Ransoms , Remittances , and Refugees : The Gatekeeper State in Eritrea

This paper draws from ethnographic research in Eritrea to explore new configurations of power and belonging in the Eritrean gatekeeper state. The gatekeeper state is a theory describing state-society relations in Africa in which the patrimonial state sits astride narrow channels of wealth creation, relying on control of the circulation of citizens, funds, and resources within and across national borders. The escape—illegal emigration—of citizens from Eritrea and the remittances sent home to families in rural areas have potentially been a source of challenge to state authority, but this paper argues that the Eritrean state has developed new gatekeeping strategies that operate in and through porous borders, transnational kinship networks, and the aspirations of citizens to escape civil service.

I introduce the contemporary gatekeeper state with a story about Amanuel, a friend who was arrested, like many other young Eritreans, trying to cross the border into Sudan. The last record I have found referring to him in field notes from my ethnographic research on resettlement in the western lowlands in 2004-2005 dated to months before I learned that he was missing. My notes, in fact, had him celebrating good news: attending the new postsecondary school of Mainefhi, he was thrilled to learn that he was one of few students in his class who would be posted to the capital city, Asmara, for the period of military service to be performed during the semester break. Mainefhi was being kick-started by the government as an alternative to the long-standing University of Asmara. Located outside the capital city, it was a militarized school, filled with students who had passed qualifying exams taken during the requisite twelfth grade spent in Sawa, the military training camp in the far western lowlands. Mainefhi students were not permitted to come and go freely, nor could family members visit, and rumors abounded that students were being jailed who failed to attend classes or were captured trying to flee the school compound and were enduring "physical education" classes that involved long hours spent digging trenches-a course ironically dubbed digology (Reid 2009). For Amanuel, semester break in Asmara meant cafes africaTODAY 60(2) and theaters, a network of friends and family, and perhaps a release from the restrictions of the academy.
Weeks later, I visited the capital city from my field site in the western lowlands, only to hear that Amanuel had been captured trying to flee the country, attempting the long, dangerous journey across the Sudanese border. He had not informed his family or friends that he was planning on leaving, and the only information they received about his welfare was a brief note smuggled out through a guard from the prison where he was being detained indefinitely. In the months following his capture, his roommates would get calls from time to time, from friends or acquaintances, asking after him. These roommates never revealed any knowledge of his whereabouts, or even that he was missing. The people calling could be anyone, they assured me, government agents, trying to figure out what we know-even people whom they remembered as Amanuel's friends. It was safer to be silent. Indeed, those suspected of assisting emigrants were often taken for interrogation, imprisoned, or pressed into military service. The families of young people who fled the country were sometimes imprisoned themselves, or forced to pay steep fines as a ransom for the seepage of human capital from a country of just over five million, in which at least 200,000 people are in active military service and all able-bodied adults are required to work in unsalaried national service-an eighteen-month period that in practice extends indefinitely (Human Rights Watch 2009; International Institute for Strategic Studies 2010).
Eritrea is currently one of the world's top refugee-producing nations (Human Rights Watch 2009). In 2010, around 3,000 Eritreans a month were fleeing their country across the Ethiopian and Sudanese borders (UNHCR 2011). This wave of migration has increased steadily over the past decade, despite the dangerous conditions facing asylum seekers-including a shootto-kill policy for those caught trying to escape, and more recently, the danger posed by human-trafficking networks targeting Eritrean refugees and asylum seekers (Monitoring Group on Somalia and Eritrea 2012; UNHCR 2013). This emigration is a reversal of the previous trend of return migration by Eritreans after independence-the intended focal point of my research on the role of the environment in refugee resettlement in the western lowlands, where as many as 200,000 Eritreans were being repatriated under state-led resettlement projects. While my research did focus on the ways in which returnees and stayees negotiated belonging and claims to resources in the resettlement community of Hagaz, I came to investigate the growing wave of migration back across the Sudanese border, which was notably linked to the inabilities of many people to participate in the promises of nation-building in ways they had envisioned on returning to their country. These frustrations were largely due to the worsening economic and political conditions since the border conflict began in 1998, including the transformation of national service into an indefinite period of forced labor (Kibreab 2009).
