Development by Dispossession: Terra Nullius and the Social-Ecology of New Enclosures in Ethiopia


  • I thank the members of the New Enclosures Research Working Group at Cornell for comments on a draft of this article, and Charles Geisler in particular for conversations over the past two years that have allowed me to clarify the arguments of the essay. Contact Fouad Makki, Cornell University, Department of Development Sociology, 107 Academic Surge A, Ithaca, NY 14850, 607–255–6237, e-mail:


One of the alarming features of the intersection of the world economic crisis with the global food and energy crises has been the tidal wave of large-scale land acquisitions it has unleashed. By enclosing the village commons and extinguishing the customary rights of smallholders, these land grabs are accelerating trends toward large-scale industrial farming and tenure rearrangements favoring international agribusiness. This article situates these developments in the social space of Ethiopia and a specific historical context of transformations in state and property relations. By critically analyzing the designation of the commons as empty or underutilized spaces awaiting redemptive development, it highlights the ways in which the legitimizing claims of terra nullius efface the complex ecologies and distinctive spatial dynamics of social reproduction in the zones designated for enclosures.


The confluence of the contemporary world economic crisis with ongoing food and energy crises has generated a frenzy of large-scale land acquisitions and a series of social and political explosions across the Global South. The “tortilla riots” in Mexico that erupted in response to an 80 percent increase in the price of corn in 2007 were followed by mass protests in more than 30 countries in the spring of 2008 when food prices spiked by 50 percent (Roger 2010). From Egypt to Indonesia and from Colombia to Haiti, this particular constellation of popular upheaval was a vivid expression of a global food crisis driven by the increased demand for staples, the reorientation of the consumption patterns and food-security strategies of some of the “emerging economies” and oil-rich countries, and the search for new zones of valorization by footloose speculative capital after the meltdown of the banking and securities markets. The global land rush is in this sense a symptom of a synchronized crisis and the desire by some states to aggressively market farmlands as lucrative sites of investment for the production of export crops and agrofuels. Their cumulative upshot has been a pronounced acceleration of transnational trends toward the consolidation of a global corporate food system at the expense of smallholder farming (Akram-Lodhi, Borras, and Kay 2007; Araghi 2000).

The colossal process of enclosures these land grabs represent, and the dispossession and marginalization of smallholders they entail, is portrayed by the dominant discourses of national and multilateral institutions as the unavoidable prerequisite for implementing the productivity-enhancing technologies of the new green revolution. As Kofi Annan, former UN secretary general and later chair of the Alliance for a Green Revolution in Africa, tellingly put it, “we have arrived at the tipping point, and are now taking Africa's Green Revolution to scale” (AGRA 1985). A critical engagement with this emerging neo-Malthusian consensus, and the ideologically charged revival of formerly discredited ideas about the “tragedy of the commons,” requires a careful unpacking of these claims through a scrutiny of the social and ecological dynamics of the new enclosures. Thus far, while increasingly complex accounts of the global determinants and contours of the land grabs have emerged, these still remain to be situated in the unevenly integrated social formations that constitute the international political order. An analysis of the new enclosures in multiple relational scales might better illuminate the profound remaking of the socio-ecological universe of agrarian societies they represent.

In this article, I examine the historically conditioned and spatially differentiated dynamics of commercialization and enclosures in Ethiopia, set within this suggestive analytic framework. I argue that these contrasting but related processes are unfolding in a complex conjuncture of state formation and the reconstitution of older sociospatial hierarchies in new social forms. In the northern highland regions where historically peasants had direct access to land, the commercialization of smallholder farming is fostering what Sara Berry calls “accumulation without dispossession” through the formal subsumption of primary producers to capital (Berry 1993). By contrast, in the geographic margins of the state, where smallholders and indigenous communities are being uprooted and displaced through enclosures, the structural conditions for the real subsumption of labor to capital are emerging. The overall effect of these contrasting patterns of agrarian transformations is the asymmetrical integration of the mass of Ethiopia's farmers into a hierarchically structured global agroindustrial complex.

This argument is developed in three interrelated parts. The first part situates the contemporary land grabs within an international historical context of colonial and capitalist enclosures. The second part examines the particular social and spatial features of commercialization and enclosures in contemporary Ethiopia, while the third part critically probes the legitimizing claims of terra nullius in relation to the modes of social life and distinctive ecologies of the zones targeted for enclosures.

