The Nexuses between Exit, Voice, and Loyalty in the Light of the Indefinite Eritrean National Service

Gaim Kibreab


For a country and society that was “rising from the ashes” of a devastating thirty-year War of Independence which on the one hand, destroyed the pre-existing rudimentary social and physical infrastructure, and on the other, brought the disparate ethno-linguistic and religious groups closer to each other than ever before to face a mightier common enemy, it was appropriate on the part of the otherwise myopic incumbents to try to (re)-build the post-independence state drawing on the experiences and values produced during the liberation struggle. The Eritrean National Service (ENS) was originally conceived as a legitimate mega project of social engineering for nation-building and common national identity construction. Initially, the majority of citizens received it enthusiastically, but after the 1998–2000 Border War, it has become open-ended. Over time, it has become a cancerous growth that has been eating into the Eritrean polity reflected in the severe haemorrhage of the country’s single most important resource—labor. This in the context of bad governance and poor economic and human rights performance engendered what Albert Hirschman in a different but similar context refers to as “objectionable state of affairs” that prompt a variety of responses. The disaffected may stay put to exercise voice and fight for political, social, and economic change; emigrate abroad in search of protection and a better life, which Hirschman refers to as the exit option; and abstain from taking any deliberate action hoping that things would get better in spite of the decline. He refers to the latter response as loyalty. The major part of the chapter is a critique of the Hirschmanian framework. The findings show that in spite of its versatility and sophistication, the dichotomisation of the three conceptual building blocks of the Hirschmanian framework—exit, voice, and loyalty—seem to limit its explanatory power when applied to the recent Eritrean exodus. When interrogated on the basis of the Eritrean data, not only are the lines between exit, voice, and loyalty fluid, blurry, and continuously shifting, but also their effects are mutually reinforcing rather than counteracting each other. The data gathered for this essay show that it is more fruitful to conceptualise the relationships between exit, voice, and loyalty in terms of nexuses rather than dichotomies.


Eritrea is seeing its future walk away. 

—Stevis and Parkinson (2015) 

For a country and society that was “rising from the ashes” of a devastating thirty-year War of Independence, which on the one hand, destroyed the pre-existing rudimentary social and physical infrastructure, and on the other, brought the disparate ethno-linguistic and religious groups closer than ever before to face a mightier common enemy, it was appropriate on the part of the otherwise myopic incumbents to try and (re)-build the postindependence state, drawing on some of the positive experiences and values produced during the liberation struggle. However, although it may sound politically incorrect to point this out, buried in these endlessly glorified experiences and core values is the culture of intolerance, hostility to democratic dialogue, and propensity to quash dissent extra-judicially. Not only has this been the single most important Achilles heel of the postindependence situation, but also of the national service. The Eritrean National Service (ENS) was originally conceived as a legitimate mega project of social engineering for nation-building and common national identity construction. Initially, the majority of citizens received it enthusiastically, but after the 1998–2000 border-war, and the introduction of the wofri Warsai-Yikealo in May 2002, it became open-ended and its popularity diminished gradually and its damaging consequences became the major drivers of forced migration and destitution. Over time, it has generated into forced labor resulting in severe hemorrhage of the country’s single most important resource—labor. 

After providing a succinct background to the Eritrean National Service, the chapter discusses briefly its deterioration into forced labor and the degree of militarization permeating the social and political landscape of the country. It presents briefly the competing theories on migration, namely: push-pull; new economics of labor migration; network theory; and the exit, voice, and loyalty framework. More importantly, it reconceptualizes the three building blocks of the Hirschmanian framework in terms of nexuses rather than dichotomies. 

Background to the National Service

Soon after it took over power in May 1991, the Provisional Government of Eritrea (PGE) was struck by the perceived differences in the level of commitment to the project of nation building between the youth who grew up in the areas controlled by the EPLF (the Eritrean People’s Liberation Front) and the Derg. The PGE and the EPLF feared that the core values that determined the successful outcome of the liberation struggle might be lost once the shooting stopped, unless a mechanism that warrants continuity in the context of change was devised. The idea of national service was conceived as a mechanism of preserving and transmitting the treasured national core values developed during the liberation struggle to the present and future generations. That was the reason the proclamation on the ENS was among the first decreed by the PGE. The preamble of the proclamation on Eritrean National Service states: 

The people of Eritrea fought a bitter war for thirty years and paid a heavy price to relieve the Country and the people from darkness of colonialism, an all-out destruction, pain and to attain freedom and sovereignty. This and future generations have a historical responsibility to fulfil the will of thousands of martyrs and ensure the continuity of the Country’s freedom and sovereignty. To enable carry out this sacred duty, it is found to be essential to promulgate and establish National Service. 

Accordingly, all Eritreans—women and men—between 18 and 40 years old are required to perform 18 months of national service.[1] However, according to Proclamation 11/1991, only a limited number of citizens, particularly the unemployed youth were targeted. Proc. No. 11/1991 was repealed and replaced by Proclamation No. 82/1995, which eliminated most of the exceptions and exemptions save the veterans of the liberation struggle, the physically unfit and mentally infirm persons. Persons with physical disability are exempted from military training, but not from national service. 

Military training was introduced for the first time in May 1994. Initially, the ENS consisted of six months military training at the Sawa Military Camp in western Eritrea and 12 months participation in nation building. During the first six months, conscripts receive military training. They also receive political socialization and indoctrination as opposed to political education. The aim of the former is to socialize the conscripts into the values of the liberation struggle, such as patriotism, sacrificial nationalism (on the latter, see Hogan 2009, 2014; Bernal 2014) and subordination of self-interest to national interest and obedience to authority. As the former Minister of Defence, Sebhat Ephrem, said, “It is through the national service that we intend to transfer the noble values developed during the armed struggle—steadfastness and dedication—to coming generations” (1995) (emphasis added). Its architects believed that the ENS would foster common national Eritrean identity and powerful commitment to the project of national unity and nation building (Isaias Afwerki 2004; Sebhat Ephrem 2008).

Militarization of Education and the ENS

When war broke out in May 1998 between Eritrea and Ethiopia, the four demobilized cohorts were remobilized and those who joined after May 1998 stayed on the grounds that war may break out (Isaias Afwerki 2004; Sebhat Ephrem 2008). In May 2002, the government also introduced the Warsai-Yikealo Development Campaign (WYDC), which officially rendered the ENS indefinite (HRW 2009; Kibreab 2009b; USDS 2010; Connell 2015). A study conducted by the author among deserters living in UK, Switzerland, Sweden, Norway, South Africa, and Kenya found that on average, they served 6 years instead of 18 months before they fled. Many of those who remained have been in the ENS for over 20 years. Many of them have family members who have been stuck in the ENS for more than 20 years. As a result, Eritrea is among the most militarized countries in the world (Tronvoll and Mekonnen 2014; Hirt and Mohammad 2013). Even the educational system is militarized (Dorman 2004; Muller 2008; Andebrehan Welde Giorgis 2014). After the ENS became open-ended, it grew unpopular and many began to flee. To pre-empt this, in 2003 the government decided to increase the duration of secondary education by one year and to relocate the final year (year 12) students to the Sawa military camp where students combine military training with final year education in which the former is prioritized (Fisher 2004). The authorities thought that relocating the students in the final year to the Sawa military camp would enable them to control their freedom of movement and monitor their activities. Berhe, a former student at Sawa told the author, 

The Warsai School is not like any other school. It is mainly a military training camp. When I went to Sawa in 2006 after completing 11th grade, I thought I was going to study and prepare myself to go to college. To my dismay, on top of the three months military training, we were required to undertake military training every morning. We could also be called at any time to undertake other tasks unrelated to our studies. We were escorted by soldiers wherever we were. There was really no time to study.[2]

Abdelkadir, another former student at Sawa said, “We were not students at Sawa. We were soldiers under the control of the Ministry of Defence.”[3] Some of those in Year 12 are under 18 years. It was because of this, UNICEF expressed concern that the ENS may constitute a violation of the rights of the child (in Fisher 2004). 

The degree of militarization in the country has intensified since the government introduced the people’s militia in March 2012 in which citizens between 50 and 70 are forced to undertake militia duties several times a week without remuneration. Such duties include working as armed guards and unpaid labor on public work projects (Tronvoll and Mekonnen 2014; Kibreab 2014; UNCOI 2015). 

