Navigation – Plan du site

Upstream Flows of Water: from the Lesotho Highlands to Metropolitan South Africa

Territorialities and Hydropolitics in Southern Africa
Yannick Rousselot
Cet article est une traduction de :
Des flots à contre-courant : des montagnes du Lesotho à la métropole sud-africaine


The aim of this article is to consider the mountains of Lesotho as suppliers of water to the Gauteng metropolitan area in South Africa. By analysing the hydropolitics of the Southern African region in the context of politics of scale, the focus is on the Lesotho Highlands Water Project (LHWP), a transboundary water transfer project between Lesotho and South Africa. The goal is to link hydropolitics with territorialities. The LHWP has resulted in a reproduction of the asymmetrical relations between and inside the metropolitan and mountain areas of both countries. Coalitions of actors have emerged from these new relations that the transfer has produced, and as such, this project should be analysed as a hydraulic assemblage in which three distinct scales of territorialities are clashing or cooperating with each other: the regional scale, made up of South Africa’s political and economic elites, who seek to direct Southern Africa’s water resources towards the thirsty region in and around Johannesburg; the scale of the Lesotho government, which has a nationalist policy of monopolising any social and spatial stakes for the benefit of the national territory; and the scale of the Maloti communities, which claim their rights as a transnational Basotho nation and autonomy in the face of interventionism from the Lesotho state.

Haut de page

Texte intégral

  • 1 Over 80% of the country lies more than 1,800 m above sea level (see the map).

1The aim of this article is to consider mountains as suppliers of water by using the case of the Lesotho Highlands Water Project (LHWP)a project of transboundary water transfer between South Africa and Lesotho, a mountainous, landlocked1 country completely surrounded by South Africa. This project was launched to provide water to the Johannesburg metropolitan area, which serves as South Africa’s economic hub and is the country’s most populous area, counting 12 million of the country’s 52 million inhabitants among its residents. The function of the LHWP is to provide irrigation but even more importantly for urban water supply.

  • 2 [W]hether it will be thanks to later phases of the Lesotho Highlands Water Project or other simila (...)

2The construction of the initial infrastructures began shortly after the LHWP’s founding treaty was signed in 1986. The project’s two most important dams, the Katse and the Mohale, were completed in 1997 and 2002, respectively. The LHWP’s hydraulic transfers that have already been built quench close to half the water needs of the Gauteng administrative district, with around 900 Mm3 of water transferred each year (Blanchon, 2008).2 Hydropower infrastructures (70 MW + 110 MW) intended for Lesotho have also been built. The water transferred from one dam to another flows through a tunnel 80 km long that has been dug out inside the Maloti-Drakensberg Mountains and reaches the Johannesburg region after a 500 km journey. When construction of the LHWP infrastructures is completed (the tentative date is set for between 2020 and 2030), dams will be constructed on every major waterway in Lesotho, and half of Lesotho’s hydrological volume will be diverted to South Africa. Given their social and environmental impacts, the LHWP facilities have generated intense opposition since the early stages of their implementation. A large array of studies have sought to evaluate these impacts, but the environmental impacts of the LHWP are beyond the scope of this paper and will not be considered here. Nevertheless, the relevance of the issue should not be underestimated, as its influence on successive steps of the project has been significant. Among others, it has contributed to a temporary delay in the first (1A) and second (1B) phases of the project and to the resizing of some parts of the project (Devitt and Hitchcock, 2010, Furlong, 2006, Hitchcock, 2012, Hoover, 2001, Horta, 1995, Nüsser, 2002, 2003, Swatuk, 2010, 2015, Thamae and Pottinger, 2006, World Bank, 1996).

  • 3 In the apartheid era, dominant actors justified harnessing water in terms of economic development, (...)
  • 4 [In the words of South African President Thabo Mbeki,] the economic destinies of our people and co (...)

3The LHWP has to be viewed within the specific context of mutual dependence between South African and Lesotho at both the economic and the political levels. This project exemplifies the powerful trend of hydraulic projects that embodied the developmental interests of the apartheid regime. In the post-apartheid area, those same policies have been extended with an eye towards the democratisation of access to water in South Africa.3 In this respect, the project was undertaken in order to ensure water availability in the long term in the Johannesburg metropolitan area and the industrial, mining and agro-industrial clusters of Gauteng province. Moreover, it has also been viewed as a way to ensure financial income for Lesotho’s political elites in the context of the decline of international development polices. As such, the two states regard the LHWP as a win-win cooperation opportunity,4 and both sides are satisfied that they stand to gain from it.

4Following Swatuk (2010, 2015) and Furlong (2006), closer scrutiny of this project’s differentiating effects between territories would put into perspective the LHWP’s discourse of legitimisation. One cannot simply regard such a substantial project organised at an interstate level as one in which there is a complete absence of any clashes between or inside the two countries implementing it. Tensions between different social groups and territories can be amplified or provoked by the newly established exchanges of materials, or differentially impacted by the transfer in terms of side-effects or advantages between and inside territories. For instance, a comparison between the Johannesburg metropolitan scale as a recipient and the Lesotho highlands scale as a provider shows that the LHWP has had very diverse social, economic and political results. The Lesotho highlands have sustained most of the negative outcomes from the construction of the LHWP infrastructures and the massive withdrawal of its water. So it would be very reductive not to try and push the analysis further, beyond a representation of this project as a mere “win-win” cooperation between the two states, as it is commonly promoted in the official statements and reports from both countries and the body in charge of implementing the project, the Lesotho Highlands Development Authority (LHDA). This is why we intend to analyse the relations with the LHWP in terms of the conflicts or alliances between territorialities that have been reordered as a result of the implementation of this project. Such an analysis will allow a consideration of the territories’ and territorialities’ differentiating processes on both sides of the transfer with an emphasis on their hydropolitical dimensions. The interplays between hydropolitics and territorialities in Southern Africa are the main issue of this paper.

