Forced Migration and “Rejected Alternatives”: A Conceptual Refinement

The “forced migration” concept can obscure how some people who migrate in this mode exercise a key form of agency. Some refugee flows occur when people reasonably reject options that might obviate the need to flee (e.g., abandoning their religious beliefs). A similar form of agency must be recognized regarding forced migration of other types: people facing severe economic difficulties sometimes become migrants by rejecting options that might secure their subsistence, and when that choice is reasonable because the alternatives amount to human rights violations, we should then describe their migration as “forced” even if it is not wholly involuntary.

This article identifies and remedies a significant elision in the way the concept of forced migration is commonly understood and used. The narrower refugee concept includes space for a crucial element of agency: people sometimes become refugees not because they have no other choice but because they insist on exercising their right to continue holding prohibited beliefs or belonging to prohibited groups. That insistence on exercising one's rights can also be important in other instances of migration that are properly described as forced-but this form of agency is typically submerged by a commonsense way of thinking about the "forcedness" of forced migration. The article thus argues for an approach to defining forced migration that provides space for that form of agency, one that can encompass a broader range of instances of migration that fit our intuitive understanding of the concept. The innovation here is to suggest that migration can be "forced" via one's rejection of local subsistence options that amount to violations of human rights. Perhaps one could avoid starvation (and thus refrain from migration) by submitting to severely degrading employment conditions, for example; migration might then appear as something other than "forced" insofar as there is an alternative. But migration in such instances is properly described as forced if it is rooted in one's insistence on not submitting to an available option that amounts to a gross violation of one's human rights-and the key point here is that this mode of evaluation is properly applied in economic terms no less than in political terms.
In certain respects, the term forced migration is relatively straight forward-particularly in the way it seems to turn on a contrast with migration typically understood as voluntary. People are sometimes forced to migrate not only by persecuting governments but also by impending famine or more generally by threats to their livelihoods emerging from the general instability and disorganization that characterize situations of chronic conflict. For empirical application, one must resist the temptation to identify instances as either voluntary or forced: most instances of migration that appear to be voluntary are also shaped by constraints (the limitations of available options, including the ability to gain entry somewhere), and forced migration can involve elements of agency that are already well understood-for example, where to go, given that one must leave (Turton, 2003). But it is well understood that the distinction between forced and voluntary migration can be conceived as forming ends of a continuum rather than a dichotomy (Richmond, 1994).
Even so, an important subtlety apparent in the refugee concept is lost when migration is understood as forced as per a commonsense notion that something external to the migrant is forcing him or her to leave. For example, Forced Migration Online, following the lead of the International Association for the Study of Forced Migration (IASFM), adopts a definition referring to "movements of refugees . . . as well as people displaced by natural or environmental disasters, chemical or nuclear disasters, famine, or development projects"; a typology then suggests that forced migration happens via displacement induced by conflict, development, and/or disaster (Forced Migration Online, 2012). Some refugees as defined by the 1951 Convention fit poorly with this sort of commonsense understanding of forced migration. As Zolberg et al. (Zolberg, Suhrke, & Aguayo, 1989) indicate, some people who face persecution on the basis of holding certain opinions or belonging to certain groups could, as an alternative to fleeing, adopt different opinions or abandon certain memberships. Their choice to flee instead is typically held to be admirable, and their insistence on remaining true to their consciences reinforces their claim to protection elsewhere. Some people become refugees via the exercise of a particular kind of agency (i.e., agency is not apparent only in decisions about where to go). That assertion might seem to sit uneasily with the notion that their migration is forced, but instead of perceiving refugees as something other than forced migrants we are better served by a more subtle understanding of forced migration conceived more broadly, an understanding that foregrounds the question of whether particular local options (including options for subsistence) may reasonably be rejected.
This article argues, then, that the broader concept of forced migration must find a place for the specific sort of agency already understood to pertain to "conventional" refugees. In many cases, migration is indeed forced in a way that accords with common sense: one is expelled or, if one does not migrate, one will very likely die. In other instances, however, survival at "home" is possible-but only by engaging in actions or submitting to conditions that constitute significant violations of one's human rights. Some "political" refugees become (forced) migrants by reasonably rejecting the alternatives that might enable them to avoid persecution. But people in other difficult situations (not covered by the 1951 Convention) also sometimes become migrants by rejecting the alternatives that might enable them to mitigate the threats to their survival; the main argument here is that, insofar as their rejection of available alternatives was justified, the act of migration is then properly described as forced, despite not being wholly involuntary.