This emigration reflects the broader and long-standing trend of economic migration from the global south, but it opens a number of questions africaTODAY 60(2) about the nature of state power in Eritrea and the webs of compliance and survivalism that bind escaping nationals to the state. Certainly, these young Eritreans are looking for economic opportunities in the context of great scarcity; they are seeking ways to support their families-a nearly impossible endeavor when forced conscription and national service removes them from urban and rural livelihoods, either as wage laborers or subsistence farmers and agropastoralists (Poole 2009). People who have been involved in the lives and frustrated aspirations of Eritrean citizens since they gained their independence from Ethiopia in the early 1990s after thirty years of warfare are prompted to ask an additional set of questions. What do we make of migration from a nation-state that has been able to claim power and legitimacy via the legacy of powerful nationalism? As Kibreab notes, flight "is contrary to expectation in view of the fact that the EPLF [Eritrean People's Liberation Front] came to power on the back of a popular struggle promising to relegate to the dustbin of history the factors that previously forced Eritreans to flee their country in search of international protection" (2009:54). In the seemingly persistent silence that many have observed among the Eritrean population, is flight the only means of dissent? Is dissent the best way to understand emigration? Does individual flight suggest that strident Eritrean nationalism is soluble after all-that collectivist national projects in countries like Eritrea, occupying a marginal position in the global economy, have begun to dissolve into individual strategies of survival?
The escape of citizens from Eritrea and the remittances sent home to families through informal mechanisms have potentially been a source of challenge to state authority; however, adapting Cooper's (2002) theory of the gatekeeper state to the Eritrean context allows us to appreciate the persistence of nationalism in the multiple paths and projects of flight across national borders. Moreover, the state in Eritrea, rather than being dismantled by the flight of its citizens, has been sustained via new strategies of gatekeeping. Along these lines, the government of Eritrea has assumed the capacity, if not to manage this migration itself, to capture the material and symbolic capital of these projects of movement, both successful and unsuccessful.

The Gatekeeper State in Africa
Theories of the African state and nationalism often seem to miss the mark when extended to the Eritrean context, where nationalism has been a defining, and seemingly enduring, feature of political life. Describing the moment of independence for African nations in the mid-twentieth century, Cooper comments on the birth of a national imagination, shaped in both discourse and practice: "The very process of claim-making had helped to define a national imaginary" (2002:198). Cooper points to the loss of this imaginary in the following decades, with the eventual slimming of political and economic spaces available to African societies. Ferguson also problematizes nationalist identifications in describing the attenuation of nationalist africaTODAY 60(2) projects in Africa in the context of an emerging neoliberal world order. During the era of decolonization, he argues, "the political form of the independent nation-state . . . obscured the continuing transnational relations that help to produce 'African poverty ' " (2006:17). When African elites, however, attempted to redefine national identity during the era hailed as the African Renaissance, these efforts were undermined by a global political economic context that selected for individuated lines of flight over collectivist projects. These observations are a critical starting point for understanding the shifting role and meaning of nationalism in African communities; however, the Eritrean context suggests that nationalist identifications, though socially constructed, may not dissolve so easily with the loss of modernist dreams, with transnational flows, and even with individuated lines of flight. Those who have left Eritrea, for instance, have pursued individual strategies to access resources at the same time that they have remained embedded in nationalist projects and have sustained state-led development efforts during a time when such a thing seemed anachronistic (Bernal 2004;Hepner 2009b). Many scholars have noted the key role of the Eritrean diaspora in negotiating the parameters of citizenship and governance after independence (Bernal 2005;Iyob 2000;Nadje, Black, and Koser 2001;Hepner 2009b).
The persistence and power of Eritrean nationalism pushes us to rethink the presumed unbundling of the hyphen linking nation to state. Unlike the 1960s, when most African states gained national independence, Eritrea emerged in the early 1990s onto a world stage in which donors preferred to work through nongovernmental organizations, the doyens of civil society, at the expense of what were considered inefficient and corrupt state bureaucracies. These transnational linkages were often understood to problematize the nation-state as a container of institutions and identities: "If the nation-state formed the basis for the projects of colonialism and nationalism, then the 'unbundled' space that is being created by new forms of governmentality . . .
