Ali’s army conquered much of Sudan and later founded Khartoum, bringing most of the White Nile Valley under Egyptian rule (Erlich 2002). Ismail, Muhammed Ali’s grandson, went further, occupying much of contemporary Eritrea and Harar in what is now eastern Ethiopia. Ismail’s pretentions to control the whole of the Nile Valley up to Lake Tana ended, however, when the Egyptians were defeated by Ethiopia in Eritrea in 1875–1876, a defeat that precipitated Egyptian colonization by the British (Erlich 2002; Verhoeven 2015).
Three European colonial powers—the British, Italians, and French—all harboured ambitions regarding Ethiopia at the end of the nineteenth and early twentieth century. Ethiopia was of particular geo-strategic importance once the Suez Canal opened in 1869 as a result of its Red Sea coastline and ports in Eritrea, in addition to providing the source of the Nile and potential influence over the river’s flow. Italy failed in its first attempt to conquer Ethiopia as a result of military defeat by Menelik’s army at the Battle of Adwa in 1896. Ethiopia agreed to cede Eritrea to Italy, but Italy retained its interest in Ethiopia, which was key to linking its colonies in Eritrea and Somalia. Meanwhile, the French occupied what is now Djibouti and focused its attention on the railway concession from Djibouti to Addis Ababa built from 1897 to 1917. For the British, the main concern was to ensure that the French were kept out of the Nile headwaters. Following the British occupation of Egypt in 1882 to secure control of the Suez Canal, the British authorities launched a series of studies of the Nile Valley with the aim of developing Egypt’s cotton industry. Indeed, protecting and enhancing the Nile’s flow to Egypt was a major driver of British colonialism in East Africa, leading to the occupation of the Sudan in 1898–1899 (Collins 1990; Verhoeven 2015). The British also established a trading post in Gambella in Ethiopia, attempting to use the White Nile as a trading route (Collins 1990; Zewde 2008b).
Having built the first Aswan Dam in Egypt in support of cotton agriculture in 1902, the British colonial authorities turned their attention to enhancing the flow of the Nile through water projects upstream. The initial focus was on the White Nile, with proposals to increase storage at Lake Albert in Uganda and reduce evaporation in Sudan’s Sudd with the Jonglei Canal (Collins 1990). However, attention soon turned to Ethiopia’s Blue Nile. The British dispatched an envoy to Ethiopia, and Menelik II signed a 1902 treaty regarding the border between Ethiopia and Sudan. Article III of the treaty, however, has resulted in longstanding controversy. While the English version states Menelik’s agreement not to interfere with the flow of the Nile without British consent, the Amharic version apparently differs, stating merely that the Nile waters could not be ‘completely stopped’ (Abtew and Dessu 2018, p. 16; McCann 1981; Zewde 1991; Erlich 2002).
Menelik consented to initial studies conducted by British engineers for the colonial-era Egyptian Irrigation Service. A study by Dupuis (1904)focused on exploiting Lake Tana as a natural reservoir in the cool Ethiopian highlands, thereby minimizing evaporation, and maximizing and regulating the flow to Egypt and Sudan. There was no consideration of irrigation or hydropower potential in Ethiopia. Meanwhile, the study doubted the viability of dams along the Blue Nile gorge itself because of the heavy sedimentation that would quickly render the dams useless (Garstin 1904). The study also seems to have been the first to propose digging a tunnel that would release water from Lake Tana into another Nile tributary, in this case the Rahad, albeit this was dismissed as prohibitively expensive and with little benefit (Dupuis 1904, p. 24). Ultimately, however, Menelik’s incapacitation after a stroke in 1906 and the internal struggle over succession prevented Ethiopian negotiation over the project (Collins 1990; Zewde 1991; Erlich 2002). Instead, the British, French, and Italians signed a 1906 agreement that divided Ethiopia into spheres of influence with western Ethiopia, including Lake Tana and the Blue Nile, recognized by the Europeans as a sphere of British influence.
The importance of enhancing the flow and reliability of the Nile grew in the early twentieth century with booming demand for cotton, leading to a series of agreements that completely disregarded Ethiopian sovereignty and interests, despite Ethiopia’s 1923 admission into the League of Nations. The 1925 agreement between Britain and Mussolini’s Italy acknowledged Ethiopia as an Italian sphere of influence in exchange for Italian support for Britain to build the Lake Tana Dam (Zewde 1991; Erlich 2002, p. 82). Meanwhile, the Egyptian declaration of independence in 1922 under British oversight meant growing contestation between Egypt and colonial Sudan over the Nile waters. The British built the Sennar Dam in 1926 to supply water for Sudan’s Gezira irrigation scheme (Collins 1990; Verhoeven 2015). However, colonial Sudan’s water use was hotly contested by Egypt, whose economy remained dominated by unprocessed cotton, despite modest industrialization efforts (Waterbury 1983). The result was the 1929 Egyptian-Sudanese Water Agreement that ignored Ethiopia and other upstream riparians and heavily favoured Egypt in its division of the Nile waters (Collins 1990).
Further missions by Egyptian and Sudanese colonial representatives in 1916 and 1920–1921 visited Lake Tana, with a view to regulating ‘the Nile in the joint interests of Egypt and the Sudan’ (Grabham and Black 1925, p. xi). The study took detailed measurements that for the first time highlighted the importance of the Blue Nile in the total Nile flow (Erlich 2002). These studies again proposed constructing a barrage at the mouth of Lake Tana to regulate the flow of the Blue Nile (Grabham and Black 1925).5 Meanwhile, the British offered Ethiopia rental fees for water rights and territory in exchange for a deal on the dam (McCann 1981). Ethiopian leaders were—rightly—concerned about European ambitions on Ethiopia, as well as the prospect that a dam that raised the height of the lake to maximize storage would flood important Orthodox monasteries located on islands in the Lake (Grabham and Black 1925; McCann 1981). Ras Tafari Mekonnen, then regent and heir to Empress Zewditu (1916–1930), also feared that the British scheme would strengthen the position of the powerful leaders of Begemder and Gojjam provinces—contenders for the throne. The Ethiopian response was to engage with the British but to delay and play for time (McCann 1981). Grabham and Black (1925, p. 31) also raised the possibility of diverting water from Lake Tana through a tunnel into the Dinder or Beles Rivers, rather than the Rahad, but like Dupuis discounted this as being too expensive and of little benefit.
In the face of growing European encroachment, Ras Tafari sought to seize the initiative. As a late-developing country, Ethiopia required external finance and expertise to tackle a dam and approached the US Government in 1927 to request assistance to develop the Nile to the benefit of Ethiopia, while tying the key provinces of Gojjam and Begemdir to Addis Ababa through infrastructural development. The result was a 1931 study produced by JG White Engineering Corporation. As with the British-sponsored studies, the proposal centred on a barrage at Lake Tana. While the British investigations appear to have taken little interest in the possibility of using the dam to benefit Ethiopian development, the JG White study did so, proposing a dam that would maintain Lake Tana at its maximum annual height, to avoid flooding important Orthodox monasteries, as well as developing small hydropower projects downstream at the Tis Isat Falls. Yet Ethiopia remained in the midst of the extended succession struggle that followed Menelik’s death. While Ras Tafari was keen to pursue the project, it was resisted by Ras Gugsa of Begemdir and Ras Hailu of Gojjam—the powerful governors of the two provinces in which Lake Tana sat (McCann 1981; Collins 1990; Zewde 1991). The regent lacked the power to impose his plans for a dam and the project stalled.