5After a theoretical development of the concepts of scales and territorialities as spatial practices framing forms, a description of the hydropolitics in Southern Africa will help to contextualise the LHWP. Finally, an analysis of the regional territorialities will serve to demonstrate how the complex regional hydropolitics have contributed to amplifying some of the differentiating process between metropolitan and mountain scales by putting in place an extractive and exportation industry of hydraulic resources through the LHWP.

Scales and territorialities

6In the light of the significant unevenness of infra- and transnational relations between Southern Africa’s territories in terms of local needs and availability of water or of population density and wealth, the region’s complexity requires a scalar geographic approach, even more so in the present case of transboundary hydraulic infrastructures that modify these relations.

7There is consensus in geography about the inherited contingency of scales as both the product and the result of social and spatial processes (Delaney and Leitner, 1997). Moore supports a discursive approach to scales as a “representational trope” or “discursive framing” (2008). According to him, scales in geography have to be considered relevant only as categories of practices and not as analytical ones. From this perspective, it is rather the performativity of the scalar tropes that should be taken into account (Kaiser and Nikiforova, 2008). Scales and territorialities are formed by and consist of the social actors’ discursive and material practices, and as such they are entirely contingent and continuously being renegotiated (Mormont, 2006). This kind of approach pertains to an analysis of social processes in terms of framing. In the same analytical framework, scalar hierarchies as “scalar narratives” frame social processes and determine what is to be considered relevant or visible.

The scale is a scalar level of action and is legitimised by a specific power that was able to build its legitimacy on this level and consequently conducts its actions of spatial organisation by and for this scalar level. Naturalised by those actions, this scale becomes legitimate and organises a space: It creates a territory. (Arnauld de Sartre, al., 2012: §5, our translation)

8This quotation puts forward the way in which a scale is formed in a territory. It could be even more accurately tied to the notion of territorialities. As such it is necessary to consider scales through spatial practices that work as processes of the co-production of territorialities.

9Agnew (1994) exposes how the hegemonic nation-state form has instituted a “territorial trap”, in other words a specifically scalar structural frame. The transnational process of universalisation of the nation-state form can be considered a paradigmatic case of the historical construction of a scalar frame. However, the nation-state scale has the specificity to claim a legitimate monopoly on the configurations of territories, places, networks and relations between scales. In other words, it claims the legitimate definition of any and all forms of territorialities that constitute or traverse its territory. As such, the nationalist processes should be apprehended as “rescaling” projects amidst other forms of territorialities.

Scalar politics and hydropolitics

10As political ecology approaches have sought to show (Kaika, 2003, Heynen & al., 2006, Linton, 2010, Swyngedouw, 2010), water is a crucial political element a hybrid that is at once historically constructed from social and natural matter and produced by social, economic and political processes in the past and the present, while inevitably related to hydrology. Given water’s vital significance, one cannot rule out political arbitration for its allocation in relation to access conditions on a given territory. A fortiori, when such a resource is harnessed through a hydraulic infrastructure, it implies (perhaps even more crucially) political arbitrations for the spatial allocation of its benefits and the negative socio-natural outputs it necessarily causes. Therefore resource allocation forms inevitably raise major political stakes in term of establishing or reproducing asymmetrical territorial relations.

11In this respect, mainstream transboundary water studies are not very well-equipped to analyse the vast array of scales impacted by this kind of a massive infrastructural project. Those studies usually fail to grasp anything outside the thin strata of intergovernmental relations, with the notion of “water wars” always on the horizon. From a geographical standpoint, it would be reductive to settle for such an analysis that endorses homogenised national territories and conceals the infra- and transnational dynamics (Furlong, 2006:445).

12Regarding those mainstream studies, there have been critical approaches that utilise inputs, for instance from the neo-Gramscian critical theories of international relations. Other critical approaches, designated as hydropolitics, have also incorporated some inputs from political ecology and political geography (Molle, 2008, Mollinga, 2008, Lebel, 2006, Zeitoun and Warner, 2006). Thus, certain analytical frameworks try to show that a cooperative situation between two state bodies implementing a transboundary project does not necessarily imply the lack of any infra- or transnational conflicts. The Southern African hydraulic scale will serve as an illustration of this kind of difficulty.

13I carried out field research during the spring of 2011 (March–April) in Lesotho. It gave me the opportunity to conduct interviews with Lesotho’s government officials, executive administrators at the LHDA and members of local NGOs involved in environmental justice issues in the country. I was given the opportunity to go into the affected area in the highlands and observe the LHWP facilities first-hand. In addition to these field investigations, considerable institutional and scholarly literature about the LHWP complemented the materials needed for my research.

14This study of the LHWP project is intended to show how hydropolitical processes’ remaking of economic, political and resource relations between different (metropolitan, rural and mountain area) scales spatially reproduces social differentiation. The study of the LHWP will also allow the way in which territorialities are co-producing themselves to develop by articulating within this scalar frame the social actors’ games of rivalry or, on the contrary, the coalition seeking to get or keep a firm grip on the hydropolitical stakes.

The hydropolitical space in South Africa

15After this short look at the theory, it is necessary to deploy the national and regional trajectories of Southern Africa in order to contextualise the LHWP and the scalar hydropolitics processes it has created. Hence, it is essential to look into the political, material and natural processes that produce the scales, and how the scales help to territorialise the hydraulic policies related to the LHWP.