[characterize] 'the postcolonial condition' " (Gupta 1998:329). In contrast to this, the Eritrean state has maintained tight control over civil society and the actions of foreign-aid and development organizations. The predominant narrative accompanying state-led development initiatives has revolved around self-sufficiency, in part drawn from a sense of isolationism during the struggle, in which global superpowers supported Ethiopia. This discourse focuses on the primacy of national identity over other forms of identification, the value of sacrifice for the good of the nation, and the central role of the developmentalist state. Isaias Awferki, president since independence, has publicly crafted a critique of foreign aid, forced privatization, and external models of democratization, arguing that foreign aid is typified by "structural flaws, crippling preconditions, and self-perpetuating tendencies" (Afwerki 1997).
Nationalist discourse in the state-run media focuses on self-sufficiency, but the sweeping projects of state-led development in Eritrea, in which a massive number of in-country citizens are mobilized in national service, persists because of remittances and taxes procured from the global diaspora, comprising as many as one-fourth to one-third of all Eritrean africaTODAY 60(2) citizens (Hepner and O'Kane 2009). Additionally, even though an isolationist stance and focus on self-sufficiency have been possible only through remittances, these ideals are used to justify authoritarian state actions that have slimmed the space available for civil-society organizations to create webbing between local and transnational organizations. 1 The government of Eritrea has expelled or placed untenable restrictions on nongovernmental organizations since 2005 "to diminish the leverage of outside powers over both its repressive policies at home and its capricious behavior in the region" (Connell 2011:424).
Fredrick Cooper's (2002) theory of the gatekeeper state offers a useful framework for understanding the ways in which contemporary Eritrean politics cannot be dismissed as exceptional, but may help to illuminate processes of power and shifting state-society relationships in Africa. Rejecting the priority often given to the moment of independence and the schism between colonial governments and liberated African states, Cooper points to what he envisions as a more fundamental frame of reference: the growth of the developmentalist state after World War II and its moment of crisis with global economic depression in the 1970s. Consequently, he develops a model of the gatekeeper state to describe the marked continuities in state-society relations throughout the late colonial and postcolonial periods. Postcolonial African states inherited not only bureaucratic infrastructures and territorial borders from colonial predecessors, but also the imperative of development spearheaded by the state, and alongside this imperative, a system of governing in which the state had weak penetration throughout communities within national borders. In addition to dependence on empowering traditional authorities through a system of indirect rule, the "coercive power" of colonial states "was more effective at staging raids and terrorizing resistors than at routinizing authority throughout a territory" (Cooper 2002:157). With the transition to independent governance, those who rose to power within African states consolidated power through similar channels, coming to occupy a space between global markets and the formal means of revenue at their disposal: What they could do was to sit astride the interface between a territory and the rest of the world, collecting and distributing resources that derived from the gate itself: customs revenue and foreign aid; permits to do business in the territory; entry and exit visas; and permission to move currency in and out. (2002:157) While it might be argued that any state performs these sorts of gatekeeping functions, the African gatekeeper state is particular in part because the means of amassing wealth is narrowed to involvement within (or control of) the state, resulting in a winner-takes-all scenario, which precludes incentives for political power sharing. While the gatekeeper state may be depicted as sitting astride society, however, state and society are deeply intertwined through social networks that take on a distinct vertical character. 2 africaTODAY 60 (2) Stressing networks between people across state spheres is one way that Cooper's analysis departs from other understandings of state-society relations in Africa, most notably Mamdani's emphasis on the dualisms between citizen and subject, and urban and rural, that he depicts as legacies from the colonial era. For Mamdani, the problems of governance in Africa-the artifacts of decentralized despotism and bifurcated rule over African rural and urban areas-emerge from colonialism. He details the ways in which indirect rule "containerized" rural populations along ethnic lines and distilled diverse sociopolitical networks into a single model of customary authority, which was "monarchical, patriarchal, and authoritarian" (Mamdani 1996:39). In urban areas, direct rule involved a racialized system of access to civil freedoms granted to citizens in civil society, leading to a bifurcated system of ethnicized subject and racialized citizen that has yet to be displaced by postcolonial attempts that have been limited to dismantling one side of the coin while upholding the other, often leading to centralized despotism over rural and urban areas. Mamdani offers an interesting critique of unilinear and structural models of African underdevelopment put forth by dependency theorists in their efforts to explain the "African predicament." Nevertheless, the question that arises about Mamdani's argument is the extent to which it solidifies social dichotomies that may long have been more fluid than rigid in lived realities outside the scope of his analysis and prescription, particularly in the context of a transnational population that has taken shape since colonialism.