It was not until the military defeat of Ras Gugsa in 1930, shortly followed by the death of Empress Zewditu, that Ras Tafari was crowned Emperor Haile Selassie (Zewde 1991). Having removed an influential competitor and centralized power, Haile Selassie was now free to pursue the Lake Tana scheme and follow-up studies were conducted by JG White in 1931–1934. However, by this time the Great Depression had contributed to falling cotton prices, while the 1929 Egypt–Sudan water agreement had guaranteed Sudan a share of the Nile waters, removing the British Empire’s sense of urgency for upstream storage and regulation (McCann 1981). Britain opted to delay discussion of the dam and failed to attend a planned conference in 1935. The invasion of Ethiopia by Mussolini’s Italy in October 1935 meant that Ethiopian and colonial ambitions to dam the Blue Nile remained unfulfilled.
Modernization and State-Building in the Awash Valley
As discussed below, securing alignment between international and domestic politics for a Blue Nile dam was no easier after Haile Selassie returned to the throne following the Italians’ expulsion in 1941. However, Haile Selassie’s attempts to modernize the Ethiopian state and economy necessitated investment in electricity generation and agricultural development. The result was that Imperial Ethiopia switched its priorities from the politically charged Blue Nile to the more straightforward challenge of developing the internal Awash River. The government sought foreign support to produce hydropower and expand irrigated agriculture along the Addis–Djibouti railway that runs through much of the river basin as part of an intertwined project of state-building and economic modernization.
Over the following three decades, Haile Selassie pursued a series of reforms to modernize the country and strengthen its independent status, expanding the bureaucracy and promoting economic development. In terms of foreign policy, the emperor sought to develop relations with a wide range of key countries, albeit the main focus was an alliance with the US as counterweight to the British colonial presence all around Ethiopia (Clapham 1988; Patman 2009). From the US perspective, Ethiopia was seen as a key anchor state in the Cold War rivalry with the USSR (Verhoeven and Woldemariam 2022). Links with the US were consolidated by providing access to the strategic Kagnew communications base in Eritrea and by sending Ethiopian troops to fight in the Korean War. Ethiopia also became the first African member of the World Bank and first recipient of support on the continent in an era in which the World Bank was the main financer of infrastructure, including dams. While Ethiopia allied to the west, the USSR came to support rival regimes in Egypt (following Nasser’s rise to power) and Somalia, particularly after Siad Barré seized power in 1969 (Patman 2009).
While the Blue Nile had dominated plans before the war and would become a major focus for Haile Selassie afterwards, it was the Awash Basin that was the main focus of Ethiopia’s dam building in this period. The Awash offered advantages over other river basins, many of these related to technical feasibility. While other river basins, such as the Baro and Omo, offer much greater potential for hydropower production, they are located further away from the economic centre and the main centres of demand for electricity—Addis Ababa and the string of towns located at the stations on the Djibouti railway (Figure 2.4). Developing sites further afield would require more challenging engineering and much greater investment in transmission lines (Waterbury 2002). Yet Ethiopia’s demand for electricity in the 1950s and 1960s was itself modest and could be met by a series of small hydroelectric plants on the nearby Awash. Moreover, the Awash is perhaps the most favourable basin to the development of irrigated agriculture (FAO 1965), providing the possibility of agricultural commercialization and reliable transport on the Ethio-Djibouti railway—either to the main domestic market in Addis or for export. Multipurpose dams could regulate the water flow, providing year-round water for large-scale irrigated agriculture.
Figure 2.4
Dams and main agricultural projects in the Awash Basin
Source: Author and Manchester Cartographic Unit.
Development of the Awash was also favoured by political factors. As the only river internal to Ethiopia, it avoided the transboundary challenges and objections from riparians that would dominate negotiation regarding the Nile and prevent Ethiopia’s access to international finance and technical expertise. Moreover, though the Upper and Middle Awash Valley are situated relatively close to Addis Ababa, in this early post-war period, much of this lowland area was avoided by highlanders as a result of malaria and the presence of supposedly ‘backward’ pastoralists (Kloos 1982). For the most part, the Awash Valley lacked the powerful sub-national elites that had earlier resisted a dam at Lake Tana. The result was that Imperial efforts focused on developing this ‘near periphery’ where the valley offered great potential for an intertwined project of state-building and economic modernization.
The potential for irrigated agriculture and hydropower in the Awash Basin was initially explored during the Italian occupation. Societa Agricoltura Industriale nel Etiopia (SAIDE) constructed a 1,600-hectare (ha) sugar plantation, with unrealized plans for a sugar factory to follow (Zewde 2008a). Furthermore, the Italians made plans for and began construction of a small hydropower dam at Koka (Last 1958). The Imperial Government’s transformation of the river basin sought to mobilize foreign investment and support. The first step in this direction was the Wenji-Shoa sugar factory and 5,000-ha plantation as a joint venture with Dutch firm HVA (Handelsvereeniging Amsterdam) in 1954 (Zewde 2008a). Following an Ethiopian Government request, the US Bureau of Reclamation (USBR) was then commissioned to carry out a study of the topography and hydrology of the Awash, particularly focused on the construction of a larger Koka Dam. At the time, the Bureau was just beginning its international strategy and, in doing so, was deeply embedded in Cold War politics and the advancement of US interests around the globe (Sneddon 2015). The close relations between the Imperial Government and the US, and the Americans’ Cold War strategy in the Horn, were key to securing technical support from the Bureau.
In 1957 construction of the first dam on the Awash began. The 43-MW Koka Dam was completed in 1960 and placed under the recently created (1956) state-owned Ethiopia Electric Light and Power Authority (EELPA). Rather than US support, however, it was Italian reparations for the occupation of Ethiopia that were used to finance dam construction, built by Italian firm Imprese Italiane all’Estero (Last 1958; see Chapter 5). Although small by later standards, the Koka Dam was at the time considered a major dam and doubled the country’s existing generation capacity. Moreover, Emperor Haile Selassie’s opening speech was an opportunity to emphasize the significance of Koka as a first step in developing the country’s water resources:
It is the duty and privilege of this generation and of posterity to conserve and develop these precious [water] resources. To fail to do so will be to fail in our God-given responsibility … This project is but the first step in a similar programme We have in mind for the other water courses of Our Empire, such as the Nile with its volume and potentialities so vastly greater, as well as the Baro the Sobat, the Akobo and the Webi Shebeli. (Haile Selassie 1958)
Koka was always intended as the regulator dam for a cascade of hydropower projects, with the World Bank-funded 32-MW Awash II (1966) and Awash III dams (1974) following afterwards. The planned fourth dam in the cascade, however, was never realized. As well as producing electricity for Addis Ababa and Dire Dawa—for a long time Ethiopia’s second largest city—the regulation provided by the dams enabled further irrigated agricultural concessions downstream (Last 1958; Nicol 2000). Also built in this period was the Legadadi Dam (1964–1970) on the Akaki River, part of the Awash Basin, to provide water for Addis Ababa. While only a small water storage dam, it is, nonetheless, noteworthy as the first dam in Ethiopia to be built by Salini Costruttori6 and designed by Studio Pietrangeli, a team that would come to fundamentally shape Ethiopia’s dam building in subsequent decades (see Chapter 5).