Enclosures, Colonialism, and the World Market

For classical political economists from Adam Smith to J. B. Say, predatory territorial expansion was the antonym of a commercial society and the exchange relations that typically characterized it. “Force and fraud,” declared Say, “were no substitute for industry,” and imperial sovereignty “does not compel a people to buy what they cannot pay for, or what is not suited to their custom; and when they are offered what is agreeable to them, they buy it without being conquered” (quoted in Jones 2004:143). These classical sentiments found an influential twentieth-century echo in the writings of Joseph Schumpeter, for whom late-nineteenth-century imperialism was an “atavistic” expression of precapitalist aristocratic residues (1951:65). Yet these idyllic representations of the genesis of capitalism and its expansionary dynamics effectively obscure the violence that has been a constitutive feature of the making of the modern world market (Federici 2004; Harvey 2003; Marx [1867] 1976). Over the past few centuries, predatory colonial and capitalist enclosures have been fundamental to the emergence and uneven consolidation of market imperatives as regulative mechanisms of social life. These enclosures entailed in the first place the uprooting of peasants from the soil and the institutionalization of a regime of absolute private property. But this primarily sociological content of primitive accumulation did not exhaust the meanings and significance of enclosures, which were never just about turning autonomous peasants into wage laborers. The separation of the direct producers from the land simultaneously separated the land from the producers and made possible their mutual reproduction in the form of what Karl Polanyi called “fictitious commodities” (1944:75–76). By dismantling the village commons and deracinating the commoner, enclosures disenchanted the historically sedimented cultural practices, collective memories, customary entitlements and modes of social reproduction that had once informed the particular relationship of communities to land and nature. The profound reconstitution of these socioecological relations has arguably been at the source of the metabolic rift that is a central feature of the contemporary global ecological crisis (Foster 2000; Marx [1867] 1976:637).

The advance of agrarian and industrial capitalism in England, where between 1793 and 1815 “more than 6.5 million acres of common fields were enclosed by acts of Parliament,” provides a paradigmatic illustration of the twined colonial-capitalist dimensions of primitive accumulation (McNally 1988:11). Examining the place of plantation slavery in England's industrialization, Robin Blackburn has suggested that the super profits generated by New World slavery served to augment enclosures within England so that the number of enclosure bills before Parliament closely followed trends in plantation profitability (1997:577). The new settler colonies, where an even more violent and extensive expropriation took place, likewise served as release valves for the resettlement of commoners evicted by enclosures and unabsorbed by the expansion of English industry. With the intensification of enclosures in England and elsewhere, there occurred what Edward Said has called “a new process of relocating England (and in France, France) within a much larger circle of the world map” (1994:83).1 And in this expanding social order, commoners in both metropole and colonies were invariably vilified as “a mischievous race of people” sustained by a commons that was maligned as a breeding-ground for “barbarians” (Thompson 1966:219).

The mid-twentieth-century collapse of the large territorial empires and the extension of an interstate system based on the norm of the nation-state did not erect any insuperable barriers to the structural extension of market relations. For the most part, processes of commodification continued to operate through the mechanisms of the world market without necessarily impinging on formal political sovereignty as such (Anderson 1974:17, 197; Wood 2003:5). In this new geopolitical context, it was possible “to command and exploit labor [and resources] located under the jurisdiction of another state” in a way that was unthinkable in previous social systems where the domains of the political and the economic were inextricably bound together (Rosenberg 1994:121, 129).

Within this global order of formally sovereign and equal states, inequalities of wealth and power were expressed in an ostensibly neutral language of development and progress that in many respects harkened back to Lockean ideas of improvement. Socioecological spaces once considered primitive or beyond the pale could henceforth be recast as undeveloped or terra nullius zones awaiting capitalist redemption and valorization. The development initiative that emerged in the postwar era provided the institutional and discursive framework for both facilitating and contesting the social and political processes associated with the extension of market relations (Cooper and Packard 1997). By the 1980s, the structural adjustments programs associated with the debt crisis decisively rolled back what little autonomous leverage postcolonial states once exercised in determining their development, trade, and fiscal policies, effectively reducing their role to maintaining macroeconomic stability through budgetary discipline, the reduction of the social budget, and the privatization of public assets. Subsidies to domestic agriculture were progressively reduced or dismantled as states were forced to renegotiate their onerous debts by opening their protectionist gates to what Farshad Araghi (2000) has called the great global enclosures of our times.

From Old to New Enclosures

The Bretton Woods institutions, and the World Bank in particular, are today at the forefront of a thinly disguised narrative of terra nullius that is deployed to designate “underutilized” or “unproductive” spaces as ideal for large-scale commercial development. Against an earlier consensus on the inverse relationship between farm size and productivity that had served as the “economic rationale for redistributive land reforms” (Dyer 1996:103), the bank has increasingly placed its institutional weight behind large-scale mechanized agriculture, albeit one in which foreign investors are regulated by a modest code of conduct (De Schutter 2011). A 2009 World Bank publication entitled Awakening Africa's Sleeping Giants posited the existence of a vast underused land reserve that could be “tapped to produce food, agricultural raw materials, and biofuels feedstocks, not only for Africa but also for other regions” (World Bank 2009:175). A year later, the bank released a companion report classifying countries according to the criteria of yield gaps, defined as the difference between the attained and possible productivity of land. It found these gaps to be especially large for sub-Saharan Africa where no country appeared to be realizing even 50 percent of its potential yield (Deininger and Byerlee 2011:182). Both reports concluded that: “in this region, low population densities and low mobility prevail, which suggests that agricultural intensification will require larger farm sizes” (Hall 2011:7).

One of the states identified by the World Bank for this form of redemptive development is Ethiopia. In 2004, the bank had prepared a confidential report on Ethiopia identifying “areas of growth potential, where increased public investment in specific geographic and development areas might make an optimal contribution to economic growth” (World Bank 2004:i). The study classified the country into four zones using 51 indicators. While the first and second zones covered the densely populated highland regions, the third and fourth zones were located in the more sparsely populated western and eastern lowlands, described as having a high potential for irrigated farming and commercial livestock production. In its essentials, the report represented a virtual blueprint for the spatially differentiated program of agricultural commercialization and enclosures. This is reflected today in the vigorous process of smallholder commercialization in the central highlands (zones 1 and 2) and the enclosure of the commons that is profoundly altering the communal and transhumant modes of life in the lowlands (zones 3 and 4).