This argument is developed via emphasis on the notion of rejected alternatives. Forced migration occurs not only when alternatives are lacking entirely but also when there is no reasonable alternative to migration. This perspective is not contentious with regard to conventional refugees (on the contrary, it is entirely consistent with the 1951 Convention, which creates a framework for protection so that people will not have to abandon their religious beliefs or political opinions); the goal here, then, is to consider what it means in regard to forced migration more broadly, particularly when the factors that induce people to leave have less to do with political persecution than with economic hardship. This way of thinking about forced migration is intuitively attractive (even if it requires a more subtle use of the word forced); the goal is to articulate key elements of that intuitive understanding so that its assumptions and assertions can be considered directly.
To a certain extent, the argument developed here is implied by Ottonelli and Torresi's recent (2013) article conceptualizing "voluntary migration." For migration to be voluntary, a condition of "sufficiency" must be met: available alternatives at home must be "good enough for the migrant" (p. 798). One might then be tempted to conclude simply that migration is forced whenever the sufficiency condition is not met. But there are good reasons to conceptualize forced migration in a more direct way, beyond a simple inversion of voluntary migration. One reason is that what counts as good enough in Ottonelli and Torresi's discussion is not well specified. In addition, the extension of forced migration to encompass specific situations of economic hardship cannot be taken as self-evident and so cannot be achieved by simple conceptual inversion.

D. Bartram
The article proceeds with a brief conceptual history indicating the emergence of forced migration as a means of escaping the confines of the refugee concept. It locates the "rejected alternatives" idea in the conventional refugees concept; the key question then becomes whether rejected alternatives is necessarily an element of forced migration conceived more broadly. The article addresses that question via discussion of the situations that have produced refugees and forced migrants in recent years from Eritrea. The conclusion raises the question of what obligations might arise for destination countries when instances of forced migration meet the conditions described here.

A BRIEF CONCEPTUAL HISTORY OF REFUGEES AND FORCED MIGRANTS
The "refugees" concept has long been burdened by its deep roots in the United Nations conventions on the Status of Refugees. The 1951 Convention established a "practical" definition governing signatories' actions. But some scholarly investigation of the refugee phenomenon in a broader sense has suffered from extensive reification of that definition: some observers treat it almost as if it were sacred, and even some attempts to transcend it arguably achieve only a higher orbit. That point is evident in the notion that some people are in "refugee-like situations." Surely such people are in fact refugees-at a minimum, "de facto refugees," per Grahl-Madsen (1983). The concept of forced migration emerged in part via the perception that the legal definition of refugees was artificially narrow, excluding people who looked very much like refugees in an intuitive sense. The 1951 Convention is centered on the notion that someone has a "well-founded fear of being persecuted" on specified grounds-and has therefore left his/her country of nationality and is unable/unwilling to rely on its protection. As is well known, this definition, rooted in a specific legal framework that arose in a particular sociopolitical context, has significant limitations as a social-science concept. Its narrow focus is rooted also in the close association between research and refugee policy agendas/institutions, which Jacobsen and Landau (2003, p. 187) argue has resulted in a "lack of rigorous conceptualisation and research design," among other problems (e.g., Bakewell's [2008] demonstration that it results in overlooking "self-settled" refugees).
Definitions matter in this sphere in part because labels can affect admissions decisions. In a context characterized by opposition to mass immigration, refugees are understood to have a special claim for entry, because their migration-motivated by the need to escape persecution-is ostensibly distinct from that of other migrants, in particular those who "merely" want to improve their economic circumstances. The migration of refugees is, in a word, forced, not freely chosen (though the point requires subtlety and elaboration, discussed further below). But migration can of course be forced by processes other than fear of persecution. If the special claims of refugees arise from the notion that their migration is forced rather than freely chosen, then that sort of claim might well arise in other instances where migration is forced rather than freely chosen. In this frame, refugee flows are a subset of forced migration (e.g., Castles, 2003) and analytical focus then shifts to the question of what makes instances of migration forced.