  • 5 (Grundling et al., 2015: 1, Gupta et al.: 2008, Nüsser, 2003: 24).

16South Africa is a very peculiar territory with regard to the geography of its water availability and needs. Gauteng province is characterised by a semi-arid climate, yet it is the most densely populated territory in the country and also the one in which most of the national wealth is concentrated: Some 20% of the inhabitants lives in the PretoriaJohannesburg metropolis, which produces 40% of the GDP and 80% of the country’s extractive outputs. Around 16% of the economic activities of sub-Saharan Africa are situated there. Lesotho, and especially the Maloti-Drakensberg mountain range that covers most of the country’s national territory, is quite another story. They benefit from the strongest rain average of the whole sub-continent (Lesotho is the source of the biggest rivers in Southern Africa), even though it has a small population with a low density and is mostly not industrialised.5

The Lesotho Highlands Water Project’s hydraulic transfers between the Vaal and the Upper Orange-Senqu catchment areas.

The Lesotho Highlands Water Project’s hydraulic transfers between the Vaal and the Upper Orange-Senqu catchment areas.
  • 6 The history of the region has also resulted in a situation whereby Southern Africa’s geography of (...)

17The geography of water in Southern Africa is defined by a radical discrepancy between its spatial distribution and the places where it is needed most. This is why water distribution is a very sensitive topic across the region, perhaps even more critically for South Africa. In this respect, a historical account of the regional hydropolitics is essential in order to denaturalise the origins of this discrepancy and inscribe it into its historical trajectory.6

  • 7 From a dusty, ad hoc settlement developed around mining in Johannesburg's population had grown to (...)

18This discrepancy between the water availability space and the most populated areas is explained by the fact that the prevailing settlement dynamics in the country have for the most part not been shaped by water availability at the local scale. Historically it finds its explanation in the extractive industries’ workforce requirements, which determined the location of Johannesburg itself. During the colonial area, these were important settlement factors and have continued to this day.7

  • 8 In southern Africa, [...] government officials exercise their political and technological power to (...)
  • 9 “Ce ‘complexe hydropolitique’ constitue un espace d’ampleur subcontinentale, où la plupart des bass (...)

19Swatuk (2010, 2007) and Blanchon (2003, 2008, 2009) conceive the high hydraulic policies of the apartheid regime to have been explicitly designed as ways and means to push forward its racist ideological agenda and to have been successful in doing so. Blanchon even speaks of a hydraulic mission that the regime appears to have pursued to protect rural Afrikaners’ way of life because its symbolic value formed part of the allegedly authentic white South African identity.8 Beyond this agenda, by systematically taming the water resources, the very abundant hydraulic projects implemented during this era have irreversibly altered the hydraulic landscape of the entire sub-continent. The overall outcome of this gigantism is the artificialisation of most of the water catchment areas across the country and even in Southern Africa (Blanchon, ibidem).9

20In the post-apartheid era, a deep shift occurred in the way South African considers its regional neighbours in favour of a more cooperative stance. It is quite a radical shift from the apartheid regimes securitisation policy with regard to its regional relations. But the significance of this change should not be exaggerated, considering the undisturbed regional supremacy of South Africa in terms of its political and economic regional weight. It has great leverage to defend its agenda against that of its neighbours. In any case, institutional, material and social-spatial inertia relating to the hydraulic and territorial policies of the past makes any attempt to alter the prevailing spatial configuration very complex (Giraut, Antheaume, 2005; Blanchon, ibidem, Turton, 2005).

  • 10 The balancing of interests in demographic and demographic terms, which implies an evaluation of the (...)
  • 11 “La direction anthropique de l’eau (l’argent et le pouvoir), l’emporte de plus en plus sur la direc (...)
  • 12 In Johannesburg where the water is being received, it is mainly directed to already well-served hi (...)

21The Southern Africa Development Community (SADC), a regional alliance, defines water as a central instrument for prosperity on the sub-continent. Hence the claimed aim of the regional hydraulic policies is to contribute to national and regional development. Accordingly the territories that are perceived to contribute more in terms of value creation and number of inhabitants, in other words the urban and industrial scales, are prioritised.10 In the context of neoliberalisation in the post-apartheid era, notably the privatisation of the water sector, the aforementioned artificialisation11 of most of the water streams in South Africa is translated into competition between infra-state and transnational scales for access to water, which contributes to exacerbating the existing socially asymmetrical spatial pattern across the sub-continent, especially throughout South Africa. The hydraulic policies from the new era actively lead to a reproduction of the very heterogeneous scalar structuration of the national territory. The implementation of the LHWP has had the paradoxical effect of accentuating the already uneven water access in the Johannesburg metropolis despite a significant increase in the absolute value of the volume of available water. 12

22This allows an understanding of how the main hydraulic projects in Southern Africa have exacerbated the social and spatial heterogeneity in the region by reproducing the existing geographical, social and economic marginalisation of certain social groups and territories and the concentration of resources in favour of other social groups and territories.

The LHWP – the invisible tragedies of the highlands in the win-win narrative

23The LHWP has peculiar significance in South Africa’s state trajectory. Simultaneously, it falls within the “high hydraulic” policies of the 1960s and 1970s and is emblematic of a take-over inside the apartheid regime during its final years in favour of the economic elites: outsiders from the ultra-nationalist core of the regime taking a stance against the securitised and racist Prime Minister P.W. Botha’s agenda. This evolution among the elites induced a strategic shift in the desecuritisation of the regional hydraulic policies (Davidsen, 2006). As a result, the LHWP was conceived to provide for the water needs of the Johannesburg metropolis, an urban and industrialised space, and no longer for rural Afrikaners, as their predecessors had done. The implementation of the LHWP is also linked to alarmist water needs projections for this metropolis made around the same period (Swatuk, 2010).