This sociospatial patterning of agrarian relations is not a straightforward reflection of demographic and geographic determinants. It is also conditioned by the modalities of incorporation of the regional peasantries into twentieth-century processes of state formation. As the only major polity in Africa whose integration into the world market was not colonially mediated, Ethiopia's social and economic transformation differed substantially from the picture associated with traditional accounts of colonial capitalism elsewhere on the continent. At the end of the nineteenth century, European encroachments into the region provoked in reaction a sustained process of territorial expansion that more than doubled the Ethiopian imperial realm. In the process, numerous smaller polities were incorporated into the expanding tributary order and their inhabitants reduced to varying degrees of servile status. A complex, culturally coded sociospatial hierarchy consequently emerged: In the core Abyssinian regions, the mahel ager, the immediate producers were by and large free with secure kin-based rights to land known as rist. Superimposed on this form of communal tenure were gult or tributary rights in which the state, the Orthodox Church, and a class of lords appropriated anywhere from a fifth to a third of the peasant output (Rahmato 2008; Tamrat 1972). In the newly incorporated southern regions of plough cultivation, on the periphery of the old core, a social order akin to serfdom was instituted. On the extreme fringes of the imperial realm, the dar ager, pastoralist or lineage-based communities considered to be outside the norms of orderly rule were subjected to periodic predatory campaigns of pillaging and plunder (Donham and James 1986; Hassen 1994).

Despite some reforms in the post-1941 period that strengthened the monarchical center at the expense of regional lords, this articulated social and political structure remained largely intact until both were transformed by the 1974 revolution. A land reform nationalized all the land as citizenship rather than kinship became the formal basis of rights to land (Halliday and Molyneux 1981; Makki 2011a; Rahmato 1984). But by vesting all the land in the state, the reform went further than simply redistributing the great estates of the nobility. It also dispossessed local communities of their customary ownership of land and abolished the institutions around the village commons that had historically afforded peasants some degree of protection against predatory lords and rulers (Donham 1986; James et al. 2002; Rahmato 2008).

As a more centralized and coercive state power replaced the earlier decentralized form of rule and surplus extraction, the threat to tenure security was never far off. This initially took the form of the imposition of compulsory grain quotas whereby peasants were expected to sell a percentage of their output to state agricultural marketing boards at fixed prices (Clapham 1988). Following the great famine of 1984, which was used to remake Ethiopia's rural communities, the regime embarked on an even more ambitious project to reorder society and nature so as to make them more legible to the state (Scott 1998:247–52). At first, this entailed promoting state farms and producer cooperatives, but it later shifted to the forced resettlement of peasants primarily from the drought-stricken and politically volatile north to grid-like villages in the south. By the end of the 1980s, “nearly 40 percent of the country's rural population, numbering about 14 million peasant farmers, had forcibly been villagized” (Berisso 2002:117). But despite these draconian measures, the regime had little success in compelling peasants to increase production, and by the end of the decade, average production levels had declined to levels below those of the late 1970s (Clapham 1988). Growing peasant disaffection fueled various rurally based, armed ethnonationalist movements that in May 1991, in the context of the end of the Cold War and withdrawal of Soviet support, besieged and defeated the regime.

The Second Republic and Market Liberalization

The new rulers reconstituted the state into a federal republic organized into ethnically defined administrative regions, and indicated a desire for a strategic reorientation away from the dirigiste policies of the former regime. The development strategy that eventually crystallized was designated “Agricultural Development-Led Industrialization.” Its initial premise was that the surplus derived from the commercialization of smallholder farming, where labor productivity has traditionally been low, would fund an industrialization drive and provide the material basis for a “democratic developmentalist state” (Zenawi 2006). This orientation was understood to be consistent with the central tenet of the 1975 land reform, which was reaffirmed in the 1994 Constitution: “ownership of rural and urban land, as well as of all natural resources, is exclusively vested in the state and peoples of Ethiopia” (FDRE 1995, Article 40). Gradually, however, under pressure from newly emergent middle classes, Ethiopian diaspora capital, and international donors, the government embraced a more concerted reliance on market mechanisms and the development of agrarian capitalism (Jemma 2001; Kibret 1998). Paradoxically, the radical land reform of 1975 proved instrumental to this reorientation in two distinctive ways. By dispossessing the landed nobility and breaking the grip of the ancien régime, the revolution had abolished the social and political power of the dominant classes that had previously constrained the emergence of new forms of production and exchange. Simultaneously, however, by nationalizing all the land, the radical land reform had in effect dispossessed peasants of their customary access to land and endowed the state with extraordinary powers to determine its allocation and use. This crucial legacy of the first republic enabled the post-1991 second republic to pursue an accelerated program of commercialization and enclosures with the ostensible aim of increasing agricultural production, supplying the raw material needs of an emergent commercial-industrial sector, and securing the cheap wage food critical to regulating the price of labor. In this sense, the development strategy of the new rulers has sought to internalize within the national framework the international division of labor that was critical to industrialization in the West (Friedmann 2005; McMichael 2005).