Some scholars have proposed a broader use of the term refugee itself, so that it is equivalent to forced migration (or forced migrant) in a more general sense (Ager, 1999); Gibney (2004) includes in his definition of refugees people who if compelled to return would face threats to their "vital subsistence needs." For this purpose, one might draw on Shue's (1980) arguments about "basic rights." Shue argues that the right to subsistence is just as basic as the right to security, in contrast to older liberal ideas about the priority of "negative" freedoms-for example, the right to be free from interference by others. The conventional definition of refugees accords with those older liberal ideas in giving priority to the threats to individual security that arise when governments persecute people for holding and acting on the wrong political views. But for Shue, if security is "basic," subsistence is no less so, in that both types of rights must be guaranteed if people are to enjoy any other rights-a view that helps reinforce the argument that the conventional refugee perspective is artificially narrow and merits expansion, so that the term refugees is used not just in its legal (Conventional) sense but much more broadly. The more common approach, however, is to position refugee flows as a subset of forced migration.
The forced migration concept has helped researchers and others escape the confines inherent in the definition of refugees as borrowed from international law (where it becomes difficult even to perceive them as migrants). But it has also given rise to a key ambiguity. With regard to (Conventional) refugees, that ambiguity consists of the loss of an insight that was readily available in earlier work on refugees by Zolberg et al. (1989). If we redefine refugees as a subset of "forced migrants"-while at the same time using a commonsense notion of the "forcedness" of forced migration-we risk overlooking an important aspect of the process that leads some people to become refugees. As noted above, people who are persecuted on the basis of holding certain religious beliefs can sometimes avoid persecution by changing their religious beliefs. Zolberg gives the example of Huguenots in France, some of whom converted to Catholicism, thus remaining in France instead of becoming refugees in England. Likewise with Jews in 15th-century Spain and Portugal: as hostility toward them intensified, some Jews fled to other countries, but others converted to Christianity under duress and remained in Spain. More recently, the observation pertains to political beliefs as well: dissidents who fled the Soviet Union might have avoided becoming refugees by ceasing to be dissidents and instead adhering to the (Communist) party line.

D. Bartram
This sort of option is by no means always available to people who become refugees, but the fact that it is sometimes available suggests that we should be careful about drawing a straightforward extension from refugees to forced migrants. As Zolberg et al. (1989) note, those who reject an option that would have allowed them to remain in their own country cannot be described unreservedly as "involuntary migrants." That point has particular resonance in key liberal-democratic destination countries, where refugees fleeing from repressive regimes are viewed as deserving protection precisely because they are exercising freedom of conscience. The United States in particular has long been more welcoming to refugees who were persecuted for expressing their rejection of totalitarian regimes and ideologies than to migrants whose situations were genuinely characterized by coerced flight in the sense that no other option was available to them. For some types of refugees, then, the label "forced migrants" might obscure the fact that their choices play a key role in the process by which they become refugees.
Even so, the term forced migration is surely here to stay, and far from asserting that refugees of this type are not forced migrants, the goal here is to clarify the conditions that make it appropriate to use the term, in part so that we can determine whether it is sensible to think in similar ways about forced migrants who are not refugees in the narrower sense (e.g., because their desperate situations involve economic deprivation not political persecution). The point of departure for what follows is (again per Zolberg et al.) that migration can properly be described as forced not because there are no alternatives to fleeing (though sometimes there are indeed none) but because adopting the alternatives that would obviate the need to flee (i.e., abandoning or suppressing one's beliefs) cannot reasonably be expected of someone who genuinely adheres to the belief in question. In the three instances described above (Huguenots, Jews, and Soviet dissidents), some individuals understandably chose to conform to the demands of oppressive regimes; the costs associated with exercising freedom of conscience are more than some people are willing to pay. But conforming is, for many observers, the less admirable choice, certainly not one that can reasonably be imposed. Rejecting the demand to conform-and then fleeing if persecution follows-is entirely a proper course of action, so much so that it leads to obligations on destination countries to offer protection; to fail to offer protection in such instances amounts to helping repressive regimes to impose conformity and to violate individuals' rights.
The term forced migration as currently used (e.g., via the Forced Migration Online/IASFM definition given earlier) does not facilitate appreciation of this element of refugee situations (still less of other types of forced migrants). It draws attention to what repressive governments do, the actions that lead to people's expulsions or escapes. By the same token, the term draws attention away from what the refugees might do to avoid becoming refugees, so that we are not in a position to consider why it is reasonable for individuals to Forced Migration and "Rejected Alternatives" 445 reject those alternatives. This difficulty is not severe in regard to refugees of the conventional type: it is relatively straightforward to discern not only what they might have done but why it was reasonable to reject those alternatives. Moreover, normative liberal theory is well developed in ways that make it unnecessary to provide (here) a detailed defense of the view that individuals are justified in becoming refugees rather than violating their consciences so that they can survive under repressive regimes. Indeed, the discussion in this section is on its own not intended as innovative; instead, it recovers insights about refugees that were available in Zolberg et al.'s work but which have become difficult to locate in the evolution of the field towards forced migration (cf. Hathaway. 2007).