Deterioration of the ENS into Forced Labor

After the 1998–2000 Border War between Eritrea and Ethiopia and the introduction of the WYDC in May 2002, the ENS has become open-ended (HRW 2009). As a result, the grand social engineering scheme that was introduced as a legitimate overarching program of nation building, post-conflict (re)-construction, and common Eritrean national identity construction has deteriorated into forced labor (Kibreab 2009b; UNCOI 2015; ILO 2010). Those who fail to comply are subjected to degrading treatment (see UNCOI 2015; AI 2004; HRW 2009; Kibreab 2013). The ENS has been characterized as forced labor by numerous analysts (Kibreab 2009b; Tsurkov 2014; Andebrahan Weldegiorgis 2014; Woldemikael 2014; Connell 2014) and human rights organizations (HRW 2009, 2013; UNCOI 2015). The Foreign and Commonwealth Office, for example, stated, “We will continue to press for an end to obligatory and indefinite national service and to compulsory and onerous civilian militia duties (such as guarding, patrolling and dam-building), all of which could amount to forced labor” (2015) (emphasis added). The ILO has also considered the question of whether the ENS constitutes forced labor and concluded that in its current form, it does (ILO 2014).

The UNCOI characterizes the ENS as forced labor and a modern form of slavery. It stated: “The Commission finds that by conscripting them [citizens] into an indefinite period of national service, the Government reduces its citizens to mere duty-bearers, negating their role as right-holders who enjoy individual rights and freedoms recognized under international human right law. The Government of Eritrea refuses to treat its citizens as human beings with rights, dignity and a free will.” (2015)[4] (emphasis added). The Commission further observes, “. . .the indefinite duration of national service; its terrible conditions and treatment including arbitrary detention, torture, sexual and gender-based violence, forced labor. . . make national service an institution where slavery-like practices occur” (ibid.).

Theoretical Perspectives

Although it is unrealistic to provide a full account of the relevant theories that may explain the recent migration of Eritrean conscripts, draft evaders, and young children; in the following only a brief reference is made to the most relevant ones. A survey of the extant literature on migration shows that there are a number of theories that could be used to explain the recent flight of large number of Eritreans. These include firstly, the push-pull theory in which migration is conceived in terms of push factors in a place or country of origin, such as poverty, unemployment, high population growth, political repression, etc. and pull factors in the place or country of destination, such as higher income, greater employment and education opportunities, greater political freedom and welfare, etc. (King 2012; Massey et al. 1993). Secondly, the new economics of labor migration in which the decision to migrate is not made by individual actors only, but rather by families or households who expect to maximize income and reduce risks (King 2012; Massey et al. 1993). Thirdly, network theory represented in sets of interpersonal ties that connect migrants and non-migrants in countries of origin and destination based on kinship, friendship and commonplace of origin. These ties “. . .increase the likelihood of international movement because they lower the costs and risks of movement and increase the expected net returns to migration. Network connections constitute a form of social capital that people can draw upon to gain access to foreign employment. . .” (Massey et al. 1993, 448). Although all these theories are evidently relevant for the explanation of the recent exodus from Eritrea, given the limited scope of the paper, it is impossible to deal with their aspects comprehensively. Instead, the focus will be on the Hirschmanian theory of exit, voice, and loyalty, which to some extent incorporates some elements of the aforementioned theories. 

Albert Hirschman’s ground-breaking theory of exit, voice, and loyalty (EVL) has inspired a flood of theoretical and empirical studies (70). The central thrust of his theory is that when participants are faced with actual or perceived deteriorating conditions, they respond either by exiting to escape a disagreeable condition or exercising voice to improve the situation. A third least understood concept in the Hirschmanian framework is loyalty, which some analysts perceive as a third behavioral response to dissatisfaction and others view it as “an affective moderating variable that influences the choice between exit and voice” (Graham and Keeley 1992, 191). 

Although the other theories presented earlier can be used to explain the migration of Eritreans, I have chosen Hirschman’s framework as enriched by some of its critics and analysts for its versatility. Among the theories mentioned in the preceding paragraphs, the EVL framework is the only one that incorporates some powerful elements that provide an opportunity to examine the ambivalent and multi-layered responses of Eritreans to the deteriorating conditions in the context of the country’s culture of resistance. In view of the limited space, an elaborate exposition of why the EVL framework is preferred to the other theories is not feasible. 

Hirschman defines voice as “. . .the act of complaining or of organizing to complain or to protest, with the intent of achieving directly a recuperation [recovery] of the quality that has been impaired.” Once the organization is made aware of its failings reflected in terms of citizens voting with their feet, it may introduce changes in order to eliminate or minimize the cause of deterioration. However, this is only plausible in countries where governments or organizations are responsive to public dissatisfaction. Hirschman states, “In all these respects, voice is just the opposite of exit. It is a far more ‘messy’ concept because it can be graduated, all the way from faint grumbling to violent protest; it implies articulation of one’s critical opinions. . . it is direct and straightforward rather than roundabout. Voice is political par excellence” (1970, 16). 

The corollary is that disaffected citizens can exercise voice at different levels from mere complaint to make the state aware of its deteriorating performance to open protest aimed at bringing pressure to bear on the incumbents. Hirschman argues, “To resort to voice, rather than exit, is for the. . . member to make an attempt at changing the practices, policies. . . of the organization to which one belongs” (1970, 30). Nevertheless, the Eritrean case demonstrates that this strategy is not pursued at any cost. Rational actors weigh the costs and benefits to be had from their actions or inactions. If the costs of complaints and protests are risky, the alternative options are either silence in oppression or “voting with one’s feet.” The latter option is not necessarily resorted to escape from politics, but rather to engage in politics safely. In the context of the current state of affairs in the country, political activity in safety and dignity is only possible abroad. 

In the Hirschmanian framework, voice refers to “. . . any attempt at all to change, rather than to escape from, an objectionable state of affairs. . . through various actions and protests” (ibid.) (emphasis added). He notes that the process of deterioration in the performance of a state is likely to trigger “certain counter forces” (1970, 15). This counter force informs the incumbents of their failings either through voice or exit. Although Hirschman later modified his theory on the relationship between exit and voice, he argues, “Easy availability of exit was shown to be inimical to voice, for in comparison with exit, voice is costly in terms of effort and time.” Moreover, to be effective, voice often requires group action and is thus subject to all the well-known difficulties of organization, representation, and free riding (1998, 12). 

In his original theory, Hirschman conceptualized exit as the antithesis of voice by arguing, “The presence of the exit alternative can. . . atrophy the development of the art of voice” (1970, 43) (emphasis in original). However, exit does not necessarily mean the end of voice. As we shall see later, it can also be a harbinger of voice. In the rapidly globalizing world, exit facilitates the exercise of voice amongst transnational communities. This is more so in fear and risk-ridden societies such as Eritrea where the exercise of voice may result in life-threatening consequences. Bert Hoffmann (2008, 4) for example, states, “The changing nature of migration in the 1980s and 1990s gave rise to the emergence of the transnationalism paradigm in migration studies.” Hepner’s (2001, 2008), Bernal’s (2014) and Conrad’s (2010) studies clearly show that among Eritrean refugees, asylum-seekers, and migrants, exit instead of atrophying voice has been the single most important instrument for the development and consolidation of vibrant Eritrean diaspora organizations and websites which are key to the exercise of voice and for coordinating protest (Bernal 2014). This has been true both among the older generation of diaspora Eritreans who fled during the independence war and those who fled the postindependence state (see Hepner 2009).

The third construct in Hirschman’s innovative work, loyalty, is ambiguous as well as less developed in his framework. He argues that loyal participants “suffer in silence, confident that things will soon get better” (1970, 38). A loyalist is a person who does not exit regardless of the degree of dissatisfaction. To Hirschman, loyalty is that unexplainable “special attachment” or feeling that supresses the propensity to quit in favor of the alternative that provides similar or better benefit. Some analysts argue that loyalty seems to be an irrational decision because it is contrary to expectation to remain within an entity that is in the spiral of decline. 