  • 13 The situation of interbasin water transfer schemes is more complex, since a transfer implies an in (...)
  • 14 [The LHWP] depends on the central government, for the good of the whole nation! [...] it’s a natio (...)

24The LHWP has reshaped relations between scales and between territorialities in Southern Africa.13 For instance, its infrastructures have established connections between local scales that had previously not been in direct relation and in so doing have put them in a state of co-dependency. For the requirements of our analysis of the hydropolitics processes, two kinds of actors are taken into consideration: the political and economic elites from both countries and the marginalised social groups. Those different actors are politically involved in the implementation of this project on behalf of their own territoriality. Nevertheless, these territories have been modified in the process and reproduced, in particular in their reciprocal relations. The dynamics are illustrated by the different stances of the main actors in relation to the LHWP: the Basotho communities of the Lesotho Highlands have transnational claims, the Lesotho government has developmental nationalism,14 and South Africa has pseudo-regionalist developmentalism.

  • 15 At the same time, however, Lesotho is an important pool of labour for South Africa’s extractive sec (...)

25For a long time, the development policy apparatus has characterised Lesotho in general and the Highlands in particular as underdeveloped and peripheral territories stuck in tradition (Ferguson, 1990). The country’s only recognised wealth is its natural resources.15 This view established the premise for the plan to move Lesotho’s economy towards an extractive sector. The LHWP was conceived from the very beginning as a means to promote this transition by offering economic opportunities from which both Lesotho and South Africa (as partners) would mutually benefit. It was conceived as a win-win project: taking water from the one side, which does not have the utility for it (in terms of value creation), in order to solve the water needs of the other (World Bank, 1998).

  • 16 This appropriation of a resource in return to royalties paid out to Lesotho can be conceived as a c (...)
  • 17 Reference: fiscal year 201516 (GoL, 2015).

26South Africa paid royalties in return for the water to the government of Lesotho, which was supposed to contribute to the development of the Lesotho.16 These royalties now amount to R780 million (about 50 million euros) per year,17 equivalent to about 5% of Lesotho’s state income outside of taxes (GoL, 2015). The 1986 treaty dispositions provided, among other things, for two parts: firstly, a scheme for direct compensation to the populations directly affected by the construction of the LHWP facilities (loss of villages and arable lands), and secondly, a development part designed in the form of a development fund. This fund had a double mandate to help balance the finances of the Lesotho government and to assist in the development of the regions most affected by the construction of the new infrastructure. Most of the second part was conceived to be directed towards the highlands. In fact, the development fund was massively diverted from its initial purposes in order to feed clientelist networks of Lesotho’s members of parliament. The misappropriation and corruption cases tied to the fund were so obvious and systematic that the World Bank, following an external audit, had stated that its dissolution would be a sine qua non condition for the launch of the next phase (1B) of the project (Matlosa, 1998). The dropping of this developmentalist part of the LHWP, while being an essential element in its justificatory apparatus, did not lead to the World Bank withdrawing from the project.

  • 18 The total project-affected population for the first two phases (IA and 1B) was estimated at 15,800(...)
  • 19 Phase 1A of construction of the roads required access to the infrastructures construction sites, f (...)

27The construction of dams and other LHWP infrastructures has had a direct impact on more than 15,000 people.18 Entire communities of the highlands were forced to abandon their lands and their villages. Their members were relocated and their dwellings rebuilt. During the first phase of the project,19 the resettled peoples were financially compensated on an individual basis for the loss of their arable lands. However, these compensations raised other difficulties, such as the fact that they are secured for a maximum period of 50 years and are not transferrable from one generation to the next. Besides, if the transport infrastructures for the LHWP have brought the highlands inhabitants “closer” to the lowlands and the capital, Maseru, and made public and commercial services more accessible, the damsreservoirs have created gaps inside the territories and separated communities that had previously been in close proximity to each other on either side of the waterways (Manwa, 2014).

28In subsequent phases of the project, the LHDA has tried to improve its communication with the highlands populations over issues relating to the project’s implementation, to go over the compensations policy once again and to establish participative processes. It has clearly emerged that the affected populations favour the replacement of the lands lost to LHWP facilities with other arable lands. However, very little arable land is available (only 11%) (Hitchcock, 2012: 322), notably as a result of altitude. Moreover, because of rural-urban migration from the highlands due to the lack of cultivable lands (an issue exacerbated by the LHWP), most of the population is already concentrated in the lowlands, the most fertile region in the country. It restricts the available relocation possibilities in these plains for the affected people looking to re-establish an agrarian way of life (Scudder, 2006). The issue of resettlement proved to be very complex, since the only available lands were to be found in the South African lowlands – in the contiguous territories on the other side of the border with Lesotho. In the context of the participative policies, many of the highlands communities on the verge of being ousted were interested in this solution – particularly because, as we shall see later on, the Basotho do not consider this space to be foreign territory but rather “conquered territories” illegitimately annexed from the Basotho territory by South Africa (Coplan, 2001).

As soon as it was taken up by the people as a serious possibility, some senior LHDA officials felt the reins of control slipping from their grasp. (Devitt & Hitchcock, 2010:82)

29The LHDA officials’ rejection of this possibility is very important. Indeed, such an inflexible response from these actors who felt their power threatened illustrates the importance of the project’s territorialist dimension over the possibility of resettlement. Resettlement means a geographic and social breaking up of the highlands’ populations. The dismantlement of their communitarian way of life was accentuated even further by the principle of more substantial compensations for those who accepted to be relocated farther from their place of origin (Devitt & Hitchcock, 2010:87, Matlosa, 1998). It also resulted in weakening their political claim of exteriority from the national realm. As such it could be understood as an attempt to “repatriate” these communities under the control of Maseru (Devitt & Hitchcock, 2010).