Even so, it is useful to consider a contemporary instance of refugee migration in these terms, in part to demonstrate how the refugee concept can diverge from forced migration when we discuss empirical cases. The main goal of this article is subsequently to develop an understanding of rejected alternatives for forced migration beyond conventional refugees. To lay the groundwork for that argument, the next section considers the extent to which refugee flows from Eritrea are properly described as forced migration. We will then be in a better position to contemplate a similar analysis for forced migrants fleeing economic deprivation and other difficulties that do not fit in the refugee paradigm.

ERITREAN ASYLUM SEEKERS: FORCED MIGRANTS?
Eritrea is a major "producer" of refugees and asylum seekers, with a higher per-capita rate of asylum-seeking than any other country and among the top three in absolute numbers (Kagan, 2010), despite a total population of just over six million. A significant number of people who have left Eritrea are indisputably refugees/asylum seekers. That conclusion gets strong support via a summary evaluation by the UNHCR (2009), which encourages a prima facie approach to claims in countries where large numbers of Eritreans seek protection.
Even if Eritrean asylum seekers generally meet the criteria for recognition as refugees, however, we should not conclude on that basis alone that they are properly described as forced migrants. To anticipate: I will conclude that most Eritrean asylum seekers are properly described as forced migrants, but I will propose doing so via an analysis that relies less on the blanket judgment of the UNHCR and more on consideration of the alternatives Eritreans might have had but rejected.
One reason to push beyond ideas rooted in "practical" perspec tives-particularly with respect to Eritrea-has to do with a catch-22 that arises in the simple act of leaving Eritrea without authorization. Leaving without permission is a criminal act, one that typically leads to severe punishment including imprisonment and torture on return. ("Return" is usually a matter of deportation from another country, not a voluntary act.) One reason the UNHCR recommends against refoulement to Eritrea is that it will surely lead to persecution; an unauthorized emigrant has very good reason to fear persecution. But the fact that individuals might meet the criteria for refugee status on those grounds does not by itself mean that the original act of migration was forced. In practice, people who leave Eritrea without authorization likely have a good reason to do so, a reason that on its own terms would qualify them for refugee status. But that reason should be considered on its own terms, instead of deducing it on the basis that return of unauthorized emigrants leads to persecution.
The analysis here focuses on Eritrean asylum seekers who leave because they are unwilling to provide the "national service" required of all Eritreans upon reaching the age of conscription. Avoidance of army/national service is the most oft-cited reason for leaving Eritrea (UNHCR, 2009;Weldaheimanot, 2009). Other grounds include persecution on the basis of political activities against the regime and belonging to a religious organization not affiliated to one of four officially recognized/sanctioned religions. One might raise the question animating this discussion about those grounds as well: Could Eritreans be expected to avoid forced migration by refraining from political dissidence and/or participating in nonsanctioned religions? But the answer to that question is obvious and requires no extensive discussion; those are the "conventional" grounds for refugee status.
The case for rejecting conscription/"national service," however, is less clear. Many countries have compulsory conscription and national service programs, and the UNHCR (2009) recognizes that in principle a requirement to provide national service is not in itself illegitimate. Many Eritrea do enlist upon reaching the age of conscription; Eritrea has a high rate of emigration leading to asylum claims, but much larger numbers of similarly placed people do not leave. One question here is, If it is possible for many people to accept conscription, why was it not possible for those who left? Why was it reasonable for the latter to reject conscription, such that their migration is then reasonably described as forced? In some cases, the answer to that question fits well within the conventional "refugee" paradigm: Members of Jehovah's Witnesses, for example, believe that rejecting military service is a matter of conscience (and they are already vulnerable for participating in a nonsanctioned religion). But most who flee from conscription/national service are not Jehovah's Witnesses. What of these others?
If we were dealing with people who were refugees on conventional grounds such as religion or political opinion, then from a conventional "refugee" perspective that question would readily appear as rooted in a false premise. The fact that some people are willing to endure violations of their freedom of conscience does not mean that others must do so. For some victims of persecution, the costs associated with exile are simply too high. But the fact that some people are unwilling to pay those costs does not mean that the circumstances that might lead them to flee are not in fact violations of their right to freedom of conscience-and so those who do flee are justified in doing so and in claiming protection elsewhere.