The continued assertion of the centrality of the state in creating the necessary political, legal, and normative conditions for the development and consolidation of an agroexport industry represents a partial refusal of the more doctrinaire propositions of the neoliberal globalizers. At first, this involved resistance to the demands of various national forces and multilateral agencies for the complete privatization of land. But it was soon extended to questions around the privatization of various other state assets and specific fiscal and monetary policies that did not accord with the prevailing prescriptions of the international financial institutions. Despite these refusals, however, the regime's relationship with bilateral donors and international agencies has been nothing but cordial, and the latter have discreetly applauded the perceptible shift toward a reliance on market indicators in the formulation of economic and development policies. They have generously showered Ethiopia with grants and loans, making it one of the largest recipients of foreign aid in sub-Saharan Africa. This accommodation between the developmentalist state and the neoliberal orientation of the multilateral institutions can be understood in terms of what John Ruggie has called “embedded liberalism,” whereby economic “multilateralism” is predicated on “domestic intervention” (Ruggie 1982:393).

Commercialization and the Mahel Ager

The emerging social form of market dependence has not been uniform across Ethiopia. It has taken a differentiated pattern of smallholder commercialization in the highlands and large-scale enclosures in the lowlands. In the former, where high population densities, small farm sizes, and intensive forms of cultivation continue to characterize tenure relations, the program of smallholder commercialization has registered real, if uneven, productivity gains while confining large capital investments to enclaves that entail little if any displacement of smallholders (Lavers 2012; MOFED 2006:3). The existing investment enclaves amount to little over 4,000 hectares and are primarily dedicated to intensive floriculture or horticulture that require limited land but considerable investments of capital and labor (Meles and Helmsing 2010; MOARD 2010). By 2011, some 85 floriculture farms had been established, and figures for the period between 2001 and 2007 indicate that the value of exported flowers increased from US$300,000 to US$113 million (Joosten 2007; Meles and Helmsing 2010). The program of smallholder commercialization occupies a critical place in the government's overall development agenda. The 2011–15 Growth and Transformation Plan envisages a tripling of the number of farmers receiving extension services and a doubling of the production of staple crops from 18.08 to 39.5 million metric tons. It anticipates a minimum annual growth rate of 8.1 percent for this agricultural sector and ambitiously proclaims its aim to ensure the country will be food self-sufficient by 2015 and a middle-income country by 2025 (MOFA 2013).

But while smallholder commercialization is in principle consistent with the redistributive land reform of 1975, the federal republic has been diluting some of its key provisions in the name of devolving the administration of land to the constituent regional governments. This has allowed for the uneven spread of informal land markets and the growing phenomenon of landlessness, which is compounding social inequalities related to the ever-decreasing size of farm plots in the face of demographic pressures.2 Titling and certification of individual use rights has been widely implemented and peasants can now lease up to 50 percent of their plot for limited periods of time (Holden and Yohannes 2001). As a result, in some districts of the Tigray and Amhara regions, a fifth to a quarter of rural households are engaged in contract farming (Gebreselassie 2006).

This land reform from above, which has been energetically promoted by the World Bank and the UN-HABITAT's Global Land Tool Network (2008), has started to foster a perceptible change that is making nonmarket access to land increasingly difficult (see Holden, Deininger, and Ghebru 2007). In conditions where 55 percent of smallholders subsist with less than a hectare of land, patriarchal gender relations and the further subdivision of farm plots have combined to increasingly exclude women and the young from access to land (Adal 2006; Holden and Tefera 2008; Tadesse 2003). Moreover, while a substantial percentage of smallholder production had historically been oriented to subsistence needs, a growing number of households are today compelled to enter into wider exchange networks in order to secure the reproduction needs of their households. Some 65 percent of smallholders already acquire two thirds of their basic food needs through the mediation of the market (Government of Ethiopia 2010:82).

Given the high population-to-land ratio, existing patterns of smallholder cultivation are undoubtedly characterized by their own intractable limits. But the primary strategy of market-oriented specialization will generate its own problems, placing traditional survival strategies and production for use at risk. The growing trend toward the commodification of land is also fostering a decline in customary reciprocal relationships, with profound implications for rural social dynamics in a context of climate change and increasing scarcity of resources. The promotion of hybrid seeds and chemical fertilizers and the growing resort to microcredits are likewise drawing peasants into new forms of indebtedness that could serve as levers of dispossession in times of drought or social misfortune (Berhanu 2009:751). The need to increase productivity by removing the structural constraints on the expansion of food and raw material production need not entail a headlong lurch into the commodification of land and labor. As an extensive study by a group of scientists from 60 countries recently concluded, smallholder agroecological farming has the potential to make agriculture more resilient and capable of improving livelihoods (IAASTD 2009).