During the 1960s, plans to develop commercial agriculture in the basin expanded. British company Mitchell Cotts, inspired by Sudan’s Gezira scheme and backed by the British authorities, approached the government to establish cotton production in the Awash (Ashami 1985; Fantini and Puddu 2016). The resulting Tendaho Plantation Share Company (TPSC) was another joint venture with the government that was granted 18,000 ha in the lower Awash which began operations in 1961 (Ashami 1985). The Imperial Government, with support from the United Nations’ Food and Agriculture Organization (FAO), subsequently created the Awash Valley Authority (AVA) in 1962, modelled on the United States’ Tennessee Valley Authority, tasked with developing irrigation and commercial agriculture. The government claimed all lands not covered by rist tenure as state domain under the control of the AVA, which granted concessions to a series of smaller projects along the river (Ashami 1985; Nicol 2000). Meanwhile, a 1965 study of the basin by the FAO proposed dams at Kessem and Tendaho to make available an additional 163,000 ha of irrigated land (FAO 1965, p. 15). Although the dams would be multipurpose, the small hydroelectric element was a secondary concern to irrigation. These plans were not realized during the Imperial era; however, commercial agriculture did expand with the opening of HVA’s Metahara 4,500-ha plantation and sugar factory downstream in 1970 alongside a number of other smaller commercial farms (Kloos 1982).
As noted above, the development of the Awash Basin was intertwined with a state-building project to integrate what had previously been a peripheral area. Supposedly ‘backward’ and ‘uncivilized’ pastoralists—Oromo in the Upper Awash Valley and Afar in the Middle and Lower Awash—who lacked centralized political authorities were easily pushed off the land targeted by the government and investors (Harbeson 1978; Ashami 1985; Zewde 2008a; Behnke and Kerven 2013; Lavers 2016).7 The exception here is the Awsa Sultanate in the Lower Awash. For several centuries farmers had practised irrigated agriculture in Awsa, sustaining a centralized authority. While Awsa was militarily defeated by Menelik in 1895, no Ethiopian administrative presence was established and the Sultanate remained largely beyond state control (Harbeson 1978; Ashami 1985; Nicol 2000). Indeed, the Sultanate resisted the AVA, retaining control of access to land and water in its territory. As such, having received Imperial approval, Mitchell Cotts had to negotiate directly with the Sultan to secure land for the Tendaho development (Ashami 1985), and the government established administrative and police posts nearby as part of the project agreement (Fantini and Puddu 2016). Moreover, an agreement was made for Awsa farmers to supply cotton to the Tendaho project, resulting in a major expansion of Awsa capitalist agriculture, enhancing the economic power of the Awsa elite, and limiting potential for further AVA land grants to outside investors (Ashami 1985).
Overall, these attempts at economic modernization and state-building were ‘an unmitigated disaster’ for local pastoralists (Markakis 2011, p. 139; Behnke and Kerven 2013). State intervention in the form of dams and reservoirs, the conversion of 52,370 ha of prime land into plantations (Kloos 1982, p. 28), and the creation of the Awash National Park restricted the access of the Afar and Jille, Kereyu, and Arsi Oromo pastoralists to key water and land resources. These developments have been linked to the catastrophic effects of the 1972–1973 famine amongst pastoralist populations, leading to the deaths of 100,000–200,000 people or 25–30 per cent of the Afar population (Kloos 1982; Ashami 1985; Markakis 2011; Behnke and Kerven 2013). Nonetheless, the government continued to frame pastoralism itself as inherently unsustainable, necessitating modernization through the expansion of settled agriculture (Rettberg 2010). The first settlement schemes in Afar were launched under the Imperial Government by the AVA just before the 1974 revolution (Harbeson 1978; Ashami 1985) and were then expanded by the Derg (Kassa 1997; Markakis 2011).
The focus on the transformation of the Awash Basin in the post-war years was therefore the result of both technical feasibility and an alignment of political interests across multiple scales of analysis. The US, World Bank, and Italy were all willing to provide finance and expertise to build dams in this internal river basin as means of consolidating alliances with a key ally. Meanwhile, this phase of dam building was enabled by and contributed to the centralization of power during Haile Selassie’s modernization drive, and the political marginalization of the largely pastoralist population of the Awash Valley.
Imperial Ethiopia’s Ambitions on the Nile
Haile Selassie had ambitions to use the Awash Basin as a stepping-stone to the development of other river basins and, in particular, the Nile. For Haile Selassie, the Nile was of key significance, both for the development of the country, and as a symbol of modernization and the legitimacy of his rule (Sneddon 2015):
It is of paramount importance to Ethiopia, and a problem of the first order, that the waters of the Nile be made to serve the life and needs both of our beloved people now living and those who will follow us in centuries to come. However generously Ethiopia may be prepared to share this tremendous God-given wealth of hers with friendly neighbouring countries for the lives and welfare of their people, it is Ethiopia’s primary and sacred duty to develop her water resources in the interest of her own rapidly expanding population and economy. (Haile Selassie 1957, cited in Erlich 2002, p. 134)
However, the alignment of technical feasibility and multi-scalar political dynamics in the Awash was not replicated elsewhere. Following the war, efforts at comprehensive basin-wide management of the Nile collapsed, leading to a series of uncoordinated national projects in Egypt and Sudan focused on meeting domestic objectives, often at the cost of basin-wide benefits. Meanwhile, Ethiopia was constrained by its inability to mobilize international support for dam building in the Nile Basin. While the Ethiopian emperor was unable to realize his ambitions before his demise in 1974, he nonetheless advanced plans for the development of the Blue Nile that would prove influential decades later.
Following World War II and in the dying days of British influence in Egypt, the Egyptian Ministry of Public Works published its famous study by Hurst, Black, and Simaika (1946) on The Future Conservation of the Nile. This study introduced the concept of ‘Century Storage’—essentially the water storage capacity required to maintain average water flow, even in the most exceptional drought years. The proposals focused on massively increasing storage in the upstream Albert and Victoria Lakes in Uganda, and Lake Tana in Ethiopia, where evaporation would be modest, and construction of an enlarged Jonglei Canal to transport the White Nile through the Sudd without major evaporation losses. The study noted the possibility of a hydroelectric plant at Tis Isat, just downstream from the proposed Tana regulator dam, as a means of appealing to the Ethiopian Government (Hurst et al. 1946, p. 98). However, hydroelectricity was not studied in any detail and the objective remained to ‘control the Nile in the interests of irrigation in Egypt and the Sudan, and also to provide flood protection for Egypt’ (Hurst et al. 1946, p. 1).