The conclusion that migration of those fleeing conscription/national service is generally forced is rooted in a similar but broader set of considerations. A sharp analysis by Kibreab (2009) concludes that demands for national service in Eritrea, and the arrangements for enforcing those demands, amount to forced labor. Depending on how these things are conceptualized and measured, the Eritrean army is among the largest in Africa (Bozzini, 2011); in per-capita terms it is exceeded only by North Korea. In certain respects its size is not surprising, given the country's history of conflict with Ethiopia; in addition, on top of direct military needs, the government's vision of the army's functions includes the overcoming of particularistic identities and the development of an encompassing national identity (a vision regarding military service shared by many governments around the world). One can also find a measure of support for national service among the Eritrean population. But that support, as with the army's legitimacy more generally, is decidedly limited-as is evident in the severe punishments required to secure the participation of significant numbers of conscripts (Kibreab, 2009). The country is also riddled with checkpoints at which soldiers check people's documents to determine whether they are complying with their service obligations (Bozzini, 2011). Most of all, the repressive nature of conscription is evident in the way government forces are prepared to shoot people trying to leave without authorization-an act which again is commonly motivated by desire to avoid conscription (Human Rights Watch, 2009).
In addition, the "labor power" of the army is regularly coopted for the benefit and enrichment of senior officers and companies owned by the ruling party and by private individuals (Kibreab, 2009). The national service aspect of conscription is connected to the government's "Warsay Ykaa lo Development Campaign," whose stated purpose is to rebuild the country to remedy the destruction caused by the war for independence and the subsequent war with Ethiopia in 1998-2000. Here as well, one might try to argue that national service for a development campaign is a legitimate enterprise, such that it is acceptable to require citizens' participation. In the Eritrean case, however, it is far from clear that the program in question actually results in "development"; on the contrary, the program is commonly used by powerful individuals to extract private gain. The conditions for the conscripts themselves-even the willing ones-are extremely difficult. Again, Kibreab concludes that this institution amounts to forced labor as per the relevant international conventions. Many people fleeing Eritrea, then, are properly described as forced migrants even leaving aside the fact that their unauthorized departure would lead to persecution if they were returned.

D. Bartram
Even so, some Eritrean asylum seekers, in interviews with researchers, relate motivations for leaving that convey an impression more consistent with "economic migration" or simply a "desire for a better life" (Humphris, 2012). One person interviewed in Israel by Furst-Nichols and Jacobsen (2011) said that he left because the government was interfering in his business. Now, perhaps the government was interfering to such a degree that his livelihood/survival was severely threatened; or perhaps the way the researchers conveyed this man's story did not do justice to an experience of serious persecution. Note that he does fall within the ambit of the international refugee protection regime, particularly as he has very good reason to fear persecution if he were returned. It is less clear, however, that he is properly described as a forced migrant in the analytical sense developed here.
There is, then, no direct equivalence between refugees and forced migrants. Usually that observation consists of noting that refugees are a particular category or subtype of forced migrants-but the analysis above indicates that one can be a refugee and yet not be a forced migrant. In broader terms, to determine whether (or to what extent) it is appropriate to describe someone as a forced migrant we should work with that concept on its own terms, rather than allowing its meaning to be dictated by its emergence in the field of refugee studies and the limitations of core concepts there. What matters is whether migration was indeed forced in some meaningful sense-and the argument here is that that descriptor makes sense not only when there were no alternatives to migration but also when any available alternatives were reasonably rejected (as with rejection of conscription, particularly for an abusive and corrupt national service program).

REJECTED ALTERNATIVES IN ECONOMIC FORCED MIGRATION
This section argues that the notion of rejected alternatives helps us define forced migration more broadly in a way that captures a sound intuitive sense of the concept, transcending the definitional limitations of the refugee concept and the carrying over of those limitations into current understandings of forced migration itself. To be useful in this way, the mode of definition developed above would have to help us understand the situation of people described as forced migrants in a more general sense, including instances that lie at some distance from conventional refugees. The possibilities for analysis in these terms are more limited, mainly because when it comes to migration of people facing "economic" threats to survival and well-being we do not have the sort of well-developed normative framework available for analysis of refugees in the more conventional sense. The value of rejected alternatives as discussed here, then, appears in the way it indicates the kinds of questions that would have to be answered in this context. Migration scholars generally note the impossibility of making clean distinctions between different types of migration (a point evident in ideas about the "migration/asylum nexus" and "mixed migration flows," e.g., Van Hear, 2012). Again, migration flows are not either forced or voluntary but forced (and voluntary) to varying degrees. Similarly, migration flows are typically not either economic ("labor migration") or political (refugees); those characterizations make sense to varying degrees, with particular instances showing elements of both (Dowty, 1987;Richmond, 1994). The displacement of people in conflict-ridden countries can be induced by their inability to grow crops in battle zones, or by the impossibility of operating markets in cities where civil war is raging. Forced migration, then, is often no less economic than it is political. Economic deprivation is itself sometimes a tool of repression, as with deliberate attempts to induce starvation in Ethiopia/Eritrea during the 1980s (Kaplan, 2003).