Enclosures and the Dar Ager

While these transformations in the central highlands are fostering an evolution toward greater inequality and the abandonment of customary survival strategies, the mechanism of social change has been greater market dependence rather than forced dispossession and displacement. In the surrounding lowlands, by contrast, state policy relies on a more naked assertion of the plenipotentiary powers of the state to allocate land and determine its use. This spatially differentiated agrarian policy is reproducing in new forms older core-periphery hierarchies within the Ethiopian social formation. For state builders in the highland core, the western and eastern lowlands had historically constituted a virtual no-man's land, and the southwest fringes in particular were the source of the gold, ivory, and enslaved captives that sustained imperial state formation (Donham and James 1986; Makki 2011b). The delimitation of state boundaries in the early part of the twentieth century did little to change this sociospatial hierarchy, and as late as 1955 the Imperial Constitution simply designated the communal and pastoral commons as state domains (Helland 2006:14). The land reform of 1975 partially redressed this by stipulating that “nomadic people shall have possessory rights over the lands they customarily use for grazing,” but this was honored more in the breach than in the observance (Rahmato 2007).

Recently, the emphasis of state policies has been on turning the lowlands into sites for large-scale, export-oriented agriculture that can also exert competitive pressure on smallholders elsewhere.3 This orientation is informed by a faith in economies of scale, for which large inputs of machinery, capital, and chemicals are a necessary prerequisite. Consequently, instead of the alliance between smallholders and the state envisioned in the highlands, the strategic alignment in the lowlands involves a pact between the state and large-scale investors. The emerging social relationship is therefore one of vertical class polarization rather than horizontal smallholder differentiation. Toward this end, the government has provided special privileges to national and foreign investors, and has undertaken infrastructural projects to facilitate large-scale commercial agriculture. This has included considerable investments in transportation and communications links, and a 45,000 kilometer network of roads currently connects all the regions to the capital. Investment codes have also been liberalized to reduce the minimum capital requirement for foreign investors, and provisions for the unrestricted repatriation of profits and asset sales have been instituted (FDRE 2003; EIU Unit 2008). Foreign direct investment has consequently been growing since 2004, when it reached US$545 million, with the share of the agricultural sector accounting for 32 percent of the total inflow (UNCTAD 2008: 254).

To manage this flow of foreign investments, in January 2009 the government established the Agricultural Investment Support Directorate, which serves as the principal agency for negotiating lease agreements on land above 5,000 hectares. The Federal Land Bank was also set up to serve as a repository of land designated for leasing. By 2011, around 3.5 million hectares—a landmass equivalent to 50 percent of the total land currently under smallholder cultivation—had been transferred to the Land Bank (MOARD 2009:11). Expectations are that another 1.5 million hectares will be transferred to the land bank in the coming years (MOFA 2013). Almost all the land so far allocated to investors has come from the four administrative regions of the western and eastern lowlands: Benishangul-Gumuz; Southern Nations, Nationalities, and Peoples; Gambella; and Afar. Exact figures on how much of this land has been leased or cultivated are hard to come by, but a 2011 World Bank report asserts that between 2004 and 2008, prior to the big global land rush, 406 investors had already been granted a total of 1.2 million hectares (Deininger and Byerlee 2011:xxxii). The investors represent a mix of private entrepreneurs, national states, and international agribusinesses producing a range of products from cut flowers, rice, cotton, and soybeans to various biofuels such as jatropha and oilseeds. While the largest international investors are Indian firms, there are also Dutch, German, Israeli, Italian, and Chinese firms operating in different regions. The bulk of the capital formation so far, however, is accounted for by Ethiopian nationals, suggesting an emergent pattern of class formation fostered by the transition to a market economy (Githinji and Mersha 2007).

To realize this program of large-scale agricultural investments, the government has embarked on a vast program of clearing the land and resettling indigenous communities in designated villages across the lowlands. Federal proclamations 455/2005 and Council of Ministers Regulations 135/2007 have empowered regional and district administrations to resettle peasants on the basis of a vaguely defined “public purpose.” Available studies indicate that almost a third of these reallocations have so far “benefited private investments rather than the public” (Deininger and Byerlee 2011:105). After the disastrous experience of the forced resettlement programs of the 1980s, however, the current program emphasizes its ostensibly voluntary and intraregional features. So while the displaced have not been completely deprived of access to land, by relocating them away from the riverbanks that are being made available to investors, the program to resettle them in villages is depriving the indigenous communities of their extensive commons and grazing land. According to Umod Ubong Olom, the president of the Gambella administration, “the region has prepared 3.2 million hectares of land” for agricultural investors and “resettled more than 30 thousand scattered households of the region into 43 resident villages.” In the 2013 budget year, he continued, it was planning on settling another 10,688 households, which amounts to a staggering two thirds of the total number of households in the region (quoted in Tariku 2013). Similarly, in the Somali region of the eastern lowlands, some “150,000 households were resettled in 21 woredas [districts] in the villagization program carried out in the past four years,” and plans are afoot to “villagize 43,000 households this Ethiopian budget year” (Walta 2013). These settlement programs will only serve to accentuate the general effects of dispossession and enclosures, and even if the promised schools, clinics, and clean water materialize, they will do little to mitigate the structural dependency and vulnerability entailed by the concentration of land in the hands of a few large-scale producers. The widespread replacement of local food crops by cash crops for external markets is only one of the symptoms of this structural precarity.