  • 20 By contrasting water-rich and economic poor Lesotho with “water-poor” and economic rich South Af (...)

30The coalition built around the LHWP (the Lesotho and South African governments, South African economic elites and international financial institutions) claims the project was intended among other things to connect the highlands to the modern world and the global market from which they had supposedly been excluded as passive spectators. With this stigmatisation, the coalition is denying the fact that Basotho scalar practices exceed Lesotho’s national territory and for that matter, the regulatory capacities of the Lesotho government – and are actually fully involved in the world. By justifying this repatriation dimension, this win-win discourse20 allows the linking of the interested parties via the LHWP. It aligns their interests by arguing for the common ground between both scalar framings: South Africa’s desire to concentrate hydraulic resources in favour of the Johannesburg metropolitan area and Lesotho’s territorialist purposes.

From a state scale perspective, the LHWP brings much needed revenues and electricity to economically stressed Lesotho, supplies equally needed water for South Africa’s industrial heartland in Gauteng, and is an exemplar of interstate water cooperation and dispute settlement. Constraining analysis to the state scale, however, obscures the uneven distribution of cost incurred and benefits accrued from the project. It also obscures the complex realities of conflict and cooperation by restricting their domain to interstate relations. (Furlong, 2006:452).

  • 21 The ordinary Basotho citizens are not gaining any benefits from the current relations between the (...)

31According to Furlong, if there is a collective that could identify itself with the win-win hegemonic concept, it is a transnational elite that certainly benefits from this project. This hypothesis makes it possible to move beyond the consensual representation of the LHWP as a shared and mutually beneficial scheme. It can be conceived as a water-appropriation device of a looting economy. This point fits with the prevailing discourse in Lesotho, especially in the highlands, that regards the LHWP as a catastrophic project with tragic consequences for the local populations: a project that has not kept its promises about the work and development that the construction of its infrastructure was supposed to bring. From this viewpoint, the LHWP could be painted as a project essentially imposed from the outside on Lesotho in order to capture the precious water from the highlands, with only the well-being of South Africa in mind.21

  • 22 The World Bank uses the failed states rhetoric symptomatic of international financial institution (...)

32Scudder (2006) argues that the compensations and development policies funding plan were insufficiently formalised in the LHWP treaty. The reason was a reluctance on the part of the South African representatives and the Lesotho representatives’ lack of engagement on the issue. However, this indecisiveness behind the treaty’s dispositions should be blamed for the compensations policies failures, as it left the door open to South African interference in the developmental dimensions of the project. Therefore, the blame for the issues arising from those policies should not be laid too quickly at the door of the LHDA or the Lesotho government’s institutional weaknesses, as the World Bank seems to allege (World Bank, 1998).22 A factor that could explain these difficulties is the reluctance of the signatory states to commit financially to the development components of the LHWP. This reluctance reveals the lack of interest of both the Lesotho and South African negotiators in the politically and symbolically marginalised populations of the highlands.

Results: controversies and conflicts of territorialities

33The main hypothesis of this study of the LHWP is that water has to be conceived of as a stake and a means of production and/or reproduction of scalar practices, namely of territorialities. Thus the LHWP is to be analysed as a material and discursive assemblage that brings together three distinctive types of territorialities. Firstly, the Lesotho nationalist territoriality seeks to monopolise any social and spatial processes in favour of the national scale. Secondly, the Basotho communities claim autonomous and transnational territoriality as the proper scale of their daily practices against the interference of Lesotho and South Africa and expect their transboundary practices to be unencumbered by borders and the bureaucracy they entail. Finally, the territoriality borne by the South African political and economic elites consists of a vast process of natural and economic resource concentration geared toward the “thirsty” Johannesburg metropolitan area to the detriment of the allegedly less productive scales (townships, rural areas, mountains, etc.).

  • 23 The Lesotho government would not have had the technical or financial means to extend its own bureau (...)
  • 24 An over-coding process of the highlands territory already raised by James Ferguson in his study of (...)

34It appears that the Lesotho government, via the LHWP, is attempting to repatriate the highlands communities, in other words to impose the nation-state scale on these populations despite their own territorial allegiances. In order to undertake this repatriation project, a stigmatising discourse was required, and the depiction of the highlands as traditional and outside the modern world was a way to perform this marginalisation. It also converged with the economism that has defined the LHWP from the outset. It justified a miserabilism regarding the highlands, notably by considering their usage of water to be unproductive because the Basotho were not valorising it to a significantly economical extent. This representation adds to a long-time blindness concerning Lesotho (Ferguson, 1990) and especially its highlands. It legitimised this massive intervention inside its territory in the name of modernisation. While the spatial practices of these populations are quite modern, they are problematised by the state because they largely slip out its control (except for the small interference created by bureaucratic border obstacles) (Coplan, 2001). Thus, this relative deterritorialisation, which relegates them to being marginalised territories requiring repatriation, was intended to justify forcing the jurisdiction of the state onto the highlands in the name of developing and modernising them. This nationalist scalar attempt was allegedly pursued for the common good. But the LHWP has incidentally performed a reterritorialisation inside the national state territory by making the redeployment of the state apparatus a financial and material possibility. Until then, state facilities were mostly concentrated in the country’s lowlands. The extension of the state presence in the highlands had been impossible before because of the weak means of the state and because this territory had no roads.23 It seems that this had been one of the Lesotho government’s main aims with regard to the LHWP, much more so than any attempt to break up an alleged traditionalist and autarkic territory. And it has been perceived as such by the Basotho populations (Thamae and Pottinger, 2006).24