It is doubtful that one needs an elaborate conceptual apparatus to appreciate the idea that migration can be forced (as per "common sense") by economic desperation: Forced migrants might be fleeing the prospect of death or abuse at the hands of a repressive government, or they might be fleeing the prospect of death or disease via starvation/malnourishment. Betts (2010) uses the term "survival migrants" to describe people fleeing an "existential threat." The prospect of death (or serious deterioration of well-being) is what forces people to move; distinctions between types of factors that create that prospect are potentially significant (i.e., political vs. economic), but they are also secondary, at least from the perspective (and motivations) of the migrants themselves. Some observers argue for maintaining a strict separation between refugees in the narrow Convention sense and other forced migrants (Hathaway, 2007). But those arguments are rooted mainly in practical considerations, not analytical ones, and it is persuasive to argue that what makes migration forced is the fact of a serious threat to life or basic well-being, regardless of whether that threat is labeled political and/or economic.
The discussion in the previous section then indicates the question, What alternatives might sometimes be available to migrants fleeing starvation and destitution? And when and on what grounds might we take the view that it was reasonable for them to reject those alternatives, so that we should conclude that their migration was in fact forced to some significant extent? As with more conventional refugees, often the answer is that there were no alternatives: The only possibility for survival consisted in flight to another location (and all too often even fleeing does not guarantee survival). Many people faced with starvation do not have the physical capacity (let alone the economic resources) to accomplish migration; their only hope might consist of food aid (itself by no means a guarantee). Even so, people can sometimes recognize an impending threat-for instance, a significant decline in the price they receive for a crop whose sale provides their main or only income. Migration might then be perceived as the only, or only acceptable, way of averting that threat.
But that slippage between "only" and "only acceptable" requires attention. As noted in the introduction, migration can be described as voluntary only if the migrant has acceptable (good enough) options in the origin country (Ottonelli & Torresi, 2013). (Voluntary migration depends on satisfaction of three additional conditions: noncoercion, exit options, and information; conversely, violation of any one of these conditions means migration is not voluntary.) Similarly, despite disagreeing about much, Abizadeh (2010) and Miller (2010) agree that assessment of migrants' actions and rights turns on whether they have adequate options in the origin country. But what makes alternatives available to migrants "acceptable"? Ottonelli and Torresi (2013) say that migrants must have had access to "basic necessities" (food, shelter, safety, and "social dimensions of human existence"); if one's "alternatives to migration involve starvation, destitution, bodily injury, or incapacitating poverty" (p. 798), then migration is not voluntary. But an enumeration of this sort is not satisfying-not least because it would not do justice to the situation of Eritreans as described above. Eritreans can secure basic necessities and avoid starvation, destitution, and so forth-they need only comply with demands for conscription and national service. Again, to make sense of forced migration generally and of situations like Eritrea, we cannot simply invert the voluntary migration concept. In particular, instead of asking about "acceptable alternatives" for this purpose, we would do better here to ask, What makes available alternatives unacceptable?
As with refugees, our analysis here is informed by a discourse of human rights. For conventional refugees, there is a good degree of clarity and consensus on the principle that people must not be forced to violate their freedom of conscience. But that formulation can be specified more precisely: What we really mean, in connection with what refugees actually experience, is that they must not be forced to violate their freedom of conscience in order to secure their well-being. The actions oppressive regimes take with regard to dissidents (imprisonment, torture, blacklisting, etc.) constitute threats to well-being; an oppressive regime seeks to suppress dissidence precisely by making one's well-being conditional on conformity to its demands.