Despite provisions in the 1994 Constitution that “pastoralists have a right not to be displaced from their own lands” and subsequent promulgations stipulating that those displaced be properly compensated, the conditions and terms under which compensation takes place are left to the discretion of the same district officials involved in the initial expropriations. In those instances where evicted peasants are compensated in cash rather than through a land swap, acquiring new farmland has proven to be difficult, belying the dominant discourse about the availability of wide expanses of empty farmland (Deininger and Byerlee 2011:108). De facto res communes, where little if any certification and titling has so far taken place, are not afforded even the minimal protections stipulated by federal law. A recent review of existing laws concludes that “communal holding rights may be subject to privatization at any time without the consent of the communities concerned, and that there are no clear legal provisions as to whether or not communally held land is to be compensated” (Tamrat 2010:14). Those displaced are therefore finding themselves caught between different forms of marginalization and exclusion, and many are compelled to seek seasonal employment on new commercial farms where their knowledge and skills are invariably discounted. A new mobile labor force is consequently emerging and will likely expand in the coming years since the large mechanized farms will at best provide full-time employment for only a small share of this floating rural labor force (Deininger and Byerlee 2011:64; Oakland Institute 2013:35).

The most important resource that has attracted large-scale investors to Ethiopia is undoubtedly water, since the mechanized farms will rely on large-scale irrigation schemes. With a national territory that covers 12 river basins, Ethiopia is relatively well endowed with water. But high dependence on rain-fed agriculture and wide spatial and temporal variations in rainfall patterns have traditionally restricted farmers to one harvest per year, exposing them to a precarious existence on the margins of frequent dry spells. In the second half of the 1990s, for instance, rainfall variability is estimated to have pushed 12 million people below the “absolute poverty” line (UNESCO 2009:274). All the key decision makers, including federal planners, regional administrations, and international funding agencies, concur on the need to expand irrigation in order to reduce the risks associated with rainfall variability. Estimates are that up to 2.7 million hectares of land have irrigation potential, but that fewer than 300,000 hectares are as yet developed. Yet the irrigation development programs so far conceived are primarily geared to large-scale projects that require extensive enclosure of water resources, which is likely to marginalize smallholders and induce water-related social conflicts. In the Awash River basin, where 150,000 hectares have been brought under commercial cultivation, the Afar pastoralists have been experiencing a complex set of transformations restricting their mobility and use of dry and wet grazing areas (Helland 2006; Markakis 1998). The Qoqa Dam constructed on the river to regulate the water also serves as a mechanism to monitor the Afar pastoralists, who are increasingly excluded from the flood plains reserved for commercial farming. This has exacerbated conflicts across the valley as enclosure-induced settlement modifies the customary norms that once regulated interclan sociospatial relations (Kassa 2001; Rahmato 2007). A similar dynamic is unfolding among the Karrayu in the eastern lowlands, where the informal land market has engendered a rapid process of commodification so that pastoralists today face the terrifying prospect of the disappearance of the pastoral commons (Gebre 2009:294).

Terra Nullius and Social Ecology

In order to justify these large-scale enclosures, the government, the World Bank, and agribusiness investors have increasingly deployed virtual notions of terra nullius (Borch 2001; Geisler 2012). Meles Zenawi, Ethiopia's former prime minister and architect of its development strategy, has repeatedly insisted that only empty and underutilized land is being made available to investors (Zenawi 2009). The grounds for this claim are the relatively sparse population of the western and eastern lowlands, averaging about 30 persons per square kilometer, and the fact that agropastoralists occupy a large proportion of the national territory even though they account for only 11 percent of the total population.

But this one-dimensional conception of the social and ecological reality of the lowlands is deeply misleading and obscures the ways in which space is constituted through social relations and material practices just as much as social relations are spatially constructed (Gregory and Urry 1985; Massey 2005; Warf and Arias 2009). Different societies organize and mobilize space in distinctive ways, and the modes of appropriating space necessarily vary across time and cultures (Lefebvre 1991; Smith 1984). The spatial organization of pastoralist societies, which occupy different parts of land at different times of the year, is profoundly different from the ways in which space is conceptualized and organized in agrarian empires, where power tends to coagulate into centers and progressively recede toward its peripheries. Agrarian empires have, in turn, a different conception of space from those of modern nation-states that are defined by their boundaries rather than their centers (Anderson 1991). Discourses of terra nullius invariably flatten and distort the variable production of space, willfully obfuscating the particular ways in which the social relations of village or pastoral communities are spatialized.

In the nomadic and agropastoral zones of the eastern and western lowlands, for instance, transhumance is generally characterized by household ownership of livestock and communal appropriation of pastures. The migratory cycles are typically wide-ranging so that wherever water fails to come to the land, people and animals move to water, and the seasonal movement from one pasture to another means that land is never a fixed possession. Social organization in these societies is consequently spatially extensive rather than intensive. Even among the more settled communities along the forested western lowlands, where shifting cultivation and agropastoralism is combined with gaming and fishing, spatial relations can be understood as relatively extensive. The livelihood of the indigenous communities in these lowlands is determined by the availability of pasture and water, and access to the village or pastoral commons is central to their subsistence and survival strategies. Unlike the sedentary village communities of the highlands, whose spatial relations are almost entirely oriented toward the control and use of land and labor within very restricted physical localities, the social and cultural orientation of the lowland agropastoral communities is toward sociocultural activity within much wider spatial categories. Their kinship patterns emphasize the breadth rather than the depth of their connections: Marriages are made to safeguard or pass resources in a territorial system of horizontal ties. The social and cultural markers of their social world are altogether set much wider.4