Hirschman conceives loyalty as a behavior, which motivates an individual to support a deteriorating organization or state. He states, “the likelihood of voice increases with the degree of loyalty” (1970, 77). Brian Barry (1974, 98) states that loyalty does not normally mean a disinclination to “leave a collectivity, but rather a positive commitment to further its welfare by working for it, fighting for it and—where one thinks it has gone astray—seeking to change it.” This view implies that voice “is built into the concept of loyalty” (ibid.). Birch on the other hand, argues that loyalty “belongs to a family of concepts which. . . includes allegiance and fidelity” (1975, 74). The corollary is that loyalty rather than increasing voice, diminishes it. 

There are analysts who perceive loyalty as a “passive constructive behavior,” such as “being quietly supportive and being patient” (Rusbult, Farrell, Rogers, and Mainous 1988). Leck and Saunders state that loyalty is “an act of waiting patiently for conditions to improve” (1992, 220). A loyal participant in an organization, firm, or state is one who is reluctant to exit in response to a decline of performance. There is an agreement among some analysts that loyalty connotes some form of positive affective attachment that binds an individual to a state, organization, or firm (Graham and Keeley 1992). It is assumed that the degree of attachment is so powerful that it outweighs the propensity to resort to the exit option. Others perceive it as an attitude, rather than as a behavior, that moderates or conditions the use of exit or voice (Withey and Cooper 1989; Leck and Saunders 1992). 

Hirschman argues that a loyalist is someone who doesn’t exit in response to dissatisfaction (1970, 38). In the Hirschmanian conception, a loyalist is an optimist who hopes that “someone will act or something will happen to improve matters” (1970, 78–79). Loyal customers or citizens try to right the wrongs of the state by exercising voice with the aim of effecting change through protest where there is freedom of expression or where citizens are willing to face the consequences. The higher the degree of loyalty, the greater the propensity to stay put to exercise voice. However, this is only possible in democratic societies. 

Although the concept of loyalty was not adequately expounded in Hirschman’s framework, its ambiguity and messiness is more apparent when used as a tool of analysis to explain Eritreans’ response to dissatisfaction. As seen earlier, although the exact number of those who flee from the dissatisfying conditions in Eritrea is unknown, as seen before, hundreds of thousands of deserters, draft evaders, and young people approaching the age of conscription have been voting with their feet to escape from repression, as well as to engage in politics in the context of safety. 

A closer scrutiny shows that there is a gap in literature on motivation for migration. The literature seeks to portray asylum-seekers and migrants as the antithesis of political engagement. Eritrea provides us an opposite example in which strategy of flight is adopted to facilitate protest and political engagement. This is greatly facilitated by information and communication technology. 

Other analysts have added a fourth response to dissatisfaction, namely neglect (Rusbult, Zombrodt and Gunn 1982). They describe neglect as “. . . refusing to discuss problems. . . just letting things fall apart” (ibid., 1231). Withey and Cooper observe that “Neglect differs from loyalty in that it is not directed at recovery of the relationship. Rather, the individual responding with neglect implicitly accepts that recovery is not going to happen” (1989, 522). This may explain the response of conscripts and others who have remained in the country instead of “voting with their feet” in spite of the deteriorating condition. 

In view of the fact that flight from postindependence Eritrea is neither an escape from repressive politics nor a political act, but both, and that those actions are multi-faceted, it is necessary to reframe the acclaimed Hirschmanian framework of exit, voice, and loyalty in terms of nexuses rather than dichotomies to explain the Eritrean exodus. When interrogated on the basis of the Eritrean data, not only are the lines between exit, voice and loyalty fluid, blurry, and ever-changing, but their effects are also mutually reinforcing rather than counteracting each other. That is why it is suggested that it is more fruitful to conceptualize the relationships between three building blocks of the Hirschmanian framework in terms of nexuses rather than dichotomies. 

In the following, a pithy description of the methods used to gather data for the chapter is presented.

A Note on Method and Data Sources

This chapter is written as part of another on-going major research project which examines, inter alia, the drivers of displacement, the transformative effects of the ENS, as well as its impact on the economy, nation-building, national identity construction, and national unity and defence capability. While researching the ENS in the past three years, I interviewed 190 respondents residing in different EU+ and two African countries and 38 key informants, 11 of whom were females, selected on the basis of chain referral or snowball sampling. Although the chapter draws on these data, the major part of the data are derived from in-depth interviews conducted with former conscripts in Geneva,[5] London, and Stockholm, who fled from the ENS. Those from Geneva were interviewed in June 2015 whilst those in London between April 2014 and January 2016. The interviews in Stockholm were conducted in November 2015. The interviewees were identified through chain referral sampling. An elder who arrived recently for a visit was also interviewed in London.[6] All the names and their characteristics are altered to insure anonymity. Muslim, Christian, female, and male names indicate the religion and sex of the interviewees. Using data gathered from personal interviews and secondary sources and drawing some insights from the Hirschmanian framework of exit, voice, and loyalty, this chapter examines how the disaffected have been responding to the disagreeable circumstances in the country. 

Conscripts’ Responses to the Open-ended ENS: Discussion and Findings

The short description of the situation in Eritrea presented at the beginning of the chapter shows the inauspicious reality facing hundreds of thousands of conscripts and nationals approaching the age of conscription, including young children far below the age of national service (see Women’s Refugee Commission 2013). As stated by Amnesty International (2015), “Eritreans, many of them children, are refugees fleeing a system that amounts to forced labor on a national scale.” The conditions are by any reasonable standard objectionable. It is therefore interesting to examine how those who are affected have been responding to the situation. It is important to bear in mind that people—including those whose livelihoods have been ruined by the open-ended national service—do not respond in the same way. The responses cannot therefore be determined a priori. Responses to the objectionable circumstances, no matter how severe are likely to be varied and wide-ranging. 

Even when Eritrea was faced with an existential threat during the War of Independence (1961–1991), in which hundreds of villages were razed to the ground, thousands of innocent civilian lives were lost, large swathes of cropped land were burned down, and tens of thousands of livestock were massacred by the marauding Ethiopian military (see Kibreab 1987a, b), those affected did not respond in the same way. A study conducted by the author among Eritrean refugees in the five land settlements in Qala en Nahal and two wage-earning settlements in Kashm el Girba and Kilo 26, eastern Sudan, show that the people in the affected areas responded in at least six different ways to the imminent threat of violence (Kibreab 1987a, b). 

Their responses included a “wait and see” stance and stayed put in spite of the imminent dangers; some moved behind the frontlines into the liberated areas; others relocated themselves to safer areas to become IDPs; many, especially the youth, joined the Eritrean Liberation Front (ELF) and the Eritrean People’s Liberation Front (EPLF) to fight; and some members of the vulnerable groups moved into the “strategic villages” established by the Ethiopian military intended to deprive the fronts of a mass base. Flight across the border was only one of these six options (Kibreab 1987a, b). The decision to flee or not is a complex process mediated by a multiplicity of inextricably and mutually reinforcing and/or counteracting factors. The implication of this finding on the present study is that in spite of the common hardship experienced by conscripts and other nationals in the onerous and the never-ending national service, their responses cannot be generalized. 

The dichotomization of the three conceptual building blocks of the Hirschmanian framework—exit, voice, and loyalty—seems to limit its explanatory power when applied to the postindependence Eritrean reality. Not only are the lines between exit, voice, and loyalty continuously shifting, but also more importantly, their effects are also mutually reinforcing rather than counteracting each other. It is more fruitful therefore to conceptualize the relationships between exit, voice, and loyalty in terms of nexuses rather than dichotomies. Given the successful history of the liberation struggle in which the Eritrean people fought together, setting aside their ethno-linguistic and religious differences to confront their giant southern neighbor, which enjoyed considerable military and financial assistance first from the United States of America and later from the Soviet Union and its allies, it is reasonable to expect those who are aggrieved by the current dissatisfying condition to stay put and fight for change and transformation. However, as the data presented below indicate, the tendency among Eritrean conscripts and draft evaders has been to exit from the dissatisfying conditions rather than exercising voice to change the situation. Therefore, the single most important question this chapter addresses is, why do Eritreans flee rather than staying to organize opposition to repression? 

Given the Eritrean government’s severe restrictions on freedom of movement and emigration, the overwhelming majority leave the country illegally and cross into Ethiopia and Sudan en route to the EU+ countries. The data elicited from the informants show that the large majority make up their mind before departing from Eritrea that their destinations are far beyond the first countries of first arrival, namely Ethiopia and Sudan. 