There is a related, but inverse, incompleteness in understandings of what makes someone a forced migrant by virtue of economic deprivation. As a matter of basic human rights in regard to economic aspects of life, individuals have a right to some minimum level of well-being (even if the level itself is not well specified), but there is less clarity in discussions of forced migration on what someone must not be compelled to do in order to secure their economic well-being.
To gain clarity regarding that lacuna, it is useful to reflect on why the concept of refugees rests in part on the premise that individuals must not be forced to violate their freedom of conscience. That element of the concept is often treated as self-evident -but for our purposes it is useful to interrogate it. The target of the discussion here is quite specific: we do not need to know why people have a right not to be persecuted but rather why they have a right not to be forced to do certain things (for refugees: profess something contrary to their true beliefs) in order to avoid being persecuted. We can then consider whether the reasons justifying that right pertain also to situations that lead people to engage in migration rooted in economic deprivation.
The reasons and justifications are by no means esoteric. The foundation of the right to freedom of conscience is the "inherent dignity" of human beings-a phrase used twice early on in the preambles to the International Covenant on Civil and Political Rights (ICCPR) and the International Covenant on Economic, Social and Cultural Rights (ICESCR), two key documents that give expression to contemporary understandings and agreements on human rights. Forcing people to profess something contrary to their beliefs, or forcing them to suppress those beliefs, violates their dignity. Seen from this angle, a core purpose of the international refugee regime is to help individuals preserve or recover their dignity. What is fundamental here, then, is not the right to protection or even the right to freedom of conscience. The foundation of these rights is individual human dignity.
It then follows that threats to dignity can lead to rights to protection for reasons other than freedom of conscience-something evident in the breadth of the UN Covenants noted just above. The 1951 Convention definition of refugees refers only to a limited set of grounds on which protection can be claimed-and as a matter of law and practice those limitations matter-but there are good reasons to set them aside in developing the concepts of refugees and forced migration for analytical purposes. If it is sensible to describe someone as a forced migrant when they flee rather than reconcile themselves to violations of their right to freedom of conscience, then the dignity that underpins that right can also inform the extension of the forced migration concept to include people who flee rather than reconcile themselves to rights violations that pose a threat to their dignity beyond matters of conscience. Violations of rights in that sense sometimes appear in the actions people might have to undertake to meet their needs for subsistence and well-being in their local setting.
Identifying alternatives that meet that description is sometimes not difficult, insofar as those alternatives amount to obvious human rights violations that threaten human dignity. Forced labor is an obvious instance, as is evident in the discussion above of Eritrean national service. One's ability to gain employment or operate a business in Eritrea is often dependent on having complied with the labor demands of an illegitimate and abusive institution (or, worse yet, on continuing to supply service/labor in a "reserves" mode), and the decision to seek subsistence elsewhere instead is then properly described as forced migration, equivalent in key respects to the migration of those who insist on exercising freedom of conscience. In both instances, what is fundamentally at stake is human dignity. Extreme poverty itself is a human rights violation that impairs dignity (see e.g., Rena, 2009, on Eritrea in this regard); but some "solutions" to poverty are also inconsistent with human dignity. In a similar vein, one should not have to rely on income derived from the labor of young children: Child labor is a direct violation of their dignity at the time the labor is performed, and it can also be expected to diminish their dignity as adults insofar as they lose the opportunity to gain the sort of education that is most effectively undertaken by children (as with basic literacy).
The line of argument pursued here might strike some as unwise or undesirable, particularly insofar as it seems to portend expansion of the forced migrant concept to such an extent that it loses rather than gains definition. But it need not do so. In certain respects the UN Covenants on rights are widely held to be overambitious, reaching beyond states' capacity for implementation; Tomuschat (2008) makes that point with reference to the ICESCR's Article 11, which promises a "continuous improvement of living conditions." But in other respects relevant to the discussion here the rights that proceed from human dignity are quite modest. In particular, the ICESCR says merely that a people may not "be deprived of its own means of subsistence." Even Article 11 refers only to an "adequate" standard of living-without specifying what adequate means beyond "freedom from hunger." Extending forced migrants to include people fleeing malnourishment and the threat of starvation is unlikely to be controversial. But even if rejected alternatives leads us to perceive forced migration when the migrants reject local options that might substitute for migration, it is not obvious that this further extension of the concept is unwarranted when those local options entail human rights violations and so are rejected on grounds similar to the insistence on exercising freedom of conscience. To argue otherwise-to hold that people should refrain from migration and adopt any local option for securing their subsistence-is precisely to argue that people should be willing to accept situations in which their human rights are violated.