In a pattern that would frequently repeat on the Nile, supposedly optimal engineering proposals for management of the entire basin foundered on the political reality of divergent national interests, however. In 1952 the revolutionary command council, led by General Nasser, seized power in Egypt and secured agreement for British withdrawal. With Britain insisting that Sudan should remain independent of Egypt, the new Egyptian Government was forced to accept the impossibility of establishing political control over the entire Nile Basin, a major blow to Century Storage (Erlich 1994). Egypt, and colonial-era Sudan and Uganda, did agree to build the Owen Falls Dam in Uganda, which opened in 1954 under the permanent watch of an Egyptian engineer. However, having built Owen Falls, Uganda would gain little from the subsequent projects required to deliver Century Storage, while facing the political challenge of large-scale land loss (Collins 1990). With Ethiopia unwilling to discuss a Lake Tana Dam, the Century Storage plan was effectively dead by the mid-1950s.
Instead, the new Egyptian Government focused on securing water supply internally through construction of the High Aswan Dam (Waterbury 1979; Collins 1990; Erlich 2002).8 A large dam at Aswan had been proposed decades before but was given little consideration by colonial-era engineers intent on finding an optimal technical solution for the management of the whole Nile Basin (Collins 1990). From the perspective of basin-wide water management, the Aswan Dam is simply ‘the wrong dam in the wrong place’ (Collins 1999, p. 3). Instead of storing water in the comparatively cool Ethiopian and Ugandan highlands to minimize evaporation, Aswan stores vast amounts of water in a flat, hot desert, resulting in massive losses through evaporation, estimated at 12–16 billion cubic metres (bcm) annually or approximately one-sixth of the river’s flow, and, to a much lesser degree, through seepage of less than 1 per cent (Hussein 2018, pp. 5–60).9 Yet for a revolutionary government seeking to build its legitimacy and confronting the fragmentation of political authority in the Nile Basin, Aswan fit the bill. The government promoted Aswan as a key nationalist symbol, with the dam providing Egypt with over-year storage capacity within Egypt’s borders (Mitchell 2002; Hanna and Allouche 2018). As such, it was an attempt to ‘free Egypt from being the hostage of upstream riparian states by providing for century storage within the boundaries of Egypt’, addressing what was increasingly regarded as a key national security issue (Collins 1990, p. 239).
The massive dam located in southern Egypt flooded extensive areas of northern Sudan and therefore required Sudanese support. Negotiations with Sudan revisited the 1929 agreement and eventually produced the 1959 Nile Waters Agreement that allocated 55.5 bcm of the annual flow of the Nile to Egypt and 18.5 bcm to Sudan, once again ignoring the water rights of all other Nile riparians (United Arab Republic and Sudan 1959). While heavily favouring Egypt, the 1959 agreement increased Sudan’s water allocation and secured Egyptian support for Sudan’s Roseires Dam, which was built in 1965. Following the failure to negotiate the Lake Tana Dam with Ethiopia, Roseires was, like Aswan, an attempt at a national solution to the failure of basin-wide planning. Roseires enabled Sudan to expand irrigation and utilize some of the increased water allocation secured in the 1959 agreement (Collins 1990, p. 251). While the US and British Governments and the World Bank had all initially explored the possibility of financing Aswan, the 1956 Suez Crisis led the British and American governments to withdraw.10 Consequently, Nasser turned to the USSR, with construction beginning in 1960 and completed by 1970 (Waterbury 1979).
Far from solving Egypt’s water problems, the Aswan Dam has enabled ever-growing demand for water in the country through efforts to expand agriculture and settlements into the desert. After 1952, Nasser’s government undertook land reform that promised to end ‘feudalism’ and dominance of the powerful landholding elite in the Egyptian economy and politics. However, the reforms stopped short of a far-reaching reorganization of agricultural production. The government established a land ceiling and redistributed 12 per cent of the land, destroying the base of the richest 2,000 landlords, but left the middle farmers and the existing system of production intact, impeding modernization of agriculture and irrigation systems (Bush 2007). Nasser’s government did attempt industrial deepening through import substitution, leading to increased employment and improved welfare. However, the government failed to discipline firms to deliver productivity improvements, eventually resulting in economic crisis amidst the high costs of subsidies (Waterbury 1983). In the absence of agricultural and industrial transformation, the consistent official narrative from this time onwards has been that population growth and land shortages along the Nile necessitate ‘greening the deserts’. In this way, the Egyptian Government sought to avoid what would be politically complex reforms to existing agriculture or disciplining industrial firms (Waterbury 1983; Mitchell 1995; Sims 2015; Hanna and Allouche 2018). The result has been a series of large-scale irrigation schemes, based on diverted Nile waters from Aswan, most of which have been spectacular failures (Sims 2015). Nasser announced the New Valley scheme in 1958 that aimed to irrigate 3 million feddans and resettle some 4 million people. However, less than 120,000 feddans were ever realized and only 57,000 people resettled (Sims 2015, p. 50). The Tahrir Province Scheme, announced in 1954, was more successful, irrigating more or less the intended 600,000 feddans but only after many years of investment and operation (Sims 2015).
The 1959 agreement provoked a fallout between Egypt and Ethiopia. Ethiopia responded in a diplomatic note verbale that rejected Egypt’s unilateral actions, signalling that this did not foreclose Ethiopia’s future right to use the Nile waters. Furthermore, the Ethiopian Orthodox Church finally separated from the Coptic Church in Alexandria in 1959, ending a connection that had lasted 1,600 years. The 1959 agreement required Egypt and Sudan to negotiate jointly with upstream countries, as and when they made demands for a share of the Nile waters. However, both countries’ stance on the Nile hardened subsequently, insisting on the protection of their ‘historic rights’ to the Nile and rejecting any reduction in their 1959 allocations (Waterbury 2002). To this end, Egypt has blocked access to international finance for any upstream projects and provoked internal instability in Ethiopia in particular (Erlich 2002). Notably, Egypt helped to form and train the Eritrean Liberation Front (ELF) in 1959 to pursue an ‘Arab’ revolution, while also promoting Somali territorial claims in Ethiopia (Erlich 1994; Woldemariam 2018).
Ethiopia’s relations with Sudan also deteriorated, not only due to the Nile but also because of the border between the two countries that was never adequately demarcated in agreements between Ethiopia and Britain. Following Sudanese independence in 1956, contestation particularly focused on the fertile Al-Fashaga triangle that attracted an influx of Ethiopian settlers connected to cotton and sesame production in the 1960s (Puddu 2017). Meanwhile, Sudanese support to the ELF in Eritrea was reciprocated by Ethiopian support for the Anyanya movement that was fighting the government in the growing civil war in Sudan (Erlich 1994). Tensions grew to the brink of war between Ethiopia and Sudan in 1967.
Meanwhile, the Imperial Government responded to the 1959 agreement and construction of Aswan by requesting that the USBR follow up its work on the Awash River with another, more substantial study of the Blue Nile. For the US Government, the project was intended to threaten Egypt, which had turned to the USSR for assistance with the High Aswan Dam, through its dependence on the Nile. This threat, however, did not translate into full-blooded support for Ethiopia or a willingness to provide the resources to construct a major dam on the Nile. Rather, the intention was to use the study to force Egypt back into the US camp in the Cold War (Sneddon 2015). As Sneddon shows, however, the Bureau of Reclamation team was certainly full-blooded in its re-imagining of the Blue Nile Basin. In line with the Bureau’s approach to the development of the western US (Reisner 1993), the Bureau envisaged
an incredible array of projects that, if built, would have radically transformed the Blue Nile into a modern basin … [providing] nearly complete control over the flows of the river and its tributaries. (Sneddon 2015, p. 90)
In total, the study proposed some 30 irrigation, hydropower, and multipurpose projects that would provide 433,754 ha of irrigated land and 8,660 MW of installed capacity (USBR 1964, p. vii).