Instead, Cooper explores the relations that become apparent "when one is not limited to seeing society divided into categories-elite versus popular classes or ethnic, racial, or gender divisions-but rather stresses relationships, and in particular vertical relationships" (2002:188). This focus on relations and networks, rather than oppositions, is closely linked to a notion of agency that reaches beyond the limits of strict social categories that would determine the position and function of groups within a larger system, instead attending to the power of political imagination to shape futures. Again redolent of a contradiction to Mamdani's categorizations that often appear to be inherited intact from the colonial experience, Cooper writes: Africans faced the constraints and the humiliations of a colonial state, but they are, above all, human beings trying to survive, form relationships, find opportunities, and make sense of the world. They cannot be reduced to stick figures in a drama with two actors, colonizer and colonized, or a story with one plot line-the struggle for the nation. What is striking about the year after the war [WWII] was how much seemed possible. (2002:38) These multiple possibilities hinge on the developments within a political imagination, formed in the dynamic interplay between individuals and state policies. Dramatized by the era of decolonization, colonial policies africaTODAY 60(2) and the response of individuals shifted the ways in which people imagined and articulated political futures. Cooper describes how within French and British colonial policies following World War II, "the desire to expand empire resources while legitimizing colonial rule . . . became the basis for a profound engagement of African and European actors, which in turn changed the meanings of 'development, ' 'citizenship,' and 'self-government' (2002:39). While subject to reimaginings, political structures and identities were also channeled, as the gatekeeper state drew from certain networks while curtailing others from forming outside state spheres. The notion of blockages is significant here, for while change may "lurch in different directions" (2002:15) within a political space of "interconnection, relatedness, and mutual influence" (2002:13), some spaces are closed off-sometimes violently-to other political imaginings.
Cooper suggests that networks existing both within and beneath the surface of state channels offer the possibility of radical social change: "Gatekeeper states in fact have something to fear from networks or collectivities able to pose a challenge or from African cultivators who use social connections to make the state irrelevant" (2002:202). In Eritrea, the escape-illegal emigration-of citizens and the remittances sent home to families in rural areas have potentially been a source of challenge to state authority.

Ransoms and Remittances: Gatekeeping in Eritrea
The contemporary government in Eritrea embodies many of the characteristics of the gatekeeper state while extending them in new directions-across national borders and through networks both material and symbolic. The flight of individuals from Eritrea often does not signal a definitive end point to their entanglement with the Eritrean state, and the financial and political pressures exerted by the Eritrean state on exiles have a long and dynamic history. The struggle for independence led to the emergence of a widespread and populous diaspora, which often remained deeply involved in funding the war, along with debating the meaning and nature of citizenship and nationhood at the time of independence. 3 Many Eritreans living in this global diaspora voted for independence in the 1993 referendum and participated in drafting the constitution, which, though it has not been fully implemented, stipulates the enfranchisement of overseas citizens (Iyob 2000). If the Eritrean gatekeeper state involves the gatekeeping of social, symbolic, and financial flows from a transnational citizenry, it has emerged out of a long history of various periods of emigration that "predate the articulation of a distinctly 'Eritrean' identity, and in fact, helped establish that identity vis-à-vis the outside" (Hepner and Conrad 2005:xi).
Eritrean government officials since independence, through political offices in major settlement countries, the operation of party-led organizations, and regular visits to the diaspora, have maintained financial and political support from a transnational constituency (Newland and Patrick 2004), africaTODAY 60 (2) and occasionally, as was recently the case with Eritrean students studying in South Africa, they have operated to quell dissent (Hepner 2009a). In this case, the Eritrean embassy in South Africa worked to dismantle a student movement by making threatening phone calls, pulling tuition and other vital resources, and revoking passports (Hepner 2009a).