But while it is not difficult to adduce some obvious types of alternatives that may legitimately be rejected for constituting human rights violations, it is more difficult to know how to approach the large "gray area" that arises in consideration of the difficult economic circumstances that lead many people to opt for migration that we might want to describe as forced. Physically difficult labor, for example, is not itself a violation of human rights. But that assertion raises questions about the way words are used to convey shades of meaning: Labor of "moderate" physical difficulty is surely not a violation, but perhaps labor of "extreme" physical difficulty crosses a line-though where that line lies cannot be determined merely through the use of adjectives that cannot be objectively defined. Even the word "dignity" does not have a universal or objective definition that enables us to deploy a straightforward category of human rights violations (McCrudden, 2008).
Even if we do not yet know how to apply these ideas in a way that will instantly gain widespread adherence, however, it does not follow that they are of no value in attempting to determine what we mean by the concept of forced migration. The primary contribution here is to identify an important lacuna in this concept as it is commonly discussed and used. Having articulated rejected alternatives as an important (but underdeveloped) element of the concept as we intuitively understand it, we are in a better position to ask more productive questions, about the concept in general and about how it can be used constructively in analysis of particular instances. As with human rights ideas more generally (McCrudden, 2008), dignity will surely play a key role in further efforts to determine when it is reasonable for individuals to reject certain alternatives to migration. CONCLUSION Migration is sometimes "forced" in the obvious sense that there is in fact no alternative. But there are many instances that might seem less obvious: There are alternatives-and yet the alternatives amount to violations of basic human rights. The notion of "rejected alternatives" is intended to facilitate development of greater clarity on aspects of this idea that form part of our intuitive understanding. In describing certain instances of migration as forced, we often have in mind situations in which people must flee because if they don't they will be shot or their land and crops will be inundated by waters rising with a new dam. But our intuitive understanding of forced migration is not exhausted by examples of that sort. In connection with refugees, migration is sometimes forced by one's reasonable insistence on exercising one's freedom of conscience. The core argument here extends that point to forced migration in a broader sense: when one's local options for gaining subsistence entail violations of basic human rights, migration is sometimes forced by one's reasonable insistence on finding elsewhere a means for subsistence that preserves a basic level of human dignity. People who flee Eritrea-facing destitution and perhaps starvation, unable to gain a livelihood for having failed to do national service-are forced migrants even though the threats to their well-being are mainly economic and (more to the point) even though it would be possible (at the cost of their human rights) to secure their economic well-being in Eritrea.
As is evident in the discussion above (and in others' work as well), the concept of forced migration cannot be merely empirical or "neutral" (e.g., Zetter, 2007). As with many social science concepts, and in particular the concepts used by migration scholars, forced migration is an inescapably normative idea. Many observers hold that states may legitimately exercise wide discretion in deciding whether to admit people whose migration falls close to the voluntary end of a continuum between forced and voluntary migration. To describe instances of migration as forced is to indicate that the range of discretion exercised by destination-country states is necessarily narrower in those instances. In connection with conventional refugees, that point carries the force of international law, even if destination-country states frequently try to evade or minimize their legal obligations in this regard.
In connection to forced migration in the broader sense, however, there is neither an established legal structure nor a well developed normative framework that helps clarify what types of forced migration in a broader sense lead to special obligations on the part of destination countries. Intuitively, the idea that special obligations arise when migration is forced seems obvious. When global warming leads to rising sea levels and the disappearance of places like Tuvalu and Kiribati, people there will have to move to other countries, and other countries-particularly the world's wealthiest countries-will, in the judgment of many, have an obligation to admit them. But when people migrate because their local subsistence options have become bound up with human rights violations, destination countries commonly reject them as "mere" economic (and therefore voluntary) migrants. The clear implication of the perspective developed here is the principles leading to recognition and protection of (conventional) refugees cannot reasonably be suspended simply because subsistence rights can be tagged as economic.
Clarifying the extent and nature of others' obligations in such instances is a task left to future work. It is not self-evident that the difficulties that lead people to migrate are always best addressed by making migration easier to accomplish: severe human rights violations are often experienced by people who are unable to migrate, not only by those who are forced to migrate, and the right to migrate is arguably less important than the right to stay (Oberman, 2011). But when local resolution cannot be accomplished in ways consistent with basic human rights and dignity, there are surely obligations of some sort. A more subtle understanding of the "forcedness" of forced migration-extending to instances in which people exercise the sort of agency described above-will help in specifying what those obligations are.