Like with the pre-war proposals, a regulator dam at Lake Tana was among the USBR projects. However, the aim of maximizing the use of the river for Ethiopian development meant that the Bureau expanded its scope to cover projects for irrigation and hydropower development across the entire river basin. The study took up the idea floated earlier by Dupuis and Grabham and Black of digging a tunnel to drain some of the water from Lake Tana into another Nile tributary, in this case the Beles River. While earlier studies found no rationale for the considerable expense involved, the USBR envisaged a major hydropower project and large-scale irrigation schemes on the Beles, projects that would subsequently be tackled by both the Derg and the Ethiopian Peoples’ Revolutionary Democratic Front (EPRDF) (see below and Chapters 4 and 5). However, the centrepiece of the USBR study was a cascade of four hydropower dams on the Blue Nile itself (Table 2.2).11 By far the largest dam was to be Karadobi, upstream in the cool highlands, which would regulate the flow for downstream dams. As such, and as is standard practice with dam cascades, the expectation was that construction would begin with Karadobi and then move on to successive downstream dams. Combined, these four Nile dams would have 73 bcm of water storage capacity and 5,570 MW of installed generation capacity, which could produce 25,149 GWh of firm energy per year. Notably, with respect to subsequent projects, each of the proposed dams was designed to have a plant factor based on firm energy generation12 over a six-year time period of around 50 per cent (USBR 1964, p. 157). As such, the designs suggest a balance between baseload and peaking power.
Table 2.2
Hydropower projects proposed on the Blue Nile by the Bureau of Reclamation
| Dam | Installed capacity (MW) | Firm energy (GWh/year) | Plant factor (per cent) | Water storage (billion cubic metres) |
|---|---|---|---|---|
| Karadobi | 1,350 | 5,835 | 49 | 32.5 |
| Mabil | 1,200 | 5,314 | 51 | 13.6 |
| Mandaya | 1,620 | 7,800 | 55 | 15.9 |
| Border | 1,400 | 6,200 | 51 | 11.1 |
Data source: USBR (1964)
The proposed dams across the basin would prove enormously influential during the Derg and, particularly, EPRDF eras. Indeed, this includes the Border Dam site, identified by the USBR study, which was eventually selected for the Grand Ethiopian Renaissance Dam (GERD). However, other than the dam’s location, the design of the GERD and Border Dam bear no resemblance to each other. The Border Dam proposal was for a relatively small dam, roughly one-seventh the water storage capacity of the GERD and dependent on upstream regulation of the water flow to enable year-round energy generation (see Chapter 7).
Despite its ambitions, the Bureau’s study acknowledged that this vast array of proposed projects was beyond the capacity or requirements of Ethiopia in the foreseeable future. The study proposed several small projects to be developed before 2000, of which the multipurpose project at Finchaa on a tributary of the Blue Nile was to be the first. The USBR study envisaged Finchaa as the first and largest in a complex of three dams on adjacent rivers—Finchaa, Amarti, and Neshi—that together would provide hydroelectric power and water regulation for a large-scale plantation (USBR 1964). The dams on the Blue Nile itself were not expected to be feasible before 2000, depending on the growth of the economy and electricity demand.
Yet US political and financial support did not follow. According to a US National Security Council member, the study was considered ‘the best hold this country [the US] could possibly have on Nasser for bargaining purposes’ rather than the starting point of a major programme of investment (Sneddon 2015, pp. 89–90). Moreover, by the late 1960s, advances in satellite communications meant the declining importance to the US of the Kagnew base and with it the US’s strategic interests in Ethiopia (Patman 2009). While the US ‘was lukewarm concerning the strategic importance of Ethiopia to broader American geopolitical goals’, Egypt remained the main US priority (Sneddon 2015, p. 97).
It was only after extensive lobbying by Haile Selassie that the US Government and World Bank finance was eventually provided to fund Italian firm Impresit’s construction of the Finchaa Dam in 1969–1972, adding 84 MW to Ethiopia’s installed capacity and extending the grid centred on Addis. By this stage the World Bank’s 1965 Operational Memorandum required all transboundary water projects to consider the impact on other riparians. In particular, World Bank staff had to state that either the transboundary issues raised were covered by arrangements between countries; that the affected riparians had stated no objection; or that the project had no impact on the riparians and their consent was therefore immaterial (Salman 2009, p. 100). Despite Egypt’s apparent protest against Finchaa (Waterbury 2002, p. 121), the World Bank’s assessment was that ‘the volume of water stored and released, when compared with the flow of the Blue Nile where the Finchaa enters its stream, would be insignificant’ (IBRD 1969, p. 5). The World Bank’s willingness to bypass Egyptian and Sudanese opposition was limited to Finchaa and would not apply to dams on the Blue Nile itself where the impacts could not be so lightly dismissed.
It should be noted that while the political reforms undertaken since World War II had considerably strengthened the position of the emperor and the bureaucracy at the expense of the landed nobility, limits to Imperial authority remained at this time. While the USBR survey team’s work was facilitated in several riverine provinces by personal connections between the emperor and provincial governors, they faced challenges in Gojjam, where the survey team felt the local population ‘had not accepted the government of Ethiopia’ (Sneddon 2015, p. 91).13 The result was that the team was accompanied at all times by four armed guards for their protection. Compensation procedures for the Finchaa Dam were also shaped by the land tenure system and prevailing agrarian structure. While the project required the resettlement of approximately 3,115 people, it was only the landholding elite that received compensation for their losses, not the tenants who actually worked their land (Müller-Mahn and Gebreyes 2019, p. 6).
The USBR study also had a lasting impact on technical capacity within Ethiopia. Although USBR engineers led the work, the massive study was seen as a means of building local capacity (Sneddon 2015). The study led to the creation in 1956 of the Water Resources Department in the Ministry of Public Works and Communications, originally staffed by just one engineer and 50 technicians (Abate 1994; Arsano 2007a; Haile 2018, p. 93). The USBR study ‘helped train an entire generation of Ethiopian hydrologists’ (Waterbury 2002, p. 116), and though the Water Resources Department was disbanded on completion of the study in 1964, the staff involved in the project provided the core of the National Water Resources Commission created in 1971.