35The Basotho claim their territory on cultural grounds: a majority of Basotho speakers inhabit this territory where the Sesotho culture prevails. They also claim it on a historical basis. Its South African side was conquered by the Boers during the Boer Wars and subsequently annexed by South Africa, which is why they call it “conquered territories”. This transnational territory in which they carry out their daily practices spreads to South Africa, far wider than the Lesotho national territory itself. At the centre of their territorial practices, in terms of work, access to commodities and services and even of children’s schooling, the border with South Africa is better described as a passage way than a boundary. In relation to this scale, the highlands are far from marginal: On the contrary, it constitutes the heart of this territory claimed by the Basotho (Coplan, 2001).

  • 25 According to these critical stances and many actors interviewed on site, the LHWP would still be to (...)

36The latter have experienced the LHWP as a major intrusion on their territory, particularly because it was designed by two authorities perceived as foreign. Furthermore, it is conceived as a violation, given that the two signatory parties of the LHWP treaty (South Africa’s apartheid regime and Lesotho’s military junta) are now viewed as illegitimate.25 It has been perceived as a means for the state, from its base in the capital, Maseru, to have direct influence on the hinterland that, until then, had been relatively autonomous. This agenda has provoked reactions in the form of daily violence against the convoys for the construction of the LHWP’s infrastructures in the highlands, including stones being thrown, insults hurled and other altercations (Scudder, 2006).

37The LHWP’s facilities have also been targeted for their symbolic value during major political mobilisations in Lesotho. These uprisings have systematically been violently repressed. The culmination occurred in 1998, when during an uprising throughout the country, opponents of the ruling government in Maseru invaded the Katse Dam’s facilities. South Africa intervened militarily in Maseru and at the Katse Dam, under legal jurisdiction from SADC, to protect the facilities of the LHWP, which resulted in the deaths of more than 100 people and heavy damage in Maseru.


38The implementation of the LHWP has led to a reassemblage of different territorialities, allowing South African to concentrate its resources in favour of the metropolitan scale and the Lesotho government to (re-)establish its national territory as the scale of practices for the Basotho populations of the highlands. These two territorialities have tended to reinforce each other at the same time as they followed a trajectory in conflict with the Basotho territoriality. It has ignited internal clashes in Lesotho and a deepening of social and spatial inequalities in Maseru and the highlands. Thus it has fuelled political instability, which has manifested as major uprisings. South Africa’s military intervention in Lesotho in 1998 is significant, as it demonstrates the convergence of South Africa’s scalar project and the one from the side of Lesotho. This event evinced how the interest of South Africa to secure the LHWP facilities in order to ensure Johannesburg’s water supply coincided with the interest of the Lesotho government to have the threatening political movement repressed. The territorialities of South Africa and Lesotho, as much as that of the Basotho, have been reworked by the emergence of this new transnational assemblage in their reciprocal relationships and in relation to this new scale. The LHWP not only amplified the differentiation between metropolitan and mountain scales but also exacerbated the social and political tensions resulting from the ever more blatant inequalities characterising the post-apartheid era.

Haut de page


Agnew, J. (1994).– The territorial trap: the geographical assumptions of international relations theory. Review of international political economy, 1(1), 53-80.

Arnauld de Sartre, X., & Gagnol, L. (2012) .– “Les échelles des territorialités”, Géographie et cultures no. 81, pp. 5-16. Laboratoire Espaces, Nature et Culture (ENEC).

Antheaume, B., & Giraut, F. (Eds.). (2005).– Le territoire est mort, vive les territoires ! : une (re) fabrication au nom du développement. IRD éditions.

World Bank (1998) .–, Request for Inspection Lesotho/South Africa: Phase 1B of Lesotho Highlands Water Project, (visited Januar 15th, 2015).

Blanchon D. (2009).– L’espace hydraulique sud-africain : le partage des eaux. Karthala Editions.

Blanchon D. (2008).– Le complexe hydropolitique sud-africain : connexions, déconnexions, reconnexions ?, Flux 2008/1, n° 71, p. 61-72.

Blanchon, D. (2003). "La nouvelle politique de l’eau en Afrique du Sud : vers une gestion environnementale des ressources ?" L’Espace géographique, 2003/1 (tome 32), pp. 21-30.

Coplan, D. B. (2001). "A river runs through it: The meaning of the Lesothofree state border". African Affairs, 100(398), 81-116.

Davidsen (2006).– The Making and Unmaking of the Politics of Exceptionality, Studying Processes of Securitisation and Desecuritisation in the Orange and Okavango River Basins, Phd,

Delaney, D., & Leitner, H. (1997).– The political construction of scale. Political geography16(2), 93-97.

Devitt, P., & Hitchcock, R. K. (2010).– Who drives resettlement? The case of Lesotho’s Mohale Dam. African Study Monographs, 31(2), 57-106.

Ekers and Loftus (2008).– The power of water: developing dialogues between Foucault and Gramsci, Environment and Planning D: Society and Space, volume 26, pages 698-718.

Ferguson, J. (1990).– The anti-politics machine: development, depoliticization, and bureaucratic power in Lesotho. CUP Archive.

Furlong, Kathryn. (2006).– Hidden theories, troubled waters: International relations, the ‘territorial trap’, and the Southern African Development Community’s transboundary waters. Political Geography, 25.4: 438-458.