Beyond quelling active dissent, the ruling party has been remarkably successful at attaining "hegemony over nationalist identity and trans/ national praxis" (Hepner 2005:79). Scholars have provided windows into the deterritorialized authority of the Eritrean state, exploring the origins of this transnational field in the 1970s and applying this analysis to a nascent civil society group in California that was stifled by the Eritrean state in the 1990s. Agents of the Eritrean regime have used multiple ways-including overt pressure and covert tactics, such as spreading rumors-to undermine an autonomous organization formed to connect Christian and Muslim Eritreans living in Southern California (Woldemikael 2005). Consequently, the PFDJ government, partly through party-aligned organizations, visits to the diaspora, government-sponsored news, and surveillance of diaspora activities, has subsumed and channeled diaspora engagement in the Eritrean political economy (Hepner 2005). Although new human-rights groups bear some potential to break up the "monopolistic domination of diaspora communities by the state" (Schmitz-Pranghe 2010:24), they operate within a framework that has been shaped by dominant political actors who have so far fractured dissent and channeled modes of transnational engagement in the country.
Diaspora Eritreans are absolutely central to the survival of the Eritrean state, as they have been one of the main sources of revenue for the government. Collecting data on the volume and nature of remittances to Eritrea is problematic. Financial data for Eritrea is somewhat opaque, given the lack of published budgets from the government, a scarcity of international economic reporting on Eritrea (Styan 2007), and a far from complete portrait of the size and shifting profile of the Eritrean global diaspora (Fessehatzion 2005). Despite these uncertainties, most scholars highlight the exceptional role of remittances in Eritrea in terms of the scale of remittances, their centrality to the Eritrean economy, and the level of government control over them. In 2007, in light of low exports and decreasing foreign aid, Eritrean foreign reserves were only enough to cover about two weeks of imports, making the country highly dependent on the diaspora as a source of foreign exchange (Styan 2007). Remittances made up nearly one third of the GDP, according to the World Bank in 2002, reaching US $1.37 billion in 2007, and according to these estimates, Eritrea ranked fifth in Africa in remittances received per capita, and first in proportion of remittances in relation to GDP (Schmitz-Pranghe 2010). In 2002, $10 came in as remittances for every dollar generated from foreign direct investment, and $40 came in as remittances for every dollar earned from exports (Fessehatzion 2005). Alongside the scale of remittances within the Eritrean economy, "the government's ability to control and channel what are-in almost all other cases in the world-private transfers is africaTODAY 60(2) worthy of attention" (Styan 2007). These remittances are collected directly, in the form of a 2 percent income tax requested of diaspora Eritreans, and indirectly, through taxes collected from rural households that depend upon remittances from family members residing abroad.