The only other development in the Blue Nile Basin at this time was the first Tis Abay hydroelectric plant, completed by Italian and Yugoslav contractors in 1964 with the remaining Italian war reparations at a site identified by the American study in 1931 (Last 1958). Throughout this period Haile Selassie maintained multiple prongs to his foreign policy strategy, expanding relations with the USSR and Soviet bloc countries, primarily with a view to coaxing greater support from the US (Patman 2009; Fantini and Puddu 2016). It is possible that Yugoslav involvement was the result of this strategy. The project relied on a weir to divert part of the Blue Nile shortly after leaving Lake Tana, and was rated at 7.7 MW, with potential for future expansion to 11.4 MW (JICA 1977). Tis Abay provided electricity for a self-contained grid centred on Bahir Dar, including a new textile factory. As a run of the river dam, Tis Abay did not create a reservoir and the project had no real impact on downstream flows. With no regulation of Lake Tana, the flow of the river and consequently electricity generation varied considerably through the year.
Unlike on the Awash, Haile Selassie never secured the necessary constellation of political forces to pursue major projects on the Nile. International support—so important for both finance and technical expertise—from the US Government was tentative, using detailed and far-reaching studies to apply pressure to Egypt, but not going so far as to damage a country of great strategic importance. The Finchaa and Tis Abay projects did reduce Ethiopian electricity dependence on the drought-prone Awash Basin. Yet by the end of Haile Selassie’s reign in 1974, 55 per cent of Ethiopian hydropower continued to be generated by a cascade of relatively small dams on the Awash.
Revolutionary Ethiopia and Cold War Realignments
The late 1960s and early 1970s brought major changes to national politics in Egypt, Ethiopia, and Sudan, as well as a complete re-alignment of Cold War allegiances in the region. Yet, these changes did nothing to improve alignment of political interests regarding a negotiated settlement over the Nile or Ethiopia’s prospects of unilaterally building a Nile dam. Although undermined by external and internal military threats, the Derg Government was able to mobilize limited external support towards dam construction outside the Blue Nile Basin.
In Ethiopia, the powerful landed elite resisted far-reaching political and economic reforms, resulting in stagnation and growing political instability (Clapham 1988). Ultimately, a ‘creeping coup’ deposed Haile Selassie following months of protests and strikes in 1974. Ethiopia was initially ruled by the Derg, a committee of junior military officials, though Mengistu Hailemariam emerged as the unchallenged leader as he gradually eliminated competitors. In 1975, the new government launched rural and urban land reforms that eradicated the economic base of the nobility (Clapham 1988). Internationally, Mengistu aggressively pursued the support of the USSR and other socialist countries in the early years following the revolution (Clapham 1988; Tekle 1989; Patman 2009). The USSR, which had been providing military aid to Siad Barré’s regime in Somalia, initially sought to reconcile fractious relations between Somalia and Ethiopia (Farer 1979; Patman 2009). However, the Ethio-Somali War (1977–1978), in which Somalia sought to claim territory occupied by ethnic Somali within Ethiopia, forced the Soviets to make a choice. In 1977, Mengistu’s emergence to supremacy led the USSR to switch sides, as a result of the greater strategic importance of Ethiopia’s Red Sea coastline and the greater prospects of the Ethiopian revolution (Patman 2009). Soviet and Cuban military support was vital to reversing initial Somali gains (Clapham 1988; Tekle 1989; Erlich 1994). While the US did not initially break ties with the new Ethiopian Government, the turn to the USSR in 1977 led to the expulsion of US officials and an end to US military support (Patman 2009).
Downstream on the Nile, Egypt and Sudan trod the reverse path by switching from Soviet to Western alliances, while also shifting security relations within the Middle East. Following Nasser’s death in 1970, and Egypt’s defeat in the 1967 and 1973 conflicts with Israel, President Sadat moved Egypt away from pan-Arabism, leading to reconciliation with the US through the Camp David agreement in 1978 and Israeli peace agreement in 1979. After 1979, Egypt abandoned Nasser’s state-subsidized import substitution, instead adopting a strategy of promoting foreign investment and export-led growth while cutting back on the state’s distributive commitments (Sallam 2022). As a result, Sadat opened up to investment from Saudi Arabia and other Gulf countries—previously dismissed as Western puppets under Nasser—while also cultivating the Muslim Brotherhood as a counterbalance to the political left that resisted liberalization (Waterbury 1983; Sallam 2022). Meanwhile, a leftist military coup in 1969 led by Ja’afar Nimeiri in Sudan aimed initially at revolutionary state-building and the promotion of national development. Following a split with the communists in 1971, however, the Nimeiri regime reached out to the West, Sadat, and the Gulf as sources of investment (De Waal 2015; Verhoeven 2015).
These realignments did not greatly increase the resources available for Ethiopian infrastructure investment. Ethiopia developed close relations with several Soviet bloc countries, sending significant numbers of Ethiopian officials for education and training (Tekle 1989). Ethiopia also developed close relations with North Korea—to some degree independently of the USSR—with Mengistu making regular visits to Pyongyang (Clapham 1988; Tekle 1989). Much like the US support to Imperial Ethiopia, however, the Soviet assessment of the strategic importance of Ethiopia seems to have been ambiguous while the Soviets grew frustrated at slow pace of the Ethiopian revolution (Patman 2009). The result was that the USSR provided just 22 per cent of Ethiopia’s aid between 1983 and 1986, to Mengistu’s considerable disappointment (Patman 2009, p. 275; Fantini and Puddu 2016). Indeed, Soviet alignment failed to translate into the kind of support that had been extended to Egypt in constructing the High Aswan Dam. While socialist countries supported the construction of several medium-sized dams outside the Nile Basin, as discussed below, significant international support for the development of the Blue Nile was not forthcoming. Meanwhile, Ethiopia did not entirely terminate its relations with the West. Trade relations were maintained with many countries in the capitalist camp, and both the European Economic Community and the Italian Government continued to provide aid in the 1980s (Tekle 1989).
From the moment it took power, the Derg faced internal insurrections, the majority of which were socialist and ethno-nationalist, favouring greater autonomy or secession for Ethiopia’s ethnic groups in contrast to the Derg’s ‘Ethiopia First’ nationalism (Clapham 1988).14 As such, the military threat to the regime did not disappear with the end of the Ethio-Somali War in 1978, but rather intensified as the Derg switched its focus to insurrections in Eritrea, Tigray, Oromiya, and elsewhere. The result was that much of Ethiopia’s internal resources and external support were directed to military expenditure rather than infrastructure development. Likewise, political instability meant that the new regime was unable to consolidate the Imperial regime’s initial efforts to build technocratic capacity through the Water Resources Commission and the AVA (Waterbury 2002; Arsano 2007b). The Awash was subsequently incorporated into the Ethiopian Valleys Development Studies Authority, but instability and regular purges of staff meant that Ethiopia did not develop the capacity to lead major water projects itself.
In the meantime, Egypt and Sudan both continued to increase their utilization of the Nile waters fuelled by Gulf finance, threatening to exceed the allocations made under the 1959 agreement. Under Sadat, Egypt maintained its push to green the desert and thereby ease population pressure in the Nile Valley, albeit now through foreign private investment rather than state projects (Sims 2015). After the return of the Sinai to Egypt in 1980, this included a plan to settle 5 million people in the peninsula based on the pumping of Nile waters. Unsurprisingly, the project was even less successful than the New Valley project (Sims 2015). Likewise, grandiose plans to build new towns and cities in the desert were poorly thought out and failed to meet high expectations (Sims 2015). Meanwhile, the key to Sudan’s economic plans was irrigated agriculture, using Nile waters, with capital-intensive foreign investments intended to make the country the breadbasket of Africa and the Middle East through a massive expansion of irrigated agriculture (Verhoeven 2015).