In December 2015, there were 131,660 Eritrean asylum-seekers and refugees in Ethiopia (UNHCR 2015). The corresponding figure in Sudan was 125,530, of whom 89,800 were old caseload from the pre-independence period (UNHCR 2015). The large majority of Eritrean asylum-seekers and refugees in the two countries can be described as birds of passage. The data gathered from hundreds of conscripts and draft evaders who fled Eritrea since 2000 via the two countries, show that the overwhelming majority regard the two transit countries as stepping-stones for further emigration rather than destinations. The figures in the two countries do not therefore show the actual number of Eritrean refugees and asylum-seekers living in the two countries. They only show the total number of Eritrean asylum-seekers who crossed into the two countries. Before the wall Israel erected on the Egyptian-Israeli border stemmed the flow of Eritrean asylum-seekers in 2013, many Eritreans tried without success to apply for asylum in Israel. 

Those in Ethiopia cross into Sudan and join their brethren in the long and dangerous sojourn to the European shores via the Sahara Desert and Libya across the Mediterranean Sea where an unknown but considerable number face the risk of death. In Israel, there are a total of 46,437 African asylum seekers and 33,899 are from Eritrea (ARDC 2016). An unknown, but undoubtedly large number have left Israel to avoid the threat of deportation and state-sponsored xenophobia (Sheen 2015). 

According to data obtained from the European Asylum Support Office (EASO) in Velletta, the total number of Eritrean asylum-seekers in the 28 EU+ countries in 2012 was 11,990, and the corresponding figures for 2013 and 2014 were 20,310 and 46,735, respectively.[7] Between 2000 and 2014, a total of 133,385 Eritreans sought asylum in the EU+ countries. Eurostat data also show that between 2008 and 2014, a total of 7,840 underage Eritrean children arrived in the EU+ countries seeking asylum.[8] The increase in the number of minors between 2013 (1,005) and 2014 (4,485) was staggering, i.e. 346 per cent over a period of one year. The increase for the total number of Eritrean asylum-seekers between 2013 and 2014 was 130 per cent. During the first half of the 2015, the number of Eritreans reaching the EU in comparison to the other five major refugee-producing countries declined considerably.[9] Between 2015 Q1 and Q4, a total of 30,120 Eritreans sought asylum in the EU member states.[10] These figures do not include those who sought asylum in Norway, Sweden, and Switzerland. 

These figures show that a considerable proportion of those who were dissatisfied with the deteriorating conditions have resorted to flight rather than staying put to exercise voice and to fight for change. However, as we shall see later, in the context of the repressive regime, protest can only be exercised from abroad. Therefore, as was mentioned earlier, it is wrong to perceive the flight of the tens of thousands as escape from politics, but rather as a means of engaging in politics from a place of safety. In spite of the fact that many have “voted with their feet” to exit from the objectionable conditions in the country, there are many others who have remained. We don’t know with certainty whether their decision to stay put is due to loyalty, fear of retaliation, neglect, or surrender. The data gathered from interviewees indicate that fear and subsistence insecurity are the major factors that limit some people’s movement. On the basis of the data elicited from informants, an attempt is made to come to grips with these complex questions in what follows. 

Why Flight instead of Voice and Protest?

When I asked the key informants, who fled from the ENS, why they resorted to flight instead of staying put to fight for change, all but one said that the whole country is in the grip of intense fear. In June 2015, I interviewed four Eritrean asylum-seekers in Geneva, who crossed the Mediterranean Sea in rickety boats, and asked them why they fled rather than remaining at home to protest. Although the interviews were conducted independently, the answers each of the interviewees gave were astonishingly similar. All said independent of each other that the exercise of voice is too dangerous where even family members are perceived to spy on each other. Abel said, “Our society is in a state of paralysis due to fear.[11]

The generalized state of fear, the complete absence of any form of autonomous civil society and/or political organization, lack of freedom of speech and expression and movement, as well as the absence of rule of law and due process were said to be the key factors that have atrophied voice inside the country. This is confirmed by the UNCOI’s report in which it is stated:

. . .the Government has created and sustained repressive systems to control, silence and isolate individuals in the country, depriving them of their fundamental freedoms. It shows how information collected on people’s activities, their supposed intentions and even conjectured thoughts is used to rule through fear in a country where individuals are routinely arbitrarily arrested and detained, tortured, disappeared or extra-judicially executed. (2015, para. 25)

This does not mean, however, that those who are frightened and silenced within the country and resort to flight have given up on the exercise of voice. On the contrary, for those who want to engage in fighting against the system, relocation in democratic countries provides golden opportunities to exercise voice from safe locations. 

Every interviewee but one said that the conditions in the country are so dangerous that any attempt at criticizing the incumbents would be suicidal. Kemal said: 

The people in power are inimical to all forms of autonomous political or civil society organizations. Fighting requires an organization. One cannot fight alone. There is no space for exercising voice or protest. Whoever tries to create a space for voice is dealt with mercilessly. Many citizens, including journalists, high and middle-ranking military officers, former ministers and many citizens have disappeared for no other reason, but for exercising voice. Whoever utters a critical voice disappears without trace. The conscripts are aware of this and hence they opt for flight rather than staying to fight.[12]

This has created a climate of fear which has stifled any attempt at protest or exercise of voice from within. Instead, those who are committed to change and transformations do so from a distance, i.e. in exile, which suggests, contrary to Hirschman’s assertion, that exit, instead of atrophying voice, enhances it. 

Haile added, under the existing circumstances, “any attempt made to criticize or to fight the government amounts to signing one’s suicide note.”[13] He further said, “In Eritrea, whoever utters a critical voice is detained and disappears. This discourages people from organizing themselves to fight for political change. The state of fear in the country is paramount. People are afraid of their own shadows.” These accounts suggest that the environment is inimical to voice. But as noted earlier, this does not mean exit is the end of voice. On the contrary, in the Eritrean reality, exit facilitates the exercise of voice. However, not all of those who resort to the exit option necessarily exercise voice or fight against the government from a distance. Therefore, it is wrong to assume that everyone who resorts to flight have lost their right to exercise voice or are necessarily ready and willing to oppose the government they fled from. The reality is messier than the neat theoretical postulation purported by the Hirschmanian framework. Hirschman states, “In all these respects, voice is just the opposite of exit” (1970, 16). In our case, there are situations in which the two are in tandem or exit is sine qua non for the exercise of voice. 

There is evidence to show that the most vocal opponents of the “objectionable state of affairs” in the country are in the diaspora. They are interconnected through the Internet and other social media channels to communicate with each other and to organize protest. For example, on June, 26 2015, between 6,000–8,000 members of the Eritrean diaspora from all over Western Europe held a mammoth demonstration in front of UN Human Rights Council’s headquarters in Geneva. Most of the demonstrators were exiters from the ENS. A corresponding rally also took place in New York on October, 29 2015 to express support for the UN Commission of Inquiry. The Eritrean government also has its own boisterous supporters among the Eritrean diaspora, including few exiters from the “objectionable state of affairs” precipitated, among other things, by the indefinite ENS. For example, on June 22, 2015, about 2,000 of them held a demonstration in Geneva in support of the Eritrean government and against the report of the UN Commission of Inquiry. On June, 21 2016, according to sources close to the government, about 7,000 diaspora Eritreans held a demonstration in front of the headquarters of the UN Human Rights Council’s in Geneva.[14] On June 23, 2016, a huge demonstration attended by over 10,000 youth,[15] the overwhelming majority of whom were deserters and draft evaders, took place in front of the UN Human Rights Council’s headquarters in support of the UNCOI’s report. According to an eyewitness interviewed by the author: 

The major difference between the two demonstrations was that the anti-government demonstration was organized and attended pre-dominantly by: young men and women who fled from Eritrea to desert from and/or to avoid the open-ended national service whilst the pro-government demonstration was organized and attended by old men and women who fled the country during the war of independence and their children born in the diaspora who are stuck in the “glorious” past.[16]

This shows that in the rapidly globalizing world, exit, instead of atrophying voice, may be instrumental in facilitating its expression. 