The negation of these more ecologically adapted forms of producing and inscribing space, and the representation of the lowland expanses as unproductive or underutilized, is but a prelude to a project of reconstituting them as emptiable spaces. That is, space emptied of social and cultural meaning, and fetishized into a commodity amenable to utilitarian calculation and rationalization. Modernization efforts of this sort have historically been accompanied by discourses that represent “traditional” social spaces as backward and stagnant, a terra nullius zone outside the invigorating dynamics of capitalist modernity. The redemption of these spaces entails a process of cultural erasure and disenchantment, and a reconfiguring of the different social forms through which the metabolism with nature was historically regulated. The essential precondition for this remaking of heterogeneous social spaces in order to construct the abstract space-time of commodity production and circulation is the sweeping away of traditional customary relations and practices, which is being advanced in Ethiopia today through forced enclosures and the fixing of mobile communities in place.

If discourses of terra nullius typically ignore the particular ways in which space is socially constructed, the surrounding natural habitat is likewise reductively conceived as an inert backstage for the unfolding of human history. The same procedures that seek to instrumentally rationalize land use abstract the land from the wider fauna and flora that constitute the ecosystem, turning nature into a fetishized and commodifiable resource. The biodiversity of the regions targeted for enclosures are consequently viewed as external to the development process and assigned an essentially passive role. Under these circumstances, the overall shift from smallholder-based polyculture to a capital- and chemical-intensive monoculture is likely to accelerate biodiversity loss while shifting the burden of environmental costs onto communities.

This is evident in what has already transpired in the Gambella region, one of the prime areas of large-scale land alienations. The region borders on the dense carbon-rich rain forest of the southwest, the source of livelihood for many indigenous communities and wildlife. This forest constitutes the second largest area of overland mammal migration in the world and is believed to be one of the most important biodiversity regions in the world, in large part because it lies at the intersection of the Sudan-Guinea savannah biome and the Somali-Masai biome. An important ecological feature of the region is its vast wetland areas and the four river systems that drain the wider basin and sustain over 110 fish species, six of which are endemic to the region. With little regard to this ecological heritage and the vital ecosystem services that wetlands and forests provide, the regional government of Gambella has ceded parts of the “formally designated national park, protected area and wildlife sanctuary” to private investors (Rahmato 2011:17). This disregard builds on a history of environmental degradation that has since 1990 resulted in the destruction of an estimated 100,000 hectares of forest in the region, with incalculable loss of biodiversity and species extinction (Behailu et al. 2011:18). Similar ecological threats exist in the other regions designated as terra nullius spaces, including the Benishangul-Gumez administrative region, where the Renaissance Dam is currently being constructed near the headwaters of the Blue Nile, and the Lower Omo Valley in the southwest, which has been designated a UNESCO World Heritage Site because of its unique cultural and ecological landscape (Oakland Institute 2013).

According to Ethiopia's Institute for Biodiversity Conservation, “the extensive introduction of genetically uniform and improved varieties has resulted in mono-cropping and has been the major threat to local crop landraces/farmers' varieties. Market oriented and cash crop cultivation has ignored farmers' varieties and the traditional as well as changing needs of communities” (IBC 2009:57). These concerns have received little if any serious consideration from the federal state beyond the bureaucratic formality that all land leases above 5,000 hectares undergo environmental impact assessments, a stipulation “often waived as sunset clauses” for lack of institutional capacity to enforce them (Deininger and Byerlee 2011:121). There is also little sign of a willingness to reconsider the large-scale leasing of land for the production of biofuels despite recent studies suggesting that more carbon dioxide is being released into the atmosphere from the production of some biofuels (Searchinger et al. 2008:1238–40). According to the MELCA Mahiber, a local nongovernmental organization that has worked with indigenous communities to promote sustainable and mixed land-use planning, by 2008 some 50 investors had been registered to cultivate biofuels on 1.65 million hectares, 300,000 hectares of which had already been leased (MELCA 2008).

The pursuit of this form of development ultimately rests on the belief that the establishment of large-scale mechanized farms, whatever their immediate social or ecological costs, will generate a long-term dynamic of self-sustaining growth essential for securing the general welfare of Ethiopians. But there are serious reasons to doubt this claim. The state's strategy of agriculture development-led industrialization is premised on the conviction that the surplus generated by commercialized farming would enable an industrialization drive without the social ills and political risks associated with the mass eviction of highland peasants and the proliferation of slums. Large-scale mechanized farming in the lowlands is meant to supplement and advance this primary strategy. But this conviction presupposes mutually reinforcing agricultural and industrial sectors within a purely national framework, one in which changes in rural productivity will directly stimulate industrial growth and the expansion of the domestic market. So long as the strategy depends on transnational agribusinesses, however, there is no reason to expect that a synergistic relationship between agriculture and industry will emerge. Corporate agribusinesses operate on an international scale and newly incorporated production sites in Ethiopia will constitute no more than low-value-added links in global commodity chains.