Egypt and Sudan’s growing water use amidst the constraints of the Nile’s annual flow and the 1959 allocations led to the revival of the Jonglei Canal as a means of reducing water losses through evaporation and thereby increasing the Nile’s flow in northern Sudan and Egypt (Collins 1990, p. 271; Verhoeven 2015). The 1972 peace agreement that ended Sudan’s 17-year civil war paved the way for the beginning of construction in 1978 (Collins 1990, p. 308). However, excavation was slow and difficult, while the canal sparked a mutiny of the Sudan People’s Liberation Army (SPLA) and the resumption of the civil war in 1984, putting an end to construction shortly before its expected completion in March 1985 (Collins 1990; Verhoeven 2015).
Bilateral relations between Ethiopia and its downstream Nile riparians, meanwhile, became ever more hostile. Mengistu issued periodic threats to reduce the flow of the Nile in response to Egyptian plans to expand irrigation (Swain 1997; Erlich 2002). Meanwhile, President Sadat’s response was that if ‘Ethiopia takes any action to block our right to the Nile water, there will be no alternative for us but to use force’ (cited in Swain 1997, p. 687). Sadat’s assassination in 1981 that brought Hosni Mubarak to power in Egypt did little to improve interactions. Ethiopian relations with Sudan were similarly problematic, with Ethiopia providing a base for the SPLA in its fight against the Sudanese government, while Sudan provided support to the Eritrean People’s Liberation Front (EPLF), Oromo Liberation Front (OLF), Tigrayan People’s Liberation Front (TPLF), and Benishangul People’s Liberation Movement (BPLM) in their fight against the Derg (Tekle 1989; RVI 2023). Clearly there was no possibility of a negotiated agreement with downstream riparians over a Blue Nile dam, something that would be required to access World Bank funds, for example. Indeed, dam planning on the Blue Nile did not progress significantly in this period, with the water development masterplan of 1990 merely replicating the USBR’s plans (WAPCOS 1990).
Despite Ethiopia’s Soviet alliance, several water projects in this period were funded by the West as it sought to recover influence in Ethiopia (Patman 2009). The Derg renewed efforts to regulate Lake Tana through a dam at the Chara-Chara cataracts, requesting a Japanese study of a weir to regulate the river, enabling the expansion of the existing hydropower plant at Tis Abay and construction of a second plant nearby (JICA 1977). Construction of the weir eventually began in 1984, but was incomplete by the time the Derg was removed from power (McCartney et al. 2009). The Chara-Chara weir was also a key part of the Tana-Beles scheme launched under the Derg and contracted to Salini Costruttori with Italian Government funding (see Chapter 5). This project aimed to realize proposals dating back to the early twentieth century to divert water from Lake Tana into the Beles River to enable hydropower generation and large-scale irrigation. Tana-Beles was framed as a response to the 1984/5 famine (Arsano 2007a; Haile 2018, p. 99), though the scheme actually became a means of forcibly resettling Tigrayan peasants and potential supporters of the TPLF from their home areas to undermine the Front. Moreover, and as with most other water projects in Ethiopia, the Tana-Beles scheme was used to strengthen state control over the area while marginalizing Gumuz-practising mobile livelihoods (Fantini et al. 2018; Fedeler 2021). Unsurprisingly, Egypt and Sudan raised strong objections to Tana-Beles (Fedeler 2021, p. 110), which would enable large-scale irrigation in the Beles Valley, extracting water and reducing the flow downstream. Despite the ambitious plans, the diversion tunnel from Lake Tana was not realized at this time and construction focused solely on infrastructure and irrigation works. Moreover, the project was abandoned as the civil war frontline advanced towards the project area shortly before Mengistu lost power (Fedeler 2021).
The Derg did build the Amarti Dam, the second of the USBR-proposed Finchaa-Amarti-Neshe complex, with European Economic Community funding (EEC 1985). Completed in 1987, the dam provides additional water storage to raise the capacity of the Finchaa plant from 84 MW to 100 MW (Müller-Mahn and Gebreyes 2019) and to irrigate a 6,000-ha state farm. Meanwhile, a Soviet study was also conducted of the Baro-Akobo Basin, a tributary of the White Nile, that identified several potential dam sites, but located far from the main demand for electricity, there was no move to develop the projects (Waterbury 2002, p. 125). The only dam built on the Baro-Akobo was the relatively small Alwero Dam, built by the state-owned Ethiopian Water Works Construction Enterprise for irrigation but unused until decades later.
While the Derg failed to realize its ambitions in the Nile Basin (Arsano 2007a, p. 174; Swain 1997), like the Imperial regime, it was more successful in tackling other river basins. The 153-MW Melka Wakena Dam in the Wabe Shebele Basin was completed in 1988, increasing national generation capacity by 70 per cent. The dam was constructed in the Bale highlands in the headwaters of the river with the support of the USSR and Czechoslovakia (Clapham 1988; Cheryachukin and Sitnin 2000; Patman 2009). The dam had significant impact downstream on farmers in Ethiopia’s Somali region (Hagmann 2007) and was also viewed with hostility by Somalia (Carr 2017). However, following military victory in the Ethio-Somali War, Somalia had little ability to resist and irritating Siad Barré probably was not a major concern to Mengistu. Completion of Melka Wakena significantly reduced Ethiopia’s dependence on the Awash Basin for power. While further plans for the Awash were advanced through feasibility studies of Tendaho and Kessem Dams, and a Basin Masterplan in 1989 (Halcrow 1989), the focus on the basin had shifted from electricity—for which little potential remained—to irrigation. As part of its land reforms, the Derg nationalized commercial plantations including sugar factories at Wenji-Shoa and Metehara in 1975, and redistributed the lands held by the Sultan of Awsa when he was forced into exile (Markakis 2011). However, little progress was made on the large-scale expansion of irrigated agriculture.
Ethiopia’s hydroelectric ambitions also shifted for the first time to the Omo Basin, one of the largest rivers by discharge and second only to the Blue Nile in hydropower potential. The initial site selected was upstream, in the highlands near Jimma and close to the existing grid. A North Korean company started developing a dam at what would become the site of Gilgel Gibe I (see Chapter 5) and North Korea provided several engineers to plan a 250,000-ha state farm in the Omo Valley (Clapham 1988, p. 235). However, neither scheme had advanced far by the time the Derg was defeated by the EPLF and TPLF, leading to abandonment of the projects.
Despite the vast hydropower potential of the Blue Nile, Baro-Akobo, and Omo Basins, by the fall of the Derg in 1991, Ethiopia’s electricity generation capacity remained very low, with 70 per cent dependence on the much more limited Awash and Wabe Shebele Basins (see Table 2.3). Likewise with irrigated agriculture, considerable potential remained unrealized and many past projects fell out of production due to the inept management of state farms (Clapham 1988).