One of the key research informants, Hagos,[17] offered a counter-narrative that lends support to Hirschman’s theory that postulates that the presence of the exit option atrophies “the art of voice” (1970, 43). Hagos, a graduate from the University of Asmara with an MSc degree from the UK, deserted national service after 12 years. He said, 

People talk about the high-risk political activity or voice represents under the present situation in Eritrea, but this is a lame excuse. No revolution takes place under convivial [safe] environment. If conditions are favourable, there is no need to fight. Our people fought and won in spite of severe adversities.

He continued, 

The reason we are not fighting to remove the dictator is because it is easier to remove ourselves than getting in serious trouble trying to get rid of him. In comparison to the dangers and imminent risks of death or indefinite incarceration that can result from exercising voice and protest; the dangers one faces in flight although potentially menacing [lethal], those who make it may lead a fulfilling life.

It is worth noting that not only do some of those who make it across the Mediterranean and farther north lead a fulfilled life, but also, they are able to send remittances to their desperate relatives at home, fuelling an irresistible drive among those who remain behind to follow suit in pursuit of the imagined nirvana, discounting or underestimating the intervening obstacles, including death, kidnapping, rape, and disappearances. 

The corollary in Hagos’ view is that had there been no opportunity for exit, the disaffected would have stayed put to fight rather than voting with their feet. He further said, “Flight is a make or break project, but the consequence of fighting or protesting in Eritrea at present amounts to slow and agonizing death.” Probably, he is referring to the 11 members of the G15 and journalists who have been languishing incommunicado in the Eiraeiro dungeon since 2001. He further said, “The threat of death and injury did not deter our fathers and mothers from fighting for independence. The reason why we are not emulating their heroic experience is because the option of flight is less costly and more rewarding for those who make it.” This, however, is a question of perception. If we were to ask the families who lost loved ones in the border crossing between Eritrea and Sudan/Ethiopia, the Sinai and the Sahara deserts, as well as the Mediterranean Sea, or whose loved ones have disappeared in between without trace, they are unlikely to think that flight is less costly than being locked up, tortured, or incommunicado detention. These are, however, afterthought reflections. 

At present, the consequence of political activity as an expression of voice is severe in the country. This can be demonstrated, inter alia, by the experience of the eleven members of the G15 who disappeared when they were picked up from their homes in early morning raids in September 2001 at the behest of the personal ruler, Isaias Afwerki, for no other reason but for exercising voice in terms of asking him to convene the long overdue regular sessions of the Central and National Councils. In an open letter sent to members of the ruling party which leaked to the private newspapers, the G15 described the crisis that had gripped the country as follows: “The problem is that the President is conducting himself in an illegal and unconstitutional manner, is refusing to consult, and the legislative and executive bodies have not performed their oversight functions properly” (G15, Open Letter). They have never been brought to court or been seen by their relatives or public since September 2001. This was mentioned as being a major disincentive and a cause of fear of government retaliatory action against the exercise of voice by most of the informants. Kahsai,[18] for example, said, “If the cruel man [president Isaias] has gleefully caused the disappearance of his closest friends and comrades in-arms, such as Haile Durué, Sherifo, and others, do you think he would hesitate to wipe out whoever criticizes his tyrannical rule?” Saleh[19] also said, “Isaias and his henchmen never forgive and forget and most Eritreans are aware of their depravity. Such knowledge is the single most important factor that has silenced their voice notwithstanding their siqiat (suffering).” 

Disappearance has become a norm. Whoever is detained is neither charged nor released. This has wreaked havoc among the youth. Fear is the major disincentive to the exercise of voice. The Report of the Commission of Inquiry on Human Rights in Eritrea for example, found that “Eritreans live in constant fear that their conduct is or may be monitored by security agents, and that information gathered may be used against them leading to arbitrary arrest, detention, torture, disappearance or death” (see also Financial Times 2009).[20]

Exit as an Instrument of Organizing Resistance and Protest

As we saw earlier, a large number of Eritreans who are within and approaching the age of conscription have been fleeing the country from 2004 onwards. Many have also been deserting national service. Although there is little doubt about the severe hemorrhage the country has been suffering, it is important to guard against misconstruing this movement as evidence of disengagement from politics. On the contrary, given the stifling political environment in the country, flight from Eritrea and relocation to democratic countries provides excellent opportunities for engaging in politics. The severe state of fear pervading the social and political landscape in the country has eliminated any opportunity for organizing protest on home soil. Those who are determined to fight for progress have therefore been “voting with their feet” to engage in political activities for democratic change. 

To the question “why don’t you organize and protest instead of fleeing,” Saleh said, “Can you clap with one hand? Voice and protest are collective action. No collective action is possible without organization and there is no room to organize”[21] in Eritrea. I met the interviewees at the demonstration in Geneva held in support of the UNCOI’s report, which demonstrates that exit has enabled them to engage in visible collective political action and to exercise voice loudly unconstrained by fear, as was the case in Eritrea. Had they stayed put, they would have not been able to exercise voice and demonstrate for change without compromising their safety. Hirschman’s claim that “voice is the opposite of exit” is not backed by the findings. Instead, they show that exit has been for the exercise of voice among the Eritrean diaspora, including those who fled to escape the open-ended national service. 

Elias, 27 years old, was conscripted through the Warsai School at Sawa when he completed 11th grade in Asmara. After receiving initial military training, he later combined it with education. At the end of the academic year, he sat his matriculation and failed. He was assigned to the army where he also participated in construction of roads, bridges, houses belonging to the government, the ruling party, and high-ranking military officers without remuneration. He said that he was only allowed to visit his parents three times throughout the period he was in the army.[22] He said he knew in great detail the potential dangers that lay ahead inside Eritrea, at the border, in eastern Sudan, in the Shagarab refugee camp, in the Sahara Desert, in Libya, and the Mediterranean Sea. 

When asked, “If you were aware of all the dangers, how could you sleep walk into them?” He said, “I had no choice. I was literally a slave to my commanders in the national service. My life was in their hands and I had nowhere to turn to seek redress against these abuses. The cruelty I suffered eliminated my ability to feel fear. I have seen the worst in the ENS. Nothing, including death, can be worse than what I have been through.” He said, “in national service, the pains and predicaments are excruciating and endless. The potential risks I expected to face in flight could be deadly, but they don’t last forever. You either die or survive.” The bitter experience of suffering and deprivation seem to have rendered him and others in his situation risk averse. 

Asked if he knew about the rescue operation in the Mediterranean Sea, he said, “Yes, but so what? There were still many of my brothers and sisters that perished in the sea in spite of Mare Nostrum and after. My decision had nothing to do with the rescue operations in the Mediterranean.” The question that arises in connection to his account is that, if the potential risks of death in flight did not deter him from fleeing, why was he unprepared to face the risks associated with exercising voice in Eritrea? He said, “If I am detained in Eritrea, it may be the end of me, but if I flee, there may be an opportunity for me to join the change-seeking organisations abroad.” This shows that he was already aware that under the existing circumstances, exit is sine qua non for exercising voice. Kahsai also said, “the only Eritreans organizing against the regime are in exile.”[23] Hirschman’s assertion, ‘‘Easy availability of exit was shown to be inimical to voice. . .” (1998, 12) is not backed by the findings of the study. 

In his original theory, Hirschman conceptualized exit as the antithesis of voice by arguing, “The presence of the exit alternative can. . . atrophy the development of the art of voice” (1970, 43) (emphasis in original). The findings here demonstrate the converse. When asked why he fled instead of exercising voice to fight for change in Eritrea, Araya said, “There was no room to organize. We did not trust each other.” The reason for this is because of the overwhelming perception that “every other person in the country is a spy.”[24] This does not need to be true. It is enough if people believe it to be so. Interestingly, he is currently one of the key leaders of a youth movement organized to bring about political change in Eritrea.

Government Supporters among Exiters

As noted earlier, whether to exit or remain to exercise voice is ambivalent and multi-layered, and exit is neither an act of surrender not disengagement from politics. On the contrary, in the Eritrean case, exit has been instrumental in the Eritrean diaspora’s political engagement and protest. However, inasmuch as it is wrong to assume that all those who have been fleeing the country have given up on their right to exercise voice, it is equally important to underscore the fact that not all those who flee the country are necessarily ready and willing to fight against the Eritrean government. Their decision to flee the country may be a manifestation of their dissatisfaction with the status quo. But the concept of dissatisfaction is subjective and ambiguous and therefore not easily amenable to objective assessment. Thus, although the situation is dissatisfying, it is wrong to assume that all Eritreans perceive it as such. It is also equally wrong to assume that all those who stay put are necessarily satisfied with the situation. By the same token, it is awry to assume that all those who exit are necessarily dissatisfied with the political system. 