The tax on Eritreans living abroad is technically voluntary. The government has a limited capacity to exact compliance with it or to enforce disclosure of income; however, refusal to pay this tax can have consequences for individuals abroad and their families in Eritrea. Some Eritreans living abroad have not had their passports renewed, or have been unable to purchase property in Eritrea for failing to pay the tax; others report that family members back in Eritrea have been punished through detention, fines, the denial of business licenses, or the confiscation of property (Human Rights Watch 2009). Many Eritreans have been able to visit their home country without having paid the tax, but they have been forced to pay back taxes in the case of requests for government services or to purchase land (Styan 2007). Studies suggest that remittances sent home tend to decrease over time, but data reported by the government of Eritrea show a steady increase in remittances from the 2 percent tax between 1997 and 2003-from $1.2 to $10.4 million (Fessehatzion 2005). These data suggest that declining political legitimacy on behalf of the ruling regime is not necessarily the determining factor in soliciting remittances (as people may be compelled to pay to access state services), or that the number of remittance senders has been steadily increasing. 4 If the aspiration of many young people fleeing the country involves finding a livelihood stable enough to send remittances back to their families, the Eritrean state has, since increasing restrictions around foreign exchange in 2005, been able to channel, monitor, and tax many of these other remittances. In March 2005, the government promulgated Legal Notice 101/2005, currency regulations restricting all domestic transactions to nakfa and specifying the potential of two years in prison and a fine of two million nakfa (more than $130,000 at the official exchange rate) for transactions involving foreign currency done without permission, including the deceitful import or export of foreign currency (Harris 2005). This measure was enacted to address a severe shortage of foreign exchange. Although the currency shortage was due to the border conflict, the collapse of trade relations with neighboring countries, and a fixed exchange rate (Harris 2005), government officials reported that the measures had been adopted to address the shortage of hard currency because of fuel subsidies needed to assist the rural poor. Fear of punishment led to the near disappearance of a black market for currency exchange at that time. It became increasingly impossible to locate foreign currency, or to exchange it for nakfa, outside the state banks and Himbol Financial Exchange, a party-controlled company. People who had relied on money sent by family members living abroad reported that they now had to receive all funds through Himbol, where it was taxed and subject to a fixed exchange rate below the informal market value. Anecdotal reports suggest that, though the black market has persisted, there have been africaTODAY 60(2) waves of restricted access. Even with the muted black market, however, informal remittances that have occurred may generate state revenue as the source of taxes for households within the country that depend upon family remittances to meet basic needs.
The government also exerted tight control over other valuables entering the country. The new decree held that all foreign currency was to be declared on entry into the country and accounted for on exit. Postal packages were frequently searched, and taxes were levied on goods. Laptops entering Eritrea were often confiscated for a period of time until proper government approval for their location and use could be provided. In addition, exit visas, including those issued to foreigners, were granted only on proof that the technology had not been sold or left behind.
The government of Eritrea issued Proclamation No. 173/2013 in February 2013, replacing the 2005 legal notice. The main difference appears to be that foreign currency needs to be declared at customs only when it exceeds US $10,000 dollars. The Ministry of Information describes the impetus behind this proclamation as a renewed focus on private industry and privatepublic partnership for economic growth, explaining the necessity of previous restrictions by referring to the state of siege that has typified governance since the border conflict broke out in 1998: It is to be noted that Eritrea has over the past 12 years been compelled to prevail over an exigent stage challenged by all acts of conspiracy ranging from flagrant invasion to a number of covert and blatant political and economic ploys, whose ultimate goal is reversing national sovereignty. Whereas the existing assets need to be employed in the most pressing priorities, such [a] state of affairs requires special handling. Instituted in relation to the scenarios Eritrea underwent, previous regulations that have been in operation are now repealed and replaced by the latest notice. (Shabait.com 2013b) Another Ministry of Information article notes that Eritreans traveling to and from the country believe that this policy will enhance investment and "encourage mobility of nationals to and fro [sic] the Homeland" (Shabait. com 2013a); however, the new legal notice maintains the restrictions on the use of foreign currency inside the country and maintains state-run or stateapproved financial organizations as the sole source of remittance processing. 5 Beyond financial flows from a well-established diaspora, gatekeeping extends through the social and family networks of those who currently attempt, successfully or unsuccessfully, to flee. Recent escapees are treated as criminals if involuntarily repatriated to the country, facing imprisonment and sometimes torture. Their families are subject to recrimination. Recall the fear on behalf of Amanuel's friends and family that state secret service would intuit prior knowledge of his attempt to flee. In this case, they were not detained or punished; however, there are many accounts of africaTODAY 60(2) family members facing stiff fines and imprisonment as a sort of ransom for the young men and women who succeed in leaving the country, or go missing from military and national service positions. A fee of 50,000 nakfa (US $3,300) was sometimes levied on families whose daughters or sons escaped the country (Styan 2007). At one point in 2005, the public school in the small town of Segenetti faced a delayed start to the school year because it was filled with at least fifty people-including elders and children-imprisoned there as punishment because family members had gone missing from nationalservice positions. Concern for family members living inside the country may dampen active political dissent by Eritreans in the diaspora, and this brings us beyond financial gatekeeping to the role of the state in channeling the traffic of political actions and the meanings of national identity.