Government of Lesotho, Budget Speech to Parliament for the 2015/2016 Fiscal Year. (visited August 8th, 2015)

Grundling, P.-L., Linström, A., Fokkema, W. & Grootjans, A.P. (2015).– Mires in the Maluti Mountains of Lesotho. Mires and Peat 15: Art. 9. (Online:

Gupta, J., & van der Zaag, P. (2008).– Interbasin water transfers and integrated water resources management: Where engineering, science and politics interlock. Physics and Chemistry of the Earth, Parts A/B/C, 33(1), 28-40.

Heynen, N., Kaika, M., & Swyngedouw, E.– Urban political ecology. In Heynen, N. C., Kaika, M., & Swyngedouw, E. (Eds.). (2006). In the nature of cities: urban political .ecology and the politics of urban metabolism (Vol. 3). Taylor & Francis, 1-20.

Hitchcock, R. K. (2012).– The Lesotho highlands water project: Water, culture, and environmental change. In Water, Cultural Diversity, and Global Environmental Change (pp. 319-338). Springer Netherlands.

Horta, K. (1995).– The Mountain Kingdom’s White Oil. The Lesotho Highlands Water Project. The Ecologist, 25(6): 227 – 231.

Hoover, R. (2001).– Pipe Dreams: The World Bank’s Failed Efforts to Restore Lives and Livelihoods of Dam-Affected People in Lesotho. International Rivers Network, Berkeley.

Kaika, M. (2003).– Constructing scarcity and sensationalising water politics: 170 days that shook Athens. Antipode, 35(5), 919-954.

Kaiser R, Nikiforova E. (2008).– The performativity of scale: the social construction of scale effects in Narva, Estonia Environment and Planning D: Society and Space 26(3) 537 – 562.

Kapa, M. Anthony. (2007).– Lesotho’s Policy Towards South Africa: Policy or Imperative? What has Changed? Africa Insight 37, no 1: 117.

Lebel, L., Garden, P., & Imamura, M. (2005).– The politics of scale, position, and place in the governance of water resources in the Mekong region. Ecology and society, 10(2), 18.

Linton, J. (2010).– What is water?: The history of a modern abstraction. UBC Press.

Manwa, H. (2014).– Impacts of Lesotho Highlands Water Project on Sustainable Livelihoods. Mediterranean Journal of Social Sciences, 5(15), 640.

Matlosa, K.– Changing Socio-Economic Setting of the Highlands Regions as a Result of the Lesotho Highlands Water Project. Transformation no 37 (1998).

Molle, F. (2008) ).– Nirvana concepts, narratives and policy models: Insights from the water sector. Water Alternatives, 1(1), 131-156.

Mollinga, P. P. (2008).– Water, politics and development: Framing a political sociology of water resources management. Water alternatives, 1(1), 7-23.

Moore, A. (2008).– Rethinking scale as a geographical category: from analysis to practice. Progress in human geography32(2), 203-225.

Mormont, M. (2006). Conflit et territorialisation. Géographie, économie, société, 8(3), 299-318.

Nüsser, M. (2003) Political ecology of large dams: a critical review. Petermanns Geographische

Mitteilungen, 147(1), 20–27.

Nüsser, M. (2002): Lesotho: Water policy and management: trade-offs between mountains and downstream areas. In: Mountain Agenda (Ed.): Sustainable Development in Mountain Areas: The Need for Adequate Policies and Instruments. Prepared for the World Summit on Sustainable Development in Johannesburg. Bern, 44-45.

Scudder, T.– Assessing the Impacts of the LHWP on Resettled Households and other Affected People 1986 – 2005 in Thamae, M. L., & Pottinger, L. (2006). On the wrong side of development: Lessons learned from the Lesotho Highlands Water Project. Transformation Resource Centre.

Swatuk, L. A. (2015).– Water conflict and cooperation in Southern Africa. Wiley Interdisciplinary Reviews: Water, 2(3), 215-230.

Swatuk, L.A .(2010).– The state and water resources development through the lens of history: A South African case study, Water Alternatives 3(3):521-536.

Swyngedouw, E. (2010).– Place, nature and the question of scale: interrogating the production of nature. Berlin-Brandenburgische Akademie der Wissenschaften.

Thamae, M.L., et L. Pottinger (eds.). (2006).– On the wrong side of development: Lessons learned

from the Lesotho Highlands Water Project. Maseru : Transformation Resource Centre.

Turton, A. R. (2005).– Water as a source of Conflict or Cooperation: the Case of South Africa and its transboundary Rivers. CSIR Report, 2005-002.

Wittfogel, K. A. (1981).– Oriental despotism: A comparative study of total power (Vol. 3). New York: Vintage books.

World Bank (1996).– World Bank Lending for Large Dams: A Preliminary Review of Impacts. Operations Evaluation Department, Précis 125. Washington, D. C.

Zeitoun, M., & Warner, J. (2006).– Hydro-hegemony-a framework for analysis of trans-boundary water conflicts. Water policy, 8(5), 435-460.

Haut de page


1 Over 80% of the country lies more than 1,800 m above sea level (see the map).

2 [W]hether it will be thanks to later phases of the Lesotho Highlands Water Project or other similar projects implemented entirely on South African territory. In total, 2,200 Mm3 per year will be extracted from the river (Blanchon, 2003:29, our translation).

3 In the apartheid era, dominant actors justified harnessing water in terms of economic development, building of the modern Western state, and keeping white South Africa safe from communism and the black peril. The dominant actors of the post-apartheid era have altered this narrative: harnessing water for economic development, poverty alleviation and justice for all (Swatuk, 2010: 522).