In the diaspora, although the large majority of the most raucous supporters of the government and the PFDJ are Eritreans who fled during the War of Independence (1961–1991) and their children born in exile, there are some who fled from the ENS and draft evaders who support the government notwithstanding the fact that they alleged that they fled from persecution when they filed their asylum-applications. They may have issues with the prevailing economic hardship and endless national service, but they do not blame this on the government. Instead, they externalize the causes by attributing them to the tesabaêti hailitat (Eritrea’s external enemies). In their eyes, the government is perceived as the victim rather than the culprit. Some of those who remained at home characterize members of the diaspora as cheerleaders of the regime. This is not far from the truth. Bettina Conrad summarizes the gist of the conversations she had with interviewees when researching her luminous PhD thesis—“We are the Prisoners of our Dreams”—in Asmara 2001 as follows:

The diaspora could use their money and influence to press for reforms, but they just clap their hands, no matter what lies the government tells them. They send their money and we have to bear the consequence (an allusion to the 1998–2000 Border War). So far away you don’t feel the heat of the fire (2010, 160). 

This is a clear indication of the fact that not all those who have stayed in spite of the disagreeable conditions are necessarily loyalists. In their view, the bulwark of loyalty is the diaspora. By the same token, it is equally awry to assume that those who are disloyal to the government, both among exiters and stayees, are disloyal to the state or to the country.[25]

Even some of the informants, including those who are deeply dissatisfied with the situation, are still reluctant to denounce the government because they feel it is unpatriotic to do so. In a long conversation with one of the key informants, Tsegai, it was easy to see that he was caught “on the horn of a dilemma.” He referred to the government as “mergem” (Summum malum or affliction) that has “wreaked havoc in the social fabric of our people.”[26] In spite of the strong language, however, he was reluctant to take a stand against it. The reason he gave was: “It is the only government we have and it is risky to abandon it when there is nothing to replace it with.” Tsegai is clearly a deserter, but is he a loyalist? If he is, his loyalty is not to the government, which he characterizes as an affliction, but rather to the state. 

There are many in the diaspora who are emotionally caught in between and this remains understudied. In light of the high price citizens paid in terms of loss of lives, property and forgone opportunities to have their own state, their attachment to the latter even when severely aggrieved by the incumbents, remains powerful. Willy-nilly, not only does this stance benefit the government, but it is also the result of its excessive obsession with securitization intended to instill anxiety and fear of the “Other.” This has engendered constant fear, bordering on paranoia, among many of its diaspora supporters. 

It is also wrong to assume that all those who exit from the apparent dissatisfying conditions are necessarily disloyal to the government. An unknown but undoubtedly a significant number of exiters who allegedly fled from persecutory treatment in the national service and at the hands of abusive commanders have turned around and supported the government. As Tanya Müller observes, “. . . choosing exit does not necessarily imply diminished loyalty or questioning the ideological legitimation of the state project” (2012, 794). However, her observation, “It [exit] is merely a sign that the political leadership has lost legitimacy. . .” is inaccurate. Most of the pro-government exiters currently residing in the EU+ countries and elsewhere do not think that the government has lost its legitimacy. As Birch argues, those who have exited, but are still loyal to the government are reluctant to criticize “the country of their birth. . . Residual feelings of loyalty frequently prevent them. . .” from doing so (1992, 75). 

In view of the severe restrictions on research activities and dearth of published official and non-official materials concerning the reality on the ground, it is difficult to state with certainty whether loyalty has anything to do with the decision of those who have not exited. This is an empirical question, which cannot be established a priori unless examined in situ, which is not possible at present. As mentioned earlier, their staying cannot be equated with loyalty. There is no evidence to show this to be the case. Rezene, an elderly man, who came to visit his relatives in London, told the author, “the people at home have lost faith in the government. Nobody trusts them anymore, but nobody seems to know what ought to be done to get rid of them. But one thing is certain, hizbi ab anqeru betsihiwo alo (people are fed up).”[27]

Other Counteracting Factors

As noted earlier, the decision of many of those who have stayed put in spite of the dissatisfying conditions may have nothing to do with loyalty. There are many possible explanations other than loyalty why citizens, including conscripts and those approaching the age of conscription, have stayed put. The “shoot to kill” policy at the Eritrea-Ethiopia and Eritrea-Sudan borders and the risk of being taken hostage by traffickers, who collaborate with smugglers, military officers in Eritrea and in the transit countries, are the major disincentives. For women, the risk of being raped throughout the displacement cycle is another disincentive. The exorbitant bribes paid to military officers in order to negotiate safe exit from Eritrea and safe entry into the transit countries, as well as the fees one needs to pay for smugglers, are exorbitant and often beyond the reach of many willing but unable exiters. Those who are able to afford such excessive fees are those who have rich families or relatives in the diaspora. Not all Eritreans are blessed with rich families and dense diaspora networks. The people are in a state of entrapment rather than exercising loyalty by staying put. 

Withey and Cooper’s study shows that loyalty, instead of being a manifestation of “supportive behaviour,” was found to be “something that resembled entrapment” (1992, 237). One of the informants, Adem, when asked “if the conscripts who have stayed are so dissatisfied, why don’t they protest or flee?”[28] said, “They are imprisoned.” This metaphor could mean anything from being trapped or gripped with fear, to having a family living from hand to mouth, or being a caretaker of elderly parents in a country where there are no social security or state and occupational pension schemes. It can also mean that people are despairing because they are deeply dissatisfied, but they don’t know how to move forward. Beyene stated the reason many conscripts in the open-ended ENS have not exited is because their families’ lives have been reduced to bare existence. He said, “If you have dependents and you live from hand-to-mouth, there is no room in your mind to think about anything else, but what you are going to put on the plate for dinner tonight.”[29] This may be exacerbated by lack of social networks that provide information and financial support to facilitate exit. For example, Kahsai said that the only families that are doing well in Eritrea are the ones who have family members abroad, not the ones whose children are stuck in the national service. It seems the expected dividends of flight are the key incentives that motivate some Eritreans to resort to this rather than staying put to exercise voice.[30] 

In view of the abject poverty and grinding deprivation permeating the economic landscape, the agential power needed to take an independent action, e.g. exiting, is diminished among those who live on a knife’s edge. Loyalty is one of the wide arrays of motives that can explain the behavior of many who have remained. There are those who may allege that the default position of most people is to stay put rather than flee or emigrate, leaving possessions, families and belonging to spatially-grounded communities. Such analysts may therefore explain the “decision” of those who have remained in spite of the dissatisfying conditions by their attachment to their country or place of origin. When conditions are bad, Eritreans have always moved to create and recreate new homes wherever opportunities exist. The strength of such an explanation is therefore weak in the context of Eritrean history. As Birch in a different but similar context theorizes, non-exiters suffer in silence because of fear of “retaliation” (Birch 1975, 73). 