Perhaps the key difference between Cooper's classic description of the African gatekeeper state and the current government in Eritrea is a refashioning of the vertical relationships that join citizen to state. Instead of political patronage, Eritrean state-society relations are linked in intricate vertical networks, characterized by the incorporation of individuals into state service through mass conscription and a social political fabric in rural and urban areas threaded through with political fear and silencing. The opacity of governance, as detailed by Riggan (this volume), along with the lack of a free press and the severity of punishments, contributes to this culture of fear. 6 That I am mostly forced to reconstruct Amanuel's story of imprisonment from memories of fieldwork conducted in Eritrea is an important piece of the story. The absence of this event from my field notes recalls a time when I was enmeshed in the political paranoia that threaded its way through many forms of communication in Eritrea-not just the wild interpretation of rumors and events missing from the only available state-run press in the country, but also through more intimate spheres: e-mails, phone conversations, casual social encounters, and public gestures, the significance of who you are spotted walking with-let alone field notes that, despite passwords and pseudonyms, felt dangerous and exposed. Eritrean social life was characterized by a pervasive sense that everything done publically, and perhaps privately, was under observation, subject to suspicious scrutiny, capricious interpretation, and punishment. And nearly everything, beneath the surface, seemed to hold within itself a double meaning, an insidious portent that seemed to mirror the tension that had existed in the stalled state of exceptionality since the border conflict broke out between Eritrea and Ethiopia in 1998. Was the university being dismantled before our eyes as a kind of punitive action against the elite urban intelligentsia, or was it being reconfigured in a visionary effort to decentralize higher education and disseminate resources to rural areas? Was it even really happening at all? Weeks before the new term was to start, when would faculty learn if any students would be admitted that year? How many young men and women were rounded up by soldiers in the capital city the day before in a seemingly random and widespread sweep of public places? By what criteria were some held and some released, and what would become of them? These whispered and wild africaTODAY 60(2) speculations were punctuated by the constant throb of fighter jets flown over the city. Did they indicate that war with Ethiopia would soon break out again? What else could these unpredictable but persistent maneuvers mean during a time when fuel scarcity in the country was so dire that petrol was unavailable even with a coveted ration coupon, stranding people for days as they waited for public buses to refuel and resume service?
These kinds of rumors, coupled with the unknowns surrounding the fate of those who attempted to flee, the fear of surveillance, and the concern that anyone could be penalized on the basis of a rumor spread about their disapproval of the government or intent to leave, provided some measure of gatekeeping on interpersonal relations, channeling dissent into flight. The effect of this situation, as much crafted by tax codes and conscription practices as a transformation of the public sphere into one of fear, has been to minimize the potential for collective dissent in the country, siphoning off a large population of youth who do have severely limited economic opportunities in Eritrea, given the stalled political and economic standoff with Ethiopia, and at the same time profiting from projects of emigration via the capacity to demand, channel, and tax remittances, and maintain tighter political control over families that remain in the country: "Used in particular by the young and educated . . . exit robs the country of those that might be able to inspire change from within. Moreover, by removing themselves as potential 'troublemakers,' and sending remittances back home, these new refugees unintentionally reduce the pressure on the regime to instigate reforms" (Conrad 2005:255).