It is also unclear how this strategy will promote domestic food security or generate the much-needed revenue to promote the diversification of the national economy. Investors can raise up to 70 percent of the operational costs of the new investments from the national bank of Ethiopia, and the incentives in place are structured so that the more export oriented the investments, the larger the tax breaks. According to the Economist Intelligence Unit, for those agribusinesses that export at least 50 percent of their production, the tax holiday is 5 years, and for those that export 75 percent of their production, no taxes will be levied at all (EIU 2008).5 Moreover, the land is being leased at incredibly low rates for 25 to 50 years and in circumstances where the government proudly proclaims that labor costs in Ethiopia are competitive with if not lower than the regional average.

Comparative studies of similar processes elsewhere in Africa cast further doubt on the underlying assumptions informing the Ethiopian state's development strategy. In a close study of southern Africa, a paradigmatic case of accumulation by dispossession, Arrighi, Aschoff, and Scully challenge the notion that this form of agrarian change is a necessary and unavoidable prerequisite for capitalist development and conclude that “such dispossession has in fact become the source of major developmental handicaps for at least some and possibly many countries of the global south” (2010:410). A recent report by the High Level Panel of the UN Committee on World Food Security has similarly questioned the likelihood that large-scale international investments in intensive agrochemical farming will in fact improve rural livelihoods: “Evidence from this land rush to date shows very few such cases. Rather, large scale investment is damaging the food security, incomes, livelihoods and environment for local people” (FAO 2011: 8).

In response, the Ethiopian government has suggested that the guiding role of the state in the economy will check potentially destructive tendencies and ensure optimal outcomes. But while the state is indeed a key actor in large-scale projects of rural transformation, state policies are essentially geared to facilitating these investments and are informed by the criterion of market efficiency, which takes precedence over all other principles and claims to resources. But market efficiency has never been a matter of merely introducing new technologies and improved seeds and agricultural inputs. It has always presumed the imperative of removing inefficient producers from the land in order to make it available to those with the capital and technology to develop it. It is precisely this logic of market efficiency that is currently being utilized to extinguish the customary rights of smallholders and commoners alike.


In the name of expanding food production and generating the surplus necessary for industrialization, extensive tracts of farmland are being annexed, rivers are being redirected, and local ecosystems are being reconfigured. These transformations are informed by policies premised on a classic conception of modernization, which presumes a steady decline in the share of agriculture in the national income and in the composition of the labor force. From this perspective, the res communes of the lowland communities constitute a vast reservoir of backwardness and a structural impediment to economic modernization. The establishment of large-scale farms, and the widespread application of modern methods of farming, will therefore require the prior mastery over these traditional communities and the conquest of nature. As Timothy Mitchell has argued, for much of the twentieth century, “the politics of national development and economic growth was a politics of techno-science, which claimed to bring the expertise of modern engineering, technology, and social science to improve the defects of nature, to transform peasant agriculture, to repair the ills of society, and to fix the economy” (2002:15). Viewed from the vantage point of the disinherited, this politics of technoscience—and the classic combination of technocratic hubris and cultural contempt it often entails—constitutes a profound attenuation of critical thought and a radical foreclosure of people's ability to self-consciously fashion alternative futures.

Discussing the inner connections between expanded reproduction and territorial enclosures, Rosa Luxemburg once argued that capital accumulation over time would require imperial expansion across space, and that the continuous incorporation of noncapitalist social spaces would make it increasingly difficult to resolve periodic crises of overproduction. Predatory capitalist expansion, she argued, would ultimately doom the civilization of capital itself (1963:453). At the end of the century, in a new context of market triumphalism and ecological attrition, Frederic Jameson was pondering the paradox that it had become easier to imagine the end of the world through ecological catastrophe than the end of historical capitalism itself (1994:xii). The deep conundrums implied by these diagnoses are today being tested on an astonishing scale across much of the continent of Africa, the ostensible “last frontier” of capitalist rationalization. The attempt to resolve the world crisis through a new process of development through dispossession is nonetheless meeting a powerful chord of resistance. This poses a profound challenge to the laissez faire doctrines that have served as the ideological cement of world politics and economics over the past few decades. Determining their outcome constitutes one of the central challenges of our times.


  1. 1

    José Harris (1993) has asserted that between 1870 and 1914, some 6 million people emigrated from the British Isles alone.

  2. 2

    According to the Atlas of Ethiopian Livelihoods (Government of Ethiopia 2010:74), “A minority of the very poor are effectively landless, while the majority have especially low land holdings and minimal livestock, and virtually no ownership of that engine of crop production: the team oxen. Their dependence on working for others, allied to their acute lack of capital assets, renders them a kind of rural proletariat.”

  3. 3

    It is not yet clear, for instance, what the implications of this emphasis on large-scale commercial farming will be on the production of Ethiopia's main export commodity, coffee, which for the most part remains in the hands of smallholders.

  4. 4

    There are considerable differences in the descent and kinship structures of the various communities of the eastern and western lowlands, and the remarks above are merely intended to highlight the spatial patterning of social and cultural relations that are occluded by the assumptions informing terra nullius discourses. For studies on the role of common descent in nomadic societies, see Khazanov (1983:138–48); for one study of exchange marriage among the Gumuz of western Ethiopia, see James (1986).

  5. 5

    A recent statement by the Ethiopian government indicates that contracts for foreign agricultural investors now include an obligation for a 40–60 divide of the produce, with 40 percent of the produce to be directed to the domestic market (MOFA 2013). In all likelihood, this was in response to pressure from various domestic and international organizations.