The potency of this explanation is evidenced by the gruesome retaliatory actions the government has been taking against those who dared to voice any form of criticism, as well as against those caught exiting. The latter is epitomized by the government’s “shoot to kill” policy of illegal exiters. The majority of interviewees referred to the ruthlessness with which the government exercises upon those who do not toe the line. They gave, independent of each other, examples of the disabled veterans whose protest was suppressed violently (see Kibreab 2009a, 74–80). They also referred to the retaliatory action the government took against the members of the mechanized brigade when they demonstrated against the government’s decision not to pay salaries to the former combatants for four years (see ibid.). All referred to the indefinite incommunicado detention of the 11 members of the G15, the journalists in the private and government-owned newspapers, as examples of retaliatory actions taken by the incumbents. Tsegai, for example said, “If the government is able to unleash so much violence with impunity against the disabled former combatants and founding members of the liberation struggle, what can stop it from taking any action against whoever exercises voice.”[31] 

When asked “If the conditions are so bad as you state they are, why aren’t people taking action against the government?” Saleh,[32] Senait,[33] and Elias,[34] independent of each other, said that people have given up. They don’t expect change can happen easily without bloodshed and they don’t think a change that happens this way can result in meaningful and sustainable change. They are also discouraged by some of the diaspora Eritreans’ betrayal. In their view, because the Eritreans in the diaspora don’t have to suffer the consequences of the government’s follies, they try to justify its endless blunders. The betrayal of the promises of the liberation struggle by the incumbents and the ruling party seems to influence the attitudes of a few exiters toward political engagement. When asked why instead of fleeing she did not stay to fight, Senait, a female with a postgraduate degree said,

If the thirty-year war of liberation, which cost 65,000 lives, displaced nearly one million and maimed nearly 100,000 citizens failed to bring about fundamental economic, political, social and cultural transformation, why would any sane Eritrean contemplate to fight? After seeing everything going wrong, every promise betrayed and every opportunity wantonly squandered, everyone I know became excessively pre-occupied with personal projects and goals.[35]

If the revolution made no difference in the lives of our people, she said, “only fools would think fighting against the regime would bring about positive change. It may be possible to overthrow the regime, but probably the one that replaces it may be as equally bad or even worse.”[36] She further said, “Look at the state Somalia, Iraq, Syria, DRC and Yemen are in. I hate the Eritrea government from the bottom of my heart, but I don’t want the country I love to fall apart. Kab zey tifelto melak, tifelto seitan (the better the devil you know).”[37] Such an attitude can easily lead to neglect and indifference. It was in this sense Rusbult, Zembrodt, and Gunn (1982) added a fourth response—neglect—to the Hirschmanian framework of dissatisfaction. Withey and Cooper observe that “Neglect differs from loyalty in that it is not directed at recovery of the relationship. Rather, the individual responding with neglect implicitly accepts that recovery is not going to happen” (1989, 522). This is precisely what the interviewees said. 

Concluding Remarks

It is an utmost irony that a country that fought a costly thirty-year war and incurred heavy losses of lives, property, and forgone opportunities, inter alia, to eliminate the root and proximate causes of population displacement and suffering, has become one of the top refugee-producing countries in the world in proportion to its population size. This development is contrary to general expectation. Another scenario that has been unfolding, which on the surface appears to be contrary to expectation, is also the decision of tens of thousands of young men and women to flee the open-ended national service and its multi-faceted negative effects rather than remaining in the country to fight against repression. Given the heroic history of the people who successfully defeated sub-Saharan Africa’s largest military (see Welch 1991) against all odds, it is reasonable to expect the youth to remain and fight against repression and the dissatisfying condition rather than “voting with their feet.” 

When those who have been fleeing the country were asked why they resorted to flight rather than staying put to exercise voice and fight for change, the overwhelming majority said that the whole country is in the grip of intense fear. The exercise of voice or protest under the circumstances is too dangerous. This does not mean however that in the Eritrean case flight is the antithesis of engagement in politics and protest. On the contrary, data gathered from deserters show that those who have been leaving the country are doing so on the one hand, to escape from repression, and on the other, to engage in politics and protest in conditions of safety. Some deserters interviewed in the study equated the exercise of voice and protest under the repressive regime in the country with the signing of one’s suicide note. 

For those who want to engage in fighting against the system, relocation in democratic countries has provided them with opportunities to exercise voice and to fight from safe locations. Therefore, it is wrong to assume that all those who resort to flight have lost their zest for fighting against repression. The most vocal opponents of the Eritrean government are exiters. This may suggest that in the rapidly globalizing world, exit instead of atrophying voice may be instrumental in facilitating its expression. That is the main reason that I have argued throughout the chapter that in light of the Eritrean data, it is more fruitful to theorize the relationships between the key building blocks of the Hirschmanian framework of exit, voice, and loyalty in terms of nexuses rather than dichotomies.

Another discernible pattern that emerges from the Eritrean case is that not all those who have fled the country are necessarily ready and willing to fight against the government. Although the large majority of the unreserved and loud cheerleaders of the government and the PFDJ, the ruling party, are among older diaspora Eritreans and their second-generation children, there are some exiters who joined their ranks. It is also equally important to underscore the fact that although the situation in the country is by any measurement dissatisfying, it is wrong to assume that all Eritreans perceive it as such. Not all those who have remained in the country are also necessarily loyalists. The reality is messier than the neat theoretical construct purported by the Hirschmanian framework. Hirschman’s assertion that voice is the opposite of exit is contradicted by the Eritrean case. The findings in the latter show that the two can operate in tandem in the sense that exit is sine qua non for the exercise of voice. 


Amnesty International. 2004. ERITREA ‘You have no right to ask’ –Government resists scrutiny on human rights. Amnesty International. Available at

———. 2015. Eritrea: Refugees fleeing indefinite conscription must be given safe haven. 1 December. Available at

Andebrahan, Welde Giorgis. 2014. Eritrea at a Crossroads: A Narrative of Triumph, Betrayal and Hope. Houston: Strategic Book Publishing and Rights Co.

Anderson, B. 1992. The New World Disorder. New Left Review 193:3–13.

ARDC. 2016. Refugees in Israel. Available at

Barry, B. 1974. Review Article: exit, voice and loyalty. British Journal of Political Science 4:79–107.

Bernal, V. 2014. Nations as Network: Diaspora, Cyberspace, and Citizenship. Chicago: The University Press of Chicago.

———. 2004. Eritrea Goes Global: Reflections on Nationalism in a Transnational Era. Cultural Anthropology 19(1):3–25.

Birch, A. H. 1975. Economic models in political science: the case of “exit, voice and loyalty”. British Journal of Political Science 5:69–82.

Commission of Inquiry. 2015. Report of the Commission of Inquiry on Human Rights in Eritrea. Geneva, June. A/HRC/29/CRP.1 2015. Available

Connell, D. 2014. “Talking Eritrea” Series—Eritreans: Migrants or Migrating Refugees. Available at

———. 2015. The Rerouted Trafficking in Eritrean Refugees. Middle East Research and Information Project 268.

———. 2015. Escaping Eritrea Why They Flee and What They Face. Middle East Research and Information Project MER264.

Conrad, B. 2010. “We are the Prisoners of our Dreams”: Long-distance Nationalism and the Eritrean Diaspora in Germany. Dissertation zur Erlangung des Grades des Doktors der Philosophie im Fachbereich Sozialwissenschaften der Universität Hamburg.

Danish Immigration Service. 2014. Eritrea – Drivers and Root Causes of Emigration, National Service and the Possibility of Return. Aug–Oct.

David Sheen. Anti-African Racism in Israel. Available at

———. 2015. Israel’s abuse of African refugees exposes Zionism’s xenophobia. Available at

Dorman, S. 2004. Past the Kalashnikov: Youth, Politics and the State in Eritrea. in Vanguard or Vandals? Youth, Politics and Conflict in Africa, edited by J. Abbink & I. van Kessel, 189–204. Leiden: Brill Academic Publishers.

Fisher, J. 2004. Eritrea rapped for “military” schooling, BBC (January). Available at 

Foreign and Common Wealth Office. 2015. Eritrea Country of Concern. Available at 

G15. 2001. An Open Letter to all Members of the PFD. (May) Available at

Graham, J. W. and Keeley, M. 1992. Hirschman’s Loyalty Construct. Employee Responsibilities and Rights Journal 5(3):191–200.

Hepner, T. R. 2001. Soldiers, Martyrs, Traitors and Exiles: Political Conflict in Eritrea and the Diaspora. Philadelphia: University of Pennsylvania Press.

———. 2008. Transnational Governance and the Centralization of State Power in Eritrea and Exile. Ethnic and Racial Studies. 31(3):476–502.

———. 2009. Seeking asylum in a transitional social field: new refugees and struggles for autonomy and human rights. in Biopolitics, Militarism, and Development, edited by, O’Kaine, D. and Hepner, T. R. New York: Berghahn Books.

Hirschman, A. 1970. Exit, Voice, and Loyalty Responses to Decline in Firms, Organizations, and States. Cambridge, Massachusetts: Harvard University Press.

———. 1993. Exit, voice, and the fate of the German Democratic republic: an essay in conceptual history. World Politics 45(2):172–202.

———. 1986. Rival Views of Market Society and Other Recent Essays. Cambridge, Massachusetts: Harvard University Press.