4 [In the words of South African President Thabo Mbeki,] the economic destinies of our people and countries are inextricably linked together ... the Lesotho Highlands Water Project [has] great potential to bring long-term benefits that could be shared by both countries (Davidsen, 2006:123). The Treaty also involves hydropower generation for Lesotho on the basis of royalties from South Africa for the next 50 years, making water Lesotho’s White Gold, creating a win-win situation for both countries (Quoted in Davidsen, 2006: 50).

5 (Grundling et al., 2015: 1, Gupta et al.: 2008, Nüsser, 2003: 24).

6 The history of the region has also resulted in a situation whereby Southern Africa’s geography of water use is virtually the inverse of its geography of water availability. [...W]hile possessing only 10% of Southern Africa’s total water resources and one third of its population, South Africa is responsible for 80% of the region’s water consumption. (Furlong, 2006:447).

7 From a dusty, ad hoc settlement developed around mining in Johannesburg's population had grown to a quarter of a million by 1914 (Swatuk, 2010: 528).

8 In southern Africa, [...] government officials exercise their political and technological power to marshal water resources for use by white farmers, a process that reinforced the social and political hegemon of the state and of the white settler elite (Forrest, 2006:151).

9 “Ce ‘complexe hydropolitique’ constitue un espace d’ampleur subcontinentale, où la plupart des bassins versants sont interconnectés, où toute action sur l’un des hydrosystèmes a des conséquences sur l’ensemble du réseau et dans lequel la notion même de bassin-versant perd de sa pertinence.” (Blanchon, 2008:67).

10 The balancing of interests in demographic and demographic terms, which implies an evaluation of the respective contribution of each scale to the common good (conceived as prosperity for the many), may very well be consistent. However, the use of GDP to justify this arbitration – while this indicator is known for its homogenising disposition regarding economic national territories – tends to veil the very same effects those policies would have, such as amplifying social and economic disparities inside and between territories and accentuating the marginalisation of rural and mountain areas.

11 “La direction anthropique de l’eau (l’argent et le pouvoir), l’emporte de plus en plus sur la direction naturelle” (Blanchon, 2009: 117).

12 In Johannesburg where the water is being received, it is mainly directed to already well-served high-income areas. In fact, due to the project’s high cost combined with user-pay full-cost recovery policies for water services, the project has actually hindered water access for Johannesburg’s urban poor (Furlong, 2006: 453).

13 The situation of interbasin water transfer schemes is more complex, since a transfer implies an increase of the spatial scale at which water is managed. Connecting physically two river basins implies a discontinuity […] socially a new connection is established between communities that hitherto did not depend on each other, at least not in terms of water (Gupta, al, 2008: 37).

14 [The LHWP] depends on the central government, for the good of the whole nation! [...] it’s a national project for the benefit of the whole country. Extract from an interview conducted by the author with the executive manager of the chieftainship directorate at the Ministry of Local Government and Chieftainship Affairs, Maseru, Lesotho, 2011.

15 At the same time, however, Lesotho is an important pool of labour for South Africa’s extractive sector, and a significant part of Lesotho’s GDP comes from its migrant workers.

16 This appropriation of a resource in return to royalties paid out to Lesotho can be conceived as a commodification of the highlands water.

17 Reference: fiscal year 201516 (GoL, 2015).

18 The total project-affected population for the first two phases (IA and 1B) was estimated at 15,800 (World Bank, 1996:10).

19 Phase 1A of construction of the roads required access to the infrastructures construction sites, for the Katse Dam and the transfer tunnel through the Maloti-Drakensberg.

20 By contrasting water-rich and economic poor Lesotho with “water-poor” and economic rich South Africa and by further claiming that the LHWP will create a win-win situation for both countries which would both be losers otherwise puts an irresistible offer on the negotiating table (Davidsen, 2006:125).

21 The ordinary Basotho citizens are not gaining any benefits from the current relations between the two countries. If anything, they argue, Basotho are forced into conditions of untold poverty and destitution by the country’s nominal independence which benefits the political and bureaucratic elite. (Kapa, 2007: 130).

22 The World Bank uses the failed states rhetoric symptomatic of international financial institutions’ liberal dogma (Ferguson, 1990).

23 The Lesotho government would not have had the technical or financial means to extend its own bureaucratic apparatus. Neither would it have had the means to improve the roads and other infrastructures on the whole territory it claims without external resources.

24 An over-coding process of the highlands territory already raised by James Ferguson in his study of a rural development project implemented in the 1970s in the same Taba-Tsake region as the LHWP (Ferguson, 1990).

25 According to these critical stances and many actors interviewed on site, the LHWP would still be to a large extent the heir of the authoritarian regime that launched the LHWP.

Haut de page

Table des illustrations

Titre The Lesotho Highlands Water Project’s hydraulic transfers between the Vaal and the Upper Orange-Senqu catchment areas.
Fichier image/png, 222k
Haut de page

Pour citer cet article

Référence électronique

Yannick Rousselot, « Upstream Flows of Water: from the Lesotho Highlands to Metropolitan South Africa », Journal of Alpine Research | Revue de géographie alpine [En ligne], 103-3 | 2015, mis en ligne le 18 janvier 2016, consulté le 29 mars 2017. URL :

Haut de page


Yannick Rousselot

PhD student, Department of Geography and Environment, University of Geneva, Uni Carl Vogt, 66 Boulevard Carl-Vogt, CH-1211 Genève 4.

Haut de page

Droits d’auteur

Licence Creative Commons
La Revue de Géographie Alpine est mise à disposition selon les termes de la licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page