Table 2.3
Hydropower generation capacity in Ethiopia in 1991
| Basin | Dam | Installed capacity (MW) | Percentage of hydropower capacity |
|---|---|---|---|
| Blue Nile | Tis Abay | 7.7 | 29 |
| Finchaa | 100 | ||
| Awash | Koka | 46 | 30 |
| Awash II | 32 | ||
| Awash III | 32 | ||
| Wabe Shebele | Melka Wakena | 153 | 41 |
| Total | 370.7 | 100 |
Data source: Author’s calculations.
The reality is that the Derg was in no position to tackle major projects on the Blue Nile or elsewhere, and no better placed than Haile Selassie in terms of the financial or technocratic capacity to tackle Ethiopia’s hydraulic mission. The government’s attention and financial resources throughout were focused on military confrontation—consolidating power following the revolution, repelling the Somali invasion, and then fighting a losing battle against growing insurrections in Eritrea and Tigray. Moreover, the Derg remained dependent on access to foreign expertise to realize its projects. Meanwhile, international support was faltering, with the USSR and socialist allies providing support, but never in a way that threatened the region’s main geo-strategic prize—Egypt.
Conclusion
This chapter examines Haile Selassie and the Derg’s efforts to launch Ethiopia’s hydraulic mission and the resulting spatial distribution of Ethiopia’s dam-building programme in this period. The analysis shows that technical inputs into dam planning and decision making were filtered through political dynamics at multiple intersecting scales of analysis and, in particular, the attempts of national governments to balance the competing pressures of international and domestic politics.
The technical feasibility of dam projects was undoubtedly a factor shaping decision making on dams, particularly by favouring sites in river basins close to the main load centre in Addis Ababa. However, political dynamics from the global to local levels played a vital role in shaping the spatial distribution of Ethiopia’s dams. Notably, the shifting global context from the colonial era to the Cold War was a major factor given Ethiopia’s dependence, as a late-developing country, on international finance and technical expertise to develop water infrastructure. These international alignments intersected with river basin politics and regional security and, in particular, growing competition between Egypt and Ethiopia over the Nile waters. Throughout this period, Egypt was considered of greater strategic importance to the key powers of the day, from Britain in the early twentieth nineteenth century to the US and USSR during the Cold War. As such, Egypt was either able to block upstream developments or the major powers refrained from taking action that would damage Egypt. In contrast, international support—shifting from the US in the Imperial era to socialist countries after the revolution—was more forthcoming for dam projects in other river basins, first the Awash and then the Wabe Shebele, or minor Nile projects that would have an insignificant impact on Egypt. The result was that by the time the EPRDF Government came to power in 1991, Ethiopian electricity generation capacity was tiny for a country of Ethiopia’s size and largely dependent on two river basins that offered among the lowest hydropower potential in the country. Meanwhile, the vast hydropower potential of the western half of the country remained largely untapped.
The EPRDF’s subsequent dam-building programme far surpassed that of the emperor or the Derg and in doing so has fundamentally altered the spatial pattern of Ethiopian dam building. However, in certain other respects these early attempts to tackle the hydraulic mission helped to establish particular continuities that shape dam development to the present. First, as a result of the physical and political geography of Ethiopia, dam development has entailed an intertwined project of state-building to consolidate the territorial gains secured under Menelik at the end of the nineteenth century and economic modernization through dams and agricultural investments. While justified in terms of the modernization of the economy and national development, benefits of this dam building have accrued to the centre, while the costs of dam and agricultural development are disproportionately borne by politically marginal populations in the periphery.
Second, early attempts to negotiate a dam on the Blue Nile consolidated the antagonistic positions taken by Ethiopia and Egypt. For Egypt, its red lines in negotiation have been its existing water infrastructure and its historic rights to use the Nile waters, while Ethiopia insisted that past unilateral Egyptian projects did not foreclose its future use of the Nile’s waters. While Ethiopia remained politically divided, economically weak, and of limited geo-strategic importance before 1991, Egypt was able to maintain what has been described as a position of ‘hydro-hegemony’ (Cascão 2008). Ethiopia’s weakness began to change, however, when the EPRDF came to power in 1991.
Notes
A phrase used by Emperor Haile Selassie (Selassie 2011, p. 576) and, according to some, originally attributed to him (Carr 2017).
The Awash is nonetheless believed to re-charge important groundwater sources in Djibouti.
The head, along with the water flow, is one of the key factors shaping the energy production potential of a dam. Head is the difference in water elevation between the reservoir level and the turbines and therefore dependent on the height of the dam and the drop in altitude of the existing watercourse.
Aquifers under the western desert are estimated to contain about 600 years of the Nile’s flow, but this deposit is the result of rainfall thousands of years ago that cannot be recharged and is therefore non-renewable (Sims 2015, pp. 40–41).
While the early proposals all focused on Lake Tana, other options were periodically considered. Major Cheesman speculated on two potential dam sites on the Blue Nile itself (Cheesman 1936, pp. 295, 360–361). The first was probably the Karadobi site and another near the Sudanese border just upstream from the site of the Grand Ethiopian Renaissance Dam (USBR 1964, p. 9).
Salini Costruttori took over Impregilo in 2012 to form Salini Impregilo. The firm re-branded itself in 2020 as WeBuild.
Puddu (2012) similarly notes how a World Bank-funded agricultural project was used by the Ethiopian Government to extend the state’s territorial reach and strengthen its claims along the Sudanese border.
For more detailed accounts of Egyptian and Sudanese dam building, see Waterbury (1979) and Collins (1990).
The 1959 agreement allows for a slightly lower amount of 10 bcm evaporation loss (United Arab Republic and Sudan 1959).
The World Bank underwent a learning process with respect to international waters in its early years, in parallel to the development of international water law. At this time, there were no established procedures. World Bank support depended on agreement between Egypt and Sudan, but not upstream riparians (Salman 2009, p. 25).
In this sense the proposals build on a study conducted during the Italian occupation (1936–1941) by Pontecorvo that rejected the idea of using Lake Tana as a reservoir and instead proposed a cascade of three dams on the Blue Nile (Pontecorvo 1938, cited in USBR 1964, p. 11).
Firm energy refers to the guaranteed energy generation regardless of the inter-annual variability of river flow. The plant load factor (or capacity factor) is the actual energy generation of a plant over a year divided by its maximum potential energy generation. A high plant factor indicates that a power plant is operating continuously at near full capacity providing baseload energy, while a low plant factor indicates that the plant operates at full capacity for only short periods, usually coinciding with peak demand.
Indeed, proposed tax reforms at this time that sought to bypass the landed elite and strengthen Imperial control over producers failed in Gojjam as a result of mass protests and rebellions (Tareke 1991).
One exception here was the Ethiopian People’s Revolutionary Party (EPRP) that fought against the Derg, prioritizing class-based revolution over ethno-nationalist organization.
Dams, Power, and the Politics of Ethiopia’s Renaissance. Tom Lavers et al., Oxford University Press. © Tom Lavers (2024). DOI: 10.1093/oso/9780192871213.003.0002
This is an open access publication, available online and distributed under the terms of a Creative Commons Attribution-Non Commercial-No Derivatives 4.0 International licence (CC BY-NC-ND 4.0), a copy of which is available at https://creativecommons.org/licenses/by-nc-nd/4.0/. Subject to this license, all rights are reserved.