———. 1998. A Propensity to Self-Subversion. Cambridge, Massachusetts: Harvard University Press.

Hirt, N. and Abdulkader Saleh Mohammad. 2013. ‘Dreams don’t come true in Eritrea’: anomie and family disintegration due to the structural militarisation of society. Journal of Modern African Studies 51(1):139–68.

Hoffmann, B. 2008. Transnational migration and political articulation: making new sense of “exit ad voice”. Available at:

Human Rights Watch. 2009. Service for Life State Repression and Indefinite Conscription in Eritrea. April. Available at

Hogan, P. 2009. Understanding Nationalism: On Narrative, Cognitive Science, and Identity. Columbus, Ohio State University Press. 

———. 2014. The sacrificial emplotment of national identity. Pádraic Pearse and the 1916 Easter uprising. Journal of Comparative Research in Anthropology and Sociology 5(1):29–47

ILO. 2015. Forced Labour Convention, 1930 (No. 29)—Eritrea. (Ratification: 2000 Direct Request (CEACR)—adopted 2014. 

Isaias Afwerki. 2004. Kale Metyik ms president Isaias Afwerki (Interview with president Isaias Afwerki). Ta’atek (Get Ready), No. 7 (Special Issue).

Jopson, B. 2009. Inside the insular and secretive Eritrea. Financial Times, 19 September.

Kibreab, G. 2009a. Eritrea: A Dream Deferred. Oxford: James Currey.

———.2009b. Forced Labour in Eritrea. Journal of Modern African Studies 47(1):41–72.

———. 1987b. Rural Eritrean Refugees in the Sudan: A Study of the Dynamics of Flight, in Refugees and Development in Africa, edited by P. Nobel, 37–44. Uppsala, Scandinavian Institute of African Studies.

———. 1987a. Refugees and Development: the Case of Eritrean Refugees. Trenton, NJ: The Red Sea Press.

———. 2013. The national service/Warsai-Yikealo Development Campaign and forced migration in post-independence Eritrea. Journal of Eastern African Studies 7(4):630–649.

King, R. 2012. Theories and Typologies of Migration: An Overview and a Primer. Willy Brandt Series of Working Papers in International Migration and Ethnic Relations. Malmö: Malmö University, Malmö Institute for Studies of Immigration, Diversity and Welfare (MIM). 

Leck, J.D. and Saunders, D.M. 1992. Hirschman’s loyalty: attitude or behaviour. Employee Responsibility and Rights Journal 5(3):219–230.

Massey, D., Arango, J. Hugo, G., Kouaouci, A. Pellegrino, A. and J.E. Taylor. 1993. Theories of International Migration: A Review and Appraisal. Population and Development Review 19(3):431–466.

Müller, T. 2012. From rebel governance to state consolidation—Dynamics of loyalty and the securitisation of the state in Eritrea. Geoforum, 43:793–803.

Reid, R. 2009. The politics of silence: interpreting stasis in contemporary Eritrea. Review of African Political Economy. 120:209–221.

Rusbult, C., Farrell, D., Rogers, G. and Mainous III, A. G. 1988. Impact of exchange variables on exit, voice, loyalty, and neglect: an integrative model of responses to declining job satisfaction. Academy of Management Journal, 31(3):599–627.

Rusbult, C. Zembrodt, and Gunn, L. K. 1982. Exit, Voice, Loyalty, and Neglect: Responses to Dissatisfaction in Romantic Involvements. Journal of Personality and Social Psychology 43(6):1230–1242.

Sebhat Ephrem. 1995. Precedence to National Sovereignty. Eritrea Profile, Nov. 18.

———. 2008. Kale Meteyik ms general Sebhat Ephrem (Interview with general Sebhat Ephrem) Ta’atek, July.

Stavis, M. and Parkinson, J. 2015. African dictatorship fuels migrant crisis: thousands flee to escape life of conscription and poverty. The Wall Street Journal, November. Available at 

Tronvoll, K. and Mekonnen, D. 2014. The African Garrison State: Human Rights and Political Development in Eritrea. Oxford: James Currey. 

Tsurkov, E. 2014. A life of forced labor: Why Israel’s Eritrean refugees fled home, 28 October. 972 Mag. Available at

UK Home Office. 2015. UK Home Office’s Country Information Guidance Eritrea: National (incl. Military) Service & Illegal Exit. March. Available at 

UNHCR. 2015. UNHCR country operations profile—Sudan. Available at

———. 2015. UNHCR Country operations profile—Ethiopia. Available at 

US Department of State. 2010. Country Reports on Human Rights Practices, Eritrea. Available at

———. 2011. Country Reports on Human Rights Practices, Eritrea. Available at

Welch, C. E. 1991. The Military and Social Integration in Ethiopia, in Ethnicity, Integration and the Military, edited by H. Dietz, 151–78. Boulder: WP.

Withey, M.J. and Cooper, W.H. 1989. Predicting exit, voice, loyalty and neglect. Administrative Science Quarterly 34:521–539.

Woldemikael, T. 2014. Introduction to Special Issue: Postliberation Eritrea. Africa Today 60(2):v–xix.

Women’s Refugee Commission. 2013.Young and Astray: An Assessment of Factors Driving the Movement of Unaccompanied Children and Adolescents from Eritrea into Ethiopia. Sudan and Beyond, May.

  1. See Article 2 of the National Service Proclamation 11/1991; Article 8 of the National Service Proclamation 82, 1995.
  2. London, August 30, 2014.
  3. Stockholm, September 25, 2014.
  4. See chapter VI, C, 2, Forced labour.
  5. Geneva, June 26, 2015.
  6. London, January 23, 2016.
  7. See also The EASO statistics wrongly show that there were zero Eritrean asylum-seekers in Switzerland between 2000 and 2008. Between 2000 and 2008, a total of 6,731 Eritreans applied for asylum. See
  8. ibid.
  9. Eurostat, Five main citizenships of first time asylum applicants, 3rd quarter 2015.png. Available,_3rd_quarter_2015.png&oldid=270682.
  11. Geneva, June 26, 2015.
  12. London, April 20, 2014.
  13. London, April 5, 2015.
  14. See "Pictures and videos of Eritrean Demonstration in Geneva Against COI Report on Eritrea," June 21, 2016. Available at
  15. "Diaspora Eritreans protest regime human rights abuses." Thousands of diaspora Eritreans march in support of UN report that accuses government of crime against humanity. June 23, 2016.
  16. Kidane, Telephone interview, Geneva, June 23, 2016.
  17. London, November 16, 2014.
  18. Stockholm, September 25, 2014.
  19. Geneva, June 24, 2015.
  20. The government vehemently denied this. But this is common knowledge in the country, including to foreign visitors. For example, after experiencing the deafening silence on the streets of Asmara, Barney Jopson (2009), of the Financial Times, told president Isaias “that there was a climate of fear on the streets,” he “responded with sarcasm. ‘It’s very important discovery on your part. You’ve been able to discover this in how many hours?’ I said I’d been in the country for two days. ‘It’s very unique. You must have a very unique brain,’ he said.” Jopson said to him, “I don’t think so.” Isaias further said, “To be able to know and read everything in this country in a matter of hours, it’s amazing. You must be a superhuman.” See Interview with Eritrea’s Isaias Afewerki, Financial Times, September 18, 2009. Available at
  21. Geneva, June 24, 2015.
  22. London, May 14, 2015 (telephone interview).
  23. Stockholm, September 25, 2014.
  24. London, January 9, 2016.
  25. Although some were unable to distinguish between where the government ends and the state begins, most of the informants said their grievances against the government had no impact on their commitment to the state. In the Tigrinya language, the distinction between the concepts of state and government is not clear unless one consistently uses mengisti (government) and hager (country).
  26. London, January 16, 2016.
  27. London, January 4, 2016.
  28. Geneva, September 24, 2015.
  29. London, May 16, 2014.
  30. Stockholm, September 25, 2014.
  31. London, January 16, 2016.
  32. Geneva, June 24, 2015.
  33. Geneva, June 24, 2015.
  34. London, May 24, 2015 (telephone interview).
  35. Geneva, June 24, 2015.
  36. ibid.
  37. ibid.


Postliberation Eritrea Copyright © 2018 by Gaim Kibreab. All Rights Reserved.

Digital Object Identifier (DOI)