The slimmed spaces for public opposition, coupled with a network of vertical relationships between people variously positioned within and around the state sphere, suggest that the gatekeeper state not only challenges periodicizations between colonial and postcolonial, but simplifies divisions between state and society. Political power in Eritrea has a capillary nature, in which intentionality may be less orchestrated by government officials than it is inferred by people who collude in the restraints placed upon them by political fear (Bozzini 2011). The experience of Amanuel's friends, and of many other people I interviewed whose friends and family had escaped the country illegally or evaded national service, certainly support this observation. Through the operation of rumors, silence, evasion, and absence, the gatekeeper state is produced within the micropolitics of daily life. Rather than a monolithic, rational actor, the state emerges from the work of a series of actors, some of whom can capitalize on resources because of the selective permeability of national borders. 7

Conclusion
Remittances compose "part of the complex web of relationships that connect diasporas with their places of origin" (Fessehatzion 2005:165). The vertical relationships that comprise the Eritrean gatekeeper state as it channels not africaTODAY 60(2) only remittances but political structures and identities extends beyond and through permeable borders, throughout transnational communities. As described above, the transnational Eritrean sphere has been constitutive of Eritrean identity, but in ways that have been channeled by the EPLF state: Diaspora engagement in Eritrea, if allowed at all, takes place in close co-operation with the government . . . the influence of the diaspora organizations supportive of the government and the "silent majority" . . . is largely limited to financial contributions to the state, which are likely to have a stabilizing effect on the current system and individual "indirect" activities, such as sending remittances and financing the escape of young relatives. (Schmitz-Pranghe 2010:30) We must be cautious about reading only resistance and individualism in flight, as Reid does in a recent article on political silence in Eritrea when he addresses emigration: "One of the more worrying developments has been a disengagement of people from the state and what the state purports to represent. Individualism is increasingly replacing the wider sense of 'national community,' which was discernible (if at times a little contrived) prior to and during the 1998-2000 war. People now care for little beyond their own circumstances and those of their immediate families" (Reid 2009:211). Cooper's concept of the gatekeeper state pushes us to think beyond a simple division between individualism and nationalism, dissent and compliance, resistance and oppression: instead, we need to appreciate people's "complex coping strategies" and "multi-sided engagement with forces inside and outside the community" (Cooper 1994(Cooper :1533, along with overlapping and perhaps conflicting commitments. Instead of interpreting dissent and individualism, it is perhaps more useful to think through the ways in which individual strategies of survival are harnessed by new forms of gatekeeping for collective purposes. Eritrea pushes us to consider the capacity of the gatekeeper state to evolve new strategies of gatekeeping around sociopolitical and financial networks that may otherwise bypass the nation-state. In doing so, the gatekeeper state has adapted to a historical moment supposedly typified by the attenuation of collectivist national projects. africaTODAY 60(2) noTeS 1. At the same time, as I have argued in contrast to Wrong's (2005) centralized model of eritrean national history, the building blocks of nationalism-memories of the struggle, nostalgia for lush landscapes damaged by colonial conflict, and shared sacrifice and martyrdom-are not completely enveloped by the state. Instead, ethnography reveals how people outside centers of power use history to imagine different futures and reconstitute themselves as historical actors.
2. The verticality of state-society relations represents an apparent contradiction to the kinds of horizontal linkages theorized within Anderson's description of imagined communities in the Western nation-state.
3. Hepner (2009b) adopts an ethnographic focus on eritrean postindependence diaspora communities to explore "how often violent efforts to construct national identity and control territory unfolded through space and time, drawing diverse and dispersed populations into a contentious but compelling political process. " 4. Fessehatzion (2005) examines the reported numbers of eritreans living in the middle east, europe, and north America against remittance data, finding it likely that many do not pay the 2 percent tax, or undervalue their salaries when they do so.
5. Revised penalties cited in Proclamation 173/2013 for illegal remits in nakfa for foreign currency received abroad, or illegal payments or exchanges of foreign currency in eritrea, involve imprisonment up to three years, or a fine of not more than 50,000 nakfa.
6. In addition to the fact that the 1997 constitution has not been implemented and there is no free press, there is no public record of the decision-making processes for proclamations issued by the executive branch, nor is there a public comment period. Riggan (this volume) explores the value of ethnographic research in understanding state power when it operates in opaque ways. See Tessema (2010) for an interesting discussion of the role that perceived lack of good governance plays in the brain drain from eritrea. 7. It is beyond the scope of this article to explore fully the recent emergence of human-trafficking networks targeting eritreans as a pressing human-rights issue. It is worth noting, however, a recent Un report that alleges the collusion of eritrean military and political officials in the lucrative cross-border smuggling of arms and people out of eritrea (monitoring Group on Somalia and eritrea 2012). Whether or not this is the case, the testimonies of eritrean emigrants in this document point to a social model of the eritrean state as performing a gatekeeping function through the selective porosity of borders.

ReFeRenCeS CITed
Afwerki, Isaias. 1997. Foreign Aid Works Best When It Is Self limiting.